1 large isoform of hepatitis delta antigen activates serum
TRANSCRIPT
8/17/00 Kato
1
Large Isoform of Hepatitis Delta Antigen Activates
Serum Response Factor-Associated Transcription
Short title: LHDAg activates SRF-associated transcription
Tadashi Goto, Naoya Kato*, Suzane Kioko Ono-Nita, HideoYoshida,
Motoyuki Otsuka, Yasushi Shiratori, and Masao Omata
Department of Gastroenterology, Faculty of Medicine,
University of Tokyo
*To whom correspondence should be addressed:
Department of Gastroenterology, Faculty of Medicine,
University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan
PHONE: +81-3-3815-5411 ext. 33070
FAX: +81-3-3814-0021
E-mail: [email protected]
Copyright 2000 by The American Society for Biochemistry and Molecular Biology, Inc.
JBC Papers in Press. Published on August 28, 2000 as Manuscript M002947200 by guest on February 7, 2018
http://ww
w.jbc.org/
Dow
nloaded from
8/17/00 Kato
2
SUMMARY
Hepatitis delta virus (HDV) infection sometimes causes severe and
fulminant hepatitis as a coinfection or superinfection along with the hepatitis B
virus. To elucidate the underlying mechanism of injury caused by HDV, we
examined whether two isoforms of the hepatitis delta antigen (HDAg) had any
effect on five well-defined intracellular signal transduction pathways: serum
response factor (SRF)-, serum response element (SRE)-, nuclear factor κB-,
activator protein 1-, and cyclic AMP response element-dependent pathways.
Reporter assays revealed that large HDAg (LHDAg) activated the SRF- and SRE-
dependent pathways. In contrast, small HDAg (SHDAg) did not activate any of
five pathways. LHDAg enhanced the transcriptional ability of SRF without
changing its DNA-binding affinity in an electrophoretic mobility shift assay. In
addition, LHDAg activated a rat SM22α promoter containing SRF-binding site
and a human c-fos promoter containing SRE. In conclusion, LHDAg, but not
SHDAg, enhances SRF-associated transcriptions. Despite structural similarities
between the two HDAgs, there are significant differences in their effects on
intracellular signal transduction pathways. These results may provide clues that
will aid in the clarification of functional differences between LHDAg and SHDAg
and the pathogenesis of delta hepatitis.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
3
INTRODUCTION
Hepatitis delta antigen (HDAg) was discovered as an antigen localized in
the nuclei of hepatocytes in a patient who had already been infected with the
hepatitis B virus (HBV) (1). HDAg is the only known protein encoded by the
hepatitis delta virus (HDV), which is a defective human pathogen whose
transmission requires the helper function of HBV (2). The two isoforms of HDAg,
large and small forms (3, 4), are identical except for additional 19 residues located
at the C-terminus of the large form. The small form of HDAg (SHDAg) consists
of 195 amino acids (aa) (molecular weights; 24 kDa), and the large form of HDAg
(LHDAg) consists of 214 aa (27 kDa) (5, 6). Both forms of HDAgs have common
functional domains: an N-terminal coiled-coil domain responsible for
oligomerization (7), a central domain responsible for a nuclear localization signal
(8, 9), and a central helix-turn-helix domain responsible for binding to the RNA
genome (10, 11). Despite these structural similarities, the two HDAgs play
complementary roles in HDV replication; SHDAg is required for genome
replication (12), but LHDAg acts as an inhibitor of replication (13). LHDAg is
also required for HDV assembly (14, 15); the formation of HDV viral particles
requires isoprenylation at the C-terminus of the LHDAg (16), although the
mechanism of LHDAg on this packaging remains unclarified.
It has been shown that SHDAg suppresses the gene expression of HBV (17).
Nevertheless, HDV infection often causes severe chronic hepatitis and liver
cirrhosis as a superinfection of chronic HBV carriers and fulminant hepatitis as a
coinfection with HBV (18). It was also reported that SHDAg possessed a
cytotoxic effect on infected hepatocytes, while LHDAg was thought to reduce this
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
4
effect (19). The reason why HDV coinfection or superinfection with HBV causes
hepatic injury, however, remains unclear. Therefore, we investigated whether the
two isoforms of HDAg had any effect on five well-defined intracellular signaling
pathways: serum response factor (SRF)-, serum response element (SRE)-, nuclear
factor κB (NF-κB)-, activator protein 1 (AP-1)-, and cyclic AMP response
element (CRE)-dependent pathways.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
5
EXPERIMENTAL PROCEDURES
Construction of HDAgs expressing plasmids
The HDV sequence used in this study was derived from pSVLD3 (kindly
provided by Dr. J. Taylor, Fox Chase Cancer Center, PA), containing a trimer of
unit-length HDV cDNA (12, 20). Both LHDAg and SHDAg regions were
amplified by polymerase chain reaction using the following primers having a ScaI
restriction site (underlined) (nucleotide positions according to the HPDGEN
sequence (21) are shown in parentheses): sense primer 5’-AAA AGT ACT ACC
ATG AGC CGG TCC GAG TCG AGG A-3’ (1598-1577) and antisense primer
5’-AAA AGT ACT TCA CTG GGG TCG ACA ACT CTG GGG A-3’ (954-978)
for the LHDAg region, and 5’-AAA AGT ACT CTA TGG AAA TCC CTG GTT
TCC CCT GA-3’ (1011-1036) for the SHDAg region. Each amplified fragment
was digested with ScaI and then cloned into pCXN2 (kindly provided by Dr. J.
