Previous Article | Next Article 
Journal of Virology, January 2009, p. 640-650, Vol. 83, No. 2
0022-538X/09/$08.00+0 doi:10.1128/JVI.01936-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
The Human Papillomavirus Type 8 E2 Tethering Protein Targets the Ribosomal DNA Loci of Host Mitotic Chromosomes
Atasi Poddar,
Shawna C. Reed,
Maria G. McPhillips,
Jonathan E. Spindler, and
Alison A. McBride*
Laboratory of Viral Diseases, NIAID, NIH, Bethesda, Maryland
Received 15 September 2008/
Accepted 4 November 2008

ABSTRACT
For many papillomaviruses, the viral protein E2 tethers the
viral genome to the host mitotic chromosomes to ensure persistent,
long-term maintenance of the genome during cell division. Our
previous studies of E2 proteins from different genera of papillomaviruses
have shown that they bind to different regions of the host chromosomes
during mitosis. For example, bovine papillomavirus type 1 (BPV-1)
E2 binds to all chromosomes as small speckles in complex with
the cellular protein Brd4. In contrast, the human papillomavirus
type 8 (HPV-8) E2 protein binds as large speckles at the pericentromeric
regions of chromosomes. Here we show that these speckles do
not contain Brd4, and unlike that of BPV-1, the N-terminal Brd4-interacting
domain of HPV-8 E2 is not required for chromosome binding. In
contrast to BPV-1 E2, the HPV-8 E2 protein targets the short
arms of acrocentric mitotic chromosomes. Furthermore, the E2
protein interacts with the repeated ribosomal DNA genes found
in this location and colocalizes with UBF, the RNA polymerase
I transcription factor. Therefore, HPV-8 E2 genome tethering
occurs by a Brd4-independent mechanism through a novel interaction
with specific regions of mitotic chromosomes. Thus, a wide range
of viruses have adopted the strategy of linking their genomes
to host chromosomes, but individual viruses use different chromosomal
targets. Characterization of these targets will enable the development
of antiviral therapies to eliminate the viral genomes from infected
cells.

INTRODUCTION
Papillomaviruses infect the basal layer of stratified epithelia.
The viral genomes are stably maintained in the nuclei of the
dividing cells as low copy number, extrachromosomal elements
for many years. Error-free replication and equal distribution
of the replicated copies of the viral DNA to the dividing cells
are key to the persistence of papillomavirus infection. The
mechanism of viral genome partitioning is different from that
of host chromosomes, as the viral DNA does not possess any sequence
equivalent to the centromere to utilize the mitotic segregation
machinery of the host. Instead, E2, a multifunctional viral
protein, attaches the viral DNA to the host chromosomes (
15,
21,
34). As the host cell chromosomes are divided equally into
daughter cells during mitotic division, the viral DNA is passively
segregated by being tethered to the host chromosomes. Chromosomal
tethering mediated by a virus-encoded protein is a common tactic
for maintaining the genomes of other persistent viruses. Gammaherpesviruses
such as Epstein-Barr virus and Kaposi's sarcoma-associated herpesvirus
use this strategy, which is mediated by the EBNA-1 and LANA
proteins, respectively (reviewed in reference
23).
Among the papillomaviruses, bovine papillomavirus type 1 (BPV-1) is best studied for chromosomal segregation. The multifunctional viral protein E2 has been shown to play an important role in this process. Both the BPV-1 E2 protein and viral genomes are localized on mitotic chromosomes as small speckles on the arms of all chromosomes, in complex with Brd4 (3, 15, 21, 34, 40). Furthermore, BPV-1 E2 mediates the attachment and segregation of plasmids containing E2 binding sites (4, 15). There is much evidence that Brd4, a cellular bromodomain protein involved in many cellular functions, including transcription regulation, is involved in the chromosomal binding complex of BPV-1 E2. Both E2 and Brd4 colocalize on mitotic chromosomes in punctate spots (3, 25, 40), and mutations in the E2 transactivation domain compromise both the interaction of E2 with Brd4 and the association of E2 with the chromosome (3, 32). Inhibition of the Brd4-E2 interaction by the overexpression of a dominant negative C-terminal domain of Brd4 inhibits the association of E2 with the chromosome, viral genome maintenance, and BPV1-induced cellular transformation (16, 24, 41). Finally, in Saccharomyces cerevisiae, E2-mediated plasmid segregation can be reconstituted only in the presence of Brd4 (4).
For most papillomaviruses, there is no good cell-based system available to study genome segregation. However, because it has been clearly shown that BPV-1 E2 partitions genomes by tethering them to mitotic chromosomes, the localization of other papillomavirus E2 proteins to the mitotic chromosome has been analyzed to determine whether they use a similar mechanism. A systematic study was carried out in our laboratory to investigate the localization of the E2 proteins of 13 animal and human papillomaviruses (HPVs) from 7 different papillomavirus genera to the mitotic chromosome (27). Many of the E2 proteins, such as those from European elk papillomavirus and HPV type 1a (HPV-1a), show colocalization with Brd4 on mitotic chromosomes similar to that of BPV-1 E2. However, the HPV-8 E2 protein was found to associate with the pericentromeric region of chromosomes as larger, often dimeric, speckles (27). The E2 proteins of the human alphapapillomaviruses HPV-11, HPV-16, and HPV-31 also showed localization patterns similar to that of HPV-8 E2 under certain fixation conditions. These E2 proteins did not colocalize with Brd4, and mutations in the HPV-31 E2 protein which abrogate Brd4 binding did not affect mitotic chromosome binding (24). In another study, the HPV-11, HPV-16, and HPV-18 alphapapillomavirus E2 proteins were reported to be associated with the mitotic spindle (37). So, while the chromosomal target of BPV has been established, the binding partners of many human papillomaviruses remain elusive, and these viruses may interact with diverse binding partners for genome segregation.
