Increased ionizing radiation sensitivity and genomic
instability in the absence of histone H2AX
Craig H. Bassing†‡, Katrin F. Chua†‡, JoAnn Sekiguchi†‡, Heikyung Suh†, Scott R. Whitlow†, James C. Fleming†,
Brianna C. Monroe†, David N. Ciccone†, Catherine Yan†, Katerina Vlasakova§, David M. Livingston§,
David O. Ferguson†, Ralph Scully¶, and Frederick W. Alt†?
†Howard Hughes Medical Institute, Children’s Hospital, Department of Genetics, Harvard Medical School, and Center for Blood Research, Boston, MA 02115;
§Department of Cancer Biology, Dana Farber Cancer Institute, Boston, MA 02115; and¶Cancer Biology Program, Division of Hematology and Oncology,
Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02115
Contributed by Frederick W. Alt, April 15, 2002
In mammalian cells, DNA double-strand breaks (DSBs) cause rapid
phosphorylation of the H2AX core histone variant (to form
?-H2AX) in megabase chromatin domains flanking sites of DNA
damage. To investigate the role of H2AX in mammalian cells, we
generated H2AX-deficient (H2AX?/?) mouse embryonic stem (ES)
cells. H2AX?/?ES cells are viable. However, they are highly sensi-
tive to ionizing radiation (IR) and exhibit elevated levels of spon-
taneous and IR-induced genomic instability. Notably, H2AX is not
required for NHEJ per se because H2AX?/?ES cells support normal
levels and fidelity of V(D)J recombination in transient assays and
also support lymphocyte development in vivo. However, H2AX?/?
ES cells exhibit altered IR-induced BRCA1 focus formation. Our
findings indicate that H2AX function is essential for mammalian
DNA repair and genomic stability.
some consists of DNA wrapped around an octamer of the four
core histones—H2A, H2B, H3, and H4 (1). The H2A histones
consist of several subfamilies that contain distinct, conserved
amino- and carboxyl-terminal amino acid sequences (2). The
H2AX subfamily contains a conserved carboxyl-terminal Ser-
Gln-Glu (SQE motif) amino acid sequence. This SQE motif
represents the consensus in vitro phosphorylation site for mem-
bers of the phosphoinositide 3-kinase related kinase (PIKK)
family that includes the protein kinases DNA-dependent protein
kinase catalytic subunit (DNA-PKcs), ataxia telangiectasia mu-
tated (ATM), and ATM and Rad3-related (ATR) (3).
The repair of spontaneous and induced DNA double-strand
breaks (DSBs) is critical for the maintenance of genomic integ-
rity. In eukaryotic cells, the two major pathways of DSB repair
are nonhomologous end-joining (NHEJ) and homologous re-
combination (HR; refs. 4 and 5). Covalent modifications of core
histones via phosphorylation, acetylation, and methylation have
been proposed to form a ‘‘histone code’’ that is read by cellular
proteins to facilitate downstream molecular events (6). In re-
sponse to DNA damage by agents that induce DNA double-
strand breaks, Mec1, the Saccharomyces cerevisiae homologue of
ATR, phosphorylates the SQE motif of H2A (7). This phos-
phorylation event is required for the efficient repair of chromo-
somal DSBs by NHEJ but does not appear to be as important for
homologous recombination (7). In mammalian cells, H2AX is
rapidly phosphorylated on the induction of DSBs by ionizing
radiation (IR) and DNA damaging agents (8, 9), resulting in
formation of ?-H2AX foci along megabase chromatin domains
flanking DNA damage sites (9).
Foci of ?-H2AX also form at the immunoglobulin heavy chain
locus during class switch recombination (CSR) in activated
mature B cells (10). CSR occurs between large, highly repetitive
S regions and also may be initiated by DSBs (10, 11) and
completed by NHEJ factors (12–15). Notably, CSR is signifi-
cantly impaired in the absence of H2AX (10). Earlier during
lymphocyte development, exons that encode immunoglobulin
he DNA in eukaryotic cells is packaged into chromatin, the
fundamental unit of which is the nucleosome. The nucleo-
and T cell receptor (TCR) variable regions are assembled by
V(D)J recombination. Formation of ?-H2AX foci occurs at the
TCR? locus during V(D)J recombination (16). V(D)J recom-
bination is initiated by the recombination-activating gene 1 and
2 proteins (RAG1 and RAG2 or RAGs), which introduce DSBs
between recombining V, D, or J segments and flanking recom-
bination signal sequences (RSs) to generate blunt, 5? phosphor-
ylated RS ends and hairpinned coding ends (17). Joining of RS
ends absolutely requires four NHEJ factors, including XRCC4,
DNA ligase IV (Lig4), Ku70, and Ku80; whereas joining of
coding ends requires these four factors plus DNA-PKcs and
Artemis (17). Thus, completion of RAG-initiated V(D)J recom-
bination in transient reporter substrates provides a strict assay
for a direct function of known factors in mammalian NHEJ. In
this context, a direct evaluation of the potential role of H2AX in
V(D)J recombination has not been reported.
