THE SPATIAL AND TEMPORAL PROGRESSION OF γ-H2AX IN UNIRRADIATED MAMMALIAN CELLS

Kirk J. McManus* and Michael J. Hendzel
Department of Oncology, Cross Cancer Institute, University of Alberta,
1560 University Ave., Edmonton, Alberta Canada T6G 1Z2
*kirkmcma@cancerboard.ab.ca 

INTRODUCTION.  Histone H2AX is a core histone variant that comprises approximately 2 to 10 percent of the total histone H2A complement in mammalian cells (1).  It is comprised of three major domains; an amino-terminal tail, a globular core domain, and a carboxy- (C-) terminal tail.  Contained within the C-terminal tail is an invariant serine/glutamine (SQ) motif located 4 residues from the C-terminus that is conserved from Giardia to humans (1).  Upon γ-irradiation, meiotic recombination, DNA replication and immunoglobulin and T cell receptor rearrangements it is the γ-serine that is rapidly phosphorylated by phosphatidylinositol 3-kinase protein kinase-like (PIKK) family members (i.e. ATM, ATR and DNA-PK) in regions flanking sites of DNA double strand breaks (DSB).

Recently, evidence has been provided suggesting a cell cycle regulation in the presence of γ-H2AX in irradiated, and to a limited degree, in unirradiated cells (2).  By combining flow cytometry with digital imaging microscopy (DIM) we were able to precisely describe the temporal and spatial progression patterns of γ-H2AX in unirradiated transformed and non-transformed mammalian cells.  Here we show a temporal progression pattern for γ-H2AX that is consistent with a role in DNA synthesis.  However, we provide additional evidence that γ-H2AX continues to increase following the completion of S-phase to attain a maximal signal intensity in prometaphase. 

METHOD.  Flow cytometry was performed to investigate the cell cycle status of γ-H2AX in a variety of mammalian transformed and untransformed cell lines an ataxia telangiectasia (AT2BE) cell line (ATM defective) and M059J (DNA-PK deficient).   Cells were fixed and permeabilized in 70% ethanol immunofluorescently labeled with anti-γ-H2AX and co-stained with propidium iodide (PI) containing RNase A to reveal the cell cycle status.  Mean γ-H2AX fluorescent intensities were normalized to mean PI-fluorescent intensities to account for variability in histone content arising from variable cell cycle status.

To further refine the precise temporal and spatial pattern of both increasing and decreasing γ-H2AX, fluorescent microscopy was employed.  Cells were co-immunostained/labeled with markers that specifically identify heterochromatin (trimethyl lysine 9of histone H3 [tMeK9]), euchromatin (trimethyl lysine 4 of histone H3 [tMeK4]), pericentromeric chromatin (anti-centromeric antigen [ACA]), transcription (fluoro-uridine labeling [FU]), DNA synthesis (bromodeoxy-uridine labeling [BrdU]), and late G2 (phosphorylated serine 10 of histone H3 [S10]).  Z-series were acquired and subjected to deconvolution as detailed elsewhere (3).

To demonstrate that the accessibility of γ-H2AX epitope throughout the cell cycle did not influence our results, Western blot analysis was performed.  Acid extracted proteins isolated from nuclei of asynchronously growing or nocodazole-arrested cells were resolved on 15% SDS polyacrylamide gels.  Proteins were transferred to PVDF membranes and probed with anti- γ-H2AX.

RESULTS.  Initial flow cytometry investigations demonstrated that γ-H2AX expression pattern in unirradiated mammalian cells that steadily increased from G0/G1 to S-phase, reaching maximal intensity at G2/M.  By normalizing the mean γ-H2AX signal intensity to the mean PI signal intensity we were able to account for histone variability arising from cell cycle variability and demonstrate that maximal γ-H2AX signal intensity occurs in G2/M. 

Digital imaging microscopy of γ-H2AX cells identified two distinct populations within unirradiated mammalian cells; a predominant population consisting of several hundred micro-foci (~30nm³) and a less prevalent population consisting of 1 to 24 larger foci per nucleus (~600nm³) which are consistent with DSB repair foci (2).  Furthermore, no obvious spatially proximal preference for euchromatin, heterochromatin or pericentromeric chromatin was observed in cells co-immunostained with tMeK4, tMeK9 or ACA, respectively, nor was a spatial relationship apparent with actively transcribing chromatin (FU labeling).  However, cells labeled with BrdU did exhibit a unique spatial relationship.  Cells in late S-phase exhibited fibrous BrdU labeling with γ-H2AX localizing to either end of the BrdU ‘fibers’.  Cells labeled with S10, a specific G2 marker, exhibited higher signal intensity than those in G1 or S-phase, but lower signal intensity than those in mitosis.  Further examination of mitotic cells revealed that γ-H2AX signal intensity reaches its maximal value in prophase/prometaphase and persists until cells enter anaphase whereby the signal is rapidly degraded until reaching basal levels by G1.   

Western blot analysis of proteins isolated from asynchronously growing or nocodazole-arrested cells demonstrated an increase in signal intensity for the nocodazole arrested population.  This supports our observations detailed above and demonstrates that the γ-H2AX epitope is not masked.

DISCUSSION.  Here we detail the temporal and spatial progression patterns of γ-H2AX throughout the cell cycle.  Our results, both the presence of high γ-H2AX signal intensity and spatial relationship with BrdU, are in agreement with a role in DNA synthesis.  However, the persistence and maximal γ-H2AX signal intensity throughout the initial stages of mitosis (i.e. prophase to anaphase) suggests the potential for an additional role for γ-H2AX in mitosis which has yet to be determined.

ACKNOWLEDGEMENTS.  This work is funded by the Canadian Institutes of Health Research (CIHR) and the Alberta Heritage Foundation for Medical Research (AHFMR).  We thank CIHR and AHFMR for financial support in the form of studentships (KJM) and scholarships (MJH).

REFERENCES. 

1. Redon, C., Pilch, D., Rogakou, E., Sedelnikova, O., Newrock, K., and Bonner, W.  (2002) Curr. Opin. Genet. Dev. 12, 162-169
2. MacPhail, S.H., Banath, J.P., Yu, Y., Chu, E., and Olive, P.L. (2003) Radiat. Res. 159, 759-767
3. McManus, K.J., and Hendzel, M.J. (2003) Mol. Cell. Biol. 23, 7611-7627