The association of HPV-8 E2 with the host chromosome is one of the most stable and distinctive of those of the papillomavirus E2 proteins that we have studied. HPV-8 is a betapapillomavirus that subclinically infects the cutaneous epithelium in normal individuals. In individuals with the genetic disorder epidermodysplasia verruciformis, infection causes abundant lesions which have a propensity to transform into carcinomas (13). Several viral types from the genus Papillomavirus have also been implicated in nonmelanoma skin cancers (11). Considering the carcinogenic potential of HPV-8 and the distinctive localization of the E2 protein of HPV-8 on mitotic chromosomes, we further investigated the mechanism of HPV-8 genome segregation.
In the present work, we show that the HPV-8 E2 protein associates with mitotic chromosomes in the absence of the N-terminal domain, which has been shown to be a critical determinant for BPV-1 E2 chromosome targeting (2, 3, 34). On human host chromosomes, HPV-8 E2 binds to the short arms of the acrocentric chromosomes, to a region containing both β-satellite DNA and the ribosomal DNA (rDNA) loci. Further mapping demonstrated that the HPV-8 E2 protein binds to the rDNA genes and colocalizes with the ribosomal transcription factor UBF1.

MATERIALS AND METHODS
Plasmids.
The pMEP-HPV-8 E2 plasmid, which expresses FLAG-tagged E2 from
the metallothionein promoter, has been described previously
(
24). DNA that encodes subdomains of HPV-8 E2 was amplified,
using PCR primers with appropriate restriction sites, and was
cloned into the pMEP4 vector. Details of the cloning procedures
will be provided on request. All E2 proteins except the HPV-8
E2 C domain, were encoded by an N-terminal FLAG epitope tag.
The tag on the C domain consisted of the FLAG tag, followed
by CBP (
calmodulin
binding
protein) and SBP (streptavidin
binding
protein) tags. The NH and HC domains of HPV-8 E2 correspond
to amino acids 1 to 404 and 206 to 498, respectively. The NC
domains encode the E2 protein with amino acids 1 to 205 fused
to amino acids 404 to 498. The N, H, and C domains correspond
to amino acids 1 to 205, 207 to 405, and 404 to 498, respectively.
The R37AI73A mutation was generated in the full-length E2 gene,
using the QuikChange multi site-directed mutagenesis kit (Stratagene).
The pMEP-9 BPV-E2 expression vector is similar to pMEP-4 BPV-E2,
except that the hygromycin resistance marker has been replaced
by the neomycin resistance gene. pA and p21β7 plasmids,
carrying rDNA and β-satellite DNA, respectively, have been
described previously (
10,
14).
Cell culture.
HPV-8- and BPV-1 E2-expressing CV-1 cells were derived, as described previously (25), in the experiment shown in Fig. 1, with double selection with hygromycin and G418. C33A cells expressing wild-type and different truncated versions of HPV-8 E2 were selected with 80 µg/ml hygromycin B (Roche) and maintained in 40 µg/ml hygromycin B.
Brd4-E2 binding assay.
The following procedure has been described previously (
3,
5).
Purified Brd4 protein (0.64 µg in 5 µl) was bound
to M2 anti-FLAG agarose by incubation in 20 mM Tris (pH 7.5),
150 mM NaCl, 1 mM EDTA, 0.1% Tween 20, 1 mM dithiothreitol,
and 1
x Complete EDTA-free protease inhibitor cocktail (Roche)
at room temperature for 1 h. E2 proteins were synthesized and
labeled with [
35S]-methionine, using Promega's TNT quick-coupled
reticulocyte lysate system, and the concentrations of each were
normalized to one another based upon the number of methionines.
Equivalent amounts of
35S-labeled E2 proteins were incubated
with Brd4 agarose beads for 1 h. Bound proteins were eluted
in 2% sodium dodecyl sulfate (SDS)-62.5 mM Tris-HCl (pH 6.8)-25%
glycerol and separated by SDS- polyacrylamide gel electrophoresis.
The amount of
35S-E2 bound to Brd4 was measured using a Typhoon
imager (Molecular Dynamics).
Indirect immunofluorescence.
pMEP plasmids containing cells were grown on Superfrost Plus slides. For mitotic synchronization, cells were blocked with 2 mM thymidine and fixed with 4% paraformaldehyde 9 h after release, as described previously (24). Four hours before fixation, E2 expression was induced with 6 µM CdSO4. For metaphase chromosome spreads, cells were treated with 30 ng/ml colchicine for 3 h before fixation and swollen in hypotonic buffer 1 (10 mM Tris [pH 7.4], 10 mM NaCl, 5 mM MgCl2) and hypotonic buffer 2 (25% phosphate-buffered saline [vol/vol]) for 15 min each at room temperature, and the slides were cytocentrifuged at 1,500 rpm for 3 min and fixed as described above (42). Cells were permeabilized and stained as described previously (24). HPV-8 E2, UBF, Brd4, and centromeres were detected with monoclonal anti-FLAG M2 antibody (1:500; Sigma), rabbit polyclonal anti-UBF antibody (1:100; Santa Cruz Biotechnologies, Inc.), rabbit polyclonal antiserum against Brd4 2290 (9), and human anticentromere antiserum (1:1,000; ImmunoVision), respectively. Fluorescent secondary antibodies were from Jackson ImmunoResearch Laboratories. Slides were mounted in Vectashield containing 1 µg/ml DAPI (4',6'-diamidino-2-phenylindole). Images were collected with a Leica TCS-SP5 laser scanning confocal imaging system. Maximal projection images of 15 to 25 0.2 µM Z-stack images were obtained, using Imaris software (Bitplane AG).
