Replication-Uncoupled Histone Deposition during Adenovirus DNA
Tetsuro Komatsu and Kyosuke Nagata
Department of Infection Biology, Faculty of Medicine and Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, Japan
unit of the chromatin structure is the nucleosome, which consists
of a histone octamer (two copies each of histones H2A, H2B, H3,
and H4) and DNA wrapping around the octamer. The deposition
of histones and/or the remodeling of nucleosome arrays is a crit-
ical process for the expression of genome functions (2), since
nucleosome packaging could be a barrier for trans-acting factors
several events on chromatin, such as transcription, DNA replica-
tion, and DNA repair.
Currently, it is known that histone deposition is carried out
dent ones, and a role of histone variants in these deposition path-
ways has been elucidated (14). In mammalian somatic cells, there
are three major histone H3 variants, H3.1, H3.2, and H3.3, and
H3.2 (which differs by only 1 amino acid [aa] from H3.1) are
expressed exclusively during the S phase, while the expression of
the variant H3.3, which differs by 4 and 5 aa from H3.2 and H3.1,
ant is called a “replication-independent” one (11). Tagami et al.
demonstrated previously that the canonical histone H3 (H3.1)
interacts with the histone chaperone CAF-1 complex and is de-
posited onto DNA in a replication-dependent manner, while
three subunits, p150, p60, and p48, and is associated with the
cellular DNA replication machinery through an interaction with
PCNA, a sliding clamp for DNA polymerases, allowing the DNA
replication-coupled deposition of histones (40, 41, 50). On the
histone chaperone in cell-free systems using Xenopus laevis egg
extracts (32), and histone variant H3.3 was shown to mark tran-
n the cell nucleus, the genomic DNA is not naked but forms a
scriptionally active genomic regions (1). Furthermore, additional
H3.3-specific chaperones were recently identified. Daxx is one of
and was reported to deposit histone H3.3 onto specific genomic
regions, such as telomeres and pericentric heterochromatin,
together with an ATP-dependent chromatin remodeler, ATRX
(10, 21). It was also reported that in Drosophila melanogaster
cells, DEK is a coactivator of a nuclear receptor and functions
as an H3.3-specific chaperone (37). Thus, the mechanistic ev-
idences for histone deposition are accumulating in the case of
The regulatory events for the chromatin structure are not lim-
ited to the cellular genome, as some viruses also have chromatin
and/or chromatin-like structures with their own genomes. The
adenovirus (Ad) genome is a linear double-stranded DNA
(dsDNA) of ?36,000 bp in length. In the virion, the Ad genome
forms a chromatin-like structure with viral basic core proteins, as
revealed by electron microscopic analyses showing that viral core
protein-DNA complexes purified from the virion show a “beads-
viral DNA-core protein complexes purified from the virion are
tems, the reactions occur at a much lower level than in the case of
naked DNA, indicating that the viral chromatin-like structure
must be remodeled to execute its genome functions (22, 23). Pre-
Received 14 February 2012 Accepted 3 April 2012
Published ahead of print 11 April 2012
Address correspondence to Kyosuke Nagata, firstname.lastname@example.org.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
June 2012 Volume 86 Number 12Journal of Virologyp. 6701–6711 jvi.asm.org
viously, we identified host cell-derived remodeling factors for Ad
chromatin by biochemical analyses (19, 22, 24, 26) and demon-
strated that TAF-I, one of these host factors, plays an important
through interactions with protein VII (15, 17, 18, 20, 27). Thus, it
was indicated that the remodeling of Ad chromatin is a crucial
genome. In addition, using chromatin immunoprecipitation
(ChIP) assays, we recently reported that in early phases of infec-
tion, cellular histones are incorporated into viral DNA-protein
VII complexes and that a histone modification occurs depending
upon the transcription states on viral chromatin, suggesting that
tin in infected cells (20).
its regulation in early phases of infection are being clarified, it is
phases of infection. In particular, since the expression of viral late
genes is dependent largely on its own DNA replication (45), the
regulation of the chromatin structure during viral DNA replica-
tion could be a key step. Therefore, in this study, we sought to
elucidate the regulatory mechanism of how the chromatin struc-
replication, in particular with respect to histone deposition. We
found that after the onset of viral DNA replication, cellular his-
tones are also incorporated into viral chromatin. We also found
that although CAF-1 is accumulated at the site of viral DNA rep-
during viral DNA replication, since the knockdown of CAF-1 did
not affect the binding level of histone H3 on viral chromatin, and
DNA even after the onset of viral DNA replication. Microscopic
analyses suggested that histones but not USF1, a transcription
factor which was shown previously to bind to and regulate tran-
scription from the viral major late promoter (MLP) (46), are ex-
cluded form the site of viral DNA replication, possibly by the
oligomerization of Ad single-stranded DNA (ssDNA) binding
these results, we propose a model whereby, unlike cellular chro-
not coupled with viral DNA replication. A feasible role of this
tion in Ad genome functions is discussed.
MATERIALS AND METHODS
infection of human adenovirus type 5 (HAdV5) were carried out essen-
tially as described previously (18, 20). Hydroxyurea (HU) was added at a
final concentration of 2 mM immediately after infection when DNA rep-
lication was to be blocked. HeLa cells stably expressing enhanced green
fluorescent protein (EGFP)-tagged histones H3.2 and H3.3 (a kind gift
from M. Okuwaki, University of Tsukuba) were also maintained as de-
scribed above. The transfection of expression plasmids was performed by
using GeneJuice (Novagen) according to the manufacturer’s protocol.
