| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
-H2AX Expression Pattern in Non-Irradiated Neonatal Mouse Germ Cells and after Low-Dose -Radiation: Relationships Between Chromatid Breaks and DNA Double-Strand Breaks1
INSERM U566-CEA-Paris 7, Laboratoire de radiosensibilité des cellules germinales, Département de Radiobiologie et Radiopathologie, Direction des sciences du vivant, Commissariat à l'Énergie Atomique, BP 6, 92265 Fontenay-aux-roses Cedex, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-H2AX, has been shown to form foci in interphase nuclei as well as megabase chromatin domains surrounding the DNA lesion on chromosomes. Using detection of -H2AX on germ cell mitotic chromosomes 2 h after -irradiation, we studied radiation-induced DSBs during the G2/M phase of the cell cycle. We show that 1) non-irradiated neonatal germ cells express -H2AX with variable patterns at metaphase, 2) -irradiation induces foci whose number increases in a dose-dependent manner, 3) some foci correspond to visible chromatid breaks or exchanges, 4) sticky chromosomes characterizing cell radiation exposure during mitosis are a consequence of DSBs, and 5) -H2AX remains localized at the sites of the lesions even after end-joining has taken place. This suggests that completion of DSB repair does not necessarily imply disappearance of -H2AX.
developmental biology, gametogenesis, spermatogenesis, stress, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A highly conserved histone H2A variant, H2AX, accounts for 10–20% of total H2A proteins of the chromatin [6]. It is found in large amounts in adult germ cells [7–10]. Most importantly, it has been shown to be rapidly phosphorylated after DSB induction. This modification is assumed to change local chromatin structure, enabling the recruitment of proteins involved in DNA repair [11]. Phosphorylated H2AX (i.e., -H2AX) can be visualized by immunocytochemistry of cell nuclei and chromosomes [12, 13]. However, accurate quantification of -H2AX foci on interphase nuclei is difficult, and to our knowledge, the only published data regarding chromosomes have been impaired by the poor quality of the metaphase spreads [13].
To our knowledge, no study has examined the effect of ionizing radiation on neonatal germ cell chromosomes, probably because of the difficulty in distinguishing germ cells from somatic cells at metaphase. We took advantage of our previous finding that the overall chromosome DNA methylation pattern distinguishes somatic cells from germ cells within testicular cell spreads [14] to study neonatal germ cell chromosome radiosensitivity during the G2/M phase of the cell cycle. This enabled us to quantify on metaphase chromosomes the -H2AX foci induced by various doses of radiation and to examine the relationships among induced DSBs, chromatid breaks, and chromatid exchanges.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Eight-day-old NMRI mice bred in our animal facility were whole-body exposed to -rays from a 137Cs source (IBL 637; CIS Bio International, France). Four doses were tested (0.2, 0.5, 1, and 2 Gy) at a dose rate of 0.652 Gy/min. Just after irradiation, i.p. injections of colcemid (Sigma, Saint Quentin Fallavier, France) were administered (1.2 mg/kg), and the animals were killed 2 h later. Two independent experiments were performed at each dose. The animals were used and maintained according to the French regulation (Ministry of Agriculture Decree 87-848). The animal installation is accredited by the Veterinary Inspectorate (A92-032-02).
Chromosome Preparations
Testes of at least three animals per dose were extirpated, and albuginea was removed. Decapsulated testes were cut in pieces and incubated in collagenase I (Serva, Heidelberg, Germany) for 20 min at 37°C in a shaking water bath. Every 5 min, testes fragments were resuspended with a pipette. After complete digestion, cells were centrifuged for 10 min at 300 x g and resuspended in PBS. Cell suspensions were divided into two batches, either for methanol/acetic acid fixation or for unfixed chromosome preparations.
Methanol/acetic acid-fixed chromosome preparations were obtained as described previously [14]. Unfixed chromosome preparations were obtained after cell cytocentrifugation on coated slides (Thermo Shandon, Cergy Pontoise, France), and the slides were incubated in 0.5% Triton/ PBS for 5 min before fixation in 3.7% formaldehyde/PBS for 5 min. The cells were again permeabilized in 0.5% Triton/PBS before being stained with 4',6'-diamidino-2-phenylindole (DAPI; 0.1 µg/ml) and mounted in Vectashield (Vector, Burlingame, CA).
