Supplemental Data


Abstract
Full Text
Genomic Instability in Mice Lacking Histone H2AX
Arkady Celeste, Simone Petersen, Peter J. Romanienko, Oscar Fernandez-Capetillo, Hua Tang Chen, Olga A. Sedelnikova, Bernardo Reina-San-Martin, Vincenzo Coppola, Eric Meffre, Michael J. Difilippantonio, Christophe Redon, Duane R.Pilch, Alexandru Olaru, Michael Eckhaus, R. Daniel Camerini-Otero, Lino Tessarollo, Ferenc Livak, Katia Manova, William M. Bonner, Michel C. Nussenzweig, André Nussenzweig

Supplementary Material

Generation and screening of mice

H2AX-/- mice were generated from H2AX+/- intercrosses and screened by Southern blot as described (1), or by PCR using the following primers: HX5: 5' CTCTTCTACCTCGTACACCATGTCCG 3', HX3 5' CGAAGTGGCTCAGCTCTTTCTGTGAGGG 3' and KXR 5' GTCACGTCCTGCACGACGCGAGC 3'. Cycling conditions were: 95 degrees 30'', 60 1', 72 2', 40 cycles. Primers HX5 and KXR give a product diagnostic of the targeted allele that is 424 bp and primers HX5 and HX3 give a wild type product of 547 bp. H2AX-/- mice were generated from a single targeted H2AX+/- ES cell clone.

Two-dimensional gels and Western blotting

Testes were processed for histone analysis as previously described (2). Thymocytes were either left untreated or subjected to 200 Gy from a 137Cs source on ice. Pellets were resuspended in 1 ml of phosphate-free RPMI 1640 containing 1 mCi 32PO4, incubated for 30 min at 37°C and processed for histone analysis. Western blot analysis was performed using polyclonal Nbs1 and Greek Letter Gamma-H2AX antibodies (3, 4).

MEF and B cell growth and irradiation induced cell cycle checkpoints

E13. 5 MEFs (105) were plated in replicate 60 mm dishes and counted every two days. Immortalized MEFs were generated by a 3T9 protocol (5). B lymphocytes were isolated using CD19 microbeads (Miltenyi Biotec) and stimulated with LPS or LPS+IL4 as described (1). To track cell divisions during class switching, cells suspensions were mixed with CFSE (5 Greek Letter MuM; Molecular Probes) and harvested at different time points after stimulation. For irradiation induced G1-S cell cycle experiments, MEFs were fixed in 75% ethanol, stained with anti-Brdu antibodies (Becton Dickenson) and then counterstained with 5Greek Letter Mug/ml propidum iodide. For intra S-phase checkpoints, MEFs and activated B cells were irradiated with different doses of Greek Letter Gamma-irradiation, and then pulsed with [3H] thymidine (5 Greek Letter MuCi ) for 2 hours. For G2-M checkpoints, MEFs and activated B cells were irradiated, stained with antibodies recognizing phosphorylated histone H3 (1:50 dilution; Upstate Biotechnology), and the percentage of mitotic cells was assessed by immunofluorescence detection (>200 cells counted) and quantified by flow cytometry as described (6). For measuring DNA repair rates, immortalized wildtype and H2AX-/- mouse embryo fibroblasts were irradiated with 80 Gy and samples were imbedded into plugs after the indicated recovery time. DNA was analyzed on a 1% agarose gel with 0.5 x TBE running buffer in a CHEF DR II apparatus (BioRad) for 65 hr, switch time 4500 sec, 0.9 V/cm at 14°C. After the gel was stained with ethidium bromide, images were captured with the ChemiImager 4400 (Alpha Innotech Corp.) and quantitated with ImageQuant software 3.3 (Molecular Dynamics Corp.).