FISH.
A combined immunofluorescence and fluorescence in situ hybridization (FISH) technique was developed for the studies described herein. After immunofluorescence detection as described above, slides were treated with methanol:acetic acid (3:1) for 10 min, 2% paraformaldehyde for 1 min, and a 70%, 90%, and 100% ethanol series for 3 min each. Fluorescein isothiocyanate-labeled
-satellite probes specific for the centromeres of chromosomes 1, 5, and 19; 13 and 21; or 14, 22, and 15 (Qbiogene) were used to identify specific chromosomes. Rhodamine-labeled
-satellite probes (2 µl) specific for chromosomes 13 and 21; 14 and 22; or 1, 5, 9, and 15 (Qbiogene) were hybridized in 8 µl hybridization buffer (Qbiogene), and the cellular DNA and probe were denatured by incubating the slides at 75°C for 5 min. Hybridization was carried out overnight at 37°C. The slides were washed with 1x phosphate-buffered detergent (Qbiogene) for 5 min at room temperature, 1x wash buffer (0.5x SSC [1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate]-0.1% SDS) for 5 min at 65°C, and 1x phosphate-buffered detergent for 5 min at room temperature and mounted as described above. Probes for rDNA and β-satellite DNA were prepared from pA and p21β7 plasmids, respectively. DNA (2 µg) was labeled with Alexa Fluor 594, using the Ulysis nucleic acid labeling kit according to the manufacturer's instructions. The probe DNA was precipitated with 4 mg/ml salmon sperm DNA and resuspended at 10 µg/ml in hybridization buffer (50% formamide, 10% dextran sulfate, 4x SSC). After indirect immunofluorescence staining and cross-linking, the slides were treated with 100 µg/ml RNaseA in 2x SSC. Subsequently, the slides were denatured in 70% formamide in 2x SSC at 72°C for 5 min, dehydrated in ice-cold 70%, 80%, and 100% ethanol for 2 min each, and rinsed with 100% ethanol for 1 min at room temperature. After denaturation, 10 µl of 10 µg/ml probe DNA was hybridized overnight to the denatured slide. For rDNA FISH, the slides were washed three times with washing solution I (50% formamide-2x SSC) for 5 min each at 45°C and three times with washing solution II (1x SSC) for 5 min each at 65°C and were rinsed briefly in washing solution III (4x SSC-0.1% Tween 20) at room temperature. For β-satellite FISH, the slides were washed two times with washing solution I (50% formamide-2x SSC) for 5 min each at 40°C and rinsed briefly in washing solution III (4x SSC-0.1% Tween 20) at room temperature. The slides were mounted in DAPI-containing Vectashield as described above. Images were collected and processed as described in the previous section.
Deconvolution and colocalization.
Three-dimensional reconstruction, maximal projection, and colocalization of the sequential Z-sections were performed with Imaris software (Bitplane AG). Collected data for analysis were deconvolved with Huygens essential software (Scientific Volume Imaging BV). Values for 2-D line profiles of maximal projections were obtained by using IPlab software (BD Biosciences Bioimaging).

RESULTS
HPV-8 and BPV-1 E2 proteins bind different regions of mitotic chromosomes.
In previous studies, the HPV-8 and BPV-1 E2 proteins showed
distinct patterns of localization on mitotic chromosomes (
27).
To analyze this further, a CV-1-derived cell line that expressed
both E2 proteins was established. The location of each protein
was determined by using an antibody against the N-terminal FLAG
tag on the HPV-8 E2 protein and a monoclonal antibody specific
for the untagged BPV-1 E2 protein. The centromeres of the host
chromosomes were identified with an antibody against the kinetochore.
As shown in Fig.
1, each E2 protein bound to distinct regions
of mitotic chromosomes, and no colocalization was observed.
The HPV-8 E2 protein was observed to always be located adjacent
to the centromeres of the host chromosomes, whereas BPV-1 E2
had a random distribution with respect to the centromere. Thus,
HPV-8 E2 has a unique chromosomal target that is distinct from
that of BPV-1 E2.
The H and C domains of HPV-8 E2 are sufficient for mitotic chromosome binding.