-H2AX Histone Immunocytochemical Detection
Slides with chromosome spreads of good quality after cytocentrifugation were selected for -H2AX immunodetection. Coverslips were removed in PBS, and slides were incubated in PBT/BR (PBS, 0.1% BSA, 0.1% Tween 20, and 0.1% purified casein protein; Roche Diagnostics, Meylan, France) for 15 min. Mouse monoclonal and rabbit polyclonal antibodies against -H2AX (Upstate, Charlottesville, VA) were used and revealed by a fluorescein isothiocyanate (FITC)-coupled anti-mouse immunoglobulin (Ig) G (Sigma) or a Cy3-coupled anti-rabbit antibody (Jackson ImmunoResearch, West Grove, PA).
Fluorescent images were captured with the same time exposure using an Olympus AX70 epifluorescent microscope equipped with charge-coupled camera (Princeton Instruments, Trenton, NJ) and IPLab software (Scanalytics, Fairfax, VA).
-H2AX Foci Counts
At least 10 well-spread metaphases per pool of three animals were analyzed for each dose of radiation exposure. For -H2AX foci counting, large and intensely stained foci were recorded separately from smaller and less intensely stained ones (see Fig. 2). Neighboring double spots on the same chromatid were recorded as a single -H2AX focus when they surrounded a break detected by DAPI counterstaining.
|
Chromosome Break Detection
Methanol/acetic acid-fixed chromosomes were stained with Giemsa to analyze chromatid breaks, gaps (achromatic region on a chromatid smaller than the chromatid width), sticky chromosomes (interchromatid links), and radial figures (chromatid exchanges) (see Fig. 4). For each independent experiment, at least 20 metaphases per dose were analyzed.
|
Identification of Germ Cell Metaphase Spreads by Immunocytochemical Detection of 5-Methylcytosine
Identification of germ cells was performed using anti-5-methylcytosine antibody either after the Giemsa or -H2AX staining procedure. Briefly, slides were exposed in PBS for 12 h to ultraviolet light with a germicidal lamp, rinsed with ice-cold PBS, and incubated in PBT (PBS, 0.1% BSA, 0.1% Tween 20) for 15 min before immunodetection with monoclonal antibody to 5-methylcytosine, which was revealed with an FITC-coupled anti-mouse IgG (Sigma). Chromosome spreads were finally counterstained with propidium iodide. In these conditions, germ cells showed heterogeneous pale staining of their juxtacentromeric regions, whereas the staining of these regions was bright and homogeneous in somatic cells. In addition, a few chromosomes were methylated along one of their chromatids (see Fig. 4 and [14]).
Statistics
Statistical comparisons were performed using Statview software (Abacus Concepts, Berkeley, CA). The results of independent experiments were grouped, because no statistical difference was found between them (P < 0.05). The t-test, Mann-Whitney U test, and linear regression were performed.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-H2AX in Non-Irradiated Germ Cell ChromosomesMost non-irradiated germ cell chromosomes exhibited a rather homogeneous staining of the DNA in both interphase nuclei and mitotic chromosomes, whereas somatic cells were not stained at all (Fig. 1). Very few somatic metaphases obtained on the same preparations were well spread in our conditions of cytocentrifugation. Long exposure times were needed to highlight the staining of the juxtacentromeric regions; the staining was pale. For germ cell chromosomes, a diffuse staining of both euchromatin and juxtacentromeric regions varied in intensity from cell to cell (Fig. 1). In approximately 20% of them, chromosome Y was the most intensely stained chromosome (Fig. 1D). At high magnification, most of the cells exhibited well-defined spots, referred to as pseudofoci because of their slightly higher intensity of staining than the overall euchromatin (Fig. 1, C and D). Their numbers and intensities were also highly variable, ranging from 1 to 10 per cell.
