The truncated C-terminal E2 (E2-TR) protein of bovine papillomavirus (BPV)
type-1 is a transactivator that modulates transcription in vivo and in vitro
in a manner distinct from the E2-TA and E8^E2 gene products
Michael J. Lacea,b,n, Masato Ushikaia, Yasushi Yamakawaa, James R. Ansona, Takaoki Ishijic,
Lubomir P. Tureka,b, Thomas H. Haugena,b
aVeterans Affairs Healthcare System, Iowa City, IA 52246, USA
bDepartment of Pathology, Roy J. and Lucille A. Carver College of Medicine, University of Iowa, Iowa City, IA 52242, USA
cDepartment of Dermatology, Jikei University School of Medicine, Tokyo, Japan
a r t i c l e i n f o
Received 11 November 2011
Returned to author for revisions
9 January 2012
Accepted 30 March 2012
Available online 1 May 2012
a b s t r a c t
The E2 open reading frame of bovine papillomavirus (BPV)-1 encodes a 410 amino acid (aa) transcriptional
activator, E2-TA, and collinear polypeptides—E2-TR (243 aa) and E8 ^E2 (196 aa). E8^E2 and E2-TR share the
DNA-binding domain of E2-TA, and both have been defined as transcriptional repressors. Although purified
E2-TR and E8^E2 proteins specifically bound E2 sites with similar affinities, only the E2-TR stimulated
transcription. Here we show that E2-TR trans-activates E2-dependent promoters 5 to 10-fold in cooperation
with cellular factors and in a dose-dependent fashion in epithelial cells and fibroblasts of animal or human
origin while E2-TA activated 4100-fold and the E8 ^E2 had no effect. However, in contrast to E2-TA, E2-TR
activated transcription from a promoter-proximal position. E2-TR also partially inhibited the BPV-1 P89 or
heterologous promoters whereas E8^E2 led to complete repression. Thus, the BPV-1 E2-TR modulates viral
gene expression in a manner distinct from other E2 proteins.
Published by Elsevier Inc.
Papillomaviruses (PVs) produce benign epithelial or fibroe-
pithelial proliferations of the skin and mucosa (Campo, 2002;
DiMaio, 1991; Doorbar, 2005) and certain human papillomavirus
(HPV) types are associated with anogenital cancers and 30% of
head and neck cancers (Psyrri and DiMaio, 2008). Although not
associated with malignant lesions, bovine papillomavirus type-1
(BPV-1) has historically served in culture as a useful model of HPV
infection and persistence.
As in HPV, early BPV gene expression is initially activated by
cellular factors, such as TEF-1 and Sp1 (Haugen et al., 2009; Sandler
et al., 1996; Ushikai et al., 1994; Vande Pol and Howley, 1992), which
are critical for establishment of viral infection. Once viral transcription
is established, viral gene products further modulate expression of the
early viral genes and support initial genome amplification and
immortalization of the host cell. The E2 gene products of papilloma-
viruses play important roles in the regulation of viral transcription
and replication as well as the establishment and maintenance of
persistent viral infection (Cripe et al., 1987; Haugen et al., 1987; Jang
et al., 2009; Lace et al., 2008a; Lambert et al., 1989; Oliveira et al.,
2006; Senechal et al., 2007; Steger et al., 1996; Stubenrauch et al.,
2000; Stubenrauch et al., 1998; Szymanski and Stenlund, 1991). The
full length E2 gene product of BPV-1 can act as a transcriptional
repressor or activator of at least three early BPV-1 promoters as well
as other heterogeneous promoters (Demeret et al., 1994; Dostatni
et al., 1991; Haugen et al., 1987; Spalholz et al., 1985; Steger and
Corbach, 1997; Thierry and Howley, 1991).
As in HPVs, the full length BPV-1 E2-TA is required for viral
replication in addition to the E1 gene product (Chiang et al., 1992;
Ustav et al., 1993; Ustav and Stenlund, 1991). The full length, 48 kD
E2-TA protein of BPV-1 consists of three functional domains (Abroi
et al., 1996; Giri and Yaniv, 1988; Winokur and McBride, 1992): the
C-terminal 101 amino acid DNA binding domain (DBD) (de Prat-Gay,
Gaston, and Cicero, 2008), the N-terminal 210 amino acid trans-
activation domain (TAD) and a central hinge domain (Ferguson and
Botchan, 1996; Giri and Yaniv, 1988; Haugen et al., 1988). The E2
DBD governs dimerization (Haugen et al., 1987; McBride et al., 1989)
and confers specific binding to conserved palindromic E2 sites,
ACC(N)6GGT (Allikas et al., 2001; Androphy et al., 1987; Sanchez
et al., 2008; Ushikai et al., 1994). Structural analysis demonstrated
that the dimer of E2 DBD forms a unique anti-parallel b-barrel when
bound to DNA (Hegde et al., 1992), inducing conformational changes
to its DNA target (Ferreiro et al., 2000).
In addition to the full length E2 protein, the E2 gene of BPV-1
encodes at least two truncated isoforms (Hubbert et al., 1988):
the 31 kD C-terminal E2-TR protein (Lambert et al., 1987) and
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/yviro
0042-6822/$-see front matter Published by Elsevier Inc.
nCorresponding author at: Veterans Affairs Healthcare System, 601 Highway
6 West, Iowa City, IA 52246, USA. Fax: þ1 319 339 7178.
E-mail address: email@example.com (M.J. Lace).
Virology 429 (2012) 99–111
a 28 kD E8^E2 spliced protein (Choe et al., 1989; Lambert et al.,
1989). The E2-TR is transcribed from P3080 (or ‘‘P5’’) located
within E2 open reading frame (ORF), contains a C-terminal
component (aa 162–210), and the E2 hinge and DBD consists of
210–410 aa, while the E8^E2 is transcribed from P890 (P3) within
the E1 ORF and consists of a short peptide leader from the E8 ORF
(a highly conserved region in many PVs) which is spliced onto the
E2 hinge and DBD region (Lambert et al., 1989; Vaillancourt et al.,
1990). Both proteins have been considered transcriptional repres-
sors through competitive binding to the E2 sites against the
full length E2-TA homodimers or through the formation of
heterodimers with the E2-TA (McBride et al., 1989). Despite the
apparent structural differences, a detailed understanding of the
functional differences between the two truncated BPV E2 proteins
has remained incomplete.
In this study, we compared the in vitro DNA binding properties
of the full-length E2-TA, E2-TR and E8^E2 proteins to their
transcriptional activation capacities in order to elucidate their
functional differences in vivo. Surprisingly, the E2-TR protein
could consistently activate transcription in a variety of assays
from a minimal tk promoter as well as the BPV-1 major early
promoter (P89), while the E8^E2 protein did not stimulate pro-
moter activity in parallel. We also compared the in vitro DNA
binding kinetics of purified E2 proteins. While these experiments
revealed a variation in protein/DNA complex stabilities, no
difference between E2-TR (aa 162–410) and E2 hinge-DBD (aa
210–410) binding affinities for consensus or native BPV E2
binding sites was observed, indicating the respective E2 binding
properties are not the sole determinant of the observed functional
differences between the E2 proteins. Experiments using vectors
expressing the GAL4 (DBD)-E2 and LEXA (DBD)-E2 chimeric
proteins showed that the part of E2 TAD encoded in the 50E2-
TR ORF (aa 162–210) is sufficient to confer transcriptional
activation capacity to the E2-TR gene product. These results show
that E2-TR functions as a transcriptional activator in a manner
distinct from the full length E2-TA protein and that the E2 TAD (aa
1–210) consists of discrete functional subdomains that cooperate
with cellular factors to modulate early viral gene expression.