ATM, and possibly DNA-PKcs, phosphorylate H2AX after IR
(18, 19). However, additional wortmannin-insensitive kinases
also have been implicated (18). ATM and DNA-PKcs are both
required for the repair of IR-induced DSBs because cells
deficient for either of these factors are hypersensitive to IR and
exhibit DNA repair defects. ATM-deficient cells also exhibit cell
cycle checkpoint defects and dramatically increased genomic
instability (20). In this context, DNA-PKcs is directly involved in
the repair of DSBs (21) whereas ATM may have a more indirect
role via phosphorylation of certain proteins involved in the DNA
damage response (20). It has been argued that ATM and related
kinases, including DNA-PKcs and ATR, may mediate some
functions via phosphorylation of H2AX (18, 19). On IR, foci of
the DNA repair proteins Mre11?RAD50?NBS1 (the MRN
complex), RAD51, 53BP1, and BRCA1 colocalize with ?-H2AX
foci (18, 22, 23). In this context, ?-H2AX may play a role in the
recruitment of BRCA1, RAD51, and perhaps other DNA repair
factors to the sites of DNA damage (18). Therefore, mammalian
H2AX may be downstream of relevant phosphoinositide 3-
kinase related kinases in the mediation of particular DNA
damage responses and, in this context, theoretically could have
a role in maintenance of genomic stability.
Materials and Methods
Targeting Constructs and Probes. The 5L?3N targeting vector was
constructed in pLNTK (24). The 5? homology arm is a 4.9-kb
NotI?BstXI genomic fragment with an loxP site inserted into a
BstXI site 5? of the H2AX promoter. The 3? homology arm is a
Abbreviations: DNA-PKcs, DNA-dependent protein kinase catalytic subunit; ATM, ataxia
telangiectasia mutated; ATR, ATM and Rad3-related; DSB, double-strand break; NHEJ,
nonhomologous end-joining; HR, homologous recombination; IR, ionizing radiation; CSR,
class switch recombination; TCR, T cell receptor; RAG, recombination-activating gene
‡C.H.B., K.F.C., and J.S. contributed equally to this work.
?To whom correspondence should be addressed. E-mail: email@example.com.
June 11, 2002 ?
vol. 99 ?
no. 12 ?
generated via PCR with primers 5?-GGAGGGATCCTGTAC-
TACGTCTACATGGGG-3? and 5?-TCTCACCTTCCAGTTC-
with primers 5?-CTCTGGATCCCGTAGAGGGCAGA-
AGG-3? and 5?-GCGCGGATCCTGATTTCAAACTGTAT-
GCCAGGG-3?. The Neo probe was generated via PCR with
primers 5?-GCAGCCAATATGGGATCGGC-3? and 5?-
GTTCGGCTGGCGCGAGCCCC-3?. The IntH2AX probe is a
300-bp AatII?BglII fragment.
Gene Targeting and Generation of Embryonic Stem (ES) Cells. The
5L?3N targeting vector was electroporated into TC1 ES cells
(25) as described (26). Targeted clones were identified by
Southern blotting with the 3?H2AX probe on HindIII-digested
DNA (7.2 kb H2AXNeo, 14.3 kb H2AXWT) and confirmed with
the 5?H2AX probe on BamHI-digested DNA (8.8 kb H2AXNeo,
14.5 kb H2AXWT) and the IntH2AX probe on HindIII-digested
DNA to detect integration of the 5?loxP site (5.0 kb). To remove
the PGK-neorgene, 2.5 ? 106cells of independent H2AXWT/Neo
clones were infected with recombinant AdenoCre. H2AXWT/Flox
clones were identified via Southern blot analysis with the
5?H2AX probe on BamHI-digested DNA (8.4 kb). Indepen-
dent H2AXWT/Floxclones were retransfected with 5L?3N.
H2AXFlox/NeoES cells were identified by Southern blotting with
the 3?H2AX probe on HindIII-digested DNA (7.2 kb H2AXNeo,
9.3 kb H2AXFlox). H2AXFlox/Flox, H2AXFlox/?, and H2AX?/?ES
cells were made from independent H2AXFlox/NeoES clones.
XRCC4?/?TC1 ES cells were obtained through sequential gene
flanked by loxP sites (C.Y., unpublished data).