The E2 protein consists of three domains. The N-terminal transactivation domain (N) and the C-terminal DNA-binding domains (C) are conserved among different papillomaviruses. These conserved domains are connected by a flexible, nonconserved hinge region (H). The N-terminal domain of BPV-1 E2 has been shown to be important for binding to mitotic chromosomes, while the C-terminal domain binds specifically to sites in the viral genome (2). The differences between HPV-8 E2 and BPV E2 chromosome localization suggested that the mechanism of chromosome binding could be different for these two proteins. To investigate the regions of the HPV-8 E2 protein required for chromosome association, FLAG-tagged proteins consisting of different domains of HPV-8 E2 (Fig. 2A) were expressed in CV-1 cells, and their localization was studied by indirect immunofluorescence. As shown in Fig. 2B, localization of the HPV-8 E2 HC protein lacking any N-terminal domain was comparable to that of the full-length protein. However, proteins consisting of NC, NH, or the individual N, H, and C domains showed no specific chromosome binding. Therefore, the hinge region and C terminal domain of HPV-8 E2 are necessary and sufficient for chromosome targeting. Alignment of the BPV-1 and HPV-8 E2 proteins is shown in Fig. 2C.
HPV-8 E2 mitotic chromosome binding does not require Brd4.
The N-terminal domain of BPV-1 E2 is critical for Brd4 binding
and hence for chromosome association. The amino acid substitutions
R37A and I73A in the N-terminal domain of BPV-1 result in the
loss of Brd4 interaction and chromosome binding in all E2 proteins
tested (
3,
24,
32). Our previous observations that BPV-1 and
HPV-8 E2 proteins bind to different regions of mitotic chromosomes
and that the N-terminal domain of HPV8-E2 is not required indicated
that HPV-8 E2 probably associates with mitotic chromosomes independently
of its interaction with Brd4. To confirm this, we investigated
the localization of Brd4 in CV-1 cells expressing the HPV-8
E2, HPV-8 E2 R37AI73A, and HPV8-E2 HC proteins. All three proteins
bound to mitotic chromosomes without any significant colocalization
with Brd4 (Fig.
3C). In vitro binding assays confirmed that
as for BPV-1 E2, the N-terminal domain of HPV-8 E2 is required
for Brd4 binding, that the R37AI73A mutation disrupts this binding,
and that HPV8-E2 HC is unable to bind Brd4 (Fig.
3A and B).
Therefore, the association of HPV-8 E2 with mitotic chromosomes
does not require Brd4 and thus is different from that of BPV-1
E2.
HPV-8 E2 binds to the pericentromeric region of a subset of mitotic chromosomes.
To further characterize the mitotic binding site of HPV-8 E2,
E2-expressing cell lines were derived from human cervical carcinoma-derived
C33A cells. Stable cell lines express detectable E2 protein
in about 20 to 30% of interphase cells after induction of the
metallothionein promoter. The E2 protein can be detected in
about 10% of mitotic cells and was localized on mitotic chromosomes
in a pattern similar to that observed previously in CV-1 cells.
To analyze this localization in further detail, chromosome spreads
were prepared, and HPV-8 E2 and centromeres were detected by
indirect immunofluorescence. E2 staining was observed as paired
or single foci of variable size, located adjacent to the centromeres
(Fig.
4A). The number of E2 signals observed per cell ranged
from three to eight signals, with an average of six signals.
The intensity of the E2 staining on each chromosome was variable;
however, each focus within a paired signal had a similar intensity.
Figure
4B shows a panel of individual, well-isolated chromosomes
with E2 and ACA staining. These chromosomes were randomly selected
from different metaphase spreads. For the majority of the E2-bound
chromosomes, both E2 and ACA staining was located very close
to the end of the chromosomes, suggesting that E2 is associated
with centromeres on the short arms of the chromosomes.
HPV-8 E2 associates with acrocentric chromosomes.
By definition, chromosomes with the centromere close to one
end are known as acrocentric chromosomes. To further prove that
E2 was binding specifically to such chromosomes, a combined
immunofluorescence/FISH technique was developed. HPV-8 E2 was
detected on metaphase chromosomes by indirect immunofluorescence
with an anti-FLAG antibody. The antibodies were cross-linked,
and FISH was performed with

-satellite probes specific for the
acrocentric chromosomes 13 and 21, 14 and 22, or 15. In mitotic
cells expressing E2, the immunofluorescence signal for E2 was
observed adjacent to the FISH signal specific for each of the
acrocentric chromosomes (Fig.
5). However, the distributions
of E2 on different acrocentric chromosomes were not identical.
E2 was localized next to the centromere of chromosomes 13 and
21 in 100% of cases (Table
1). In contrast, only 43% of chromosomes
14 and 22 and 73% of chromosomes 15 showed E2 binding. To further
prove the specificity of E2 binding to acrocentric chromosomes,
similar immuno-FISH experiments were carried out with probes
specific for the nonacrocentric chromosomes 1, 5, and 19. E2
staining was never observed adjacent to the FISH signal of these
chromosomes, confirming the specificity for acrocentric chromosomes
(Fig.
5D).
HPV-8 E2 binds adjacent to, but does not overlap with, β-satellite DNA on the short arms of acrocentric chromosomes.
The short arms of the acrocentric chromosomes contain different
families of repetitive DNA sequences arranged in tandem. The
tight foci of the HPV-8 E2 immunofluorescence signals at equivalent
positions on each sister chromatid prompted us to investigate
whether E2 binds to these sequences on the mitotic chromosomes.
β-Satellite DNA, one of the best characterized repetitive
DNAs, is located on the short arms of acrocentric chromosomes
(
14). Approximately 250 to 500 copies of higher-order repeats
of β-satellite DNA are estimated to be present per haploid
genome. Two subfamilies of β-satellite DNA are organized
on the short arms of acrocentric chromosomes. The p21ß7
subfamily is localized to the distal portion of the short arms,
and the p21ß2 subfamily is located both at the distal
end of the chromosomes and proximal to the centromeric DNA.