The E2-TR gene product is a transcriptional trans-activator in vivo
and in vitro
Both C-terminal E2 proteins, E2-TR and E8^E2, have been reported
to function as competitive repressors of E2 transactivation (Haugen
et al., 1987; Lambert et al., 1989; Lambert et al., 1987; Monini et al.,
1993) but potential functional differences between the two forms
have not been thoroughly examined. In previous studies, E2
activities were determined by transfection of a range of plasmid
expression constructs with varying efficiencies (Cripe et al., 1987;
Giri and Yaniv, 1988). To compare the ability of transcriptional
activation among the three forms of E2 gene products, we used the
full length pCG E2-TA, truncated pCG E2-TR and pCG E8^E2 in which
each E2 ORF is subcloned downstream of the constitutively active,
E2-independent, CMV enhancer/promoter (Fig. 1A) capable of effi-
cient, comparable expression of the E2-TR and E8^E2 from ORFs
encoding gene products of predicted size [Fig. 1C and (Ushikai et al.,
1994; Ustav and Stenlund, 1991)]. We cotransfected increasing
quantities of each pCG E2 expression plasmid with a minimal
E2-responsive reporter in our in vivo transfection assays. The
E2?2-Sp1?2-tk (-38)-cat contains two E2 and two Sp1 sites 50of
the HSV-1 tk (-38) promoter (Fig. 1A). This promoter responds
strongly to the full-length PV E2 activators, yet unlike the complex
native BPV promoters driven by cellular transactivators, has little
baseline activity, thereby permitting us to discretely examine
E2 mechanisms of transactivation independent of cellular transacti-
Consistent with an earlier report (Steger et al., 1995), the
E2-TR could activate transcription of the minimal E2-dependent
promoter in a dose-dependent manner in CAT reporter assays
(Fig. 1B), though it was weaker than the E2-TA (5-20 fold vs.
?100 fold) (Table 1). The amount of pCG-E2-TR expression
plasmid which yielded the maximum activation was similar to
that reported for pCG-E2 (Fig. 1B and Ushikai et al., 1994,
respectively), with both exhibiting a similar biphasic activation
profile as excess levels of the expressed E2 transactivators led to
squelching of promoter activity – similar to the activation kinetics
of described cellular factors (Xiao et al., 1991). In contrast, the
pCG E8^E2 did not stimulate transcription in parallel (Fig. 1B).
Since all three proteins have previously been shown to be
E2-independent enhancer–promoter (CMV) (Ustav and Stenlund,
1991), the differential transcriptional activation observed with
the E2 expression plasmids is therefore not solely a function of
varying promoter efficiency of E2 gene product expression.
Furthermore, we demonstrated that the E2-TR (similar to, but to
a more limited degree, the E2-TA (Ushikai et al., 1994)) quantita-
tively stimulated steady-state levels of transcripts generated from
the same template while the E8^E2 functioned as a repressor
in RNase protection assays (Fig. 1E) and as a repressor and
E2-inhibitor in subsequent reporter assays (Fig. 7).
To confirm the in vivo transcription results, we also tested the
capacity of purified E2 proteins to stimulate an E2-responsive
promoter in in vitro transcription assays. We complemented our
RNase protection assay results which measured E2-dependent activa-
tion of steady-state mRNA levels (a function of both transcript
initiation and mRNA stability) with G-free in vitro transcription assays
as a defined system to quantify transcription initiation from the
respective E2-responsive G-free templates (Fig. 1D). In contrast to
previous studies which employed E2 proteins expressed in non-
mammalian (i.e. insect, bacterial or yeast) systems, we expressed the
E2 isoforms in mammalian cells to insure requisite post-translational
modification critical for their function. We synthesized the correctly
modified recombinant E2-TA (aa 1–410), E2-TR (aa 162–283) and
E2-hinge-DBD (aa 210–410) proteins using a vaccinia expression
system. Vaccinia-expressed E2-TR and E2 hinge-DBD products were
extracted and purified by ion exchange and sequence-specific DNA
affinity columns described (Ishiji et al., 1992; Ushikai et al., 1994).
Aliquots of the purified E2 proteins were then resolved on a 10%
denaturing SDS polyacrylamide gel and stained by silver nitrate. The
E2-TR protein displayed a single dominant band with an apparent
size of approximately 32 kD, compared to standard molecular
weight markers, which is consistent with the predicted molecular
weight for the E2-TR (Fig. 1C).
Increasing quantities of the purified E2-TR and E2 hinge-DBD
proteins were incubated with E2-responsive transcription templates
and HeLa whole cell extracts. For the target template we used E2?
2-Sp1?1-tk (-38)-G-free (380) which is identical to E2?2-Sp1?
2-cat except that the cat reporter gene was replaced with a 380 nt
G-free cassette to provide a means of quantifying transcripts initiating
from the authentic target promoter. The tk (-38)-G-free (250) plasmid,
which lacks E2 and Sp1 sites, served as an internal control. We also
tested the C-terminal, DNA-binding domain of the E2 protein, E2
hinge-DBD (aa 210–410), which was synthesized in the same manner
as the E2-TR protein (Ushikai et al., 1994). The E2-TR protein
stimulated the E2?2-Sp1?1-tk promoter in vitro up to ?3 fold in
a dose-dependent manner (Fig. 1D, lanes 1–4), while the control tk
(-38)-G-free (250) was not affected by a similar titration of E2-TR
(aa 162–410). Since activation by the full length E2 in this reporter
clone depends on the cooperation of Sp1 (Ushikai et al., 1994), this
M.J. Lace et al. / Virology 429 (2012) 99–111
result implies that E2-TR can also cooperate with cellular factors, such
as Sp1. In contrast, the E2 hinge-DBD was ineffective at stimulating
transcription (Fig. 1D, lanes 5–8). One possible explanation for the
observed residual activation by the E2 hinge-DBD is that it could
function as an anti-repressor by inhibiting the binding of some
repressor proteins, such as histones, in vitro. Taken together, these
results show that the E2-TR can function as a transcriptional
transactivator both in vivo and in vitro.
Furthermore, we tested the ability of the E2-TR to transactivate
transcription in a variety of transfected fibroblast and epithelial cell
Fig. 1. BPV-1 E2-TR is a transcriptional trans-activator in vivo and in vitro. (A) Schematic of E2 expression vectors and reporter clones. In the pCG expression vectors, the
full length E2-TA, E2-TR and E8^E2 were expressed from the human CMV enhancer–promoter and 30rabbit globin poly(A) sequences. pCG E2 has the point mutation in the
AUG at nt 3091 to prevent the synthesis of the E2-TR. The E2?2-Sp1?2-tk(-38)-cat contains two consensus E2 sites and two Sp1 sites adjacent to the minimal HSV tk
promoter. (B) HeLa cells were cotransfected with increasing amounts of pCG E2, pCG E2-TR and pCG E8^E2 as indicated. Total amounts of pCG plasmid were held constant
by reciprocal pCG-neo addition. Relative CAT activities were derived from 2 to 3 independent transfections and expressed as fold above normalized pCG-neo in a log-linear
plot. (C) Silver-stained SDS-PAGE analysis of E2-TR and E2 hinge-DBD proteins. (D) A total of 40 ng of the reporter plasmid DNA was preincubated with purified
recombinant E2-TR and E2 hinge-DBD proteins (0, 0.3 1 and 2 ml) then incubated with 10 ml of HeLa whole cell extract in G-free in vitro transcription assays. (E) E2-TR-
dependent activation of steady-state transcription levels was demonstrated in RNase protection assays in HeLa cells. The SV2-cat plasmid was used as a transfection control.
All values are expressed as fold above normalized controls and represent averages of 2–3 independent experiments.
M.J. Lace et al. / Virology 429 (2012) 99–111
lines derived from bovine, other animal and human sources
(Table 1) to determine if the functional phenotypes of the E2
gene products were cell type-dependent. As we previously noted
(Ushikai et al., 1994), the E2-TA is a robust transactivator of
E2-responsive targets in diverse cell types. The E2-TR, however,
also consistently transactivated an E2-dependent reporter plas-
mid while the E8^E2 did not, regardless of the cell type trans-
fected;this indicates that candidate
components targeted by the E2-TR are present in all cell types
tested and not specific to bovine fibroblasts, as previously shown
for E2-TA (Rank and Lambert, 1995). These results also confirm
that the promoter assays performed in HeLa cells provide a
suitable model for measuring E2-dependent transactivation levels
comparable to the phenotypic trends observed in other cell types.
Taken together, these results show that the function of E2-TR is
distinct from E8^E2 as it drives transcriptional activation in contrast
to its initial characterization as a transcriptional repressor. Further-
more, in contrast to the robust transactivation observed with the E2-
TA, E2-TR comparatively serves as a quantitatively ‘‘weaker’’ acti-
vator of E2-responsive promoters.
Persistent overexpression of E2-TA has been previously shown
to inhibit both transient and long term HeLa cell growth in colony
reduction assays by presumably repressing E6–E7 expression
from the integrated HPV-18 genome and inducing apoptosis
(Hwang et al., 1993; Stubenrauch et al., 2007). We therefore
monitored cell growth in HeLa cultures cotransfected with opti-
mal concentrations of the pCG E2 expression plasmids as defined
in the previous experiments (Fig. 1 and Table 1). We noted no
differential growth or morphological changes in cells transfected
with pCG-neo, pCGE2-TA, pCG-E2-TR or pCG E8^E2 within the
same transient time period used in our previous transactivation
assays (Fig. 2). These results confirmed that the observed com-
parative transactivation profiles of the E2 gene products in these
assays were not simply a function of altered cell growth.