Spectral Karyotyping, FISH and Sister chromatid exchange (SCE)

Primary MEFS (passage 3), immortalized (Ku80-/-, H2AX+/+ and H2AX-/-) MEFs, and T cells stimulated with soluble anti-CD3 and -CD28 antibodies (5 Greek Letter Mug/ml each; Pharmingen) were arrested at mitosis with Colcemid (Gibco/BRL) treatment at 0.1 Greek Letter Mug/ml for 0.5-3 hrs. Mitotic chromosome spreads were prepared following standard procedures, and Spectral Karyotype (SKY) analysis was performed as described (7). For FISH analysis, metaphases were hybridized with chromosome painting probes generated by DOP-PCR using flow sorted single chromosomes (7), and a BAC probe containing the TCRGreek Letter Alpha locus (4). For SCE analysis, activated lymph node T cells were plated with BrdU (10 Greek Letter MuM) and grown for 48-72 h. Metaphase spreads were stained with 1 Greek Letter Mug/ml Hoechst 33342 in PBS and SCE events were directly visualized under a fluorescent microscope with a UV filter. In some experiments, mitomycin C (10ng/ml) was added for 24 hours prior to generating metaphase spreads.

Flow cytometric analysis of lymphocyte development

For analysis of lymphocyte development, single cell suspensions were prepared from thymus, bone-marrow, spleen and peritoneal cavity. Cells were stained with monoclonal antibodies specific for cell surface markers. The following antibodies (all from Pharmingen) were used for cell phenotyping: FITC anti-CD43, anti-CD4, anti-B220 and anti-IgD; PE anti-CD25, anti-CD23 and anti-CD5; APC anti-B220 and anti-CD8; and biotin anti-IgM, anti-CD44, and anti-CD21 visualized with streptavidin-RED670 (Gibco-BRL). Double negative thymocytes were enriched by depleting thymocytes with anti-CD4 and anti-CD8 beads (Dynal).

Semiquantitative PCR analysis of T cell receptor rearrangments

Serial dilutions of genomic DNA isolated from wild type and mutant thymi (approximately 250, 50 and 10 ng) were used to amplify TCR gene rearrangements with specific primers. Primers for TCR Greek Letter Gamma and Greek Letter Delta gene rearrangements have been described (8). TCRGreek Letter Beta rearrangements were detected by using a VGreek Letter Beta8 gene specific (9) and a JGreek Letter Beta2.5 gene specific primer. TCRGreek Letter Alpha rearrangements were detected by using a combination of VGreek Letter Alpha2, VGreek Letter Alpha10, JGreek Letter Alpha50 and JGreek Letter Alpha26/27 specific primers, which were designed based on public DNA sequences. The VGreek Letter Alpha2 and VGreek Letter Alpha10 primers are expected to amplify multiple members of these gene families. RAG-2 PCR reactions (1.6kb size) served as control for DNA quantity and integrity (8). TCRGreek Letter Beta, Greek Letter Gamma and Greek Letter Delta products were hybridized to specific, radioactive labeled probes (the TCRGreek Letter Beta probe hybridizes only to the JGreek Letter Beta2.1-2.4 genes) and analyzed on Phosphorimager (Molecular Dynamics). The TCRGreek Letter Alpha and RAG-2 PCR products were visualized by SybrGreen (Molecular Probes) staining and analyzed on Fluorimager (Molecular Dynamics). Signal joint PCR reactions were performed with primers located at the 3' and 5' flanking regions of the VGreek Letter Alpha2, 10 and JGreek Letter Alpha50 genes respectively. Primer sequences not published before are as follows (5' to 3'): JGreek Letter Beta2.5: GAGCCGAGTGCCTGGCCCAAAGTA;VGreek Letter Alpha2: GCCGGATCCAGGAGAAACGTGACCAGCAG; VGreek Letter Alpha10: AGCGAATTCCCGCGTCCTTGGTTCTGCA; VGreek Letter Alpha10 signal joint: CCTGGATCCAGAATTCTACCAATACARGAAAG, where R=A or G; JGreek Letter Alpha26/27: CCTGGATCCTTACTGTCATATATCGAA