|
Expression Pattern of -H2AX in Irradiated Germ Cell Chromosomes
The overall background staining of euchromatin was considerably lower in the irradiated compared to the non-irradiated cells (Compare Figs. 1 and 2; the pictures were captured with the same time exposure). At 0.2 Gy, the background staining remained, making it difficult to distinguish between radiation-induced foci and pseudofoci (not shown). For cells irradiated at doses of 0.5 Gy or greater, it was necessary to increase the capture time of the camera to highlight the pale staining of chromosomes. In this case, -H2AX foci were no longer countable, because the fluorescent signals were overexposed. All germ cell metaphases exhibited -H2AX foci of variable size and intensity (Fig. 2). In contrast to pseudofoci in non-irradiated cells, the number of -H2AX foci was fairly constant from one cell to another for a given dose of radiation (Table 1). Some foci corresponded to visible radial figures, chromatid breaks, and gaps, whereas others did not (Fig. 2). The relationship between -H2AX foci and chromatid alterations remained difficult to quantify with accuracy, because the spreading of metaphases after cytospin is of lower quality than that in classical chromosome preparations. However, in the metaphases of highest quality, it appeared to be very likely that a low percentage of -H2AX foci colocalized with visible chromatid aberrations (Fig. 2, B, D, and F). Table 1 indicates the average numbers of large and small foci and their total per cell for two independent experiments. Compared to the control, -H2AX foci numbers were significantly increased, even at 0.2 Gy (Table 1). Statistically significant differences were also found between -H2AX foci numbers at all radiation doses tested (Table 1). A dose-dependent increase in the average number of -H2AX foci was found as well. This relationship follows the linear equation Y = 4.74 + 18.8X obtained by linear regression with a correlation coefficient (r2) of 0.9 (Fig. 3). Symmetrical foci on sister chromatids represented, on average, 14%, 14.6%, and 22.3% of all foci per cell at 0.5, 1, and 2 Gy, respectively. Foci were also frequently detected at telomeric regions in similar proportions at the three doses (18%, 16.7%, and 20% of all foci at 0.5, 1, and 2 Gy, respectively).
|
|
Break Counts in Irradiated Germ Cell Chromosomes by Giemsa Staining
Chromosome breaks were recorded after Giemsa staining of methanol/acetic acid-fixed chromosomes obtained from the same testicular cell suspension as that used for unfixed chromosome preparations (see Materials and Methods). The percentages of metaphases without any detectable anomaly were 64.4%, 34.2%, 29.3%, 20.7%, and 1.6% at 0, 0.2, 0.5, 1, and 2 Gy, respectively. This decrease was dependent on the dose (r2 = 0.99). Most of the chromosome lesions were either chromatid breaks, interchromosome links referred to as "sticky chromosomes" by Al Achkar et al. [5], and gaps (Fig. 4). Classical radial figures were detected in 4–6% of the cells. These results are consistent with chromosome alterations induced by radiation in the late S/G2/M phases of somatic cells [15]. As shown by -H2AX foci, chromatid gaps were considered to result from one DSB, whereas isochromatid gaps and those chromatid exchanges visualized as sticky chromosomes were counted as resulting from two breaks (Table 2). Significantly more breaks were observed in irradiated than in non-irradiated cells at all doses (P < 0.0001). The differences between each group of irradiated cells were also significant (P < 0.001), except between 0.2 and 0.5 Gy (P = 0.38). On average, -H2AX immunodetection was approximately fivefold more sensitive than the classical cytogenetic approach to characterization of the effect of -irradiation on germ cells (Y = 3.9 + 5.3X; r2 = 0.91) (Fig. 5).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-H2AX. Our cytochemical approach allowed us to show different patterns of expression at metaphase among non-irradiated germ cells: strong and homogeneous staining or heterogeneous staining with or without intense staining of chromosome Y. In addition to the rather homogeneous staining of the chromatin, some foci could also be detected. This variable expression pattern may depend on the stage of differentiation of spermatogonia in 8-day-old mice, which cannot be distinguished at metaphase even by their DNA methylation pattern. Adult spermatogonia have also been found to express -H2AX [9], but this was observed with immunohistochemistry, without detailed information on nuclear and chromosome distribution. Hamer et al. [9] found that type-A spermatogonia were homogeneously stained by the anti--H2AX antibody, whereas type-B spermatogonia exhibited foci. As described by histology, type-A and -B spermatogonia constitute 63% and 37%, respectively, of all spermatogonia at 8 days post partum (dpp) [16]. These proportions do not fit with our findings regarding metaphases with or without foci on chromosomes, but they may roughly correspond to the overall staining of chromosome Y in approximately 20% of the metaphases. Interestingly, the natural and global labeling of chromosomes was lowered after irradiation, but the difference was less marked at 0.2 Gy. It remains to be specified which mechanisms are involved in the concomitant, dose-dependent decrease in overall chromosome labeling and the increase in foci numbers. Our results also suggest that the accurate counting of foci after low-dose irradiation (<0.2 Gy) cannot be performed on spermatogonia.