In vitro binding kinetics of the BPV-1 E2-TR protein
The relativebinding affinities ofpurified recombinant
E2-TA, E2-TR and E2 hinge-DBD proteins were compared in
mobility shift experiments using a high-affinity synthetic E2
consensus site (ACCGATATCGGT) as a DNA probe in the presence
of a range of concentrations of unlabeled E2 site oligonucleotide
competitors (Fig. 3A). Dissociation constants (KD), defined as
competitor DNA concentrations required for 50% reduction of
DNA probe-protein complex formation, of E2-TR (aa 162–410) for
BPV E2#10 and the E2 consensus site were found to be 1.1 nM
and 0.6 nM, respectively, and 0.8 nM and 0.7 nM for E2 hinge-
DBD (aa 210–410) (Fig. 3B and Table 2). We had previously
demonstrated that the relative affinities of the full length E2-TA
(aa 1–410) for the native BPV-1 E2#10 and consensus E2 sites
were similar – 0.5 nM and 1.0 nM, respectively (Ushikai et al.,
1994). Thus, these results indicated that relative binding affinities
of the E2-TA, E2-TR and E2 hinge-DBD for both consensus and
native BPV E2 motifs are similar.
To characterize E2–DNA complex stabilities in vitro, we analyzed
the dissociation rates, or ‘‘off-rates’’ (T1/2), defined as the time it takes
for one half of a preformed complex to dissociate in the presence of
excess unlabeled competitor E2 consensus site DNA. Excess amounts
of unlabeled oligonucleotide were incubated with DNA probe, E2-TR
or E2 hinge-DBD and the complex formation examined by EMSA at
various time points (Fig. 3B). In contrast to the comparable binding
affinities, each protein displayed varying dissociation rates (Table 2).
The E2-TR-DNA complex was much more stable at both E2 sites than
those formed with the full-length E2 or E2 hinge-DBD even though all
three proteins encompass an identical binding domain. Clearly, E2
structural domains outside of the DBD appear to influence the overall
E2–DNA complex stabilities in these assays. For instance, the total
length of protein and its final conformation could likely influence the
steady-state stabilities of the respective DNA–protein complexes.
However, it is unclear if the observed differences in complex
stabilities among the E2 isoforms observed in vitro directly result in
biologically significant functional differences in vivo. Taken together,
these results show that the in vitro binding kinetics of the E2 protein:
DNA complexes alone are not the sole determinant of the observed
variation in transcription functions of the E2 isoforms.
The E2 TAD harbors discrete functional sub-domains
Even though the E2-TR contains only a portion (aa 162–210) of
the broader defined E2 TAD (aa 1–210) encoded in the full length
E2-TA, we have demonstrated that it functions as a transcrip-
tional transactivator in vivo and in vitro. This is consistent with
a previous report (Steger et al., 1995) that proposed a modular
structure for the transactivation domain (TAD) – one comprised of
Fig. 2. E2 Expression is not accompanied by significant short term growth
inhibition. Optimal concentrations of pCG expression plasmids encoding the E2-TA,
E2-TR, E8^E2 and neo control were cotransfected with the E2x2Sp1?2 tk-38-cat
plasmid in HeLa cells and cell growth monitored for the same 72 h period as used in
transcription assays defined in this study (Fig. 1 and Table 1). CMV-b-gal plasmid was
cotransfected as an additional control in parallel while ‘‘*’’ indicates a representative
untransfected culture. Results represent cell density averages of triplicate cultures in
parallel with growth rates expressed as PD (population doublings) in days.
BPV-1 E2-TR can function as a transcriptional trans-activator in a variety of
Expression plasmid Relative CAT activitya
Primary human fibroblasts
Primary human keratinocytes
aAll values are expressed as maximum fold induction of the E2?2 Sp1?2 tk
(-38)-cat reporter (1 mg) by respective pCG E2 plasmid (100 to 1000 ng) addition –
above or below normalized transfection with pCG neo alone and represent
averages of 2–3 independent experiments.
bAs previously reported (Ushikai et al., 1994).
M.J. Lace et al. / Virology 429 (2012) 99–111
functionally independent subdomains. To further test this possi-
bility, we synthesized clones that express chimeric proteins
containing the GAL4 DBD (aa 1–147) linked to the E2-TR (aa
162–283), the reciprocal portion of the TAD, E2 (aa 1–162), or the
intact E2 (aa 1–210) TAD of the full length E2 (Fig. 4). The GALx5-
Sp1?2-tk (-38) cat plasmid which has five 17 mer-GAL binding
sites and two Sp1 sites followed by a minimal tk promoter linked
to a cat reporter gene served as a reporter clone in HeLa cell
cotransfections (Fig. 4). The pCG-GAL4-E2 and pCG-GAL4-E2 TAD
expression plasmids, which contain the full length E2 or E2 TAD
respectively, showed the strongest activation (up to 100 fold) as
expected from the results that pCG E2 strongly activated the
minimal tk promoter when E2 binding sites were present (Ushikai
et al., 1994). Interestingly, both the C-terminal portion of the TAD
contained in the E2-TR ORF or the reciprocal N-terminal portion
were able to transactivate though at a weaker level than the
entire E2 TAD (3–5 fold vs. ?100 fold). Furthermore, these
reciprocal subdomains of the E2 TAD could also cooperate with
one another when cotransfected at equimolar ratios in similar
reporter assays to achieve greater transactivation than observed
with either sub-domain alone (data not shown).
Our results show that GAL4 E2 (aa 162–283) chimeric protein,
which has the partial TAD encoded within the E2-TR, has the
ability to transactivate, but much more weakly than the more
Fig. 3. BPV-1 E2-TR can bind to conserved E2 motifs with the same affinity as the E2-TA gene product. (A) Purified E2-TR and E2 hinge-DBD proteins were incubated with
BPV E2#10 or E2 consensus probes in the presence of increasing amounts (0, 1, 3, 10, 30 and 100 nM) of homologous unlabeled competitors in mobility shift assays.
Relative binding affinities (expressed as dissociation constants (Kd)) of E2-TR and E2 hinge-DBD for the E2 sites were determined by the amount of competitor at 50%
complex reduction. (B) Dissociation rates (or ‘‘off rates’’) for preformed E2 protein: DNA complexes were measured in mobility shift assays and expressed as the incubation
time (over a defined range – 0, 1, 10, 30, 60 and 90 min) required for 50% reduction of each complex. All values were determined by scanning densitometry and represent
averages of 2–3 independent experiments.
Comparison of relative binding kinetics of the E2-TR and E2 DBD proteins.
E2-TA (1–410)E2-TR (162–410) E2-H-DBD (210–410)
BPV-1 E2 #10/caaACCGTCTTCGGTgct 0.518 1.1
490 0.8 11
E2 consensus/tcgACCATATCGGTcga 1.0
aIn vitro binding kinetics of complexes formed with purified E2 proteins and BPV-1 or consensus E2 sites were determined in 2–3 EMSAs.
M.J. Lace et al. / Virology 429 (2012) 99–111
complex intact E2-TA (aa 1–410) product (Fig. 4, compare clones a
and e). Since the reciprocal N-terminal portion of E2 TAD (aa 1–162)
also showed only weak activation (Fig. 4, clone d), the overall
capacity of the intact E2 TAD for transactivation is markedly
decreased upon functional dissection, although most components
retain some functional activity. These results confirm that discrete
functional domains of the E2 TAD, such as that encoded in the E2-TR
gene, can cooperatively activate transcription. To confirm that the
observed E2 subdomain activities were not dependent on residual
transactivation activity from the GAL4 DBD, we performed similar
assays with the same panel of corresponding chimeric LEX-E2
expression constructs. While lower overall basal activities were
noted with these constructs compared to the GAL-E2 activities, the
pattern of LEX-E2 subdomain activation phenotypes was similar
The BPV-1 E2-TR can cooperate with cellular trans-activating factors
in a manner distinct from other E2 gene products
We then compared the abilities of the exogenously expressed E2
proteins to cooperate with endogenous cellular factors to transacti-
vate a heterologous promoter. Defined binding motifs for a variety of
cellular factors were inserted immediately downstream of the E2
binding sites, linked to the minimal tk promoter/cat reporter. Con-
sistent with protocols used in our previous E2-TA study (Ushikai
et al., 1994), addition of both the E2-TA and E2-TR expression
plasmids activated promoters containing defined consensus binding
sites for Sp1, USF, Oct-1, AP-1 and NF-1 (Fig. 5A, clones a–f)while the
E8^E2 had no effect – similar to transfection with pCG-neo. Similar
patterns of E2 and E2-TR activation of complex enhancer/promoter
reporter plasmids were also noted in parallel (Fig. 5A, clones i and j).