Antibodies. Antibodies used in this study are as follows: rabbit anti-
histone H3 (catalog no. ab1791; Abcam), rabbit anti-histone H4 (catalog
no. 04-858; Millipore), rabbit anti-histone H2A (catalog no. ab18255;
Abcam), mouse anti-HIRA (catalog no. 04-1488; Millipore), mouse anti-
FLAG M2 (catalog no. F3165; Sigma), rat anti-hemagglutinin (HA)
(3F10; Roche), and mouse anti-?-actin (Sigma) antibodies. Rabbit anti-
histone H2A-H2B, mouse anti-CAF-1 p150, and mouse anti-DBP anti-
bodies were kindly provided by M. Okuwaki (University of Tsukuba), A.
Verreault (University of Montreal), and W. C. Russel, respectively. Rat
anti-protein VII antibody was described elsewhere previously (17).
Vector construction. To construct the expression vectors for USF1,
full-length DBP, and its deletion mutant (DBP?C, which lacks the C-ter-
by PCR; digested with BamHI and EcoRI; and cloned in frame into the
gene (pCHA-puro vector; kindly provided by K. Kajitani, University of
Tsukuba). The resulting vectors were designated pCHA-puro-USF1,
pCHA-puro-DBP, and pCHA-puro-DBP?C, respectively. Similarly, for
5=-GTTTAGGATCCCATATGAAGGGGCAGCAG-3= and 5=-GGGCCG
AATTCTTAGTTGCTGTCATTCTTG-3= for USF1 cDNA, 5=-AAAGGA
TCCATGGCCAGTCGGG-3= and 5=-GCGGAATTCTTAAAAATCAAA
GGGGTTCTG-3= for DBP cDNA, 5=-AAAGGATCCATGGCCAGTCGG
G-3= and 5=-CCCGAATTCTTAGTTGCGATACTGG-3= for DBP?C
cDNA, and 5=-AAAGGATCCATGTTCGAGGCGC-3= and 5=-ATCGTC
GACCTAAGATCCTTCTTC-3= for PCNA cDNA.
For the preparation of cells stably expressing HA-PCNA, HeLa cells
2 ?g/ml puromycin for 2 weeks.
fragment of histone H3.1 was amplified by PCR, digested with NcoI, and
from pBS-H3.1-FLAG by digestion with BamHI and EcoRI and cloned
into the pcDNA3 vector (pcDNA3-H3.1-FLAG). The primers used here
were as follows: 5=-AAAACCATGGCGCGTACTAAGCAG-3= and 5=-TT
The expression vectors for FLAG-tagged histones H3.2 and H3.3
(pcDNA3-H3.2-FLAG and pcDNA3-H3.3-FLAG, respectively) and HA-
S. Saito, respectively (University of Tsukuba).
Indirect immunofluorescence assays. Indirect immunofluorescence
(IF) assays were carried out essentially as previously described (18). The
mouse IgG conjugated with Alexa Fluor 488, anti-mouse IgG conjugated
with Alexa Fluor 568, and anti-rabbit IgG conjugated with Alexa Fluor
568; Invitrogen). DNA was visualized by staining with TO-PRO-3 iodide
(Invitrogen). Labeled cells were observed by confocal laser scanning mi-
croscopy (LSM5 Exciter; Carl Zeiss), using argon laser (488 nm) and
He/Ne laser (546 and 633 mm) lines.
ChIP, RT-PCR, siRNA-mediated knockdown, and Western blot as-
says. ChIP, reverse transcription (RT)-PCR, small interfering RNA
(siRNA)-mediated knockdown, and Western blot assays were carried out
essentially as described previously (20). siRNA targeted for CAF-1 p150
was purchased commercially (Stealth siRNA; Invitrogen). Primers used
and RT-PCR assays were described elsewhere previously (20). In ChIP
were quantitatively measured by qPCR, and mean values with standard
deviations (SD) were obtained from three independent experiments.
Cellular histones are bound with viral chromatin in both early
phases of infection (before the onset of viral DNA replication),
viral chromatin is composed of both viral core protein VII and
cellular histones and that this “chimeric” chromatin functions as
the template for transcription (20). To examine whether histones
Komatsu and Nagata
jvi.asm.org Journal of Virology
are also bound with viral chromatin after the onset of viral DNA
replication, we performed ChIP assays using antibodies against
histones and protein VII (Fig. 1). Viral DNA replication starts at
after viral DNA replication, HeLa cells infected at an MOI (mul-
tiplicity of infection) of 100 were harvested at 6 and 12 hpi for
ChIP assays. We chose five regions for ChIP assays, four viral
this condition, the amount of viral DNA was increased by ?20-
fold through viral DNA replication (Fig. 1B). At 6 hpi, all core
histones were bound with viral chromatin but at a low binding
level compared with that of cellular chromatin (histones H3, H4,
and H2A-H2B) (Fig. 1C). This was in good agreement with our
previous observations (20). At 12 hpi, core histones were also
found to be associated with viral chromatin. The binding level of
but slightly lower than that on cellular chromatin. This is consis-
ing that viral genome DNA purified from infected cells at late
phases of infection has nucleosome-like particles, which are less
binding level of protein VII was drastically decreased after the
onset of viral DNA replication (Fig. 1C), suggesting that newly
synthesized viral DNA is associated mainly with cellular histones.