Preparation and Characterization of H2AX Antibodies. CK-
ATQASQEY and CKATQAS*QEY (the asterisk denotes phos-
phoserine) peptides were synthesized, coupled to keyhole limpet
hemocyanin (Pierce), and used to generate high titer polyclonal
antisera in rabbits. Affinity-purified antibody samples recog-
nized a predominant ?17-kDa band in histone preparations
from human or wild-type mouse cells but not from mouse
H2AX?/?ES cells. Only the Abs specific for S139-phosphory-
lated H2AX revealed a dose-dependent increase in immunoblot
signal and characteristic nuclear foci by immunostaining after IR
exposure of cells.
Histone Extraction and Western Blot Analysis. Histone preparations
were made as previously described (8). Western analysis was
performed with anti-H2AX rabbit polyclonal antisera (0.1 ?g?
ml) and antibodies to histone H4 (Cell Signaling).
IR Sensitivity and Genomic Instability Assays. ES cells passaged off
feeder cells were plated onto gelatinized plates. For IR sensi-
tivity assays, cells were irradiated 18 h later by the indicated
doses of ?-rays, cultured for 7 days, then stained and counted.
For genomic instability assays, cells were irradiated (150 rad)
24 h later, then cultured for 48 h. Metaphases were prepared and
analyzed as described (27).
V(D)J Recombination Assays. All ES cell assays were performed
and analyzed as described (28).
FACS Staining of H2AX?/?Lymphocytes.Lymphocytes isolated from
H2AX?/?RAG chimeric mice (29) were stained with FITC-
conjugated anti-CD8, anti-B220 as well as phycoerythrin-
conjugated anti-CD4 and anti-IgM antibodies (PharMingen)
and analyzed via a FACScan (Becton Dickinson).
Immunostaining and Confocal Microscopy. ES cells were plated on
gelatinized cover slips, and immunostaining was performed as
described (30). Cells were stained with phospho-H2AX-specific
polyclonal (1 ?g?ml), mouse BRCA1-specific monoclonal (1:5
dilution), and RAD51 polyclonal (Oncogene Science; 1:5 dilu-
tion) antibodies, and rhodamine- and FITC-conjugated second-
ary antibodies (Jackson ImmunoResearch; 1:50 dilution). Nuclei
were stained with To-pro-3 (Molecular Probes). Images were
collected by confocal microscopy (Bio-Rad Radiance 2000) and
processed by using Adobe PHOTOSHOP (Adobe Systems, Moun-
tain View, CA) software.
Generation of H2AX?/?ES Cells. To directly investigate potential
roles of H2AX in mammalian cells, we have generated and
characterized H2AX-deficient mouse ES cells. We generated
H2AX-deficient ES cells with a vector (5L?3N) designed to
create conditional H2AX null alleles (Fig. 1A). Because ex-
pressed drug resistance markers can have negative effects on
transcription of adjacent genes, we used a loxP-Neorcassette to
delete the drug resistance gene after targeting. We first gener-
ated targeted TC1 ES clones with the H2AX gene flanked by a
5?loxP site and a 3?loxP-Neorcassette (H2AXWT/Neo; Fig. 1B,
lane 2). Then, we used transient Cre expression to delete the
and H2AXFloxand H2AX?alleles. The H2AX promoter, exon, and polyadenyl-
show the relative locations of the 5?H2AX, IntH2AX, and 3?H2AX probes.
Restriction site designations: BX, BstXI; BH, BamHI; H3, HindIII. (B) Southern
blot analysis of H2AXWT/Neo(lane 2), H2AXWT/Flox(lane 1), H2AXFlox/Neo(lane 3),
and H2AX?/?(lane 4) HindIII-digested DNA probed with 3?H2AX. The sizes of
the bands are indicated. (C) Western blot analysis of TC1, H2AXFlox/?(no. 40),
and H2AX?/?(no. 45) ES cells with anti-H2AX and anti-H4 antibodies (loading
control). (D) Unirradiated and irradiated (15 min post 20 Gy) H2AXFlox/?and
H2AX?/?cells immunostained for ?-H2AX foci (red).
Generation and characterization of H2AX-deficient ES Cells. (A)
www.pnas.org?cgi?doi?10.1073?pnas.122228699Bassing et al.