Assuming that one -H2AX focus corresponds to one DSB [1], we found an average number of 18.8 DSB Gy–1 cell–1 in germ cells of 8-day-old mice. This suggests that postnatal germ cell radiation sensitivity may be approximately half that of somatic cells, which has been roughly estimated to be 25–40 DSB Gy–1 cell–1 [3, 17]. However, these data correspond to a time point at 2 h after in vivo irradiation. In somatic cells, H2AX has been shown to be phosphorylated within a few minutes following in vitro irradiation, giving rise to foci on both nuclei and chromosomes [13]. Thereafter, these foci grow in both number and size, reaching a maximum between 0.5 and 1 h after irradiation [13, 18]. Ongoing studies by our group aim to specify the kinetics of disappearance of -H2AX foci in germ cells following in vivo irradiation and to compare the data to those of somatic cells in the same experiment. Such studies will afford interesting insights regarding potential differential radiation sensitivities to DSB induction and repair between these cell types [19]. Symmetrical staining on sister chromatids has been detected and seemed to increase depending on the dose. The number of symmetrical -H2AX foci on both sister chromatids was non-negligible and may reflect an ongoing process of homologous recombination using sister chromatids as a template for DSB repair [20]. Colocalization at these sites of rad51 protein, which is a specific component of the homologous recombination DSB repair pathway [21], would help to characterize these symmetrical signals on sister chromatids.
After exposure to clastogenic agents, such as ionizing radiation, the type of chromosome aberration observed at first metaphase depends on the time elapsed since exposure. Chromatid-type aberrations, such as gaps, breaks, and exchanges (radial figures), are typical postreplicative lesions (i.e., induced during the late S or G2/M phase). A small proportion of chromosome-type (isochromatid) lesions may also occasionally occur after late S or G2/M phase exposure. In germ cells, we could show that the cell cycle duration is approximately 24 h, which is an average duration for somatic cells in culture. In such conditions, the G2/M phase lasts for approximately 2.5–3 h. Thus, all the metaphases harvested 2 h after irradiation and colcemid treatment corresponded to cells exposed in the G2/M phase. Some of them were irradiated at the moment when chromosome compaction was engaged. For such cells, it was shown that another alteration could occur, namely "sticky chromosomes." These were characterized by chromosomes of apparently normal morphology but were linked by filaments [5], the origin of which was unclear. Such chromosomes were also found in the present study.
It is generally assumed that among DNA lesions, DSBs are those involved in chromosome- or chromatid-type aberrations by illegitimate end-joining [22, 23]. However, neither qualitative nor quantitative relationships between DSBs and chromosome aberrations remained totally elucidated.
The majority of -H2AX foci occurred on apparently normal chromosomes or chromosome segments. This suggests that most DSBs were repaired without visible consequence at the cytogenetic level. Another alternative is that DNA breaks remained but were masked by chromosome compaction and sister chromatid cohesion.
