1. Hepatitis C virus and alcohol: Same mitotic targets but
different signaling pathways
Anna Alisi1,2
, Monica Ghidinelli1
, Alessandro Zerbini3
, Gabriele Missale3
, Clara Balsano1,4,⇑
1
Laboratory of Molecular Virology and Oncology, Fondazione A. Cesalpino, University of Rome, Rome, V. le del Policlinico 155, 00161 Rome, Italy;
2
Units of Metabolic and Autoimmune Liver Diseases, ‘‘Bambino Gesù’’ Children’s Hospital and Research Institute, P. le S. Onofrio 4, 00165 Rome,
Italy; 3
Laboratory of Viral Immunopathology, Azienda Ospedaliero, Universitaria di Parma, Via Gramsci 14, 43100 Parma, Italy; 4
Department of Internal Medicine (M.I.S.P), University of L’Aquila, University of L’Aquila, Piazzale Salvatore Tommasi 1, 67100 L’Aquila, Italy
Background & Aims: Chromosomal aberrations are frequently
observed in hepatitis C virus (HCV)- and alcohol-related hepato-
cellular carcinomas (HCCs). The mechanisms by which chromo-
somal aberrations occur during hepatocarcinogenesis are still
unknown. However, these aberrations are considered to be the
result of deregulation of some mitotic proteins, including the
alteration of Cyclin B1 and Aurora kinase A expression, and
the phosphorylation of gamma-tubulin. Our study aims at inves-
tigating changes in expression of the above mentioned proteins
and related intracellular pathways, in in vitro and in vivo models
of both HCV- and alcohol- dependent HCCs.
Methods: In this study, the molecular defects and the mechanisms
involved in deregulation of the mitotic machinery were analyzed in
human hepatoma cells, expressing HCV proteins treated or not with
ethanol, and in liver tissues from control subjects (n = 10) and patients
with HCV- (n = 10) or alcohol-related (n = 10) HCCs.
Results: Expression of Cyclin B1, Aurora kinase A, and tyrosine-
phosphorylated gamma-tubulin was analyzed in models repro-
ducing HCV infection and ethanol treatment in HCC cells. Inter-
estingly, HCV and alcohol increased the expression of Cyclin B,
Aurora kinase A, and tyrosine-phosphorylated gamma-tubulin
also in tissues from patients with HCV- or alcohol-related HCCs.
In vitro models suggest that HCV requires the expression of PKR
(RNA-activated protein kinase), as well as JNK (c-Jun N-terminal
kinase) and p38MAPK (p38 mitogen-activated protein kinase)
proteins; while, ethanol bypasses all these pathways.
Conclusions: Our results support the idea that HCV and alcohol
may promote oncogenesis by acting through the same mitotic
proteins, but via different signaling pathways.
Ó 2011 Published by Elsevier B.V. on behalf of the European
Association for the Study of the Liver.
Introduction
Each year, 550,000 new patients are diagnosed with hepatocellu-
lar carcinoma (HCC) worldwide, thus, at present, liver cancer is
considered the fifth most frequent neoplasm and, because of its
poor prognosis, the third leading cause of cancer death [1]. One
of the most important clinical challenges associated with chronic
hepatitis rises from the evidence that a considerable number of
patients develop end-stage liver disease and HCC. HCC develop-
ment and progression may depend on several etiologic factors,
including hepatitis B virus (HBV), hepatitis C virus (HCV), alcohol
abuse, and non-alcoholic steatohepatitis (NASH) [2,3]. HCV and
alcohol consumption often coexist and may act synergistically,
leading to a more rapid progression of liver disease and increas-
ing the risk of HCC [4–6]. Accordingly, several studies from differ-
ent geographic areas demonstrate that the risk for developing
HCC is higher in patients with HCV infection and alcohol abuse
than in patients with hepatitis C alone [6–8].