Neorcassette to generate ES clones with a wild-type and a floxed
H2AX allele (H2AXWT/Flox; Fig. 1B, lane 1). H2AXWT/FloxES
cells were retargeted with 5L?3N vector to generate clones with
the H2AX gene floxed on one allele and flanked by a 5?loxP site
and a 3?loxP-Neorcassette on the second allele (H2AXFlox/Neo;
Fig. 1B, Lane 3). Transient Cre expression from a recombinant
AdenoCre vector in H2AXFlox/Neoclones resulted in the gener-
ation of ES cells with deletion of the H2AX gene on both alleles
(H2AX?/?; Fig. 1B, Lane 4). This same AdenoCre infection also
generated H2AXFlox/Floxand H2AXFlox/?ES cells that serve as
controls for potential effects of Cre expression. Western blotting
with an antibody specific for nonphosphorylated H2AX, the
major pool of H2AX in cells (8), confirmed that H2AX?/?ES
cells lack detectable H2AX protein (Fig. 1C). Thus, H2AX is
not required for ES cell viability; however, potential growth
defects remain to be examined. Consistent with complete dele-
H2AX-Deficient ES Cells Are Hypersensitive to Ionizing Radiation. To
evaluate the role of H2AX in DNA DSB repair, we used a colony
formation assay to assess the IR sensitivity of H2AX mutant vs.
normal ES cells. Two independent H2AX?/?ES cell subclones
were highly sensitive to IR as compared with wild-type TC1 ES
cells, but apparently less sensitive than XRCC4?/?ES cells (Fig.
2A). However, H2AXFlox/?ES cells did not exhibit increased IR
sensitivity (Fig. 2A). Two additional H2AX?/?ES subclones
from a second independently targeted ES cell, but neither
H2AXFlox/?nor H2AXFlox/FloxES cells from the same clone, were
also highly IR sensitive (data not shown). H2AX?/?ES cells also
showed increased sensitivity to the radiometric drug bleomycin
(data not shown). However, like S. cerevisiae with H2A lacking
the SQE motif, H2AX?/?ES cells were not obviously hypersen-
sitive to UV irradiation (data not shown). Therefore, H2AX
function contributes to the survival of mouse ES cells in the
presence of ionizing radiation and other DNA damaging agents
that induce chromosomal DSBs.
H2AX-Deficient ES Cells Exhibit Genomic Instability. The potential
role of H2AX in suppressing genomic instability was examined
by using spectral karyotyping (SKY) to examine levels and types
of chromosomal aberrations in early passage H2AX?/?ES cells
in comparison with H2AXFlox/?cells cultured for the same time.
SKY chromosomal painting allows unambiguous identification
of every mouse chromosome (31). We observed that 17–29% of
metaphases from three independently derived H2AX?/?cells
lines (nos. 45, 85, and 91) exhibited chromosomal abnormalities,
as opposed to 3–8% of metaphases from H2AXFlox/?control cell
lines (nos. 40 and 79; Table 1). The types of chromosomal
aberrations observed comprised fragments, detached centro-
meres, fusions, and translocations (Fig. 2B). Stable, high level
Cre expression in mouse embryonic fibroblasts can also lead to
genomic instability (32, 33). However, in our studies, Cre
expression was transient; moreover, increased instability was
observed in H2AX?/?vs. H2AXFlox/?lines that were generated
from the exact same AdenoCre infections of H2AXFlox/Neo
parental cells. AdenoCre-infected H2AXFlox/Floxcontrol cells
and other infected or uninfected ES cell lines show baseline
instability that, at the resolution of our current assays, was not
clearly different from that of H2AXFlox/?cells (ref. 34; D.O.F.
and J.C.F., unpublished observations). We conclude that early
passage H2AX?/?ES cells exhibited significantly increased
spontaneous chromosomal instability compared with controls.
However, without additional studies, we cannot rule out a minor
effect of the retained loxP sites or of H2AX haploinsufficiency
on maintenance of genomic stability.
We also examined karyotypic consequences of exposure to IR
in H2AX?/?, XRCC4?/?, and wild-type ES cells. Early passage
ES cells were exposed to 150 rad of IR and allowed to recover
for 48 h. SKY analyses revealed a significant increase in the
percentage of metaphases with chromosomal abnormalities in
the H2AX?/?and XRCC4?/?cells (70% and 55% of metaphases
with at least 1 aberration, respectively) compared with wild-type
controls (15% with abnormalities; Table 2). Furthermore, the
total number of abnormalities observed in irradiated H2AX-
deficient and wild-type ES cells differed by ?8-fold (1.7 per
metaphase vs. 0.2 per metaphase, respectively). XRCC4-
deficient ES cells exhibited a similar, albeit lower (?1.0 per
metaphase), level of IR-induced chromosomal aberrations than
H2AX-deficient ES cells. (A) IR sensitivity of wild-type (TC1), XRCC4?/?,
Data are plotted as the percentage of colonies that grew out at a given ?-ray
dose over unirradiated cells. The plotted numbers were obtained from trip-
licate data points of a representative experiment. (B and C) Metaphase
spreads of unirradiated (B) and ?-irradiated (C) H2AX?/?ES cells. (Left) 4?,6-
diamidino-2-phenylindole (DAPI)-stained chromosomes. (Right) SKY analysis.