Since the E2 signal was observed at a single locus on each chromosome,
we used a FISH probe corresponding to the p21ß7 subfamily
to detect β-satellite sequences. Combined immunofluorescence/FISH
was performed as described above to detect HPV-8 E2 and β-satellite
DNA, respectively. On chromosome spreads from C33A cells, the
E2 signal was observed adjacent to the β-satellite signal
(Fig.
6). When individual chromosomes were examined, the β-satellite
signal was located at the distal end of the chromosome, while
HPV-8 E2 was located toward the long chromosome arm (Fig.
6A).
To map the location of E2 with respect to other chromosomal
sequences, double immunofluorescence staining was done with
anti-FLAG and ACA antibodies together with FISH by using β-satellite
probes. E2 was found to be located between the centromere DNA
and the β-satellite DNA on acrocentric chromosomes (Fig.
6).
HPV-8 E2 binds to the rDNA loci on mitotic chromosomes.
The genes carrying 5.8S, 18S, and 28S rRNA are organized as
tandemly repeating units on the short arms of acrocentric chromosomes
(
14,
22). There are approximately 400 copies of these repeat
units present per diploid genome. The p21ß7 subfamily
is known to be localized to the distal end of the rDNA gene
cluster (
14). Localization of HPV-8 E2 relative to β-satellite
DNA indicated that rDNA could be the potential chromosomal target
of HPV-8 E2. To confirm this, HPV-8 E2, kinetochores, and rDNA
were located on chromosome spreads by indirect immunofluorescence
with anticentromeric ACA and anti-FLAG antibodies and by FISH
with a probe specific for rDNA. The probe was generated from
the pA plasmid that contains a 7.3-kb EcoRI fragment spanning

200 nucleotides of the 18S rRNA gene, the internal transcribed
spacer 1 region, sequences carrying the 5.8S rRNA gene, the
internal transcribed spacer 2, and

4,500 nucleotides of the
upstream end of the 28S rRNA gene. We could detect between six
and nine FISH signals per cell, representing the rDNA loci.
The majority of the E2 signal was found to be colocalized with
these rDNA signals (Fig.
7A). Analysis of randomly selected,
well-isolated chromosomes from different mitotic chromosome
spreads showed that both HPV-8 E2 and rDNA colocalize on the
mitotic chromosomes and are located adjacent to the centromere
DNA, as detected by ACA staining (Fig.
7B). The relative position
of each locus was determined by plotting the relative intensities
of each fluorescence signal along a line drawn through the signals
on the acrocentric chromosome (Fig.
7C). This further demonstrated
that the peak of the intensity of the E2 signal coincided with
that of rDNA. The fluorescence intensity peak for the centromere
(ACA) was distinct and did not overlap with either E2 or rDNA
signals.
HPV-8 E2 colocalizes with the rDNA transcription factor UBF on mitotic chromosomes.
UBF is a transcription factor required for transcription of
rDNA by RNA polymerase I (
17). UBF has been shown to associate
with the chromosomal region corresponding to rDNA and remains
bound in all stages of mitosis, although transcription is silenced
in mitosis (
28,
29). Colocalization of E2 with UBF with reference
to the centromere was investigated, using anti-FLAG, anti-UBF,
and anti-ACA antibodies. HPV-8 E2 was found to associate with
the chromosome at the same position as UBF (Fig.
8A). Colocalization
was observed as yellow foci, equivalently positioned on the
sister chromatids of the mitotic chromosomes. The numbers of
UBF foci per cell ranged from four to eight, with an average
of seven. Like that for E2, UBF staining could be seen as either
single foci or paired dots, and the intensities of different
UBF signals were variable. Immunostaining with ACA, anti-FLAG,
and anti-UBF showed that E2 and UBF colocalize adjacent to the
centromere (Fig.
8B). This data was confirmed by the fluorescence
line scans of the E2, UBF, and ACA signals (Fig.
8C).

DISCUSSION
In this study, we demonstrated that the chromosomal target of
the HPV-8 E2 tethering protein is quite different from that
of the BPV-1 E2 protein, which binds to the chromosome arms
in complex with the cellular Brd4 protein (
3,
40). In contrast
to BPV-1 E2, the N-terminal domain of HPV-8 E2, which is required
for association with Brd4, is not required for chromosomal association.
Rather, both the C-terminal DNA binding and dimerization domain
and the hinge region are required. Furthermore, Brd4 does not
colocalize on mitotic chromosomes with the HPV-8 E2 R37AI73A
and HPV-8 E2 HC proteins that are unable to interact with Brd4.
Thus, as reported previously, Brd4 is not required for genome
partitioning of all papillomaviruses, although it does seem
to be required for transcriptional activation and repression
(
16,
24,
31,
32,
39). Further characterization of the binding
site of HPV-8 E2 on human chromosomes revealed that this protein
binds to the rDNA genes located between β-satellite DNA
and centromeric DNA on the short arms of acrocentric chromosomes
(Fig.
9).
There are several reasons why rDNA might be an attractive target
as a chromosomal tethering region for some papillomaviruses.