We also cotransfected chimeric expression plasmids encoding various
cellular and viral TADs fused to the GAL4 DBD with equimolar
quantities of the E2 expression plasmids in HeLa cells to similarly
assess the range of E2/cellular factor cooperativity (Fig. 5B). As we
had shown previously (Ushikai et al., 1994), the E2-TA was able
to cooperate with multiple acidic and non-acidic transactivation
domains, such as GAL4, Sp1, TEF-1, Ad E1a, and a ligand-dependent
(i.e. beta-estradiol-responsive) estrogen receptor transactivation
domain (ER-EF) (Fig. 5B, clones a–g and i). The E2-TR was also able
to cooperate with many of the same TADs to a degree (Fig. 5B, clones
a, f, g and i) comparable to that of the E2-TA with the exception of the
GAL-Sp1 constructs (Fig. 5B, clones b–e). Since the E2-TR can
cooperate with endogenous Sp1 in previous experiments (Fig. 1),
the differential effect noted with the chimeric GAL-Sp1 domain
expression plasmids may reflect the abilities of the respective
E2-TA versus E2-TR proteins to interact with the potentially altered
conformations of the chimeric GAL-Sp1 proteins. The E8 ^E2, however,
was unable to cooperate with any of the TAD constructs tested. These
results show that the E2 proteins, including the E2-TR, are critical yet
distinct, trans-acting regulators with varying capacities to interact
with cellular factors to cooperatively activate transcription – a mode
of transcription regulation consistent with the complex structures of
the native BPV promoters, such as major early, P89 (P2), promoter.
Since the E2-TA has been described as a repressor when bound to
sites immediately upstream of the transcription start sites in HPV
promoters, we examined the ability of E2-TR to function from the
same promoter position in a defined heterologous promoter. In
contrast to the E2-TA which inhibited transcription at this position,
the E2-TR further activated an enhancer/promoter construct already
Fig. 4. C-terminal domain of E2 TAD is sufficient to activate transcription. Schematic of chimeric GAL4-E2 TAD expression vectors and reporter clones. The entire E2, E2
TAD, partial TAD contained in E2-TR (aa 162–283) and the reciprocal portion of E2 TAD (aa 1–162) were connected to GAL4 DBD (aa 1–147), and expressed as GAL4
chimeric proteins. The GAL4?5-Sp1?2-tk (-38)-cat, which has 5 GAL4 site and 2 Sp1 sites followed by minimal tk promoter and cat reporter gene. Cotransfection
experiment with the clones which express GAL4 DBD-E2 TAD chimeric proteins. A total of 1 mg of reporter plasmid was cotransfected with the indicated chimeric GAL or
LEX expression plasmid in HeLa cells, respectively. Relative CAT activities are expressed as fold above the corresponding normalized GAL4 DBD or LEX DBD (clone h) alone
and represent averages of 2–4 independent experiments.
M.J. Lace et al. / Virology 429 (2012) 99–111
driven by cellular factors from E2 binding sites immediately
upstream of the transcription start site (Fig. 6A).
We also tested the ability of purified E2-TR to activate or
potentially inhibit transcription, depending on its binding posi-
tion, from a minimal promoter driven by Sp1 in in vitro transcrip-
tion assays. E2-TR supported in vitro transcription when the E2
sites were upstream of the Sp1 site (Fig. 6B, template a). Addition
of excess quantities of unlabeled competitor oligonucleotides
binding Sp1 or E2 decreased transcription from this template
(Fig. 6B, lanes 1–8). When the E2 and Sp1 sites were placed in the
inverted orientation (i.e. E2 sites placed proximal to the start
site (Fig. 6B, template b)), E2-TR did not inhibit Sp1-dependent
transcription (Fig. 6B, lanes 9–16). The E2?2-tk(-38)-G-free plasmid
(template c) was unaffected by E2-TR in parallel, serving as a
negative control (Fig. 6B, lanes 17–22). Taken together, these results
show that the E2-TR can activate transcription from a promoter-
The BPV-1 E2-TR antagonizes E2-TA dependent transcriptional
Finally, in contrast to previous studies, we extended our
analyses to quantify the relative abilities of the truncated E2
proteins to inhibit E2-TA-dependent transcriptional activation of
both a minimal heterologous promoter and the native BPV-1 P89
early promoter. The E2-TR effectively inhibited activation of both
promoter targets by E2-TA in a dose-dependent manner as
increasing concentrations of E2-TR rapidly reduced E2-activated
transcription (Fig. 7A). However, at higher E2-TR levels, the
inhibition did not lead to full repression, but instead reached
Fig. 5. BPV-1 E2-TR can cooperate with cellular trans-activating factors. (A) Reporter plasmids harboring defined binding sites for specific cellular factors were
cotransfected with pCG E2 expression plasmids in HeLa cells. All activities represent fold above or below the normalized E2?2 tk-cat (þneo) baseline (clone h) as indicated
by the open box. (B) Chimeric GAL-TAD expression plasmids (300 ng) harboring functional domains of cellular transactivators were cotransfected with optimal quantities
of pCG E2 expression plasmids (1 ng pCG-E2; 100 ng pCG E2-TR or pCG E8^E2) and a minimal tk promoter reporter plasmid (1 mg) containing adjacent GAL and E2 binding
sites. Relative CAT activities are expressed as fold above the normalized GAL4 DBD alone (clone j). Activities represent averages of 2 to 7 independent experiments. ‘‘*’’
indicates representative value consistent with previously reported phenotypes (Ushikai et al., 1994).
M.J. Lace et al. / Virology 429 (2012) 99–111
a plateau of ?10% of the original E2-induced activity. The E8^E2,
however, was more efficient at inhibiting both E2-TA activated
targets as well as inhibiting targets activated by E2-TR (Fig. 7B
and C, respectively). These results demonstrate that the truncated
E2 gene products are not functionally redundant in BPV-1 but
represent alternate modes of differential modulation of E2-TA
This study demonstrates that the E2-TR gene product is a
transcriptional activator, capable of cooperating with cellular factors
to modulate BPV early gene expression in a manner distinct from
that of other E2 isoforms. In contrast to the potent E8^E2 repressor,
the E2-TR only partially inhibits E2-TA activation of the BPV-1 major
early promoter (P89), potentially preventing a complete shutdown
of transcription by the E8^E2 protein. BPV early gene regulation is
therefore determined by the complex interplay of all three E2 gene
The BPV-1 E2-TR gene product functions as a transcriptional
transactivator in vivo and in vitro
The E2 gene of bovine papillomavirus type-1 encodes three
proteins: a full-length E2 transactivator gene product and two
C-terminal forms, E2-TR and E8^E2. Although both truncated E2
proteins have been reported to function as transcriptional repressors,
Fig. 6. BPV-1 E2-TR cooperates with cellular factors as a promoter–proximal transactivator in vivo and in vitro. (A) Increasing quantities of pCG-E2 and pCG-E2-TR
expression plasmids (0.3, 1.0 and 3.0 mg) were cotransfected with an E2-responsive target template (16 mg) and the SV2-cat plasmid as a transfection control in HeLa cells
as in Fig. 1C. Transcription from the minimal tk promoter was analyzed by RNase protection assays. Activities were determined by scanning densitometry and represent
fold above or below normalized controls and represent averages of 2–3 independent experiments. (B) Purified recombinant E2-TR protein was incubated with transcription
templates harboring E2 and Sp1 binding sites in G-free in vitro transcription assays. Unlabeled Sp1, E2 and non-specific oligonucleotide competitors were also incorporated
to gauge transcription specificity. Template b, containing the reversed orientation of Sp1 and E2 sites, was tested in parallel as well as a template (c) containing only E2
sites. A template (d) containing an Ad MLP G-free 110 nt template (with no E2 sites) was included as an internal control.
M.J. Lace et al. / Virology 429 (2012) 99–111
in part due to the competitive displacement of the E2-TA homo-
dimers bound at E2 sites or via heterodimer formation, the functional
differences between E2-TR and E8^E2 had been incompletely defined.