We do not exclude the possibility that protein VII remains asso-
ciated with a small population of viral chromatin, because the
binding level of protein VII on viral chromatin was still higher
than that on cellular chromatin even at 12 hpi. The ratio among
core histones bound on viral chromatin was almost the same as
that among core histones bound on cellular chromatin both at 6
hpi (20; data not shown) and at 12 hpi (Fig. 1D), indicating that
viral chromatin contains the canonical nucleosome structure.
CAF-1 and PCNA are not involved in histone deposition
onto newly synthesized viral DNA. It is known that during the
DNA replication of the cellular genome and some DNA virus ge-
histone chaperone (41, 43). CAF-1 is associated with the DNA
replication machinery through interactions with PCNA, thereby
enabling the replication-coupled deposition of histone H3-H4
complexes (40). Thus, it was worthwhile to examine whether
newly synthesized Ad DNA, although there is no definitive evi-
dence that these two are involved in Ad DNA replication. To test
tagged PCNA (HA-PCNA) to examine the relationship among
viral DNA replication, PCNA, and CAF-1 (Fig. 2A). By using an
antibody against DBP, an Ad ssDNA binding protein involved in
nated “viral DNA replication foci” (here referred to as “VDRF”),
can be visualized (31). HeLa cells or cells stably expressing HA-
PCNA were infected at an MOI of 50, and at 18 hpi, the cells were
subjected to IF assays using anti-DBP and anti-HA antibodies. In
mock-infected cells, HA-PCNA was localized throughout the nu-
as reported previously (29). At 18 hpi, VDRF was observed as a
donut-like signal by using an anti-DBP antibody, and we found
that HA-PCNA, showing a punctate localization, was accumu-
from infected cells at 6 and 12 hpi. The amount of viral DNA was quantitatively measured by qPCR using primers for the E1A promoter region. The amount of
infected cells at 6 and 12 hpi. Immunoprecipitation was carried out by using the indicated antibodies and an anti-FLAG antibody (as a negative control). The
against that obtained in a negative control (anti-FLAG antibody). (D) Binding levels of core histones. Based on the results of ChIP assays, shown in panel C, the
binding levels of histone H4 and H2A-H2B were normalized to that of histone H3.
DBP and Histone H3.3 in Adenovirus DNA Replication
June 2012 Volume 86 Number 12jvi.asm.org 6703
of CAF-1 inside VDRF (see Fig. 4A). These results suggest that
PCNA and CAF-1 are recruited together to the site of viral DNA
Next, to investigate a role of CAF-1 in histone deposition onto
viral DNA, the siRNA-mediated knockdown of CAF-1 p150, the
largest subunit of the CAF-1 complex, was carried out (Fig. 2B to
for CAF-1 p150 (siCAF-1), infected with HAdV5 at an MOI of
100, and harvested at 6 and 12 hpi. First, we examined the knock-
down efficiency of CAF-1 p150 by RT-qPCR assays (Fig. 2B). The
of that in control cells, although at 12 hpi, the level was slightly
increased, possibly due to the S-phase-like environment induced
by Ad infection (30). In contrast, the mRNA level of glyceralde-
hyde-3-phosphate dehydrogenase (GAPDH) was almost unaf-
fected by siRNA treatment and Ad infection. Under this condi-
tion, the binding level of histone H3 on viral chromatin was
examined by ChIP assays (Fig. 2C). The binding level of H3 on
viral chromatin was not decreased by the CAF-1 knockdown and,
rather, slightly increased (but not statistically significantly) at 12
hpi (E1A pro and MLP) (Fig. 2C). Note that the binding level of
H3 on cellular chromatin was also unaffected by the CAF-1
knockdown (rRNA gene) (Fig. 2C, and see Discussion). In addi-
viral DNA replication levels (Fig. 2D). Taken together, these re-
sults suggest that it is not likely that CAF-1 is involved in the
histone deposition onto viral chromatin during viral DNA repli-
cation, although CAF-1 was accumulated at VDRF together with
rated into viral chromatin. It is known that among histone H3
variants, histones H3.1 and H3.2 are deposited onto DNA by
CAF-1 during DNA replication, while H3.3 is deposited indepen-
dently of DNA replication (11, 14). If CAF-1 is not involved in
histone deposition during viral DNA replication, histone H3.3,
formed ChIP assays using FLAG-tagged histones H3.2 and H3.3
(Fig. 3). HeLa cells were transfected with expression vectors for
H3.2 and H3.3, and at 24 hpt (hours posttransfection), the cells
were infected at an MOI of 100. We first studied cells at early
phases of infection (before the onset of viral DNA replication)
(Fig. 3A). Infected cells were harvested at 2 and 6 hpi as well as at
10 hpi in the presence of HU to block viral DNA replication (20)
FIG 2 Localization and role of PCNA and CAF-1 during viral DNA replication. (A) IF assays. HeLa cells and cells stably expressing HA-PCNA grown on
coverslips were mock infected or infected with HAdV5 at an MOI of 50. At 18 hpi, the localization patterns of HA-PCNA and DBP were analyzed by IF using
anti-HA and anti-DBP antibodies. DNA was visualized by TO-PRO-3 iodide staining. Merged images are also indicated. Higher-magnification images of the
sets for CAF-1 p150 and GAPDH mRNAs. The mRNA levels relative to those of control cells at 12 hpi were graphed. (C) ChIP assays. siRNA-treated cells were
amounts of viral DNA. Viral DNA was purified from lysates for ChIP assays (C) and subjected to qPCR using the primer set for the E1A promoter. The DNA
amounts at 12 hpi relative to those at 6 hpi are shown.