Miyazaki, University of Osaka, Japan), a mammalian expression plasmid having a
β-actin based CAG promoter (22) to generate pCXN2-DLm and pCXN2-DS
expressing SHDAg. Because pCXN2-DLm contained the nucleotide A at position
1012 according to the sequence of HPDGEN (21), we substituted G for A to make
a construction of LHDAg expressing plasmid pCXN2-DL using site-directed
mutagenesis (Quick change site-directed mutagenesis kit, Stratagene, La Jolla,
CA).
For an electrophoretic mobility shift assay (EMSA), pCXN2-FlagDL,
which expresses Flag-tagged LHDAg, was constructed by adding the Flag
sequence into the 5’ terminal of the LHDAg region.
For evaluation of the transcriptional activation of LHDAg, pM-GAL4DL,
which expresses LHDAg protein fused to the yeast GAL4 DNA binding domain
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
6
(DNA-BD), was constructed by subcloning the LHDAg region into the pM vector,
a mammalian expression plasmid encoding the GAL4 DNA-BD gene (Clontech,
Palo Alto, CA).
All cloned plasmids were purified using the Endofree plasmid kit (Qiagen,
Hilden, Germany) and sequenced using an autosequencer (PE Applied
Biosystems, Foster City, CA) by the dye-termination method as described
previously (23) to confirm the integration of HDAg genes.
Construction of Elk1- and SRF-expressing plasmids
Using the T7Elk 1-428 (kindly provided by Dr. R. Treisman, Imperial
Cancer Research Fund, UK) (24) as a template, the full-length Elk1 region (1-428
aa) was amplified by polymerase chain reaction with Elk1 region-specific primers
having an XhoI restriction site. Amplified products were digested with XhoI and
then cloned into the XhoI site of pCXN2 to generate pCXN2-Elk1.
pFA-GALElk (Stratagene) expresses the activation domain of Elk1 (307-
427 aa) fused to the GAL4 DNA-BD. A plasmid expressing SRF (10-508 aa)
fused to the GAL4 DNA-BD, pSG5-GALSRF (10-508), was a gift from Dr. M.
Fujii (Kanazawa University, Japan) (25).
Reporter plasmids of intracellular signal transduction pathways
Five reporter plasmids containing the Photinus pyralis (Firefly)-luciferase
reporter gene driven by a basic promoter element (TATA box) plus a defined
inducible cis-enhancer element were utilized (PathDetect cis-reporting systems,
Stratagene, La Jolla, CA). Each Firefly-luciferase gene in the reporter plasmid
was controlled by the following synthetic enhancer sequences (the binding site for
the transcription factor is capitalized): 5 repeats of the binding sites for SRF
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
7
(gtCCATATTAGGac, pSRF-Luc), 5 repeats of SRE
(AGGATgtCCATATTAGGacatct, pSRE-Luc), 5 repeats of the binding sites for
NF-κB (tGGGGACTTTCCgc, pNF-κB-Luc), 7 repeats of the binding sites for
AP-1 (TGACTAA, pAP1-Luc), and 4 repeats of CRE (agccTGACGTCAgag,
pCRE-Luc). A control plasmid expressing Renilla reniformis (Seapansy)-
luciferase driven by the herpes simplex virus thymidine kinase promoter, pRL-TK,
was used to correct the efficiency of transfection (Promega, Madison, WI). As a
positive control for the activation of SRF-associated pathways, pFC-PKA
(Stratagene), which expresses the catalytic subunit of cAMP-dependent protein
kinase (protein kinase A; PKA) driven by a CMV promoter, was utilized.
In addition to these reporter plasmids which contain synthetic promoters,
SM22α-Luc, a luciferase reporter plasmid with a rat smooth muscle-specific gene
SM22α promoter (-1507 to +32) containing the binding sites for SRF, generally
called the CArG box (kindly provided by Dr. M. E. Lee, Brigham and Women’s
Hospital, MA) (26), and HF456, a luciferase reporter plasmid with a human c-fos
promoter (-456 to +47) containing SRE (kindly provided by Dr. T. Ishikawa,
University of Tokyo, Japan) (27), were utilized. Specific mutations were
introduced into the SM22α-Luc and HF456 plasmids using site-directed
mutagenesis (Stratagene). The specific base changes were chosen based on their
ability to disrupt SRF binding. As for the SM22α-Luc, the CArG box 1 at bp -162
to bp -153 was converted from CCAAATATGG to AAAAATATGG (SM22α
m1-Luc); the CArG box 2 at bp -283 to bp -266 was converted from
CCATAAAAGGTTTTTCCC to CCATAAAAAATTTTTCCC (SM22α m2-Luc).
Combinations of mutations were generated by performing mutagenesis reactions
on a template already containing a mutation (SM22α m1m2-Luc). As for the
HF456, the CArG box at bp -314 to bp -305 was converted from CCATATTAGG
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
8
to CCATATTATT (HF456 m1). The sequences and nucleotide positions are
according to the previous reports (28, 29). Altered nucleotides are indicated in
bold type. All mutations were verified by sequencing.
pFR-Luc (Stratagene) has a Firefly-luciferase gene controlled by five yeast
GAL4 upstream activation sequences, which was utilized as a reporter to examine
transcriptional activation of GAL4 DNA-BD fused Elk1 or SRF by HDAg.
Cell culture
HeLa cells (human cervical carcinoma cell line), HuH-7 cells (human
hepatocellular carcinoma cell line), and HepG2 cells (hepatoblastoma cell line)
were obtained from the Riken cell bank (Tsukuba science city, Japan). Cells were
cultured in Dulbecco’s modified Eagle medium (DMEM, Gibco BRL,
Gaitherburg, MD) supplemented with 10% fetal bovine serum (FBS) at 37 oC in
5% CO2.
Transfection and luciferase assays
Approximately 4 x 105 HeLa, HuH7, or HepG2 cells were plated into a 6-
well tissue culture plate (Iwaki Glass, Chiba, Japan) 24 hours before transfection.