Arrow points to chromosomal translocations between two different chromo-
somes (B and C) and a dicentric chromosome (C).
Increased ionizing radiation sensitivity and genomic instability of
Bassing et al.PNAS ?
June 11, 2002 ?
vol. 99 ?
no. 12 ?
H2AX-deficient cells (Table 2). After IR treatment, dramatic
chromosomal fragmentation, detached centromeres, fusions,
and translocations were observed in the H2AX-deficient cells
(Fig. 2C). Furthermore, irradiated H2AX-deficient cells exhib-
ited a significant increase in the percentage of metaphases with
aneuploidy (83%) in comparison with XRCC4?/?(23%) and
wild-type controls (20%) (Fig. 2C).
and cells deficient in checkpoint proteins often exhibit increased
chromosomal instability (35). Similar to wild-type ES cells (36,
37), H2AX?/?ES cells, as well as XRCC4?/?ES cells, arrest at
G2on exposure to 1,000 rad of IR (Fig. 4, which is published as
supporting information on the PNAS web site, www.pnas.org).
Therefore, the H2AX gene and, by inference, the formation of
?-H2AX foci, is not required for the normal DNA damage
checkpoint of ES cells. Consequently, the increased genomic
H2AX Is Not Required for V(D)J Joining in a Transient Assay. Defi-
ciencies in all known NHEJ proteins have major effects on the
joining of RAG-liberated coding and?or RS ends in the context
of transient V(D)J recombination substrates. Thus, to further
examine potential roles of H2AX, we examined the ability of
H2AX-deficient ES cell lines to support coding and RS joining
within extrachromosomal V(D)J recombination substrates by
using an established transient assay (28, 38). Two independently
derived H2AX?/?ES cell lines (nos. 45 and 85), as well as two
independent H2AXFlox/?(nos. 40 and 79), were assayed. V(D)J
recombination percentages observed in wild-type, H2AXFlox/?,
and H2AX?/?cells were quite similar (Table 3), thus demon-
strating that the efficiency of coding and RS joining within
plasmid substrates is not significantly impaired in H2AX-
deficient ES cells. Structures of coding joins from plasmids
recovered from H2AX?/?ES cells also were indistinguishable
from those of wild-type cells (Fig. 5, which is published as
supporting information on the PNAS web site). RS joins from
H2AX-deficient cells, like those of wild-type ES cells, were
precise (Table 3). Thus, H2AX is not required for normal
processing and ligating of either blunt or hairpinned, RAG-
mediated DSBs in extrachromosomal V(D)J substrates.
H2AX Is Not Required for Chromosomal V(D)J Recombination. Mice
are severely impaired in lymphocyte development because of
inability of progenitor lymphocytes to complete V(D)J joining
(17). To further examine potential roles of H2AX in V(D)J
recombination and lymphocyte development, we examined ef-
fects of H2AX deficiency on lymphocyte development via
RAG-2-deficient blastocyst complementation (RDBC; ref. 29).
Analysis of RAG-deficient chimeras generated from H2AX?/?
ES cells revealed the accumulation of significant numbers of
mature (CD4?, CD8?and B220?, IgM?) peripheral T and B
lymphocytes (Fig. 6A, which is published as supporting infor-
mation on the PNAS web site), although the absolute numbers
were significantly reduced compared with those of wild-type
mice. However, we cannot make quantitative conclusions re-
garding the role of H2AX in lymphocyte development because
of inherent limitations of RDBC. Thus, H2AX, in marked
contrast to all known NHEJ factors, is not required for chro-
mosomal V(D)J recombination or lymphocyte development. We
also PCR-amplified and sequenced TCR? coding joins from an
H2AX?/?thymus and found that they were indistinguishable
from those of wild-type (Fig. 6B). Thus, H2AX is not required
absolutely for the normal ligation of RAG-liberated DSBs in
Altered IR-Induced BRCA1 and RAD51 Focus Formation in H2AX?/?ES
Cells. The formation of ?-H2AX foci is a very early event after
induction of DSBs and has been proposed to function in
recruiting downstream DNA repair factors, including homolo-
gous recombination factors BRCA1 and RAD51, to DNA
damage sites (18). To assess BRCA1 and RAD51 focus forma-
tion in H2AX?/?ES cells, we examined IR-induced foci by
immunofluorescence. In the absence of irradiation, both
H2AXFlox/?and H2AX?/?cells exhibited on average one or two
BRCA1 and RAD51 foci (Fig. 3A). Within 6 h postirradiation,
nearly all wild-type and H2AXFlox/?cells had at least five foci per
cell (Fig. 3B). Notably, in H2AX?/?cells, a much more modest
induction of BRCA1 foci was observed on irradiation, although,
when viewed at higher gain, some focal staining was clearly
detectable (Fig. 3B). In contrast, similar numbers of RAD51 foci
were observed 6 h postirradiation in H2AX?/?and H2AXFlox/?