Komatsu and Nagata
jvi.asm.org Journal of Virology
and subjected to ChIP assays with an anti-FLAG antibody. As
was observed at all the regions that we tested, with a gradual in-
crease as infection proceeded. This finding is consistent with data
from a recent report of a helper-dependent Ad vector (HdAd)
indicating that histone H3.3 is specifically deposited onto HdAd
DNA by HIRA, an H3.3-specific histone chaperone (34). Since
state of HdAd may reflect that of wild-type Ad in early phases of
infection (13, 34).
Next, we performed ChIP assays at 12 hpi to examine which
H3 variant is deposited onto newly synthesized viral DNA (Fig.
3B). We found that histone H3.3 but not H3.2 was associated
with viral chromatin at this time point, as observed in the early
phases of infection. The expression levels of both H3 variants
were comparable (Fig. 3C), and both variants were associated
with cellular chromatin with a similar binding level (rRNA
gene) (Fig. 3A and B). These findings strongly suggest that this
result was not due to some technical issues. Furthermore, we
obtained the same results by using FLAG-tagged histone H3.1
instead of H3.2 (Fig. 3D and E). Thus, these results suggest that
replication-independent histone variant H3.3 is selectively de-
posited onto not only incoming but also newly synthesized
viral DNA in infected cells.
from the site of viral DNA replication. To further investigate the
FIG 3 Incorporation of histone H3 variants into viral chromatin. (A and B) ChIP assays with FLAG-tagged histone H3 variants. HeLa cells were
transfected with the pcDNA3 empty vector, pcDNA3-H3.2-FLAG, or pcDNA3-H3.3-FLAG and infected with HAdV5 at an MOI of 100 at 24 hpt (hours
posttransfection). At 2, 6, and 10 hpi (A) or at 12 hpi (B), ChIP assays were carried out by using anti-FLAG and anti-HA (as a negative control) antibodies.
Note that at 10 hpi, HU was added to block viral DNA replication. The results were graphed as relative enrichments. (C) Western blot analyses. At 24 hpt,
subjected to 15% SDS-PAGE, followed by Western blot analyses using anti-FLAG (top) and anti-?-actin (bottom) antibodies. (D) Western blot analyses.
assays. HeLa cells transfected with the pcDNA3 empty vector (lanes 2, 3, 9, and 10), pcDNA3-H3.1-FLAG (lanes 4, 5, 11, and 12), or pcDNA3-H3.3-FLAG
(lanes 6, 7, 13, and 14) were infected with HAdV5 at an MOI of 100. At 10 hpi (left, lanes 1 to 7) or at 12 hpi (right, lanes 8 to 14), ChIP assays were carried
out by using anti-FLAG (lanes 3, 5, 7, 10, 12, and 14) and anti-HA (as a negative control) (lanes 2, 4, 6, 9, 11, and 13) antibodies. At 10 hpi, HU was added
to block viral DNA replication. The immunoprecipitated DNAs were amplified by semiquantitative PCR using the indicated primer sets. PCR products
were separated on a 7% polyacrylamide gel and visualized by staining with ethidium bromide (EtBr). Input DNAs (lanes 1 and 8) were purified from 0.5%
of lysates of cells transfected with the empty vector.
DBP and Histone H3.3 in Adenovirus DNA Replication
June 2012 Volume 86 Number 12jvi.asm.org 6705
fluctuation in histone levels during viral DNA replication, we per-
formed IF analyses using HeLa cells stably expressing EGFP-tagged
top). VDRF was observed at 18 hpi, as described above, and H3.3-
EGFP was found to be excluded from VDRF. Similar results were
obtained by using cells stably expressing EGFP-tagged histone H3.2
nous histone using anti-histone H2A antibody (Fig. 4C). This local-
ization pattern was specific for the late phases of infection, since the
localization of EGFP-tagged H3 variants was not changed in early
phases of infection (Fig. 4D). We also performed IF analyses using
anti-CAF-1 p150 antibody and observed that CAF-1 was accumu-
observed for HA-PCNA (Fig. 2A). In summary, these results indi-
FIG 4 Localization of histones in late phases of infection. (A) IF analyses using cells stably expressing histone H3.3-EGFP. HeLa cells stably expressing histone
H3.3-EGFP grown on coverslips were mock infected or infected at an MOI of 50, and at 18 hpi, the cells were subjected to IF assays using anti-DBP (top) and
phases of infection. HeLa cells stably expressing histone H3.2-EGFP were mock infected or infected with HAdV5 at an MOI of 50, and at 18 hpi, the cells were
using anti-histone H2A and anti-DBP antibodies. Higher-magnification images of the regions marked by squares are shown below. (D) Histone localization in
4 hpi, the cells were subjected to IF analyses using an anti-protein VII antibody.