Immunization and ELISA assay

6 week-old mice were immunized intraperitoneally with 100 Greek Letter Mug TNP-KLH in aluminum sulfate. Serum samples were collected prior to immunization and on days 5, 15 and 21 post immunization. Total and TNP-specific serum Ig were assayed by ELISA. For TNP-specific Ig, plates were incubated at 4°C overnight with 2.5 Greek Letter Mug of TNP-OVA capturing antigen. Serial dilutions of serum samples were added to the wells for 1 hour at 37° C. Plates were washed, incubated with horseradish peroxidase-conjugated goat-anti-mouse IgGreek Letter Alpha, IgGreek Letter Mu, IgGreek Letter Gamma3 or IgGreek Letter Gamma1 at 1:4000 (Southern Biotechnology Associates), developed with ABTS Microwell Peroxidase Substrate System, and the optical density at 405nm was recorded by an ELISA plate reader. Titers are expressed as the serum dilution that gave half maximum optical density.

Testes histology, meiotic chromosome spreads and immunofluorescence

Testes were dissected and fixed in buffered 10% formalin and paraffin sections were stained with hematoxylin-eosin or fixed in Bouin's fixative and stained with Periodic Acid Schiff reagent. Apoptotic nuclei were detected with TdT-mediated dUTP-biotin nick labeling. Chromosome spreads were prepared as described (10). Primary antibodies used for immunofluorescence were mouse Greek Letter AlphaScp3 (1:500; provided by Peter Moens), rabbit Greek Letter AlphaScp1 (1:200, provided by Christa Heyting), rabbit anti Greek Letter Gamma-H2AX (1:500), rabbit Greek Letter AlphaRad51/DMC1 (1:50; Santa Cruz Biotechnology) mouse Greek Letter AlphaBrca1 (1:5 provided by S. Ganesan) , mouse Greek Letter AlphaMlh1 (1:10; Pharmingen). All secondary antibodies conjugated with Rhodamine Red-X or FITC (Jackson Immunoresearch laboratories) were used at 1:200. Immunofluorescence staining of B cells with Greek Letter AlphaNbs1, anti-Greek Letter Gamma-H2AX, Greek Letter AlphaBrca1, Greek Letter AlphaRad51 antibodies has been described (1). 53bp1 antiserum (used at 1:1000) was provided by Junjie Chen. In situ fractionation and immunofluorescent staining using Greek Letter AlphaNbs1 and mouse Greek Letter AlphaPCNA antibodies (1:250; Santa Cruz) was performed as described (11).

Generation of H2AX-/- ES cells, gene targeting frequencies and MMC sensitivity.

To generate doubly deficient H2AX-/- ES cells, an H2AX+/- ES cell line was transfected with a H2AX targeting construct containing the hygromycin selectable marker. Loss of H2AX in H2AX-/- ES cells was confirmed by two-dimensional gel electrophoresis and Western blot analysis as described above. To measure gene targeting frequencies, we transfected H2AX+/+, H2AX+/- and H2AX-/- ES cells with a linear construct containing the puromycin resistance selectable marker flanked by either the mouse CSB or Rad54 locus, as described (12, 13). All ES cell lines yielded similar number of puromycin resistant clones. Genomic DNA from individual clones was analyzed by Southern blot analysis, which was used to distinguish random from homologously targeted integration (12). ES cell sensitivity to interstrand cross-links was measured by incubating exponentially growing ES cells to graded doses of mitomycin C for 3 hours, and then plating different dilutions onto pre-gelatinized 60 mm dishes. Cells were fixed and stained with crystal violet 10-12 days later and colonies were counted for clonogenic survival.


Supplemental Figure 1. Inactivation of mouse H2AX. A. Structure of the targeted H2AX mutation. The region coding for the H2A conserved core (amino acid residues 50-56) between the Asc I and Xho I site (indicated in bold) was deleted and replaced by a neo cassette. B-D, Two dimensional gel analysis of histones extracted from testes (B) and thymus (C-D) from H2AX +/+(left panels) and H2AX-/- (right panels) mice. B. Presence of intact core histones in the preparations (upper panel) and enlargement of the region containing H2AX (lower panel). C. H2AX-containing region of 2-D gels from unirradiated thymocytes. Upper panel: Coomassie stained gel, lower panel : 32PO4 incorporated gel. D. H2AX-containing region of 2-D gels from 200 Gy irradiated thymocytes. Upper panel: Coomassie stained gel, lower panel: 32PO4 incorporated gel. The position of Greek Letter Gamma-H2AX is noted with a solid (present) or stippled (predicted) line. E. Western blot analysis of Nbs1 (upper panel) and Greek Letter Gamma-H2AX (lower panel) proteins in control (C) and 10 Gy irradiated (IR) thymocytes from H2AX+/+, H2AX+/- and H2AX-/- mice. The position of phosphorylated Nbs1 relative to the unphosphorylated form is noted with arrows (upper and lower respectively).