Using the Effectene transfection reagent (Qiagen, Hilden, Germany), cells were
transiently cotransfected with 0.2 µg of reporter plasmids and 0.2 µg of pCXN2,
pCXN2-DL, or pCXN2-DS.
For evaluation of the transcriptional ability of Elk1 or SRF, cells were
transiently cotransfected with 0.15 µg reporter plasmids, 0.05 µg pFA-GALElk or
pSG5-GALSRF, and 0.2 µg pCXN2, pCXN2-DL, or pCXN2-DS. For evaluation
of the transcriptional ability of LHDAg, 0.2 µg reporter plasmids and 0.2 µg pM
vector or pM-GAL4DL were used. For the assay using HF456 containing the c-
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
9
fos promoter, HeLa cells were cultured in DMEM supplemented with 0.5% FBS
to give low background.
Forty-eight hours after transfection, whole cell lysates were examined for
luciferase activity (PicaGene dual Sea Pansy system, Toyo ink, Tokyo, Japan)
with a luminometer (Lumat LB9507, EG&G Berthold, Bad Wildbad, Germany).
Firefly-luciferase activity was normalized for transfection efficiency based on
Seapansy-luciferase activity. The luciferase activity of the cells which were
transfected with the reporter plasmid plus pCXN2 was set arbitrarily at 1.0, and
then the relative luciferase activity was compared to this established value. Assays
were performed at least in triplicate.
Concentration of cells transiently transfected with plasmids
As only a small percentage of the cells were transfected by a transient
transfection method, we utilized the MACSelect system (Miltenyi Biotec,
Germany) to concentrate transiently plasmid-transfected cells. The concentration
of the plasmid-transfected cells was achieved by magnetically isolating the
transfected cells via a surface marker, a truncated mouse H-2 k molecule, which
was expressed from the cotransfected plasmid, pMacsKk. HeLa cells were
cotransfected with pCXN2, pCXN2-FlagDL, or pFC-PKA together with pMacsKk.
After 36 hours, cells were treated with 0.05% trypsin and dispersed by being
pipetted into single-cell suspensions after addition of 100 µl FBS. The cells were
resuspended with 600 µl of PBE buffer (phosphate-buffered saline supplemented
with 0.5% bovine serum albumin and 2 mM EDTA) containing 80 µl of
micromagnetic beads conjugated with a monoclonal antibody against mouse H-2 k
and incubated for 15 min at room temperature. Then, magnetically labeled cells
were recovered by the magnetic separation column and used for an electrophoretic
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
10
mobility shift assay (EMSA).
Electrophoretic mobility shift assay (EMSA)
Annealed oligonucleotides for the CArG box (5’-GGATGTCCATATTA
GGACATC-3’) were end-labeled with [α-32P] ATP using T4 polynucleotide
kinase. Nuclear extracts were obtained from HeLa cells transfected with pCXN2,
pCXN2-DL, or pFC-PKA as previously described (30). The protein concentration
of nuclear extracts was measured by a Micro BCA protein assay reagent kit
(Pierce, Chester, UK), then adjusted to give equal concentration. Four micrograms
of the nuclear extracts were incubated with 0.035 pmol of the CArG box-
radiolabeled probe. Anti-SRF antibody (Santa Cruz Biotechnology, Santa Cruz,
CA) were used for the supershift assay. The CArG box-unlabeled competitor was
added at a 100-fold molar excess to confirm site specific binding. The binding
reaction was performed at room temperature for 30 minutes in a 10 µl mixture
consisting of 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM dithiothreitol,
50 mM NaCl, and 0.5 µg of poly(dI-dC)·(dI-dC). DNA-protein complexes were
then loaded onto a chilled 4% nondenaturing acrylamide gel. Gel electrophoresis
was executed in 0.25 x Tris borate-EDTA at 4oC. The gel was dried, and
autoradiography was performed using a Fujix bio-imaging analyzer BAS 2000
(Fuji Photo Film, Tokyo, Japan).
Western blotting analysis
LHDAg and SHDAg expression was confirmed by Western blotting (ECL-
plus, Amersham, Buckinghamshire, UK) using the soluble protein extracts of
HeLa cells which were transfected with pCXN2, pCXN2-DL, or pCXN2-DS (31).
HDAg expression was examined using serum obtained from a patient with
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
11
chronic HDV infection. Using the soluble protein extracts of HeLa cells
transfected with pCXN2-Elk1 and pCXN2-DL cultured in DMEM supplemented
with 0.5% FBS, the phosphorylation of Elk1 was examined by anti-Elk1 antibody
and anti-phospho-specific Elk1 (Ser 383) antibody [PhosphoPlus Elk1 (Ser 383)
antibody kit, New England Biolabs, Beverly, MA] according to the
manufacturer’s instructions.
Statistics
Data are expressed as the means + S.D. Statistical analysis was performed
using the t test. A p-value of less than 0.05 was considered statistically significant.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
12
RESULTS
Activation of SRF- and SRE-dependent signal transduction pathways by
LHDAg
Each of the five reporter plasmids (pSRF-Luc, pSRE-Luc, pNF-κB-Luc,
pAP1-Luc, or pCRE-Luc) was transiently cotransfected into HeLa cells with
pCXN2, pCXN2-DL, or pCXN2-DS. LHDAg activated the SRF- and SRE-
dependent signal transduction pathway at a value 4.0 + 1.2 (mean + SD)-fold and
2.5 + 1.0-fold higher than the control, respectively (Fig. 1A). There was no
significantly increased activation in the remaining three pathways by LHDAg. In
addition, SHDAg did not activate any of the five pathways (Fig. 1A). Expression
of LHDAg and SHDAg were confirmed by Western blotting at the expected size:
LHDAg at a molecular weight of 27 kDa, and SHDAg at a molecular weight of
24 kDa (5).