Komatsu and Nagata
jvi.asm.org Journal of Virology
cated that histones are localized reciprocally to VDRF (and CAF-1/
To gain more insights into the accessibility of other nuclear
proteins to VDRF, we again performed IF analyses (Fig. 5). First,
IF analyses were carried out by using antibody against HIRA, an
H3.3-specific histone chaperone, and it was observed that the lo-
calization of HIRA was not drastically changed in both the early
and late phases of infection (Fig. 5A). We also performed IF anal-
yses using cells transiently transfected with expression vectors for
HA-tagged DEK and USF1 (Fig. 5B and C). DEK is a cellular
and USF1 is an E-box binding transcription factor and was re-
ported previously to bind to and regulate transcription from the
tors, and at 24 hpt, the cells were subjected to Western blot anal-
yses (Fig. 5B) or infected at an MOI of 50. At 18 hpi, the localiza-
tion of DBP, HA-DEK, and HA-USF1 was visualized by IF
analyses using anti-DBP and anti-HA antibodies (Fig. 5C). In
mock-infected cells, both HA-tagged proteins showed a nuclear
localization, and in the case of HA-DEK, a strong signal was ob-
served at the nuclear periphery. At 18 hpi, HA-DEK appeared to
be excluded from VDRF (Fig. 5C). However, in contrast to HA-
DEK, we observed that HA-USF1 could be localized inside VDRF
allow the selective access of cellular nuclear proteins, and at least
one of the transcription factors, USF1, is able to access VDRF.
Oligomerization of DBP is critical for histone exclusion
from VDRF. To investigate the mechanism of histone exclusion
from VDRF, we hypothesized that DBP may play a role, since an
abundant amount of DBP was associated with Ad DNA in VDRF.
extension at its C terminus (Fig. 6A), and this C-terminal “arm”
hooks onto the next DBP molecule, resulting in the oligomeriza-
tion of DBP (47). It was also reported previously that the
oligomerization of DBP mediated by the C-terminal arm enables
DBP but not the deletion mutant that lacks the C-terminal arm
fore, to examine a role of DBP and its oligomerization in histone
fected with the expression vectors for HA-tagged full-length DBP
or DBP?C, and at 36 hpt, the cells were subjected to Western
blotting and IF assays using anti-HA and anti-DBP antibodies
(Fig. 6B and C). The expression levels of both DBP proteins were
almost the same, as indicated by Western blotting (Fig. 6B). By IF
analyses, we observed that full-length DBP forms foci like VDRF
FIG 5 Localization of nuclear proteins in late phases of infection. (A) IF analyses using anti-HIRA antibody. HeLa cells were mock infected or infected with
HAdV5 at an MOI of 250 (for 4 hpi) or 50 (for 18 hpi) and subjected to IF analyses using anti-protein VII and anti-HIRA antibodies. (B) Western blot analyses.
were subjected to 10% SDS-PAGE, followed by Western blot analyses using anti-HA (top) and anti-?-actin (bottom) antibodies. (C) Localization of HA-DEK
with HAdV5 at an MOI of 50, and at 18 hpi, the cells were subjected to IF assays using anti-HA and anti-DBP antibodies.
DBP and Histone H3.3 in Adenovirus DNA Replication
June 2012 Volume 86 Number 12jvi.asm.org 6707
was excluded from these foci, as observed for infected cells (HA-
out the nucleus and did not form such foci (Fig. 6C). Taken to-
critical role in histone exclusion from VDRF.
In this study, we showed that replication-independent histone
variant H3.3 is deposited onto both incoming and newly synthe-
sized Ad DNA (Fig. 3). These results, together with the results
from knockdown experiments with CAF-1 (Fig. 2) and micro-
scopic analyses (Fig. 4 to 6), indicated that the histone deposition
onto the replicated virus genome is most likely uncoupled with
viral DNA replication. Based on these results, together with data
to the fluctuation of the viral chromatin structure during the in-
fection cycle (Fig. 7). In virions, viral DNA is tightly packed with
viral core proteins (13). After entry into the cell, cellular histones
and histones functions as the template for viral early gene expres-
sion (20). In this process, histone H3.3 is specifically deposited
onto viral DNA, possibly by a histone chaperone, HIRA (34). As
oligomerization of DBP establishes a “histone-free” environment
for viral DNA replication. Newly synthesized viral DNA is then
associated with histone H3.3 in a replication-uncoupling fashion
and might be acting as the template for viral late gene expression
outside VDRF (31). In later phases of infection (?24 hpi), both
histones and newly synthesized core proteins VII and V are asso-
eny virion assembly (35). Since histones are not included in viri-
ons, histones must be removed and replaced with newly
aging mechanism of progeny viral DNA during virion assembly
remains unclear, we have reported the involvement of nucleolar
tin structure during progeny virion assembly (35, 36).
The mechanistic details of the histone deposition after viral
FIG 6 Role of DBP oligomerization in histone localization. (A) Schematic diagrams of full-length DBP and the C-terminally deleted mutant DBP?C. DBP of
HAdV5 consists of 529 aa, and the C-terminal 17 aa (aa 513 to 529) function as an “arm” for oligomerization. DBP?C lacks the C-terminal 17 aa. (B) Western
blot analyses. HeLa cells stably expressing histone H3.3-EGFP were transfected with the pCHA-puro empty vector (lane 1), pCHA-puro-DBP (lane 2), or
(middle), and anti-?-actin (bottom) antibodies. (C) IF analyses. At 36 hpt, cells grown on coverslips were subjected to IF analyses using anti-HA (left) and
anti-DBP (right) antibodies (B). Higher-magnification images of the regions marked by squares are shown below.