In this study, we analyzed the transcriptional functions of E2-TR
in vivo and in vitro. Consistent with a previous report (Steger et al.,
1995), the E2-TR showed the ability to activate transcripts from a
minimal promoter in reporter assays but in contrast to previous
results was also capable of activating the complex native major early
promoter (P89) of BPV-1 in vivo. In comparison, the E8^E2 did not
activate either promoter in parallel. Recombinant E2-TR protein,
purified by sequence-specific chromatography, also stimulated tran-
scription initiation in G-free in vitro transcription assays. Further-
more, functional analysis of GAL4 chimeric proteins containing
discrete segments of the E2 TAD revealed that the C-terminal portion
of the E2 activation domain encoded in the E2-TR is sufficient to
drive transcriptional activation in vivo and in vitro.
In contrast to previous reports which relied on representative
promoter assays, this study quantitatively determined that the
E2-TR protein functions as a transcriptional activator in a variety
of assays. Since the activation by E2-TR (Fig. 1E) was weaker than
the full length E2 (?3 fold vs. 100 fold) (Ushikai et al., 1994), it is
still reasonable to predict that the E2-TR gene product function-
ally serves as a transcriptional inhibitor of the activation by the
full length E2 through competitive binding to the E2 site or
heterodimer formation. In contrast to E2-TR, the E8^E2 protein did
not activate transcription, indicating that the E2-TR and E8^E2
are functionally distinct; this is consistent with their respective
structures since the C-terminal portion of the E2-TAD is encoded
Fig. 7. E2-TR competitively inhibits E2-dependent activation in vivo. Increasing amounts (as indicated) of the (A) pCG E2-TR and (B) pCG E8^E2 expression plasmids were
cotransfected in HeLa cells with the E2?2-Sp1?2-tk(-38)-cat reporter (left panels) or the native BPV promoter reporter P89-cat, (right panels) in the absence or presence
of optimal indicated concentrations of the pCG E2 expression plasmid. (C) Increasing amounts of pCG E8^E2 expression plasmid were cotransfected with the E2?2-Sp1?
2-tk(-38)-cat reporter (left panel) or the native BPV promoter reporter P89-cat, (right panel) in the absence or presence of optimal indicated concentrations of the pCG
E2-TR expression plasmid. Binding sites within the P89 promoter are indicated by open triangles. Relative CAT activities are expressed as fold above the normalized
baseline activities of the respective reporter plasmid and represent averages of 3 to 6 independent experiments.
M.J. Lace et al. / Virology 429 (2012) 99–111
in E2-TR ORF, but not in E8^E2 ORF. Although the precise
mechanisms of E2-TR modulation of the native BPV-1 promoters,
such as P89, are still unclear, it can be predicted that E2-TR, but
not E8^E2, can act as a weak transcriptional activator for other
E2-responsive BPV promoters and that varying levels of the E2
isoforms (Hubbert et al., 1988) likely play a role in modulating
BPV-1 gene expression throughout the viral life cycle.
In this study we showed that the C-terminal portion of E2 TAD
contained in E2-TR can support transcriptional activation in
assays using GAL4 DBD chimeric proteins. This result not only
confirmed our results that E2-TR is a transcriptional activator, but
further indicates that the E2 TAD is composed of multiple
functional subdomains, similar to other transcriptional activators
(Courey and Tjian, 1988) that may act via synergistic cooperation.
Differential mechanisms of transcriptional modulation by the BPV-1
E2 gene products
The BPV-1 E2 has been shown to cooperate with a wide range
of cellular TADs to activate transcription (Ham et al., 1991; Li
et al., 1991; Rank and Lambert, 1995; Ushikai et al., 1994; Ustav
et al., 1993). Comparison of the capacities of the E2 isoforms to
cooperate with cellular TADs revealed that the E2-TR could also
cooperate with the same extended range of heterologous cellular
factor TADs but to varying degrees when compared to the E2-TA.
The E8^E2, however, did not exhibit any cooperative interaction
with the same range of TADs tested in parallel.
We have also demonstrated that E2-TR can activate transcription
from E2 sites proximal to the transcription initiation site, while the
full length E2-TA or E8^E2 repress transcription from the same
position. The E2-TA can activate transcription only from the distal
E2 sites, as demonstrated in HPV-16 (Lace et al., 2008b). The E2-TA
has been shown to inhibit BPV promoter activity by displacement of
cellular factors bound to the constitutive enhancer (Vande Pol and
Howley, 1990). Similarly, we have noted that the HPV-16 E2-TA can
displace cellular transactivators bound to overlapping motifs and
directly inhibit transcription when bound to an E2 site proximal to
the HPV-16 major early promoter via displacement or direct inhibi-
tory interaction (i.e. quenching) potentially as a result of varying
steric effects of E2-dependent promoter conformations (data not
Clearly, the E2-TR targets some of the same cellular components
but functions in a manner distinct from the E2-TA when bound
adjacent to a transcription initiation site. The E2-TR can function as a
transcriptional activator in this position while the E2-TA is inhibi-
tory when bound to the same site in parallel. The E2-TR gene
product may therefore interact with different cellular targets, such
as the TFIID and/or TFIIB components of the general transcription
complex (Rank and Lambert, 1995), in a manner distinct from other
E2 gene products, potentially as a result of conformational con-
straints which vary from that of the DNA/E2-TA complex. The same
structural determinants that permit E2-TR protein to function as an
activator of the general transcription machinery in this context may
conversely limit its ability to interact directly with cellular TADs
bound to proximal cis elements in vivo.
We also noted that complexes formed between the E2-TR
protein and the BPV E2 site in vitro displayed greater stability (i.e.
greater off rates) than that of the E2-TA/BPV E2 site complexes in
parallel, suggesting that the differential complex kinetics may
provide an additional mechanism by which the E2-TR can antag-
onize E2-TA-dependent regulation.
The role of multiple E2 gene products in viral persistence
The E2-TA of BPV and HPVs has been shown to interact with
cellular BRD4 to facilitate partitioning of the viral genome to daughter
cells, promote stability of the viral episomes and sequester the
genome in transcriptionally active chromatin (Jang et al., 2009; You
et al., 2005) and further stabilize the E2-TA protein by preventing
ubiquitin-dependent degradation (Zheng et al., 2009). The E2-TR and
E8^E2 gene products, however, do not share this capacity to associate
with mitotic chromatin (Oliveira et al., 2006) and potentially interfere
with this mode of E2-TA dependent maintenance of replicating
extrachromosomal PV genomes. The E2-TA is necessary for viral
replication as well as gene regulation; in contrast, previous reports
have shown that the E2-TR homodimers alone cannot support
replication and may antagonize maintenance of the viral plasmid
genome in persistently infected cells (Pepinsky et al., 1994). However,
the regulation of viral DNA replication may also involve heterodimers
of E2-TA with E2-TR and/or E8 ^E2 (Kurg et al., 2009).
The E8^E2 gene product appears to be a highly conserved genetic
component in PVs and serving critical functions in the viral life
cycles (Jeckel et al., 2003; Stubenrauch et al., 2000). We recently
demonstrated that the HPV-16 E8^E2 inhibits replication by limiting
the expression of limiting levels of E1 replicase (Lace et al., 2008a)
yet was not required for HPV plasmid genome maintenance in
keratinocytes. The HPV-31 E8^E2 and other high risk HPV E8^E2 gene
products similarly require interaction with cellular corepressors to
inhibit transcription and long term cell growth (Ammermann et al.,
2008; Fertey et al. 2011; Francis et al., 2000) but whether or not the
BPV-1 E8^E2 or its genetic equivalents in other PVs utilize the same
mechanisms to modulate viral transcription and replication through-
out the respective viral life cycles is still unclear.