Cell line independent and dose-dependent activation of SRF-dependent pathway
by LHDAg
We investigated whether LHDAg activated the SRF-dependent pathway in
other cell lines. Results similar to those in HeLa cells were obtained in HuH-7
cells (4.3 + 2.4-fold higher than control) and HepG2 cells (3.3 + 1.0-fold) (Fig.
1B). In addition, LHDAg activated the SRF-dependent signal in a dose-dependent
manner in HeLa cells (Fig. 1C).
No additional effect of SHDAg on the activation of the SRF-dependent pathway
by LHDAg
Since LHDAg and SHDAg play complementary roles in genome viral
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
13
replication, we examined whether SHDAg had influence on SRF-dependent
transcription by LHDAg. Although HeLa cells were transiently transfected with
pSRF-Luc and pCXN2-DL in combination with various amounts of pCXN2-DS,
the presence of SHDAg had no influence on activation of the SRF-dependent
pathway by LHDAg (Fig. 2).
Enhancement of transcriptional ability of SRF by LHDAg
Because LHDAg is a nuclear protein, we examined whether LHDAg itself
possessed transactivation activity. pM-GAL4DL, a GAL4 DNA-BD-LHDAg
fusion protein expression plasmid, was cotransfected into HeLa cells with pFR-
Luc. The GAL4 DNA-BD-LHDAg fusion protein, however, did not activate
transcription from pFR-Luc (1.0 + 0.1-fold higher than control), suggesting that
LHDAg itself had no transcriptional ability in mammalian cells.
To elucidate how LHDAg activates the SRF-dependent signal transduction
pathway, we examined the influence of LHDAg on: 1) transcriptional ability of
SRF by a reporter assay, 2) SRF binding to SRF binding site, the CArG box
[CC(A/T)6GG], by EMSA, and 3) a rat SM22α promoter containing two CArG
boxes.
Transcriptional activation of SRF: we initially examined whether LHDAg
enhanced the transcriptional activation of SRF. HeLa cells were transiently
transfected with pFR-Luc, pSG5-GALSRF(10-508) and pCXN2-DL. The cells
were then assayed for the luciferase activity. As shown in Fig. 3A, the
transcriptional ability of SRF increased by approximately 2.5-fold with LHDAg.
LHDAg did not increase SRF binding to the CArG box: next, we examined
whether LHDAg modified SRF binding to the CArG box by EMSA using nuclear
extracts of HeLa cells transfected with pCXN2, pCXN2-FlagDL or pFC-PKA and
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
14
concentrated by MACSelect system. After concentration by the MACSelect
system, we could confirm that more than 70% of collected cells were transfection
positive by immunostaining with anti-Flag M2 antibody (Upstate Biotechnology,
Lake Placid, NY) (data not shown). LHDAg did not influence SRF binding to the
CArG box, however binding increased by the catalytic subunit of PKA (Fig. 3B).
Moreover, LHDAg was not contained in the DNA-protein complex because
mobility of DNA-SRF complex did not change in the presence of Flag-tagged
LHDAg and no supershifted band was observed by adding anti-Flag M2 antibody
(data not shown).
The effect of HDAgs on the rat SM22 promoter: we examined whether
LHDAg activated a rat promoter containing the CArG box. By use of SM22α-Luc,
the reporter plasmid with a SM22α promoter containing two CArG boxes,
luciferase activity increased by a value 3.3 + 0.5-fold higher than the control with
LHDAg (Fig. 4B), while there was no increase with SHDAg (data not shown). To
determine whether this activation by LHDAg was mediated by the CArG box, we
mutated two CArG boxes, either alone or combinations, and then measured
luciferase activity with LHDAg or without LHDAg. Mutations of each CArG box,
which abolish SRF binding, were introduced into the indicated elements in the
context of the wild type reporter construct (Fig. 4A), and their influence were
examined in transient transfection assays. As shown in Fig. 4B, the effect of
LHDAg on the SM22α promoter were decreased when the CArG boxes were
mutated, indicating that activation of the SM22α promoter by LHDAg is
dependent on enhancement of the transcriptional ability of SRF.
LHDAg activates the SRE-dependent pathway through SRF, not Elk1
LHDAg activated not only the SRF-dependent pathway but also the SRE-
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
15
dependent pathway. SRE is composed of the CArG box bound SRF and nearby
Ets motif bound ternary complex factors (TCFs) such as Elk1. Elk1, regulated by
mitogen-activated protein kinase (MAPK) (32-34), is phosphorylated by MAPK
at a cluster of Serine/Threonine motifs located at its C-terminus. Phosphorylation
at these sites, particularly Ser 383, is critical for transcriptional activation (24, 35).
Therefore, we examined the influence of LHDAg on: 1) transcriptional activation
of Elk1 protein by a reporter assay, 2) phosphorylation of Elk1 protein by
Western blotting, and 3) the human c-fos promoter containing SRE.
Transcriptional activation and phosphorylation of Elk1 protein: HeLa cells
were transiently transfected with pFR-Luc, pFA-GALElk, and pCXN2-DL. Cells
were then assayed for luciferase activity. The transcriptional ability of ELK1 was
not increased by LHDAg (Fig. 3C). Western blotting analysis of Elk1 also
indicated that LHDAg did not enhance the phosphorylation of Elk1 (Fig. 3C),
thereby suggesting that LHDAg has no effect on the transcriptional ability of
Elk1.