Komatsu and Nagata
jvi.asm.org Journal of Virology
DNA replication still remain unclear. First, what factor(s) is in-
volved in histone deposition at late phases of infection? HIRA is a
potential candidate for this process as well as in early phases of
infection (34). However, we did not perform knockdown experi-
ments for HIRA, since even if we could observe some effect of an
HIRA knockdown on viral chromatin in late phases of infection,
we could hardly distinguish whether the knockdown affects the
effect is derived indirectly from earlier events on incoming viral
chromatin. IF analyses showed that the localization of HIRA was
not drastically changed during the infection cycle (Fig. 5A). Re-
cent reports indicated that Daxx, a component of PML bodies, is
also an H3.3-specific histone chaperone (10, 21). However, Daxx
seems not to function in the H3.3 deposition onto viral DNA,
because during Ad infection, some components of PML bodies,
including Daxx, are relocalized by the viral protein E4orf3, possi-
bly for the inactivation of the components (6, 42). Indeed, it was
shown previously that the Daxx-mediated antiviral response is
antagonized by E4orf3 (48). It was also revealed that Daxx nega-
dent degradation during Ad infection (39). Furthermore, most
of the capsid proteins, binds to and counteracts Daxx, at least
partly by displacing it from PML bodies (38). Those reports
strongly suggest that Daxx is inactivated entirely throughout the
be a chaperone for histone H3.3 in Drosophila cells (37), but it is
chaperone or not. Our IF analyses indicated that exogenously ex-
pressed DEK is excluded from VDRF (Fig. 5B and C). Further
phases of infection.
In this study, we could not observe a role of CAF-1 in histone
deposition onto viral DNA, while the accumulation of CAF-1 at
VDRF was observed (Fig. 2 and 4A). The CAF-1 knockdown did
down efficiency of CAF-1 is not sufficient under the conditions
cells exhibit aberrant cell shapes (data not shown) and that the
knockdown affected viral gene expression (see below). Second,
histone H3.3 is selectively incorporated into viral chromatin (Fig.
3), while CAF-1 generally functions as a chaperone for H3.1 and
H3.2 (43). Third, although CAF-1 was reported previously to be
able to be associated with H3.3 under some specific conditions
H3.3 during Ad infection, at least under our experimental condi-
tions (data not shown). In addition to their roles during DNA
reported that the DNA damage response pathway is only partially
activated during Ad infection and that some related factors, such
as ATRIP and TopBP1, are accumulated at VDRF (5). Therefore,
CAF-1 (and PCNA) might localize at VDRF in the course of this
limited DNA damage response. Recently, it was reported that
FANCD2, one of the factors involved in the DNA damage re-
sponse, is accumulated at VDRF and that the loss of this protein
results in lower expression levels of viral late, but not early, genes
(8). Similarly, we observed that the CAF-1 knockdown affected
mRNA levels of viral late genes without any effect on viral DNA
replication (our unpublished observations). Thus, factors related
mechanisms are unknown. Under our conditions, the CAF-1
knockdown did not affect the binding level of histone H3 on cel-
lular chromatin (rRNA gene) (Fig. 2C). This is consistent with a
previous report that the loss of CAF-1 impairs the replication-
coupled deposition of histones but that the formation of nucleo-
some arrays on genomic DNA was still observed in the absence of
CAF-1 (44). In addition, a recent report demonstrated that a de-
fect of histone H3.1 deposition by CAF-1 depletion could be res-
cellular chromatin, an alternative histone deposition pathway(s)
could rescue the loss of CAF-1 function.
It remains to be clarified what is the biological/virological sig-
nificance of histone deposition uncoupled with viral DNA repli-
cation. On cellular chromatin, a replication-dependent histone
chaperone, CAF-1, is associated with the DNA replication ma-
FIG 7 Hypothetical model for the viral chromatin structure during the infection cycle. For details, see Discussion.
DBP and Histone H3.3 in Adenovirus DNA Replication
June 2012 Volume 86 Number 12 jvi.asm.org 6709
chinery and deposits histone H3.1-H4 (and H3.2-H4) complexes
during DNA replication (14, 40, 43). This DNA replication-cou-
pled system of histone deposition is thought to also be utilized by
some DNA viruses. For instance, the DNA replication of simian
virus 40 (SV40) is dependent largely on the cellular replication
machinery, and indeed, CAF-1 was originally identified by using
cell-free DNA replication systems of SV40 (41). For cytomegalo-
virus infection, it was reported previously that cellular histones,
(25). In the case of herpes simplex virus 1, it was shown that
and that H3.1 then becomes associated with viral DNA accompa-
tional link between DNA replication and histone deposition en-
ables the transfer of “epigenetic memory,” such as histone
modifications, to the daughter DNA strands (43). Thus, some
DNA viruses might take advantage of this system for late gene
expression, which generally occurs after viral DNA replication.