During HCC development, unrespectable to its etiologic ori-
gin, hepatic cells, committed to proliferate, may accumulate
mutations in key cell cycle genes or bear epigenetic changes,
microsatellite instability, and several chromosomal aberrations,
including aneuploidy [9,10]. Chromosomal aberrations are fre-
quently discovered in dysplastic nodules of cirrhotic livers
and are often present in HCC, suggesting the occurrence of
chromosomal defects at the early stages of hepatocarcinogene-
sis [9–11].
Although the mechanisms by which chromosomal aberrations
arise during hepatocarcinogenesis are still unknown, it is known
that deregulation of some mitotic events, usually found in other
tumors, occurs. Alteration of expression and/or activity of mitotic
regulatory components (cyclin B1, Aurora kinase A, etc.), as well
as deregulation of continuous cycles of microtubule polymeriza-
tion/depolymerization and phosphorylation of structural proteins
(i.e. tubulins), radically influence the control of centrosome mat-
uration and separation, bipolar spindle assembly, chromosome
alignment and segregation, and cytokinesis in human cells, lead-
ing to aneuploidy [12–14]. Cyclin B1 was found over-expressed in
HCC with advanced stage, portal invasion, and intrahepatic
metastasis [15]. Recently, the role of aurora kinase A in HCC
has moved into the focus of preclinical research. In fact, the up-
regulation of the Aurora-A gene in HCC nodules compared with
nontumorous liver tissues has been demonstrated [16].
Journal of Hepatology 2011 vol. 54 j 956–963
Keywords: Aurora kinase A; Cyclin B1; Ethanol; HCC; HCV.
Received 4 January 2010; received in revised form 28 July 2010; accepted 15 August
2010; available online 22 December 2010
⇑ Corresponding author. Address: Fondazione A. Cesalpino, c/o I Clinica Medica,
Policlinico Umberto I, Viale del Policlinico n.155, 00161 Rome, Italy. Tel.: +39 6
49975144; fax: +39 6 49975145.
E-mail address: clara.balsano@uniroma1.it (C. Balsano).
Abbreviations: HCCs,
hepatocellular carcinomas; HBV, hepatitis B virus; HCV, hepatitis C virus; NASH,
non-alcoholic steatohepatitis; p38MAPK, p38 mitogen-activated protein kinase;
PKR, RNA-activated protein kinase; JNK, c-Jun N-terminal kinase; NF-kB, nuclear
factor-kB; FAK, focal adhesion kinase; PI3K, phosphatydil-inositol-3 kinase; siRNA,
small interference RNA.
Research Article

2. Moreover, modifications of beta-tubulin isotypes were found in
liver cancer [17]. The close association between mitotic defects
and hepatocarcinogenesis is also supported by in vivo and vitro
studies demonstrating that both HCV and ethanol are able to
interfere with the regular control of mitosis in hepatocarcinoma
cell lines [18–20]. HCV is able to induce an accumulation of cells
in G2/M phase and, concomitantly, an increase of cyclin B1/cdk1
complex nuclear translocation, through its effects on p38 mito-
gen-activated protein kinase (p38MAPK) and RNA-activated pro-
tein kinase (PKR), both molecules implicated in the control of G2/
M phase progression [18,19]. Animal and human studies suggest
that alcohol may be involved in initiation, promotion, and pro-
gression of HCC, through the activation of various molecular
mechanisms including oxidative stress, changes in DNA methyla-
tion, immunosuppression, and genetic susceptibility [21,22].
Interestingly, it has been reported that HepG2 cells stably
expressing alcohol dehydrogenase showed 6-fold increase in
the percentage of cells in G2/M phase after ethanol exposure.
The impairment in cell-cycle progression was due to the accumu-
lation of the phosphorylated inactive form of Cdk1 [20,23].
Unfortunately, regarding the pathogenesis of HCV-associated
HCC, it still remains controversial whether the virus plays a direct
or an indirect role, and how alcohol operates in the acceleration
of HCC development [24,25].
We postulate that HCV, acting in synergism with alcohol
abuse, might promote mitotic aberrations through the activa-
tion/inhibition of different intracellular pathways; therefore, in
this study we analyzed the changes in expression of some impor-
tant mitotic regulators in HCC tissues derived from alcoholics or
HCV infected patients.