The effect of HDAgs on a human c-fos promoter: by use of HF456, a reporter
plasmid with a c-fos promoter containing SRE, the luciferase activity increased by
a value 1.8 + 0.3-fold higher than the control with LHDAg (Fig. 4C), while it did
not increase with SHDAg (data not shown). To determine whether this activation
by LHDAg was mediated by the CArG box, we mutated the CArG box and then
measured luciferase activity with LHDAg or without LHDAg. Mutation of the
CArG box was introduced into the indicated element in the context of the wild
type reporter construct (Fig. 4A) and the influence was examined. As shown in
Fig. 4C, the effect of LHDAg on the c-fos promoter was decreased when the
CArG box was mutated. Since LHDAg had no effect on Elk1, this activation of
the c-fos promoter occurred through the transcriptional activation of SRF by
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
16
LHDAg. These results suggest that activation of the SRE-dependent signal by
LHDAg is dependent on enhancement of the transcriptional ability of SRF, not
Elk1.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
17
DISCUSSION
In this study, LHDAg clearly activated SRF-associated pathways by
enhancing the transcriptional ability of SRF without changing its DNA-binding
affinity. We confirmed this LHDAg-induced activation not only by using
plasmids with synthetic enhancers, but also by using plasmids with the rat SM22α
promoter and the human c-fos promoter as well as a plasmid with the yeast GAL4
upstream activation sequence. SRF was identified as a critical factor involved in
mediating serum and growth factor-induced transcription from the c-fos proto-
oncogene (36). The SRF binding site in the promoters of immediate-early genes c-
fos and pip92 (37) is called SRE, which is composed of two elements, the CArG
box (CCATATTAGG) and the nearby Ets motif (AGGAT). SRF binds to the
CArG box as a dimer whereas Elk1 can not do so by itself, but it binds to the Ets
motif by making a ternary complex with SRF (34, 36). Although the activation of
c-fos SRE through the MAPK cascade is mainly dependent on Elk1 (32-34), SRF
can solely activate c-fos SRE in response to serum growth factors and intracellular
activation of heterotrimeric G protein (29).
Several viral transforming proteins such as human T-cell leukemia virus
type I (HTLV-I) activator protein Tax, the polyomavirus middle-T antigen, and
HBV X protein (HBx) target c-Fos induction (25, 38-40). Tax and the
polyomavirus middle-T antigen enhance c-Fos induction through SRF (38, 39).
Tax directly binds to SRF to activate transcription. On the other hand, middle-T
antigen activates transcription via Rac activation in the cytoplasm, but the
pathway from Rac to SRF is unknown. Although LHDAg is a nuclear protein, we
could not observe an interaction between LHDAg and SRF by
immunoprecipitation (data not shown). It has been shown that stimulation by
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
18
growth hormone, angiotensin II, and HMG-I induce the transcriptional activation
of SRF by enhancing binding activity of SRF to the CArG box (26, 41, 42).
LHDAg, however, did not increase SRF binding to the CArG box, while PKA
enhanced the binding (Fig. 3B). Although the mechanism underlying activation of
the SRF-dependent signaling pathway has not been completely elucidated, there is
mention that PKA is required for SRF nuclear import (43). PKA has also been
shown to activate the CRE-dependent pathway. Therefore, this evidence together
with a lack of activation of the CRE-dependent pathway by LHDAg (Fig.1A)
leads to the conclusion that SRF activation by LHDAg seems to be PKA-
independent. These results suggest that LHDAg activates SRF-associated
transcription through a novel mechanism.
In the case of HBx, results similar to those of LHDAg have been noted (44).
Although HBx can not bind double strand DNA directly and does not act as a
transcriptional activator (45), it can transactivate many cellular genes. HBx
activates transcription through its interaction with a variety of transactivators
(46,47) and transcription factors such as TFIIB, TFIIH, and RNA polymerase
subunit 5 (48,49). Therefore, a possible mechanism underlying the enhancement
of transcriptional activation of SRF by LHDAg could be that LHDAg functions as
a co-transactivator. Since HDAgs have no RNA polymerase activity, HDV
requires host cellular proteins such as RNA polymerase II for replication of its
genome (50,51). Since both HDAgs are known to participate in HDV replication,
it is not difficult to speculate that HDAgs may recruit a transcription factor or
RNA polymerase II related proteins for its viral replication. In fact, HDAg
interacts with nuclear proteins such as DIPA, nucleolin, and karyopherin alpha2
(52-54). Overexpression of nucleolin (nucleolin expressing plasmid pCGN-Nu
was kindly provided by Dr. S. C. Lee, National Taiwan University, Taiwan (55))
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
19
activated the SRF-dependent pathway at a value two times higher than that with
LHDAg alone (data not shown). Although there is no evidence that DIPA or
karyopherin alpha2 mediates transcription, nucleolin has been implicated in
rDNA transcription, rRNA maturation, ribosome assembly, and nucleo-
cytoplasmic transport (56). Therefore, it is possible that nucleolin plays an
important role in the enhancement of transcriptional activation of SRF by
LHDAg.
Recently, Wei et al. showed that LHDAg, but not SHDAg, could activate
transcription from several promoters, including the AP-1 binding site (57). Their
findings are consistent with our results that only LHDAg can activate
transcription, although Wei et al. did not investigate the mechanism of how
LHDAg activates the pathway. Although transfection of 0.2 µg pCXN2-DL did
not activate the AP-1-dependent pathway, transfection of 0.4 µg pCXN2-DL
activated the pathway by approximately 2.5-fold higher than control (data not
shown). Thus, LHDAg may be a potent activator of the AP-1-dependent pathway
when expressed in a relatively large amount. On the other hand, Lo et al.
demonstrated that both LHDAg and SHDAg inhibited SP1-activated and basal
RNA polymerase II transcription (58). Although we did not examine the effect of
HDAgs on SP-1-associated or basal RNA polymerase II transcription, it may be
possible that LHDAg inhibits this transcription by withholding the key
transcriptional factors necessary for the activation of SRF-dependent pathways.