Chromatid breaks and gaps were associated with -H2AX labeling on both telomeric and centromeric ends. This demonstrates that gaps, the physicochemical substratum of which was not clear, correspond to DNA DSB, just as chromatid breaks do. Chromosomes involved in radial figures were also labeled near their chromatid exchanges. Their distortion indicates they are strongly linked, very likely by covalent linkage of their DNA molecule. This was confirmed by the study of their derivative chromosomes at the next cell generations [24]. Their anti--H2AX labeling indicates that the phosphorylation of H2AX can remain after DNA ligation or that cytological observation of end-joining does not mean that DSB DNA repair is completed at the molecular level. This suggests that the conclusions of Nazarov et al. [25] should be modified in that H2AX dephosphorylation may not strictly correlate with achievement of DNA damage repair.
For sticky chromosomes, a clear labeling of both the filament joining the two chromosomes involved and the chromatids at the site of the filament attachment also occurs. Thus, as for radial figures, the sticky chromosomes correspond to achieved chromatid exchanges, with persistent H2AX phosphorylation. Because they principally occur after irradiation just before mitosis, this indicates that the illegitimate DNA linkage occurred at the moment when chromosome compaction was already engaged. Except at the filament, this did not prevent full chromosome compaction. Thus, the apparent normal morphology of the chromosomes involved would be only caused by protein (condensin and cohesin) assembly.
Estimated radiation sensitivity of germ cells from 8-day-old mice was, on average, fivefold greater using -H2AX foci detection than that using the classical cytogenetic approach. This difference may be related to the highest technical sensitivity of the immunodetection of the -H2AX approach. Assuming that -H2AX is retained at sites that have already been repaired, this difference may also indicate the proportion of chromatid breaks that are left unrepaired.
Altogether, these findings imply that -H2AX is a useful tool to investigate in vivo postnatal spermatogonia radiation sensitivity to the induction of chromatid breaks. Because the chromosome DNA methylation pattern differs considerably between postnatal proliferating germ cells and is also very different in somatic cells, immunodetection of both -H2AX and DNA methylation will help to find potential variations in radiation sensitivity during postnatal testis development in mice. However, considering doses less than 0.2 Gy, the persistent heterogeneous staining and the presence of rather numerous foci in non-irradiated cells hinders accurate detection of radiation-induced DSBs in neonatal spermatogonia. This is not the case for testicular somatic cells which do not express -H2AX at such levels. Thus, the immunodetection of -H2AX on somatic cell metaphases would be a more sensitive approach to estimate radiation-induced DNA DSBs than chromatid break counts at low doses. Finally, colocalization of -H2AX foci with either Rad51 [26] or DNA ligase IV on such preparations of irradiated chromosomes may help to define the contribution, in the late S/G2/M phases, of homologous recombination and non-homologous end-joining DNA repair pathways [12, 23, 27, 28], respectively.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: J. Bernardino-Sgherri, CEA/DSV/SEGG/LRCG, BP 6, 92265 Fontenay-aux-roses Cedex, France. FAX: 33 0 1 46 54 99 06; jacqueline.bernardino{at}cea.fr
Received: 15 January 2004.
First decision: 5 February 2004.
Accepted: 9 April 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-H2AX during repair of DNA double-strand breaks in mammalian cells and its inhibition by calyculin A. Radiat Res 2003 160:309-317[Medline]This article has been cited by other articles:
|
|
 |
|
  A. Forand, P. Fouchet, J.-B. Lahaye, A. Chicheportiche, R. Habert, and J. Bernardino-Sgherri Similarities and Differences in the In Vivo Response of Mouse Neonatal Gonocytes and Spermatogonia to Genotoxic Stress Biol Reprod, May 1, 2009; 80(5): 860 - 873. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chicheportiche, J. Bernardino-Sgherri, B. de Massy, and B. Dutrillaux Characterization of Spo11-dependent and independent phospho-H2AX foci during meiotic prophase I in the male mouse J. Cell Sci., May 15, 2007; 120(10): 1733 - 1742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Adiga, M. Toyoshima, T. Shimura, J. Takeda, N. Uematsu, and O. Niwa Delayed and stage specific phosphorylation of H2AX during preimplantation development of {gamma}-irradiated mouse embryos Reproduction, February 1, 2007; 133(2): 415 - 422. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