Although several mitotic regulators have been implicated in
cancer development and progression, we chose to study cyclin
B1 and aurora kinase A as they form a network of interactions
regulating the onset of mitosis, the centrosome biogenesis and
microtubule nucleation, and the cytokinesis. Moreover, gamma-
tubulin is a little known common partner of these two proteins
[26,27].
Here, we identify the intracellular mechanism(s) potentially
involved in mitotic aberrations, by the use of a cell-based in vitro
model.
Materials and methods
Cells, plasmids, and polyclones
Human hepatocarcinoma cell lines (HepG2 and Huh7) were grown in Dulbecco’s
modified Eagle’s medium (DMEM), supplemented with 10% fetal bovine serum
(FBS), 100 U/ml penicillin, and 100 lg/ml streptomycin at 37 °C in a 5% CO2 incu-
bator. Cells were grown to 70–80% confluence in 100-mm dishes and then trans-
fected, using Lipofectamin 2000 reagent (Invitrogen, Carlsbad, CA, USA),
according to the manufacturer’s protocols, with two different plasmids: an empty
vector (pcDNA3-FLAG) or a unique construct encoding for entire HCV products
(p34-9,7-HCVwt 1a) (gently provided by Dott. La Monica Nicola, IRBM-Pomezia,
Italy) [28].
Control polyclones (with the empty vector) and polyclones stably expressing
all HCV proteins were obtained by 21 days geneticin-selection of transfected cells.
Polyclones were cryo-preserved at À80 °C and used at the occurrence.
Polyclones selected after transfection with the empty vector were used as
internal experimental controls: preliminary experiments were performed to
exclude the presence of substantial differences in comparison with non-transfec-
ted cells (see Supplementary Fig. 1).
For all reported experiments, polyclonal cells were alternatively synchronized
in G1/S (by 2 mM thymidine) or G2/M transition (by 200 ng/ml nocodazole). See
Supplementary Table 1 for FACS analysis.
Mitotic index
The mitotic index was calculated for G1/S-synchronized polyclones using double
thymidine block [29]. After block, G1 cells were released to progress through the
cell cycle over the next 15 h, and then treated or not with ethanol. Mitotic index
was calculated by DAPI (40–60-diamino-2-phenylindole) staining, as already
described [18]. After staining, the slides were extensively washed and mounted
in Vectashield (Vector Laboratories Inc., Burlingame, CA, USA), before examination.
The mitotic index was expressed as the mean (%) of positive cells with respect
to the total cell population per field, for a total of 20 fields. Cells were counted by
fluorescence microscope (Nikon Eclipse E600 microscope, Nikon, Italy).
Western blotting
For Western blot analysis, equal amounts of proteins from tissue or cell extracts
obtained after lysis in Ripa buffer (50 mM Tris (pH 7.5), 150 mM NaCl, 1% Nonidet
P-40, 5 mM EDTA, SDS 0,1%, 2 mM phenylmethilsulfonyl fluoride, 1 g/ml aproti-
nin, 1 g/ml of leupeptin and phosphatase inhibitors), were electrophoresed on
SDS–PAGE. Proteins were transferred to PVDF membrane (Millipore, Marlbor-
ough, MA, USA) and treated with specific primary antibodies overnight at 4 °C.
Filters were then washed four times with PBS-Tween 20 and newly incubated
with peroxidase-coupled secondary antibodies for 1 h at RT. After incubation,
the blots were visualized by ECL (Amersham Pharmacia Biotechnology, Freiburg,
Germany).
Immunoprecipitation
For immunoprecipitation, 0.1 mg of total lysate was incubated with 0.1 lg of
anti-phosphotyrosine antibody (Santa Cruz) overnight at 4 °C. Samples were then
incubated with protein A agarose (Amersham Pharmacia) for 1 h at 4 °C, washed
three times with Ripa buffer, and resuspended in 10 ll of SDS-sample buffer. Each
sample was then electrophoresed on SDS–PAGE, and Western blot to reveal
gamma-tubulin was performed as described above.