The tandemly arranged repeating rDNA units (400 per diploid
cell) increase the local concentration of the binding target,
thereby increasing the likelihood of interaction with the E2
protein complex. Because of its repetitive nature, the rDNA
locus has specialized mechanisms for replication, cohesion,
condensation, and segregation (
8,
12,
19,
35,
36). The rDNA
transcription units are also poised for immediate RNA synthesis
as cells exit mitosis, and transcription factors such as UBF
remain bound throughout mitosis (
26). The resulting unique chromatin
structure could provide unique targets distributed among the
host chromosomes. In interphase cells, all papillomavirus E2
proteins are detected in a nuclear, but nucleolar-excluded,
pattern. In addition, the HPV-8 E2 protein can be observed in
SC35-splicing speckles by virtue of the serine-arginine-rich
hinge sequences (A. A. McBride, unpublished data), as has been
shown for the related HPV-5 E2 protein (
20). This separation
in localization at different stages of the cell cycle could
be advantageous to the virus, as the E2 protein would be free
during interphase to participate in viral transcription, replication,
and splicing; the high-affinity nucleolar-organizing region
(NOR) targets would be sequestered in the nucleoli until the
nucleolar envelope broke down in early mitosis. It is not yet
known if E2 plays any functional roles at the rDNA locus other
than tethering the viral genomes to the chromosome, but several
viruses are known to modify rDNA expression or factors (
1,
6,
30).
Notably, only 8 or 9 acrocentric chromosomes were detected by rDNA FISH instead of the expected 10 per chromosome spread. This is due to aneuploidy of the C33A cell line (Table 2). Transcription of each rDNA locus results in the formation of a nucleolus, and thus each region is termed a NOR. Within each cell, some NORs are silent, and only active NORs are associated with polymerase I transcription factors (26). Correspondingly, on average only seven UBF foci and six E2 signals were detected per metaphase spread, and each protein bound acrocentric chromosomes in the preferential order of 13, 21, >15, >14, 22 (Table 3). This implies that the E2 protein binds only to active mitotic NORs and that the E2 target may be a transcription factor or chromatin component associated with these sites. Although E2 and UBF colocalize at the NORs, no direct interaction has yet been detected by coimmunoprecipitation (data not shown), and it remains to be determined whether they have a functional interaction or simply bind to the same chromosomal locus.
There are also connections between other viral tethering proteins
and nucleoli or NORs. The cellular target of the Epstein-Barr
virus tethering protein, EBNA, is a protein termed p40 nucleolar
proliferation antigen, or hEBP2, and is involved in pre-RNA
processing (
33). hEBP2 is nucleolar in interphase (
7) but is
distributed along condensed chromosomes in mitosis (
38). Recently,
the C-terminal domain of the LANA tethering protein of KSHV
has been shown to bind as paired dots to pericentromeric and
telomeric regions of certain metaphase chromosomes (
18), some
of which we speculate might be acrocentric chromosomes. We have
previously demonstrated that E2 proteins from the
Alphapapillomavirus genus can also be observed binding to pericentromeric regions
of chromosomes, although only in certain fixation conditions
(
27). Thus, the rDNA loci may be an attractive tethering target
for several viruses, and further characterization of this interaction
could result in the development of specific antiviral therapies.

ACKNOWLEDGMENTS
This research was supported by the Intramural Research Program
of the NIAID, NIH.
We thank Juraj Kabat for help with image processing.

FOOTNOTES
* Corresponding author. Mailing address: Laboratory of Viral Diseases, NIAID, NIH, 4 Center Dr. MSC 0445, Bethesda, MD 20892. Phone: (301) 496-1370. Fax: (301) 480-1497. E-mail:
amcbride{at}nih.gov 
Published ahead of print on 12 November 2008. 

REFERENCES
1 - Banerjee, R., M. K. Weidman, S. Navarro, L. Comai, and A. Dasgupta. 2005. Modifications of both selectivity factor and upstream binding factor contribute to poliovirus-mediated inhibition of RNA polymerase I transcription. J. Gen. Virol. 86:2315-2322.[Abstract/Free Full Text]
2 - Bastien, N., and A. A. McBride. 2000. Interaction of the papillomavirus E2 with mitotic chromosomes. Virology 270:124-134.[CrossRef][Medline]
3 - Baxter, M. K., M. G. McPhillips, K. Ozato, and A. A. McBride. 2005. The mitotic chromosome binding activity of the papillomavirus E2 protein correlates with interaction with the cellular chromosomal protein, Brd4. J. Virol. 79:4806-4818.[Abstract/Free Full Text]