uncoupling mechanism, as shown here. Like other DNA viruses,
Ad late genes are expressed only after the onset of viral DNA rep-
lication. Thomas and Mathews demonstrated previously that Ad
late gene expression requires its DNA replication in cis (45), al-
matin structure during DNA replication could be an important
process for late gene expression. In general, the histone/nucleo-
such as transcription factors. In this view, DBP is an attractive
candidate for the key regulatory factor for the DNA replication-
the “histone-free” environment, which could be an opportunity
window for transcription factors to access the viral DNA for the
activation of viral late genes. Our IF analyses showed that the
transcription factor USF1, which binds to the MLP region after
viral DNA replication (46), is not excluded from VDRF (Fig. 5B
with a previous report that DBP enhances the binding of USF1 to
by DBP oligomerization, at least partly, and plays a role in the
DNA replication-dependent activation of viral late gene expres-
has been shown to require DNA replication (12). However, the
regulation mechanism of “DNA replication-dependent gene ex-
pression” remains to be determined. As Ad has late genes, the
expressions of which are DNA replication dependent (45), this
virus could be a good model for analyses of such regulations.
Therefore, this study might give a clue to an understanding of the
functional relationship between DNA replication and transcrip-
tion on cellular and/or viral chromatin.
We thank M. Okuwaki, S. Saito, A. Verreault, W. C. Russel, and K. Kaji-
tani for their kind gifts.
This work was supported in part by grants-in-aid for scientific re-
search from the Ministry of Education, Culture, Sports, Science, and
Technology of Japan (to K.N.) and the University of Tsukuba research
infrastructure support program (to T.K.).
1. Ahmad K, Henikoff S. 2002. The histone variant H3.3 marks active
chromatin by replication-independent nucleosome assembly. Mol. Cell
2. Bell O, Tiwari VK, Thoma NH, Schubeler D. 2011. Determinants and
dynamics of genome accessibility. Nat. Rev. Genet. 12:554–564.
3. Beyer AL, Bouton AH, Hodge LD, Miller OL, Jr. 1981. Visualization of
4. Burg JL, Schweitzer J, Daniell E. 1983. Introduction of superhelical turns
into DNA by adenoviral core proteins and chromatin assembly factors. J.
5. Carson CT, et al. 2009. Mislocalization of the MRN complex prevents
ATR signaling during adenovirus infection. EMBO J. 28:652–662.
6. Carvalho T, et al. 1995. Targeting of adenovirus E1A and E4-ORF3
7. Chatterjee PK, Vayda ME, Flint SJ. 1986. Adenoviral protein VII pack-
ages intracellular viral DNA throughout the early phase of infection.
EMBO J. 5:1633–1644.
8. Cherubini G, et al. 2011. The FANC pathway is activated by adenovirus
infection and promotes viral replication-dependent recombination. Nu-
cleic Acids Res. 39:5459–5473.
9. Dekker J, et al. 1997. Multimerization of the adenovirus DNA-binding
strand displacement synthesis. EMBO J. 16:1455–1463.
10. Drane P, Ouararhni K, Depaux A, Shuaib M, Hamiche A. 2010. The
death-associated protein DAXX is a novel histone chaperone involved in
the replication-independent deposition of H3.3. Genes Dev. 24:1253–
11. Elsaesser SJ, Goldberg AD, Allis CD. 2010. New functions for an old
variant: no substitute for histone H3.3. Curr. Opin. Genet. Dev. 20:110–
12. Fisher D, Mechali M. 2003. Vertebrate HoxB gene expression requires
DNA replication. EMBO J. 22:3737–3748.
13. Giberson AN, Davidson AR, Parks RJ. 2012. Chromatin structure of
14. Groth A, Rocha W, Verreault A, Almouzni G. 2007. Chromatin chal-
lenges during DNA replication and repair. Cell 128:721–733.
15. Gyurcsik B, Haruki H, Takahashi T, Mihara H, Nagata K. 2006. Binding
of the adenovirus chromatin. Biochemistry 45:303–313.
16. Hake SB, Allis CD. 2006. Histone H3 variants and their potential role in
indexing mammalian genomes: the “H3 barcode hypothesis.” Proc. Natl.
Acad. Sci. U. S. A. 103:6428–6435.
17. Haruki H, Gyurcsik B, Okuwaki M, Nagata K. 2003. Ternary complex
formation between DNA-adenovirus core protein VII and TAF-Ibeta/
SET, an acidic molecular chaperone. FEBS Lett. 555:521–527.
18. Haruki H, Okuwaki M, Miyagishi M, Taira K, Nagata K. 2006. Involve-
ment of template-activating factor I/SET in transcription of adenovirus
early genes as a positive-acting factor. J. Virol. 80:794–801.
19. Kawase H, et al. 1996. NAP-I is a functional homologue of TAF-I that is
required for replication and transcription of the adenovirus genome in a
chromatin-like structure. Genes Cells 1:1045–1056.
20. Komatsu T, Haruki H, Nagata K. 2011. Cellular and viral chromatin
proteins are positive factors in the regulation of adenovirus gene expres-
sion. Nucleic Acids Res. 39:889–901.
21. Lewis PW, Elsaesser SJ, Noh KM, Stadler SC, Allis CD. 2010. Daxx is an
H3.3-specific histone chaperone and cooperates with ATRX in replica-
U. S. A. 107:14075–14080.
22. Matsumoto K, Nagata K, Ui M, Hanaoka F. 1993. Template activating
replication. J. Biol. Chem. 268:10582–10587.