Antibodies
The following antibodies were used: anti-core protein monoclonal antibody
(Affinity Bioreagents, Denver, CO, USA); anti-NS5A mouse monoclonal antibody,
anti-actin goat polyclonal antibody, anti-cyclinB1 mouse monoclonal antibody
and anti-PKR rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz,
CA, USA); anti-phosphotyrosin mouse monoclonal antibodies (Upstate Biotech-
nology, Lake Placid, NY, USA); anti-aurora A rabbit polyclonal antibody, and
anti-gamma-tubulin rabbit polyclonal antibody (GeneTex Inc., San. Antonio, TX,
USA); peroxidase-conjugated goat anti-rabbit, goat anti-mouse, and rabbit anti-
goat IgG (Santa Cruz).
Inhibitors treatment
Cells were pre-treated for 30 min with 10 lM SP600125 or SB203580 (Sigma–
Aldrich, Milano, Italy) to inhibit the activity of c-Jun N-terminal kinase (JNK)
and p38MAPK, respectively.
siRNA transfection
A cocktail of siRNA directed against several regions of PKR (siPKR) was designed
by New England Biolabs (Beverly, MA, USA) and transfected using Lipofectamin
2000 reagent (Invitrogen), according to the manufacturer’s protocols. Dosing
experiments showed that optimal silencing was achieved using 10 nM PKR siRNA
(see Supplementary Fig. 2). Cells were also transfected with siRNA double target-
ing GFP gene (Quiagen, Germantown, MD, USA), as a negative control.
Patients and liver samples
The study was performed on 60 archived liver tissues obtained from 10 patients
with HCV-related HCC, 10 patients with alcohol-associated HCC, and control non-
cirrhotic subjects obtained from 10 surgically treated patients during the course
of routine clinical care at the Surgical Departments, University of ‘‘La Sapienza
‘‘Rome’’ and at the Azienda Ospedaliero-Universitaria di Parma. At the time of
surgical resection the tumor area was separated by the surrounding tissue. Then,
part of the resected sample was fixed in formalin and embedded in paraffin for
histological diagnosis, another part of all tissues were snap frozen in liquid nitro-
JOURNAL OF HEPATOLOGY
Journal of Hepatology 2011 vol. 54 j 956–963 957

3. gen and then stored at –80 °C for molecular analysis. This study was approved by
the local ethical committee and all samples were obtained with the patient’s
informed consent.
Statistical analysis
Results are expressed as mean ± standard deviation (SD) of four samples from
at least three independent experiments. In particular, we performed an ANOVA
among groups for repeated measures, followed by Bonferroni’s correction. p
value < 0.05/n (where n is the number of comparisons) was considered to indi-
cate a statistically significant difference vs control. ⁄p <0.001, ⁄⁄p <0.01,
⁄⁄⁄p <0.05.
Results
HCV and ethanol impair mitosis in HCC cells
Since both HepG2 and Huh7 polyclones showed similar results, for
brevity we reported only the results of experiments performed on
HepG2 cells. Here, we investigated whether HCV and ethanol alter
mitosis in cell-based models of HCV infection and alcohol abuse.
Control and HCV polyclones, established as described in Materials
and methods, were firstly treated with a 25 mM minimum toxic
concentration of ethanol (see Supplementary Fig. 3), or PBS alone
for 24 h, and then synchronized in G1/S phase with double thymi-
dine block. The mitotic index was evaluated with DAPI staining,
after 0, 24, and 48 h from synchronization. As reported in Table
1, the accumulation of cells in M phase reached the maximal value
at 48 h in polyclones expressing HCV proteins. Ethanol treatment
induced a statistically significant increase of M phase cells in the
control, but not in HCV polyclones.
Considering these results, we analyzed the expression or the
activity of some molecules specifically involved in the control of
mitosis, such as cyclin B1, Aurora A kinase, and gamma-tubulin.
Forthis purpose, control and HCV polyclones,treated or not with eth-
anol, were synchronized in G2/M phase with nocodazole and, after
24 h, total proteins were extracted to perform Western blotting.