Cell lines harboring mutations within the ORFs encoding the E2
gene products have been reported in animal and human PVs and
exhibit a variety of phenotypes which indicate the truncated E2
gene products may play additional critical roles in viral transforma-
tion (Jeckel et al., 2003; Lace et al., 2008a; Lambert et al., 1990;
Vande Pol and Howley, 1995; Zemlo et al., 1994). For example,
mutations preventing expression of either truncated BPV-1 E2 gene
product alone did not prevent immortalization of C127 host cells,
while simultaneous loss of both gene products abrogated BPV-1
immortalization capacity. These results suggest that the E2-TR and
E8^E2 may confer complementary roles in viral persistence which are
revealed only when both gene products are disrupted, yet the role of
the E2-TR and E8^E2 in modulating critical gene expression and
replication in the establishment phase may be distinct from that
in the persistence phase. Whether similar E2-TR gene products
are expressed by HPVs, however, remains to be demonstrated. For
example, we have recently mapped the functional cistron encoding
an HPV-16 E8^E2 (Lace et al., 2008a) but have not detected an E2-TR
activity in HPV-16 to date (Lace et al., unpublished data). Further-
more, while truncated BPV E2 isoforms are abundantly expressed
and readily detectable in fibroblasts, comparative levels of all HPV
replication factors appear to be very limiting during the establish-
ment and persistence phase of HPV infection in human keratino-
cytes. However, it will prove interesting to determine whether or
not HPVs utilize a similar complex program of early gene expression
modulated by an E2-TR gene product, as demonstrated in BPV-1, or
if this capacity represents another example of divergent mechan-
isms governing critical early events in the life cycles and potentially
influencing establishment and persistence of bovine and human
Materials and methods
Plasmid designs are illustrated in the respective figures. The
BPV-1 pCG-E2, pCG-E2-TR, pCG-E8^E2 expression plasmids were a
generous gift from Tanaka and Herr (1990), Ustav and Stenlund
M.J. Lace et al. / Virology 429 (2012) 99–111
(1991). The pCG-E2 has a point mutation in the AUG at nt 3091 to
avoid the synthesis of C-terminal E2-TR form as previously
described (Ushikai et al., 1994). The GAL4-E2 (aa 1–410), GAL4-E2
(aa 4–283), GAL4-E2 (aa 1–162) and GAL4-E2 (aa 162–283) express
the full length E2-TA, the TAD of the E2-TA, TAD of the E2-TR and the
reciprocal 50portion of the TAD, respectively in-frame with the GAL4
DBD (aa 1–147). The GAL4 DBD (aa 1–147) and LEX DBD (aa 1–202)
expression plasmids were kind gifts from Green and Ptashne.
Heterologous GAL-chimeric factor constructs and the design of the
tk(-38)-cat clones was previously described (Ushikai et al., 1994). The
HSV-1 tk core promoter (nt ?38 to þ56) was inserted between the
Xba I and BamH I sites in pUC-cat. Synthetic double stranded
oligonucleotides of the motifs for cellular DNA binding factors
(Ushikai et al., 1994) were then inserted between the Xho I and an
upstream Xba I sites, separated by ?1 helical turn upstream of the tk
TATAA motif. Plasmids containing the G-free 250 nt, 320 nt or 380 nt
cassettes were constructed as previously described (Lace et al., 2009).
The Ad MLP G-free 110 nt plasmid was a gift from Roeder.
Transient transfection assays and cell culture
RNase protection and CAT enzymatic reporter assays were
performed in transfected cells as described (Cripe et al., 1990;
Haugen et al., 1987; Ushikai et al., 1994). Cell extracts for CAT
assays were prepared from sub-confluent transfected cultures
72 h post-transfection. All RNA samples were isolated from
duplicate 150 mm2HeLa culture plates using guanidine thiocya-
nate extraction of RNA up to 24 h post transfection (Ushikai et al.,
1994). Protected fragments detected in RNase protection assays
were visualized by autoradiography after resolution on a 10%
polyacrylamide-8M Urea gel. As an internal control, we used the
SV2N-cat-del-13 plasmid which contains the strong SV40 enhan-
cer/promoter and a deletion at the 50untranslated end of the cat
gene cassette yielding a 56 nt protected fragment. Sequencing
ladders and labeled DNA size makers were run in parallel to
determine the apparent sizes of protected fragments. HeLa cells
as well as fibroblast cell lines of bovine, human, murine or simian
origin (B1, GM-3, 3T3 and CV-1) were cultured as previously
described (Ushikai et al., 1994). Primary human keratinocytes
were cultured in E-media as described (Lace et al., 2008b).
E2-dependent growth curves (calculated as per (Lace et al.2011))
were measured for HeLa cells (200,000 cells per replicate MP6 wells)
transfected with pCG E2 expression plasmids and E2 responsive
targets at the same optimized concentrations used in parallel assays
shown in Fig. 1 and Table 1 assays and cultured for same time period.
Transfection efficiencies were confirmed to be 50% or higher as
determined by cotransfection of CMV-b-gal plasmid (2mg) and
analysis by standard staining methodologies.
Expression and purification of recombinant E2 proteins
Synthesis and isolation of recombinant vaccinia viruses
expressing E2 proteins were prepared as previously described
(Ushikai et al., 1994). E2 proteins were expressed in HeLa cells
coinfected with both vaccinia-E2 recombinant and vTF73 which
contains the bacteriophage T7 RNA polymerase under the control
of the strong vaccinia promoter (Moss, 1991). E2 proteins were
extracted from harvested HeLa cells with 4 volumes of 800 mM
NaCl and 20 mM Tris-HCl, pH¼8.0.
Crude E2 proteins, produced in Hela cells via vaccinia expres-
sion, were purified by Heparin agarose column elution followed
by 2 cycles of adsorption and elution using specific DNA affinity
chromatography as previously described (Ishiji et al., 1992;
Ushikai et al., 1994). All fractions were monitored in gel mobility
shift assays for the presence of E2 proteins. The purified fractions
were concentrated ?7 fold on a Centricon ultrafiltration device
(molecular weight cutoff¼30 kD; Amicon, Waltham, MA, USA).
Silver-stained SDS-PAGE gels demonstrated that E2 preparations
contained 75–90% pure E2 proteins of the correct apparent size
and comparable concentration of 0.3–1.2 ng/ml (as per (Ushikai
et al., 1994)).
In vitro transcription assays
The in vitro transcription assays were performed with DNA
templates containing the 320 nt or 380 nt G-free cassette in the
presence of HeLa whole cell extract as previously described (Lace
et al., 2009; Ushikai et al., 1994). The tk (-38)-G-free (250)
containing 250b nt G-free cassette was served as an internal
control. DNA templates were preincubated with E2 proteins for
30 min at 25 1C followed by an additional preincubation after the
addition of HeLa extract. Transcription was initiated by adding
32P-UTP, GTP, ATP and 30O-methyl GTP to final concentration of
0.5 mM and performed at 25 1C for 45 min. Transcription reac-
tions then were digested by 10 U of RNase T1 at 37 1C for 30 min
and stopped by the addition of 200 ml of 0.2% SDS, then extracted
with phenol/chloroform. The transcribed RNA was precipitated
with ethanol, and visualized by autoradiography after resolution
on a 10% polyacrylamide-6M urea gel.
Mobility shift assays
Synthetic oligonucleotides encompassing binding motifs used in
this study included: The BPV-1 E2#10 and E2 consensus sites
(Ushikai et al., 1994) (see Table 2), an Sp1 consensus site (Forsberg
and Westin, 1991) (50-CCGGCCCCGCCC-30) and a GAL4 site (Carey
et al., 1989) (50-CGGAGGACTGTCCTCCG-30). Purified E2 proteins
were incubated with a32P[g-ATP]-labeled double stranded oligonu-
cleotides in the presence of 0.2 to 0.5 ng/ml of bovine serum albumin
and 30 ng of synthetic poly-dI.poly-dC polynucleotide in mobility
shift assays as previously described (Ishiji et al., 1992; Lace et al.,
2009; Ushikai et al., 1994). Binding reactions were performed at
30 1C for 30 min, then resolved on a 6% polyacrylamide denaturing
gel at 4 1C. DNA binding complexes were visualized by autoradio-
graphy and quantified by scanning densitometry. To determine the
binding affinities, unlabeled competitor oligonucleotides were added
to binding reactions at a range of concentrations before adding E2
proteins (Lace et al., 2009; Ushikai et al., 1994). To determine the
dissociation rate (or ‘‘off rate’’), excess quantities of unlabeled
competitors were added after DNA probe: protein complexation
and incubated over a range of time periods (1–90 min).
The authors wish to thank B. Moss for vaccinia expression
vectors, M. Ustav and A. Stenlund for pCG E2 plasmids and
I. Davidson for assistance with in vitro transcription protocols.
This work was supported by merit awards to LPT and THH from
the Department of Veterans Affairs, the NIH (CA-49912) and the
University of Iowa Diabetes and Endocrinology Research Center.
Abroi, A., Kurg, R., Ustav, M., 1996. Transcriptional and replicational activation
functions in the bovine papillomavirus type 1 E2 protein are encoded by
different structural determinants. J. Virol 70, 6169–6179.
Allikas, A., Ord, D., Kurg, R., Kivi, S., Ustav, M., 2001. Roles of the hinge region and
the DNA binding domain of the bovine papillomavirus type 1 E2 protein in
initiation of DNA replication. Virus Res 75, 95–106.
Ammermann, I., Bruckner, M., Matthes, F., Iftner, T., Stubenrauch, F., 2008.