Medium version | Full size version


Supplemental Figure 2. Growth retardation in H2AX-/- mice. Weights of male and female H2AX wild-type (black circles), heterozygous (grey circles) and knockout (open circles) mice at 5 weeks of age are compared. The average weight of males (+/+: 23.7 g, +/-: 21.3 g, -/-: 18 g) and females (+/+: 17.8 g, +/-: 16.8 g, -/-: 12.5 g) is indicated by a bar.


Medium version | Full size version


Supplemental Figure 3. Chromosomal aberrations in H2AX-/- MEFs and T cells.A. DAPI stained metaphase from a H2AX-/- MEF culture contains chromatid breaks and a chromosome rearrangement (arrows). B. Typical chromosome aberrations from SKY analysis of H2AX-/- MEFs. From left to right, chromatid break, centromere-telomere and Robertsonian translocations, and a complex rearrangement. C. Example of a reciprocal translocation in H2AX-/- T cells in which the breakpoint was near the T-cell receptor alpha locus (TCRGreek Letter Alpha; arrow). Chromosome 14 (red) and 6 (blue) painting probes were used for FISH analysis together with a TCRGreek Letter Alpha BAC probe (4). Arrowheads indicate normal chromosomes. D. Typical chromosome aberrations of H2AX-/- T cells analyzed by DAPI, SKY or FISH. Chromatid break, reciprocal translocation (top panels; SKY hybridization: chromosomes 4 (green) and 7 (red)), quadriradial (bottom left panel; chromosome 14 painting probe (red)) and complex chromosome rearrangement (bottom right panel; chromosome 14 (red) and 15 (yellow) painting probes).


Medium version | Full size version


Supplemental Figure 4. Irradiation-induced checkpoints in B cells.A. B cells were stimulated with LPS+IL4 for 43 hours, irradiated with 500 cGy and the cell cycle distribution was determined at different times postirradiation. Both H2AX wildtype and knockout cells accumulate in S phase and then G2/M, after which they die by apoptosis (population with sub 2N DNA content). B. Irradiation induced intra-S phase checkpoint. The [3H]thymidine incorporation in unirradiated cultures was set to 100%. C. Irradiation induced G2-M checkpoint. B cells in mitosis were identified by immunofluorescence using Greek Letter Alpha-phospho histone H3 antibody. Percentage of mitotic cells was normalized relative to unirradiated controls.