Table 1. Increased spontaneous genomic instability in H2AX?/?ES cells
GenotypeH2AXFlox/?no. 40H2AXFlox/?no. 79H2AX?/?no. 45 H2AX?/?no. 85H2AX?/?no. 91
Total anomalies (%)†
Percentage with abnormalities‡
*One translocation was reciprocal.
†The percentage represents events per 100 metaphases.
‡Percentage of metaphases with at least one aberration.
Table 2. Increased IR-induced genomic instability in H2AX?/?
Genotype Wild-type TC1 XRCC4?/?
Total anomalies (%)†
Forty-eight hours post 150 rad.
*One translocation was reciprocal.
†The percentage represents events per 100 metaphases.
‡Percentage of metaphases with at least 1 aberration.
www.pnas.org?cgi?doi?10.1073?pnas.122228699Bassing et al.
cells. However, these RAD51 foci appeared smaller in H2AX?/?
cells than H2AXFlox/?cells (Fig. 3B). The differences seen in the
pattern of RAD51 foci could reflect induction of different
subpopulations of RAD51 foci in H2AX?/?vs. H2AXFlox/?cells.
Alternatively, the total numbers of RAD51 molecules recruited
to individual DNA damage sites may be reduced in H2AX?/?
cells. Regardless, our data indicate that H2AX contributes to
BRCA1 and, apparently, RAD51 focus formation responses
H2AX-Deficient ES Cells Exhibit Genomic Instability. We have dem-
onstrated a critical role for the H2AX core histone in the cellular
response to chromosomal DSBs. H2AX-deficient ES cells have
increased IR sensitivity and elevated levels of both spontaneous
and IR-induced genomic instability. Because H2AX is not
required for the normal IR-induced DNA damage checkpoint,
increased genomic instability in H2AX-deficient ES cells is likely
due to a DNA repair defect. In S. cerevisiae, the H2A core
histone also functions in repair of DSBs (7). In addition, a
parallel gene-targeted mutation study reached similar conclu-
sions regarding mammalian H2AX function (39). Therefore, the
essential function of a core histone in the cellular response to
chromosomal DSBs is conserved throughout evolution, from
yeast to mammals. Given the role of H2AX in suppression of
genomic instability and its role in recruitment of BRCA1 to
DNA damage sites, H2AX also may be predicted to function as
a tumor suppressor.
H2AX Is Not Required for Catalysis of V(D)J Recombination. Absence
of H2AX had no effect on V(D)J recombination (either effi-
ciency or quality) within extrachromosomal V(D)J recombina-
tion substrates in ES cells and allowed generation of normal
V(D)J coding joins in H2AX-deficient thymocytes. Because
deficiencies in known NHEJ factors severely impair V(D)J
recombination in both contexts, our findings indicate that H2AX
does not directly participate in catalysis of this reaction and, by
extension, in catalysis of classical NHEJ. Given that H2AX is not
directly involved in V(D)J recombination, localization of RAG-
dependent ?-H2AX foci to the TCR? locus may function more
Table 3. Analysis of signal and coding joining in H2AX?/?ES cells
Cell lines (Ampr? Camr)?Ampr
% Relative levelFidelity (%)
pJH290 (coding joining)
pJH200 (signal joining)
H2AXFlox/?ES cells. Immunofluorescence of H2AXFlox/?and H2AX?/?cells unir-
radiated (A) or 6 h postirradiation with 20 Gy (B) to visualize foci of BRCA1
(rhodamine, red) and RAD51 (FITC, green). Nuclei appear blue (Topro-3). In
the merged images, overlapping foci appear yellow.
Formation of IR-induced BRCA1 and RAD51 foci in H2AX?/?vs.
Bassing et al. PNAS ?
June 11, 2002 ?
vol. 99 ?
no. 12 ?
generally to effect a DNA damage response to RAG-initiated
DSBs (16), possibly in the context of ATM-mediated surveil-
lance (40). However, H2AX still may be required to effect
efficient NHEJ-mediated repair of other chromosomal DSBs
perhaps via accessibility?recruitment functions. In this context,
the observations that G1-specific ?-H2AX and NBS foci are
linked to the IgH locus during CSR and that CSR is substantially
impaired in H2AX?/?lymphocytes (10) clearly implicate H2AX
in CSR, a recombination process likely completed by NHEJ.