LHDAg activated a rat SM22α promoter having a CArG box. The CArG
box has been identified as a constituent in the promoters of a number of muscle-
specific genes, including SM22α (28), α-smooth muscle actin (α-SMA) (59), and
the cardiac and skeletal muscle actin (60, 61). LHDAg may enhance the
transcription of these muscle-specific genes containing a CArG box. For example,
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
20
α-SMA is expressed by activated hepatic stellate cells, a kind of hepatic
sinusoidal cell that causes hepatic fibrosis (62). Therefore, there may be a
possibility that LHDAg upregulates transcription of α-SMA through SRF, causes
activation of hepatic stellate cells, and induces hepatic fibrosis. LHDAg also
activated the c-fos promoter through SRF. A number of experiments have
suggested that c-Fos plays a critical role in the response to growth factors and that
aberrant production of c-Fos protein can lead to oncogenesis. Interestingly,
hepatitis delta patients with primary hepatocellular carcinoma were younger than
patients infected by HBV alone (63). It is tempting to speculate that LHDAg may
promote cell proliferation through the induction of c-Fos protein.
In summary, LHDAg, but not SHDAg, activates SRF-associated pathways.
This observation may provide clues to help clarify the functional differences
between LHDAg and SHDAg and the pathogenesis of delta hepatitis.
Acknowledgments-We would like to thank Drs. Taylor J, Miyazaki J, Treisman R,
Lee ME, Lee SC, Ishikawa T, and Fujii M for plasmids.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
21
REFERENCES
1. Rizzetto, M., Canese, M. G., Arico, S., Crivelli, O., Trepo, C., Bonino, F., and
Verme, G. (1977) Gut. 18, 997-1003
2. Rizzetto, M. (1983) Hepatology 3, 906-912
3. Wang, K.-S., Choo, Q.-L., Weiner, A.-J., Ou, J.-H., Najarian, R. C., Thayer, R.
M., Mullenbach, G. T., Denniston, K. J., Gerin, J. L., and Houghton, M. (1986)