In polyclones expressing HCV proteins, the expression of
cyclin B1 and Aurora kinase A was up-regulated at all time points
(Fig. 1A). Thus, we investigated the effect of ethanol on the same
mitotic regulatory molecules. As shown in Fig. 1B, ethanol treat-
ment caused a further increase in cyclin B1 and Aurora kinase A
protein expression levels, as compared to HCV expressing cells. A
semiquantitative densitometric analysis of immunoblots was
performed (Fig. 1C).
Moreover, we analyzed, in cell-based models, the pattern of
tyrosine-phosphorylation of gamma-tubulin at G2/M transition.
As shown in Fig. 2A, HCV was able to induce a significant and
sustained increase in tyrosine-phosphorylation of gamma-tubu-
lin. Noteworthy, ethanol further increased gamma-tubulin phos-
phorylation levels compared to HCV expressing cells (Fig. 2B and
C). Newly, the ethanol effects were additive to those observed in
untreated HCV polyclones (Figs. 1C and 2C).
Expression of mitotic regulatory molecules in HCC tissues
To verify the in vivo relevance of results obtained in in vitro mod-
els, we analyzed the expression profile of cyclin B1 and Aurora A
kinase, and the phosphorylation rate of gamma-tubulin in liver
tissues derived from 30 subjects: 10 with HCV-related HCC, 10
with alcohol-associated HCC, and 10 with normal liver. In partic-
Table 1. Mitotic index values in HepG2 polyclones. p value refers to each condition vs. control.
C
C + EtOH
HCV
HCV + EtOH
*
**
* *
*
Cyclin B1
Aurora A
β-Actin
Cyclin B1
Cyclin B1
Aurora A
Aurora A
β-Actin
β-Actin
Arbitraryunits
0
100
200
300
400
500
C HCV
0 2 6 24 0 2 6 24
C HCV
A
B
C
Et-OH - + - +
Fig. 1. HCV and ethanol affect the expression of cyclin B1, and Aurora kinase
A in HepG2 cells. (A) Expression levels of cyclin B1 and Aurora kinase A in the
control and HCV polyclones at 0, 2, 6, and 24 h after release from G2/M blocking.
Beta-actin is present as the control of equal loading. (B) Expression levels of cyclin
B1 and Aurora kinase A in the control and HCV polyclones, treated or not with
ethanol (EtOH), were analyzed 24 h after nocodazole release from G2/M blocking.
To demonstrate equal loading, membranes were re-probed with the beta-actin
antibody. Immunoblots are representative of at least three independent exper-
iments. (C) Semiquantitative densitometric data are reported in the histograms as
mean values of four independent experiments ± SD (bars). p <0.001; ⁄⁄p <0.01
versus control polyclones.
Research Article
958 Journal of Hepatology 2011 vol. 54 j 956–963

4. ular, the expression levels of cyclin B1 and Aurora kinase A were
analyzed in control normal livers, in HCC tissues (HCC), and in
surrounding tumor tissues (sHCC). As shown in Fig. 3A, cyclin
B1 and Aurora kinase A were significantly over-expressed in
HCC tissues (lanes 2 and 3). We also examined the expression
of the tyrosine-phosphorylated form of gamma-tubulin. As
reported in Fig. 3B, the level of the tyrosine-phosphorylated form
of gamma-tubulin is up-regulated in HCC (lanes 2 and 3) and in
sHCC (lanes 4 and 5) and seems to be higher in alcohol-related
compared to HCV-related HCC. Quantitative data, obtained by
densitometric analysis of immunoblotting of the studied mole-
cules, is shown in Table 2. In the same table we reported clinic-
o-pathological findings of examined liver tissues.
PKR is involved in HCV-related mitotic defects
Then, we investigated which intracellular mechanism(s) might be
responsible for the interesting results we obtained both in vivo
and in vitro.