Function of H2AX in DNA Repair. We find that H2AX deficiency
causes a dramatic decrease in BRCA1 focus formation after IR,
and a more subtle decrease in the size of IR-induced RAD51
foci. These effects could reflect inefficient propagation of spe-
cialized chromatin tracts around a break, a reduction in the
or both. Therefore, by virtue of its molecular associations, the
H2AX protein likely has a role in modulating DNA repair via
homologous recombination as well as NHEJ. Consistent with
this notion, H2AX-deficient ES cells appear hypersensitive to
the cross-linking agent mitomycin C (ref. 39; C.H.B. and H.S.,
unpublished data), a phenotype shared with ES cells deficient in
the homologous recombination factor RAD54 (41), but not
NHEJ-deficient ES cells (J.S., unpublished data). Phosphoryla-
tion of H2A (in yeast) and, presumably, H2AX (in mammals)
causes an alteration in chromatin structure that may facilitate
DNA repair (7). Therefore, H2AX may function to promote
‘‘accessible’’ chromatin and thus facilitate the kinetics through
which DNA repair factors associate with and repair DSBs.
Finally, the IR-dependent physical association between ?-H2AX
and 53BP1 (23) suggests that ?-H2AX may also function as a
docking site for factors involved in the cellular response to DNA
damage. Overall, H2AX?/?ES cells should provide an important
model system for the molecular analysis of the precise nature of
H2AX function in the repair of chromosomal DSBs.
We thank Barbara Woodman and Laurie Davidson for technical assis-
for critical review of this manuscript. C.H.B. is a Research Associate of
the Howard Hughes Medical Institute. J.S. is a Special Fellow of the
Leukemia and Lymphoma Society. K.F.C. is a Fellow of the Jane Coffin
Childs Memorial Fund for Medical Research. F.W.A. is an Investigator
of the Howard Hughes Medical Institute. This work was supported by
National Institutes of Health Grant AI35714 (F.W.A.), National Cancer
Institute (NCI) Grant CA92625 (F.W.A.), and a NCI Howard Temin
1. Wolffe, A. P. (1998) Chromatin Structure and Function (Academic, New York).
2. Redon, C., Pilch, D., Rogakou, E., Sedelnikova, O., Newrock, K. & Bonner, W.
(2002) Curr. Opin. Genet. Dev. 12, 162–169.
3. Abraham, R. T. (2001) Genes Dev. 15, 2177–2196.
4. Hoeijmakers, J. H. (2001) Nature (London) 411, 366–374.
5. Cromie, G. A., Connelly, J. C. & Leach, D. R. (2001) Mol. Cell 8, 1163–1174.
6. Strahl, B. D. & Allis, C. D. (2000) Nature (London) 403, 41–45.
7. Downs, J. A., Lowndes, N. F. & Jackson, S. P. (2000) Nature (London) 408,
8. Rogakou, E. P., Pilch, D. R., Orr, A. H., Ivanova, V. S. & Bonner, W. M. (1998)
J. Biol. Chem. 273, 5858–5868.
9. Rogakou, E. P., Boon, C., Redon, C. & Bonner, W. M. (1999) J. Cell Biol. 146,
10. Petersen, S., Casellas, R., Reina-San-Martin, B., Chen, H. T., Difilippantonio,
M. J., Wilson, P. C., Hanitsch, L., Celeste, A., Muramatsu, M., Pilch, D. R., et
al. (2001) Nature (London) 414, 660–665.
11. Kenter, A. & Wuerffel, R. (1999) Ann. N. Y. Acad. Sci. 870, 206–217.
12. Manis, J. P., Gu, Y., Lansford, R., Sonoda, E., Ferrini, R., Davidson, L.,
Rajewsky, K. & Alt, F. W. (1998) J. Exp. Med. 187, 2081–2089.
13. Casellas, R., Nussenzweig, A., Wuerffel, R., Pelanda, R., Reichlin, A., Suh, H.,
Qin, X. F., Besmer, E., Kenter, A., Rajewsky, K. & Nussenzweig, M. C. (1998)
EMBO J. 17, 2404–2411.
14. Rolink, A., Melchers, F. & Andersson, J. (1996) Immunity 5, 319–330.
15. Manis, J. P., Dudley, D., Kaylor, L. & Alt, F. W. (2002) Immunity 16, 607–617.
16. Chen, H. T., Bhandoola, A., Difilippantonio, M. J., Zhu, J., Brown, M. J., Tai,
X., Rogakou, E. P., Brotz, T. M., Bonner, W. M., Ried, T. & Nussenzweig, A.