Nature 323, 508-514
4. Makino, S., Chang, M.-F., Shieh, C.-K., Kamahora, T., Vannier, D. M.,
Govindarajan, S., and Lai, M. M. C. (1987) Nature 329, 343-346
5. Bergmann, K. F., and Gerin, J. L. (1986) J. Infect. Dis. 154, 702-706
6. Imazeki, F., Omata, M., and Ohto, M. (1990) J. Viol. 64, 5594-5599
7. Wang, J.-G., and Lemon, S.M. (1993) J. Virol. 67, 446-454
8. Xia, Y.-P., Yeh, C.-T., Ou, J.-H., and Lai, M. M. C. (1992) J. Virol. 66, 914-
921
9. Chang, M.-F., Chang, S. C., Chang, C.-I., Wu, K., and Kang, H.-Y. (1992). J.
Virol. 66, 6019-6027
10. Chang, M. F., Baker, S. C., Soe, L. H., Kamahora, T., Keck, J. G., Makino, S.,
Govindarajan, S., and Lai, M. M. C. (1988). J. Virol. 62, 2403-2410
11. Lin, J. H., Chang, M.-F., Baker, S.C., Govindarajan, S., and Lai, M. M. C.
(1990) J. Virol. 64, 4051-4058
12. Kuo, M. Y. P., Chao, M., and Taylor, J. (1989) J. Virol. 63, 1945-1950
13. Taylor, J. (1993) Prog. Clin. Biol. Res. 382, 1-4
14. Chao, M., Hsieh, S.-Y., and Taylor, J. (1990) J. Virol. 64, 5066-5069
15. Chang, F.-L., Chen, P.-J., Tu, S.-J., Wang, C.-J., and Chen, D.-S.(1991) Proc.
Natl. Acad. Sci. USA 88, 8490-8494
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
22
16. Glenn, J. S., Watson, J. A., Havel, C. M., and White, J. M. (1992) Science 256,
1331-1333
17. Wu, J.-C., Chen, P.-J., Kuo, M.-Y., Lee, S.-D., Chen, D.-S., and Ting, L.-P.
(1991) J. Virol. 65, 1099-1104
18. Hoofnagle, J. H. (1989) J. Am. Med. Assoc. 261, 1321-1325
19. MacNaughton, T. B., Gowans, E. J., Jilbert, A. R., and Burrell, C. J. (1990)
Virology 177, 692-698
20. Bichko, V., Barik, S., and Taylor, J. (1997) J. Virol. 71, 512-518
21. Kuo, M. Y. .P., Goldberg, J., Coates, L., Mason, W. S., Gerin, J., and Taylor, J.
(1988) J. Virol. 63, 1855-1861
22. Niwa, H., Yamamura, K. and Miyazaki, J. (1991) Gene 108, 193-199
23. Togo, G., Toda, N., Kanai, F., Kato, N., Shiratori, Y., Kishi, K., Imazeki, F.,
Makuuchi, M., and Omata, M. (1996) Cancer Res. 56, 5620-5623
24. Marais, R., Wynne, J., and Treisman, R. (1993) Cell 73, 381-393
25. Fujii, M., Chuhjo, T., Minamino, T., Nomura, M., Miyamoto, K., and Seiki,
M. (1995) Oncogene 11, 7-14
26. Chin, M. T., Pellacani, A., Wang, H., Lin, S. S. J., Jain, M. K., Perrella, M. A.,
and Lee, M.-E. (1998) J. Biol. Chem. 273, 9755-9760
27. Ishikawa, T., Igarashi, T., Hata, K., and Fujita, T. (1999) Biochem. Biophys.
Res. Commun. 254, 566-571
28. Kemp, P. R., Osbourn, J. K., Grainger, D. J., and Metcalfe, J. C. (1995)
Biochem. J. 310, 1037-1043
29. Hill, C.S., and Treisman, R. (1995) EMBO J. 14, 5037-5047
30. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic
Acids Res. 17, 6419
31. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular cloning: A
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
23
Laboratory Manual, 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring
Harbor, NY
32. Hill, C. S., Marais, R., John, S., Wynne, J., Dalton, S., and Treisman, R.
(1993) Cell 73, 395-406
33. Gille, H., Kortenjann, M., Thomae, O., Moomaw, C., Slaughter, C., Cobb, M.
H., and Shaw, P. E. (1995) EMBO J. 14, 951-962
34. Treisman, R. (1994) Curr. Opin. Geneti Dev. 4, 96-101
35. Kortenjann, M., Thomae, O., and Shaw, P. E. (1994) Mol. Cell. Biol. 14,
4815-24
36. Treisman, R., and Ammerer, G. (1992) Curr. Opin. Geneti Dev. 2, 221-226
37. Latinkic, B. V., and Lau, L. F. (1994) J. Biol. Chem. 269, 23163-23170
38. Fujii, M., Tsuchiya, H., Chuhjo, T., Akizawa, T., and Seiki, M. (1992) Genes
Dev. 6, 2066-2076
39. Urich, M., Senften, M., Shaw, P. E., and Ballmer-Hofer, K. (1997) Oncogene
14, 1235-1241
40. Avantaggiati, M. L., Natoli, G., Balsano, C., Chirillo, P., Artini, M., DeMarzio,
E., Collepardo, D., and Levrero, M. (1993) Oncogene 8, 1567-1574
41. Liao, J., Hodge, C., Meyer, D., Ho, P. S., Rosenspire, K., and Schwartz, J.
(1997) J. Biol. Chem. 272, 25951-25958
42. Hautmann, M. B., Thompson, M. M., Swartz, E. A., Olson, E. N., and Owens,
G. K. (1997) Circ. Res. 81, 600-610
43. Gauthier-Rouviere, C., Vandromme, M., Lautredou, N., Cai, Q. Q., Girard, F.,
Fernandez, A., and Lamb, N. (1995) Mol. Cell. Biol. 15, 433–444
44. Lin, Y., Tang, H., Nomura, T., Dorjsuren, D. Hayashi, N., Wei, W., Ohta, T.,
Roeder, R., and Murakami, S. (1998) J. Biol. Chem. 273, 27097-27103
45. Yen, B. T. S. (1996) J. Biomed. Sci. 3, 20-30
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
24
46. Maguire, H. F., Hoeffler, J. P., and Siddiqui, A. (1991) Science 252, 842-844
47. Lin, Y., Nomura, T., Yamashita, T., Dorjsuren, D., Tang, H., and Murakami,
S. (1997) Cancer Res. 57, 5137-5142
48. Haviv, I., Vaizel, D., and Shaul, Y. (1996) EMBO J. 15, 3413-3420
49. Cheong, J. H., Yi, M., Lin, Y., and Murakami S. (1995) EMBO J. 14, 143-150
50. MacNaughton, T. B., Gowans, E. J., McNamara, S. P., and Burrell, C. J.
(1991) Virology 184, 387-390
51. Fu, T.-B., and Taylor, J. (1993) J. Virol. 67, 6965-6972
52. Brazas, R., and Ganem, D. (1996) Science 274, 90-94
53. Lee, C. H., Chang, S. C., Chen, C. J., Chang, M. F. (1998) J. Biol. Chem. 273,
7650-7656
54. Chou, H. C., Hsieh, T. Y., Sheu, G. T., Lai, M. M. (1998) J. Virol. 72, 3684-
3689
55. Yang, T., H., Tsai, W., H., Lee, Y., M., Lei, H., Y., Lai, M., Y., Chen, D., S.,
Yeh, N., H., and Lee, S., C. (1994) Mol. Cell. Biol. 14, 6068-6074
56. Ginisty, H., Sicard, H., Roger, B., and Bouvet, P. (1999) J. Cell Sci. 112, 761-
772
57. Wei, Y., and Ganem, D. (1998) J. Virol. 72, 2089-2096
58. Lo, K., Sheu, G.-T., and Lai, M.M.C. (1998) Virology 247, 178-188.
59. Min, B. H., Foster, D. N., and Strauch, A. R. (1990) J. Biol. Chem. 265,
16667-16675
60. Muscat, G. E., Gustafson, T. A., and Kedes, L. (1988) Mol. Cell. Biol. 8,
4120–4133
61. Walsh, K. (1989) Mol. Cell. Biol. 9, 2191–2201
62. Friedman SL. (1993) N. Engl. J. Med. 328, 1828-35
63. Oliveri, F., Brunetto, M. R., Baldi, M., Piantino, P., Ponzetto, A., Forzani, B.,
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
25
Smedile, A., Verme, G., and Bonino, F. (1991) Prog. Clin. Biol. Res. 364, 217-
222
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
26
The abbreviations used are:
HDAg, hepatitis delta antigen; HBV, hepatitis B virus; HDV, hepatitis delta virus;
SHDAg, small hepatitis delta antigen; LHDAg, large hepatitis delta antigen; SRF,
serum response factor; SRE, serum response element; NF-κB, nuclear factor κB;
AP-1, activator protein 1; CRE, cyclic AMP response element; EMSA,
electrophoretic mobility shift assay; DNA-BD, DNA binding domain; PKA,
protein kinase A; DMEM, Dulbecco’s modified Eagle medium; FBS, fetal bovine
serum; TCFs, ternary complex factors; MAPK, mitogen-activated protein kinase;
α-SMA, α-smooth muscle actin; HTLV1, human T-cell leukemia virus type 1;
HBx, hepatitis B virus X protein.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
27
FIGURE LEGEND
Fig. 1. LHDAg activates SRF-associated signal transduction pathways.