4 - Brannon, A. R., J. A. Maresca, J. D. Boeke, M. A. Basrai, and A. A. McBride. 2005. Reconstitution of papillomavirus E2-mediated plasmid maintenance in Saccharomyces cerevisiae by the Brd4 bromodomain protein. Proc. Natl. Acad. Sci. USA 102:2998-3003.[Abstract/Free Full Text]
5 - Cardenas-Mora, J., J. E. Spindler, M. K. Jang, and A. A. McBride. 2008. Dimerization of the papillomavirus E2 protein is required for efficient mitotic chromosome association and Brd4 binding. J. Virol. 82:7298-7305.[Abstract/Free Full Text]
6 - Cavender, J. F., C. Mummert, and M. J. Tevethia. 1999. Transactivation of a ribosomal gene by simian virus 40 large-T antigen requires at least three activities of the protein. J. Virol. 73:214-224.[Abstract/Free Full Text]
7 - Chatterjee, A., J. W. Freeman, and H. Busch. 1987. Identification and partial characterization of a Mr 40,000 nucleolar antigen associated with cell proliferation. Cancer Res. 47:1123-1129.[Abstract/Free Full Text]
8 - D'Amours, D., F. Stegmeier, and A. Amon. 2004. Cdc14 and condensin control the dissolution of cohesin-independent chromosome linkages at repeated DNA. Cell 117:455-469.[CrossRef][Medline]
9 - Dey, A., J. Ellenberg, A. Farina, A. E. Coleman, T. Maruyama, S. Sciortino, J. Lippincott-Schwartz, and K. Ozato. 2000. A bromodomain protein, MCAP, associates with mitotic chromosomes and affects G2-to-M transition. Mol. Cell. Biol. 20:6537-6549.[Abstract/Free Full Text]
10 - Erickson, J. M., C. L. Rushford, D. J. Dorney, G. N. Wilson, and R. D. Schmickel. 1981. Structure and variation of human ribosomal DNA: molecular analysis of cloned fragments. Gene 16:1-9.[CrossRef][Medline]
11 - Feltkamp, M. C., R. Broer, F. M. di Summa, L. Struijk, E. van der Meijden, B. P. Verlaan, R. G. Westendorp, J. ter Schegget, W. J. Spaan, and J. N. Bouwes Bavinck. 2003. Seroreactivity to epidermodysplasia verruciformis-related human papillomavirus types is associated with nonmelanoma skin cancer. Cancer Res. 63:2695-2700.[Abstract/Free Full Text]
12 - Freeman, L., L. Aragon-Alcaide, and A. Strunnikov. 2000. The condensin complex governs chromosome condensation and mitotic transmission of rDNA. J. Cell Biol. 149:811-824.[Abstract/Free Full Text]
13 - Fuchs, P. G., T. Iftner, J. Weninger, and H. Pfister. 1986. Epidermodysplasia verruciformis-associated human papillomavirus 8: genomic sequence and comparative analysis. J. Virol. 58:626-634.[Abstract/Free Full Text]
14 - Greig, G. M., and H. F. Willard. 1992. β satellite DNA: characterization and localization of two subfamilies from the distal and proximal short arms of the human acrocentric chromosomes. Genomics 12:573-580.[CrossRef][Medline]
15 - Ilves, I., S. Kivi, and M. Ustav. 1999. Long-term episomal maintenance of bovine papillomavirus type 1 plasmids is determined by attachment to host chromosomes, which is mediated by the viral E2 protein and its binding sites. J. Virol. 73:4404-4412.[Abstract/Free Full Text]
16 - Ilves, I., K. Maemets, T. Silla, K. Janikson, and M. Ustav. 2006. Brd4 is involved in multiple processes of the bovine papillomavirus type 1 life cycle. J. Virol. 80:3660-3665.[Abstract/Free Full Text]
17 - Jantzen, H. M., A. Admon, S. P. Bell, and R. Tjian. 1990. Nucleolar transcription factor hUBF contains a DNA-binding motif with homology to HMG proteins. Nature 344:830-836.[CrossRef][Medline]
18 - Kelley-Clarke, B., M. E. Ballestas, T. Komatsu, and K. M. Kaye. 2007. Kaposi's sarcoma herpesvirus C-terminal LANA concentrates at pericentromeric and peri-telomeric regions of a subset of mitotic chromosomes. Virology 357:149-157.[CrossRef][Medline]
19 - Kobayashi, T., D. J. Heck, M. Nomura, and T. Horiuchi. 1998. Expansion and contraction of ribosomal DNA repeats in Saccharomyces cerevisiae: requirement of replication fork blocking (Fob1) protein and the role of RNA polymerase I. Genes Dev. 12:3821-3830.[Abstract/Free Full Text]
20 - Lai, M. C., B. H. Teh, and W. Y. Tarn. 1999. A human papillomavirus E2 transcriptional activator. The interactions with cellular splicing factors and potential function in pre-mRNA processing. J. Biol. Chem. 274:11832-11841.[Abstract/Free Full Text]
21 - Lehman, C. W., and M. R. Botchan. 1998. Segregation of viral plasmids depends on tethering to chromosomes and is regulated by phosphorylation. Proc. Natl. Acad. Sci. USA 95:4338-4343.[Abstract/Free Full Text]
22 - Long, E. O., and I. B. Dawid. 1980. Repeated genes in eukaryotes. Annu. Rev. Biochem. 49:727-764.[CrossRef][Medline]
23 - McBride, A. A., J. G. Oliveira, and M. G. McPhillips. 2006. Partitioning viral genomes in mitosis: same idea, different targets. Cell Cycle 5:1499-1502.[Medline]