Inhibition of transcription and DNA replication by the papillomavirus
E8-E2C protein is mediated by interaction with corepressor molecules. J. Virol.
82 (11), 5127–5136.
M.J. Lace et al. / Virology 429 (2012) 99–111
Androphy, E.J., Lowy, D.R., Schiller, J.T., 1987. Bovine papillomavirus E2 trans-
activating gene product binds to specific sites in papillomavirus DNA. Nature
Campo, M.S., 2002. Animal models of papillomavirus pathogenesis. Virus Res. 89
Carey, M., Kakidani, H., Leatherwood, J., Mostashari, F., Ptashne, M., 1989. An
amino-terminal fragment of GAL4 binds DNA as a dimer. J. Mol. Biol. 209,
Chiang, C.M., Ustav, M., Stenlund, A., Ho, T.F., Broker, T.R., Chow, L.T., 1992. Viral E1
and E2 proteins support replication of homologous and heterologous papillo-
maviral origins. Proc. Natl. Acad. Sci. USA 89, 5799–5803.
Choe, J., Vaillancourt, P., Stenlund, A., Botchan, M., 1989. Bovine papillomavirus
type 1 encodes two forms of a transcriptional repressor: structural and
functional analysis of new viral cDNAs. J. Virol 63, 1743–1755.
Courey, A.J., Tjian, R., 1988. Analysis of Sp1 in vivo reveals multiple transcriptional
domains, including a novel glutamine-rich activation motif. Cell 55, 887–898.
Cripe, T.P., Alderborn, A., Anderson, R.D., Parkkinen, S., Bergman, P., Haugen, T.H.,
Pettersson, U., Turek, L.P., 1990. Transcriptional activation of the human
papillomavirus-16 P97 promoter by an 88-nucleotide enhancer contain-
ing distinct cell-dependent and AP-1-responsive modules. New. Biol. 2,
Cripe, T.P., Haugen, T.H., Turk, J.P., Tabatabai, F., Schmid, P.G., D¨ urst, M., Gissmann,
L., Roman, A., Turek, L.P., 1987. Transcriptional regulation of the human
papillomavirus-16 E6-E7 promoter by a keratinocyte-dependent enhancer,
and by viral E2 trans-activator and repressor gene products: implications for
cervical carcinogenesis. EMBO J. 6, 3745–3753.
de Prat-Gay, G., Gaston, K., Cicero, D.O., 2008. The papillomavirus E2 DNA binding
domain. Front Biosci. 13, 6006–6021.
Demeret, C., Yaniv, M., Thierry, F., 1994. The E2 transcriptional repressor can
compensate for Sp1 activation of the human papillomavirus type 18 early
promoter. J. Virol 68, 7075–7082.
DiMaio, D., 1991. Transforming activity of bovine and human papillomaviruses in
cultured cells. Adv. Cancer Res 56, 133–159.
Doorbar, J., 2005. The papillomavirus life cycle. J. Clin. Virol. 32 (Suppl), 7–15.
Dostatni, N., Lambert, P.F., Sousa, R., Ham, J., Howley, P.M., Yaniv, M., 1991. The
functional BPV-1 E2 trans-activating protein can act as a repressor by
preventing formation of the initiation complex. Genes Dev. 5, 1657–1671.
Ferguson, M.K., Botchan, M.R., 1996. Genetic analysis of the activation domain
of bovine papillomavirus protein E2: Its role in transcription and replication.
J. Virol 70, 4193–4199.
Ferreiro, D.U., Lima, L.M., Nadra, A.D., Alonso, L.G., Goldbaum, F.A., de Prat-Gay, G.,
2000. Distinctive cognate sequence discrimination, bound DNA conformation,
and binding modes in the E2 C-terminal domains from prototype human and
bovine papillomaviruses. Biochemistry 39, 14692–14701.
Fertey, J., Hurst, J., Straub, E., Schenker, A., Iftner, T., Stubenrauch, F., 2011. Growth
inhibition of HeLa cells is a conserved feature of high-risk human papilloma-
virus E8^E2C proteins and can also be achieved by an artificial repressor
protein. J. Virol 85 (6), 2918–2926.
Forsberg, M., Westin, G., 1991. Enhancer activation by a single type of transcrip-
tion factor shows cell type dependence. EMBO J 10, 2543–2551.
Francis, D.A., Schmid, S.I., Howley, P.M., 2000. Repression of the integrated
papillomavirus E6/E7 promoter is required for growth suppression of cervical
cancer cells. J. Virol 74, 2679–2686.
Giri, I., Yaniv, M., 1988. Structural and mutational analysis of E2 trans-activating
proteins of papillomaviruses reveals three distinct functional domains. EMBO J
Ham, J., Dostatni, N., Arnos, F., Yaniv, M., 1991. Several different upstream
promoter elements can potentiate transactivation by the BPV-1 E2 protein.
EMBO J 10, 2931–2940.
Haugen, T.H., Cripe, T.P., Ginder, G.D., Karin, M., Turek, L.P., 1987. Trans-activation
of an upstream early gene promoter of bovine papilloma virus-1 by a product
of the viral E2 gene. EMBO J 6, 145–152.
Haugen, T.H., Lace, M.J., Ishiji, T., Sameshima, A., Anson, J.R., Turek, L.P., 2009.
Cellular factors are required to activate bovine papillomavirus-1 early gene
transcription and to establish viral plasmid persistence but are not required
for cellular transformation. Virology 389 (1-2), 82–90.
Haugen, T.H., Turek, L.P., Mercurio, F.M., Cripe, T.P., Olson, B.J., Anderson, R.D.,
Seidl, D., Karin, M., Schiller, J., 1988. Sequence-specific and general transcrip-
tional activation by the bovine papillomavirus-1 E2 trans-activator require an
N-terminal amphipathic helix-containing E2 domain. EMBO J. 7, 4245–4253.
Hegde, R.S., Grossman, S.R., Laimins, L.A., Sigler, P., 1992. Crystal structure at 1.7˚A
of the bovine papillomavirus-1 E2 DNA- binding domain bound to its DNA
target. Nature 359, 505–512.
Hubbert, N.L., Schiller, J.T., Lowy, D.R., Androphy, E.J., 1988. Bovine papilloma
virus-transformed cells contain multiple E2 proteins. Proc. Natl. Acad. Sci. USA
Hwang, E.S., Riese II, D.J., Settleman, J., Nilson, L.A., Honig, J., Flynn, S., DiMaio, D.,
1993. Inhibition of cervical carcinoma cell line proliferation by the introduc-
tion of a bovine papillomavirus regulatory gene. J. Virol. 67, 3720–3729.
Ishiji, T., Lace, M.J., Parkkinen, S., Anderson, R.D., Haugen, T.H., Cripe, T.P., Xiao, J.H.,
Davidson, I., Chambon, P., Turek, L.P., 1992. Transcriptional enhancer factor
(TEF)-1 and its cell-specific co-activator activate human papillomavirus-16 E6
and E7 oncogene transcription in human keratinocytes and cervical carcinoma
cells. EMBO J. 11, 2271–2281.
Jang, M.K., Kwon, D., McBride, A.A., 2009. Papillomavirus E2 proteins and the
host BRD4 protein associate with transcriptionally active cellular chromatin.
J. Virol. 83 (6), 2592–2600.
Jeckel, S., Loetzsch, E., Huber, E., Stubenrauch, F., Iftner, T., 2003. Identification of
the E9/E2C cDNA and functional characterization of the gene product reveal a
new repressor of transcription and replication in cottontail rabbit papilloma-
virus. J. Virol. 77 (16), 8736–8744.
Kurg, R., Uusen, P., Sepp, T., Sepp, M., Abroi, A., Ustav, M., 2009. Bovine
papillomavirus type 1 E2 protein heterodimer is functional in papillomavirus
DNA replication in vivo. Virology 386 (2), 353–359.
Lace, M.J., Anson, J.R., Klussmann, J.P., Wang, D.H., Smith, E.M., Haugen, T.H., Turek,
L.P., 2011. Human papillomavirus type 16 (HPV-16) genomes integrated in
head and neck cancers and in HPV-16-immortalized human keratinocyte
clones express chimeric virus-cell mRNAs similar to those found in cervical
cancers. J. Virol. 85 (4), 1645–1654.
Lace, M.J., Anson, J.R., Thomas, G.S., Turek, L.P., Haugen, T.H., 2008a. The E8–E2
gene product of human papillomavirus type 16 represses early transcription
and replication but is dispensable for viral plasmid persistence in keratino-
cytes. J. Virol. 82 (21), 10841–10853.