A, The effect of HDAgs on various synthetic promoters. Reporter
plasmids, pSRF-Luc, pSRE-Luc, pNF-κB-Luc, pAP1-Luc, and pCRE-Luc were
cotransfected into HeLa cells with pCXN2, pCXN2-DL, or pCXN2-DS. The
results are expressed as the fold of luciferase activity above that induced from the
reporter plus pCXN2 plasmid. Plasmid expressing seapansy luciferase was used
as an internal control for transfection efficiency. Data shown are the average +
S.D. of more than three independent experiments. Western blotting shows the
expression of LHDAg (lane 1) and SHDAg (lane 2). The left bar indicates the
position at 27 and 24 kDa and the expected size of LHDAg and SHDAg,
respectively.
B, Activation of SRF by LHDAg in various cell lines. HeLa, HuH7, and
HepG2 cells were transfected with pSRF-Luc and the pCXN2 or pCXN2-DL.
Cells were assayed for luciferase activity. The results are expressed as the fold of
luciferase activity above that induced in the absence of HDAg.
C, LHDAg activates SRF-dependent pathway in a dose-dependent
manner. Increasing amounts of LHDAg expressing plasmid (0, 0.05, 0.1, 0.2 µg)
were cotransfected into HeLa cells with pSRF-Luc. pCXN2 was added to each
transfection to keep the total amount of DNA constant (0.4 µg). * p<0.05 vs.
control.
Fig. 2. The presence of SHDAg has no influence on activation of the SRF-
dependent pathway by LHDAg.
0.1 µg of pSRF-Luc was transiently cotransfected into HeLa cells with 0.1
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
28
µg of pCXN2-DL and with: lanes 2-4, pCXN2-DS, 0.033, 0.1, 0.2 µg,
respectively. pCXN2 was added to each transfection to keep the total amount of
DNA constant (0.4 µg).
Fig. 3. LHDAg enhances transcriptional activation of SRF, but not Elk1.
A, LHDAg enhances transcriptional activation of SRF. pFR-Luc having
the firefly-luciferase gene controlled by the yeast GAL4 upstream activation
sequence, pRL-TK and either GAL4-SRF expressing plasmid or empty vector
were cotransfected into HeLa cells with pCXN2 or pCXN2-DL. The results of
luciferase assay are expressed as the fold of luciferase activity above that induced
from the reporter plus pCXN2 without GAL4-SRF expressing plasmid. * p<0.05
B, No increase in SRF binding to the CArG box by LHDAg. Nuclear
extracts of HeLa cells transfected only with pCXN2 (lane 2), or 0.3 (lane 3, 5, and
6) or 0.9 µg (lane 4) of pCXN2-DL were incubated with 0.035 pmol of the CArG
box-radiolabeled probe. Lane 5 showed a supershifted band with use of the anti-
SRF antibody. The CArG box-unlabeled competitor was added at a 100-fold
molar excess (lane 6). Nuclear extracts of HeLa cells transfected with pFC-PKA
(lane 1) were incubated with 0.035 pmol of the CArG box-radiolabeled probe.
C, LHDAg does not enhance transcriptional activation of Elk1. pFR-
Luc, pRL-TK and Gal4-Elk1 expressing plasmid was cotransfected into HeLa
cells with pCXN2 or pCXN2-DL. The results of luciferase assay are expressed as
the fold of luciferase activity above that induced from the reporter plus pCXN2
without GAL4-Elk1 expressing plasmid. The right upper Western blotting showed
Elk1 and phosphorylated Elk1. HeLa cells were transiently transfected with
pCXN2-Elk1 and pCXN2 or pCXN2-DL in DMEM supplemented with 0.5%
FBS. Forty-eight hours after transfection, whole cell lysates were harvested for
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
8/17/00 Kato
29
Western blotting analysis. Elk1 and phosphorylated Elk1 were detected by anti-
Elk1 antibody and anti-phospho-specific Elk1 (Ser 383) antibody.
Fig. 4. The effect of LHDAg on the natural promoters.
A, Schematic representation of reporter constructs used. Site-specific
mutations that disrupt the CArG box (see “Experimental Procedures”) were
introduced into SM22α-luc and HF456 having a luciferase gene under the human
c-fos promoter.
B, The effect of LHDAg on the SM22 promoter. The indicated reporter
constructs were cotransfected into HeLa cells with pCXN2 or pCXN2-DL. The
results of luciferase assay are expressed as the fold of luciferase activity above
that induced from the SM22α-luc plus pCXN2 plasmid. * p<0.05 vs. control.
C, The effect of LHDAg on the c-fos promoter. HF456 having a
luciferase gene under the human c-fos promoter containing SRE, or HF456 m1
having mutated CArG box was cotransfected into HeLa cells with pCXN2 or
pCXN2-DL. A luciferase assay was performed and the fold of luciferase activity
is shown. * p<0.05 vs. control.
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from
Yasushi Siratori and Masao OmataTadashi Goto, Naoya Kato, Suzane Kioko Ono-Nita, Hideo Yoshida, Motoyuki Otsuka,
Factor-Associated TranscriptionLarge Isoform of Hepatitis Delta Antigen Activates Serum Response
published online August 28, 2000J. Biol. Chem.
10.1074/jbc.M002947200Access the most updated version of this article at doi:
Alerts:
When a correction for this article is posted•
When this article is cited•
to choose from all of JBC's e-mail alertsClick here
by guest on February 7, 2018http://w
ww
.jbc.org/D
ownloaded from