24 - McPhillips, M. G., J. G. Oliveira, J. E. Spindler, R. Mitra, and A. A. McBride. 2006. Brd4 is required for E2-mediated transcriptional activation but not genome partitioning of all papillomaviruses. J. Virol. 80:9530-9543.[Abstract/Free Full Text]
25 - McPhillips, M. G., K. Ozato, and A. A. McBride. 2005. Interaction of bovine papillomavirus E2 protein with Brd4 stabilizes its association with chromatin. J. Virol. 79:8920-8932.[Abstract/Free Full Text]
26 - McStay, B. 2006. Nucleolar dominance: a model for rRNA gene silencing. Genes Dev. 20:1207-1214.[Free Full Text]
27 - Oliveira, J. G., L. A. Colf, and A. A. McBride. 2006. Variations in the association of papillomavirus E2 proteins with mitotic chromosomes. Proc. Natl. Acad. Sci. USA 103:1047-1052.[Abstract/Free Full Text]
28 - O'Sullivan, A. C., G. J. Sullivan, and B. McStay. 2002. UBF binding in vivo is not restricted to regulatory sequences within the vertebrate ribosomal DNA repeat. Mol. Cell. Biol. 22:657-668.[Abstract/Free Full Text]
29 - Roussel, P., C. Andre, C. Masson, G. Geraud, and D. Hernandez-Verdun. 1993. Localization of the RNA polymerase I transcription factor hUBF during the cell cycle. J. Cell Sci. 104:327-337.[Abstract]
30 - Samad, M. A., M. Okuwaki, H. Haruki, and K. Nagata. 2007. Physical and functional interaction between a nucleolar protein nucleophosmin/B23 and adenovirus basic core proteins. FEBS Lett. 581:3283-3288.[CrossRef][Medline]
31 - Schweiger, M. R., J. You, and P. M. Howley. 2006. Bromodomain protein 4 mediates the papillomavirus E2 transcriptional activation function. J. Virol. 80:4276-4285.[Abstract/Free Full Text]
32 - Sénéchal, H., G. G. Poirier, B. Coulombe, L. A. Laimins, and J. Archambault. 2007. Amino acid substitutions that specifically impair the transcriptional activity of papillomavirus E2 affect binding to the long isoform of Brd4. Virology 358:10-17.[CrossRef][Medline]
33 - Shire, K., D. F. Ceccarelli, T. M. Avolio-Hunter, and L. Frappier. 1999. EBP2, a human protein that interacts with sequences of the Epstein-Barr virus nuclear antigen 1 important for plasmid maintenance. J. Virol. 73:2587-2595.[Abstract/Free Full Text]
34 - Skiadopoulos, M. H., and A. A. McBride. 1998. Bovine papillomavirus type 1 genomes and the E2 transactivator protein are closely associated with mitotic chromatin. J. Virol. 72:2079-2088.[Abstract/Free Full Text]
35 - Sullivan, M., T. Higuchi, V. L. Katis, and F. Uhlmann. 2004. Cdc14 phosphatase induces rDNA condensation and resolves cohesin-independent cohesion during budding yeast anaphase. Cell 117:471-482.[CrossRef][Medline]
36 - Torres-Rosell, J., F. Machin, S. Farmer, A. Jarmuz, T. Eydmann, J. Z. Dalgaard, and L. Aragon. 2005. SMC5 and SMC6 genes are required for the segregation of repetitive chromosome regions. Nat. Cell Biol. 7:412-419.[CrossRef][Medline]
37 - Van Tine, B. A., L. D. Dao, S. Y. Wu, T. M. Sonbuchner, B. Y. Lin, N. Zou, C. M. Chiang, T. R. Broker, and L. T. Chow. 2004. Human papillomavirus (HPV) origin-binding protein associates with mitotic spindles to enable viral DNA partitioning. Proc. Natl. Acad. Sci. USA 101:4030-4035.[Abstract/Free Full Text]
38 - Wu, H., D. F. Ceccarelli, and L. Frappier. 2000. The DNA segregation mechanism of Epstein-Barr virus nuclear antigen 1. EMBO Rep. 1:140-144.[Medline]
39 - Wu, S. Y., A. Y. Lee, S. Y. Hou, J. K. Kemper, H. Erdjument-Bromage, P. Tempst, and C. M. Chiang. 2006. Brd4 links chromatin targeting to HPV transcriptional silencing. Genes Dev. 20:2383-2396.[Abstract/Free Full Text]
40 - You, J., J. L. Croyle, A. Nishimura, K. Ozato, and P. M. Howley. 2004. Interaction of the bovine papillomavirus E2 protein with Brd4 tethers the viral DNA to host mitotic chromosomes. Cell 117:349-360.[CrossRef][Medline]
41 - You, J., M. R. Schweiger, and P. M. Howley. 2005. Inhibition of E2 binding to Brd4 enhances viral genome loss and phenotypic reversion of bovine papillomavirus-transformed cells. J. Virol. 79:14956-14961.[Abstract/Free Full Text]
42 - Zheng, P. S., J. Brokaw, and A. A. McBride. 2005. Conditional mutations in the mitotic chromosome binding function of the bovine papillomavirus type 1 E2 protein. J. Virol. 79:1500-1509.[Abstract/Free Full Text]
Journal of Virology, January 2009, p. 640-650, Vol. 83, No. 2
0022-538X/09/$08.00+0 doi:10.1128/JVI.01936-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
This article has been cited by other articles:
-
Zheng, G., Schweiger, M.-R., Martinez-Noel, G., Zheng, L., Smith, J. A., Harper, J. W., Howley, P. M.
(2009). Brd4 Regulation of Papillomavirus Protein E2 Stability. J. Virol.
83: 8683-8692
[Abstract]
[Full Text]
-
Kelley-Clarke, B., De Leon-Vazquez, E., Slain, K., Barbera, A. J., Kaye, K. M.
(2009). Role of Kaposi's Sarcoma-Associated Herpesvirus C-Terminal LANA Chromosome Binding in Episome Persistence. J. Virol.
83: 4326-4337
[Abstract]
[Full Text]