Lace, M.J., Anson, J.R., Turek, L.P., Haugen, T.H., 2008b. Functional mapping of the
human papillomavirus type 16 E1 cistron. J. Virol. 82 (21), 10724–10734.
Lace, M.J., Yamakawa, Y., Ushikai, M., Anson, J.R., Haugen, T.H., Turek, L.P., 2009.
Cellular Factor YY1 Downregulates the Human Papillomavirus (HPV)-16 E6/E7
Promoter, P97, In Vivo and In Vitro from a Negative Element Overlapping the
Transcription Initiation Site. J. Gen. Virol.
Lambert, P.F., Hubbert, N.L., Howley, P.M., Schiller, J.T., 1989. Genetic assignment
of multiple E2 gene products in bovine papillomavirus-transformed cells.
J. Virol. 63, 3151–3154.
Lambert, P.F., Monk, B.C., Howley, P.M., 1990. Phenotypic analysis of bovine
papillomavirus type 1 E2 repressor mutants. J. Virol. 64, 950–956.
Lambert, P.F., Spalholz, B.A., Howley, P.M., 1987. A transcriptional repressor
encoded by BPV-1 shares a common carboxy-terminal domain with the E2
transactivator. Cell 50, 69–78.
Li, R., Knight, J.D., Jackson, S.P., Tjian, R., Botchan, M.R., 1991. Direct interaction
between Sp1 and the BPV enhancer E2 protein mediates synergistic activation
of transcription. Cell 65, 493–505.
McBride, A.A., Byrne, J.C., Howley, P.M., 1989. E2 polypeptides encoded by bovine
papillomavirus type 1 form dimers through the common carboxyl-terminal
domain: transactivation is mediated by the conserved amino-terminal
domain. Proc. Natl. Acad. Sci. USA 86, 510–514.
Monini, P., Blitz, I.L., Cassai, E., 1993. Cooperative DNA binding of the bovine
papillomavirus E2 transcriptional activator is antagonized by truncated E2
polypeptides. J. Virol 67, 5668–5676.
Moss, B., 1991. Vaccinia virus: A tool for research and vaccine development.
Science 252, 1662–1667.
Oliveira, J.G., Colf, L.A., McBride, A.A., 2006. Variations in the association of
papillomavirus E2 proteins with mitotic chromosomes. Proc. Natl. Acad. Sci.
USA 103 (4), 1047–1052.
Pepinsky, R.B., Androphy, E.J., Corina, K., Brown, R., Barsoum, J., 1994. Specific
inhibition of a human papillomavirus E2 trans-activator by intracellular
delivery of its repressor. DNA Cell Biol. 13, 1011–1019.
Psyrri, A., DiMaio, D., 2008. Human papillomavirus in cervical and head-and-neck
cancer. Nat. Clin. Pract. Oncol 5 (1), 24–31.
Rank, N.M., Lambert, P.F., 1995. Bovine papillomavirus type 1 E2 transcriptional
regulators directly bind two cellular transcription factors, TFIID and TFIIB.
J. Virol 69, 6323–6334.
Sanchez, I.E., Dellarole, M., Gaston, K., de Prat Gay, G., 2008. Comprehensive
comparison of the interaction of the E2 master regulator with its cognate
target DNA sites in 73 human papillomavirus types by sequence statistics.
Nucleic Acids Res. 36 (3), 756–769.
Sandler, A.B., Baker, C.C., Spalholz, B.A., 1996. Sp1 is critical for basal and
E2-transactivated transcription from the bovine papillomavirus type 1 P89
promoter. J. Gen. Virol. 77, 189–198.
Senechal, H., Poirier, G.G., Coulombe, B., Laimins, L.A., Archambault, J., 2007. Amino
acid substitutions that specifically impair the transcriptional activity of papillo-
mavirus E2 affect binding to the long isoform of Brd4. Virology 358 (1), 10–17.
Spalholz, B.A., Yang, Y.C., Howley, P.M., 1985. Transactivation of a bovine
papilloma virus transcriptional regulatory element by the E2 gene product.
Cell 42, 183–191.
Steger, G., Corbach, S., 1997. Dose-dependent regulation of the early promoter of
human papillomavirus type 18 by the viral E2 protein. J. Virol 71, 50–58.
Steger, G., Ham, J., Lefebvre, O., Yaniv, M., 1995. The bovine papillomavirus 1 E2
protein contains two activation domains: one that interacts with TBP and
another that functions after TBP binding. EMBO J 14, 329–340.
Steger, G., Ham, J., Yaniv, M., 1996. E2 proteins: modulators of papillomavirus
transcription and replication. Methods Enzymol. 274, 173–185.
Stubenrauch, F., Hummel, M., Iftner, T., Laimins, L.A., 2000. The E8E2C protein, a
negative regulator of viral transcription and replication, is required for
extrachromosomal maintenance of human papillomavirus type 31 in kerati-
nocytes. J. Virol 74, 1178–1186.
Stubenrauch, F., Lim, H.B., Laimins, L.A., 1998. Differential requirements for
conserved E2 binding sites in the life cycle of oncogenic human papillomavirus
type 31. J. Virol 72, 1071–1077.
Stubenrauch, F., Straub, E., Fertey, J., Iftner, T., 2007. The E8 repression domain can
replace the E2 transactivation domain for growth inhibition of HeLa cells by
papillomavirus E2 proteins. Int. J. Cancer 121 (10), 2284–2292.
M.J. Lace et al. / Virology 429 (2012) 99–111
Szymanski, P., Stenlund, A., 1991. Regulation of early gene expression from the Download full-text
bovine papillomavirus genome in transiently transfected C127 cells. J. Virol 65,
Tanaka, M., Herr, W., 1990. Differential transcriptional activation by Oct-1 and
Oct-2: interdependent activation domains induce Oct-2 phosphorylation. Cell
Thierry, F., Howley, P.M., 1991. Functional analysis of E2-mediated repression of
the HPV18 P105 promoter. New Biol 3, 90–100.
Ushikai, M., Lace, M.J., Yamakawa, Y., Kono, M., Anson, J., Ishiji, T., Parkkinen, S.,
Wicker, N., Valentine, M.E., Davidson, I., Turek, L.P., Haugen, T.H., 1994. trans
activation by the full-length E2 proteins of human papillomavirus type 16 and
bovine papillomavirus type 1 in vitro and in vivo: Cooperation with activation
domains of cellular transcription factors. J. Virol. 68, 6655–6666.
Ustav, E., Ustav, M., Szymanski, P., Stenlund, A., 1993. The bovine papillomavirus
origin of replication requires a binding site for the E2 transcriptional activator.
Proc. Natl. Acad. Sci. USA 90, 898–902.
Ustav, M., Stenlund, A., 1991. Transient replication of BPV-1 requires two viral
polypeptides encoded by the E1 and E2 open reading frames. EMBO J 10,
Vaillancourt, P., Nottoli, T., Choe, J., Botchan, M.R., 1990. The E2 transactivator of
bovine papillomavirus type 1 is expressed from multiple promoters. J. Virol.
Vande Pol, S.B., Howley, P.M., 1990. A bovine papillomavirus constitutive enhancer
is negatively regulated by the E2 repressor through competitive binding for a
cellular factor. J. Virol 64, 5420–5429.
Vande Pol, S.B., Howley, P.M., 1992. The bovine papillomavirus constitutive
enhancer is essential for viral transformation, DNA replication, and the
maintenance of latency. J. Virol 66, 2346–2358.
Vande Pol, S.B., Howley, P.M., 1995. Negative regulation of the bovine papilloma-
virus E5, E6, and E7 oncogenes by the viral E1 and E2 genes. J. Virol 69,
Winokur, P.L., McBride, A.A., 1992. Separation of the transcriptional activation and
replication functions of the bovine papillomavirus-1 E2 protein. EMBO J 11,
Xiao, J.H., Davidson, I., Matthes, H., Garnier, J.M., Chambon, P., 1991. Cloning,
expression, and transcriptional properties of the human enhancer factor TEF-1.
Cell 65, 551–568.
You, J., Schweiger, M.R., Howley, P.M., 2005. Inhibition of E2 binding to Brd4
enhances viral genome loss and phenotypic reversion of bovine papilloma-
virus-transformed cells. J. Virol. 79 (23), 14956–14961.
Zemlo, T.R., Lohrbach, B., Lambert, P.F., 1994. Role of transcriptional repressors in
transformation by bovine papillomavirus type 1. J.Virol. 68, 6787–6793.
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 (17), 8683–8692.
M.J. Lace et al. / Virology 429 (2012) 99–111