Medium version | Full size version


Supplemental Figure 5. T and B cell development in H2AX-/- mice.A. Flow cytometric analysis of thymocytes in 6 week-old H2AX wild-type (+/+) and knockout (-/-) mice. Dot plots show CD4 and CD8 expression on total thymocytes (top), and CD25 and CD44 expression on gated CD4-CD8-double negative thymocytes (bottom). Average thymus cellularity was 1.4 x 108 ± 3.2x107 for H2AX+/+ vs. 7.13 x 107±2.9x107 for H2AX-/- (n=8 for each genotype). B. Bone marrow B cell precursor phenotype. Dot plots show B220, CD43 or IgM expression on bone marrow cells. Average number of bone marrow cells was 3.9 x 107±9x106for H2AX+/+ vs. 2.75 x 107±1x107 for H2AX-/- mice (n=3 for each genotype). C. IgD and IgM expression on total splenocytes (upper panel) and CD21 and IgM expression on gated B220+CD23-(middle panel) and B220+CD23+ (lower panel). Early (T1) and late (T2) transitional B cell, marginal zone (MZ), and mature (M) B cell populations are indicated. Average number of CD19+ splenic B cells was 4.9 x 107±1x107 for H2AX+/+ vs. 2.4x107 ±1x107for H2AX-/- (n=8 for each genotype). D. Peritoneal cavity (PC) B cell phenotype. Dot plots show IgD and CD5 expression on gated B220+IgM+ PC B cells. E. PCR amplification of T cell receptor gene rearrangements from serially diluted thymus DNA of H2AX mutant (-/-) and wild type (+/+) littermate animals. Gene specific primers were used to amplify VGreek Letter Alpha-JGreek Letter Alpha coding joint (left), VGreek Letter Alpha10-JGreek Letter Alpha50 signal joint (left bottom), VGreek Letter Beta-JGreek Letter Beta, VGreek Letter Gamma-JGreek Letter Gamma and VGreek Letter Delta-JGreek Letter Delta (right) rearranged products. Amplification of the RAG-2 gene served as a loading control (right bottom).


Medium version | Full size version


Supplemental Figure 6. Impaired switching to IgG3 in H2AX-/- mice. Surface expression of IgG3 in B cells after stimulation with LPS for 4 days.


Medium version | Full size version


Supplemental Figure 7. Fragmentation and illegitimate associations between sex chromosomes and autosomes in H2AX-/- spermatocytes. Scp1 (red) and Brca1 (green) staining of H2AX wildtype (A) and knockout (B-D) pachytene cells. Brca1 is localized to unsynapsed regions of X and Y chromosomes (14). X and Y chromosomal fragmentation (B) and association with autosomal chromosomes (C and D) is observed in H2AX-/- cells. Magnification of X-autosome interaction is shown in D (middle and right panels). Arrow indicates the X chromosome and arrow-head shows the Y chromosome. Scale bar is 10 Greek Letter MuM.


Medium version | Full size version


Supplemental Figure 8. Spontaneous and induced chromatid exchanges and MMC sensitivity in H2AX-/- cells.A. Histogram displaying the number of sister chromatid exchanges in T cells from H2AX wild-type (filled bars; average number of SCE per cell =5) and knockout (open bars; average number of SCE per cell=8.8) mice. B. Induction of chromosome aberrations (radial structures and fusions) in T cell cultures after treatment with mitomycin C (10 ng/ml). Percentage of H2AX+/+ (filled bars; n=50) and H2AX-/- (open bars; n=50) metaphases exhibiting aberrations is plotted. C. Clonogenic survival of H2AX wild-type (black circles), heterozygous (grey circles) and knockout (open circles) ES cells after treatment with mitomycin C.


Medium version | Full size version


Supplementary References

1. S. Petersen et al., Nature414, 660 (2001).

2. E. P. Rogakou, D. R. Pilch, A. H. Orr, V. S. Ivanova, W. M. Bonner, J Biol Chem273, 5858 (1998).

3. E. P. Rogakou, C. Boon, C. Redon, W. M. Bonner, J Cell Biol146, 905 (1999).

4. H. T. Chen et al., Science290, 1962 (2000).

5. G. J. Todaro, H. Green, J. Cell Biol.17, 299 (1963).

6. B. Xu, S. Kim, M. B. Kastan, Molecular & Cellular Biology21, 3445 (2001).

7. M. Liyanage et al., Nat Genet14, 312 (1996).

8. F. Livak, A. Wilson, H. R. MacDonald, D. G. Schatz, Eur J Immunol27, 2948 (1997).

9. F. Livak, M. Tourigny, D. G. Schatz, H. T. Petrie, J Immunol162, 2575 (1999).

10. P. J. Romanienko, R. D. Camerini-Otero, Mol Cell6, 975 (2000).

11. R. S. Maser et al., Mol Cell Biol21, 6006 (2001).

12. J. Essers et al., Cell89, 195 (1997).

13. L. J. Niedernhofer et al., Embo J20, 6540 (2001).

14. R. Scully et al., Cell88, 265 (1997).