(2000) Science 290, 1962–1965.
17. Bassing, C. H., Swat, W. & Alt, F. W. (2002) Cell 109, S45–S55.
18. Paull, T. T., Rogakou, E. P., Yamazaki, V., Kirchgessner, C. U., Gellert, M. &
Bonner, W. M. (2000) Curr. Biol. 10, 886–895.
19. Burma, S., Chen, B. P., Murphy, M., Kurimasa, A. & Chen, D. J. (2001) J. Biol.
Chem. 276, 42462–42467.
20. Shiloh, Y. (2001) Biochem. Soc. Trans. 29, 661–666.
21. Jeggo, P. A. (1998) Radiat. Res. 150, S80–91.
22. Schultz, L. B., Chehab, N. H., Malikzay, A. & Halazonetis, T. D. (2000) J. Cell
Biol. 151, 1381–1390.
23. Rappold, I., Iwabuchi, K., Date, T. & Chen, J. (2001) J. Cell Biol. 153, 613–620.
24. Gorman, J. R., van der Stoep, N., Monroe, R., Cogne, M., Davidson, L. & Alt,
F. W. (1996) Immunity 5, 241–252.
25. Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A. & Leder, P. (1996) Cell 84,
26. Sleckman, B. P., Bardon, C. G., Ferrini, R., Davidson, L. & Alt, F. W. (1997)
Immunity 7, 505–515.
27. Ferguson, D. O., Sekiguchi, J. M., Chang, S., Frank, K. M., Gao, Y., DePinho,
R. A. & Alt, F. W. (2000) Proc. Natl. Acad. Sci. USA 97, 6630–6633.
28. Gu, Y., Jin, S., Gao, Y., Weaver, D. T. & Alt, F. W. (1997) Proc. Natl. Acad.
Sci. USA 94, 8076–8081.
29. Chen, J., Lansford, R., Stewart, V., Young, F. & Alt, F. W. (1993) Proc. Natl.
Acad. Sci. USA 90, 4528–4532.
30. Scully, R., Chen, J., Ochs, R. L., Keegan, K., Hoekstra, M., Feunteun, J. &
Livingston, D. M. (1997) Cell 90, 425–435.
31. Liyanage, M., Coleman, A., du Manoir, S., Veldman, T., McCormack, S.,
Dickson, R. B., Barlow, C., Wynshaw-Boris, A., Janz, S., Wienberg, J., et al.
(1996) Nat. Genet. 14, 312–315.
32. Silver, D. P. & Livingston, D. M. (2001) Mol. Cell 8, 233–243.
33. Loonstra, A., Vooijs, M., Beverloo, H. B., Allak, B. A., van Drunen, E., Kanaar,
R., Berns, A. & Jonkers, J. (2001) Proc. Natl. Acad. Sci. USA 98, 9209–9214.
34. d’Adda di Fagagna, F., Hande, M. P., Tong, W. M., Roth, D., Lansdorp, P. M.,
Wang, Z. Q. & Jackson, S. P. (2001) Curr. Biol. 11, 1192–1196.
35. Dasika, G. K., Lin, S. C., Zhao, S., Sung, P., Tomkinson, A. & Lee, E. Y. (1999)
Oncogene 18, 7883–7899.
36. Schmidt-Kastner, P. K., Jardine, K., Cormier, M. & McBurney, M. W. (1998)
Oncogene 16, 3003–3011.
37. Hirao, A., Kong, Y. Y., Matsuoka, S., Wakeham, A., Ruland, J., Yoshida, H.,
Liu, D., Elledge, S. J. & Mak, T. W. (2000) Science 287, 1824–1827.
38. Hesse, J. E., Lieber, M. R., Gellert, M. & Mizuuchi, K. (1987) Cell 49, 775–783.
39. Celeste, A., Petersen, S., Romanienko, P. J., Fernandez-Capetillo, O., Chen,
H. T., Sedelnikova, O., Reina-San-Martin, B., Coppola, V., Meffre, E.,
Difilippantonio, M. J., et al. (2002) Science 296, 922–927.
40. Perkins, E. J., Nair, A., Cowley, D. O., Van Dyke, T., Chang, Y. & Ramsden,
D. A. (2002) Genes Dev. 16, 159–164.
41. Essers, J., Hendriks, R. W., Swagemakers, S. M., Troelstra, C., de Wit, J.,
Bootsma, D., Hoeijmakers, J. H. & Kanaar, R. (1997) Cell 89, 195–204.
www.pnas.org?cgi?doi?10.1073?pnas.122228699Bassing et al.