23. Matsumoto K, et al. 1995. Stimulation of DNA transcription by the
replication factor from the adenovirus genome in a chromatin-like struc-
ture. J. Biol. Chem. 270:9645–9650.
24. Nagata K, et al. 1995. Replication factor encoded by a putative oncogene,
Komatsu and Nagata
jvi.asm.orgJournal of Virology
25. Nitzsche A, Paulus C, Nevels M. 2008. Temporal dynamics of cytomeg-
alovirus chromatin assembly in productively infected human cells. J. Vi-
26. Okuwaki M, Iwamatsu A, Tsujimoto M, Nagata K. 2001. Identification
proteins. J. Mol. Biol. 311:41–55.
27. Okuwaki M, Nagata K. 1998. Template activating factor-I remodels the
chromatin structure and stimulates transcription from the chromatin
template. J. Biol. Chem. 273:34511–34518.
28. Placek BJ, et al. 2009. The histone variant H3.3 regulates gene expression
during lytic infection with herpes simplex virus type 1. J. Virol. 83:1416–
29. Polo SE, Roche D, Almouzni G. 2006. New histone incorporation marks
sites of UV repair in human cells. Cell 127:481–493.
30. Polo SE, et al. 2004. Chromatin assembly factor-1, a marker of clinical
value to distinguish quiescent from proliferating cells. Cancer Res. 64:
31. Pombo A, Ferreira J, Bridge E, Carmo-Fonseca M. 1994. Adenovirus
replication and transcription sites are spatially separated in the nucleus of
infected cells. EMBO J. 13:5075–5085.
32. Ray-Gallet D, et al. 2002. HIRA is critical for a nucleosome assembly
pathway independent of DNA synthesis. Mol. Cell 9:1091–1100.
33. Ray-Gallet D, et al. 2011. Dynamics of histone h3 deposition in vivo
reveal a nucleosome gap-filling mechanism for h3.3 to maintain chroma-
tin integrity. Mol. Cell 44:928–941.
34. Ross PJ, et al. 2011. Assembly of helper-dependent adenovirus DNA into
chromatin promotes efficient gene expression. J. Virol. 85:3950–3958.
35. Samad MA, Komatsu T, Okuwaki M, Nagata K. 15 February 2012,
posting date. B23/nucleophosmin is involved in regulation of adenovirus
chromatin structure at late infection stages, but not in its replication and
transcription. J. Gen. Virol. [Epub ahead of print.] doi:10.1099/
36. Samad MA, Okuwaki M, Haruki H, Nagata K. 2007. Physical and
functional interaction between a nucleolar protein nucleophosmin/B23
and adenovirus basic core proteins. FEBS Lett. 581:3283–3288.
37. Sawatsubashi S, et al. 2010. A histone chaperone, DEK, transcriptionally
coactivates a nuclear receptor. Genes Dev. 24:159–170.
38. Schreiner S, et al. 2012. Transcriptional activation of the adenoviral
genome is mediated by capsid protein VI. PLoS Pathog. 8:e1002549. doi:
39. Schreiner S, et al. 2010. Proteasome-dependent degradation of Daxx by
the viral E1B-55K protein in human adenovirus-infected cells. J. Virol.
40. Shibahara K, Stillman B. 1999. Replication-dependent marking of DNA
by PCNA facilitates CAF-1-coupled inheritance of chromatin. Cell 96:
41. Smith S, Stillman B. 1989. Purification and characterization of CAF-I, a
human cell factor required for chromatin assembly during DNA replica-
tion in vitro. Cell 58:15–25.
42. Stracker TH, et al. 2005. Serotype-specific reorganization of the Mre11
complex by adenoviral E4orf3 proteins. J. Virol. 79:6664–6673.
43. Tagami H, Ray-Gallet D, Almouzni G, Nakatani Y. 2004. Histone H3.1
and H3.3 complexes mediate nucleosome assembly pathways dependent
or independent of DNA synthesis. Cell 116:51–61.
44. Takami Y, Ono T, Fukagawa T, Shibahara K, Nakayama T. 2007.
assembly for DNA replication and cell division in vertebrate cells. Mol.
Biol. Cell 18:129–141.
45. Thomas GP, Mathews MB. 1980. DNA replication and the early to late
transition in adenovirus infection. Cell 22:523–533.
46. Toth M, Doerfler W, Shenk T. 1992. Adenovirus DNA replication facil-
promoter within infected cells. Nucleic Acids Res. 20:5143–5148.
47. Tucker PA, et al. 1994. Crystal structure of the adenovirus DNA binding
protein reveals a hook-on model for cooperative DNA binding. EMBO J.
49. Vayda ME, Rogers AE, Flint SJ. 1983. The structure of nucleoprotein
cores released from adenovirions. Nucleic Acids Res. 11:441–460.
50. Verreault A, Kaufman PD, Kobayashi R, Stillman B. 1996. Nucleosome
assembly by a complex of CAF-1 and acetylated histones H3/H4. Cell
51. Zijderveld DC, d’Adda di Fagagna F, Giacca M, Timmers HT, van der
by transcription factor USF is enhanced by the adenovirus DNA binding
protein. J. Virol. 68:8288–8295.
DBP and Histone H3.3 in Adenovirus DNA Replication
June 2012 Volume 86 Number 12 jvi.asm.org 6711