As already reported, PKR is one mediator of HCV core protein
effects on G2/M progression [18,19]. Thus, here, we firstly
evaluated whether PKR could be involved in the HCV-related
deregulation of cyclin B1, Aurora kinase A, and tyrosine-
phosphorylated gamma-tubulin. To this aim, HCV and control
polyclones were transfected with a mix of small interference
(si)RNAs (10 nM) against PKR, or alternatively, with a siRNA
against GFP as the control. Three hours later, we treated the con-
trol cells with 25 mM ethanol or an equal amount of PBS for 24 h.
G2/M synchronized cells were collected immediately after
release and at 24 h.
Silencing of PKR abrogated all mitotic HCV-related effects,
while it was unable to revert the ethanol-dependent alterations
of the mitotic molecules (Fig. 4A). Again, in HCV expressing cells,
gamma-tubulin tyrosine phosphorylation was affected by PKR
silencing (Fig. 4B). Densitometric analysis was performed and
reported as fold changes in protein levels compared to the control
considered as 1 after normalization against beta-actin (Fig. 4C).
Our results clearly indicate that PKR is involved in HCV- but
not in ethanol-dependent mitotic protein deregulation. The rele-
vance of PKR in HCV-dependent effects was also reinforced by the
analysis of the mitotic index in HCV polyclones in the presence of
siPKR (see Supplementary Fig. 4).
The effect produced on mitotic proteins by PKR silencing in
HCV polyclones stimulated with ethanol was comparable to that
observed in untreated HCV polyclones (data not shown).
Emodin counteracts the ethanol-dependent mitotic effects
To analyze the mechanisms by which ethanol induces mitotic
deregulation, we treated polyclonal cells with several different
drugs inhibiting important signal transduction intracellular path-
ways. In particular, we used SP600125 (10 lM) to inhibit JNK and
SB203580 (10 lM) to specifically inhibit p38MAPK, two impor-
tant intracellular pathways activated by ethanol treatment. As
reported in Fig. 5A and B (lanes 7–9), both inhibitors were inef-
C
C + EtOH
HCV
HCV + EtOH
P-γ-Tubulin
P-γ-Tubulin
β-Actin
P-γ-Tubulin
β-Actin
β-Actin
Arbitraryunits
0
100
200
300
400
500
C HCV
0 2 6 24 0 2 6 24
C HCV
A
B
C
- + - +Et-OH
P
S
IP: P-Tyr
P
S
IP: P-Tyr
*
*
*
Fig. 2. HCV and ethanol alter tyrosine-phosphorylation of gamma-tubulin in
HCC cells. (A) Tyrosine-phosphorylation levels of gamma-tubulin in the control
and HCV polyclones at 0, 2, 6, and 24 h after release from blocking in G2/M with
nocodazole. (B) Tyrosine-phosphorylation levels of gamma-tubulin in the control
and HCV polyclones treated or not with ethanol (EtOH) for 24 h after release from
blocking in G2/M. Panels P (upper) show the pellets containing the tyrosine-
phosphorylated form of gamma-tubulin; while panels S (lower) show the beta-
actin levels in supernatants as loading controls. Immunoblots are representative
of at least three independent experiments. (C) Semiquantitative densitometric
data are reported in the histograms as mean values of four independent
experiments ± SD (bars). p <0.001 versus control polyclones.
A
B
P-γ-Tubulin
β-Actin
P
S
IP: P-Tyr
Cyclin B1
Aurora A
β-Actin
Et-OH Et-OHC HCV HCV
Et-OH Et-OHC HCV HCV
HCC sHCC
HCC sHCC
Fig. 3. Expression levels of cyclin B1, Aurora kinase A, and phosphorylated
gamma-tubulin in liver tissues. (A) Cyclin B1 and Aurora kinase A protein
expression levels in liver tissue of control subjects (C), and in HCC (HCC) and
surrounding tissue (sHCC) of patients with hepatocarcinoma associated with HCV
infection (HCV) or alcohol abuse (EtOH). Beta-actin is present as the control of
equal loading of proteins. (B) Levels of gamma-tubulin expressed in pellets
resulting from the immunoprecipitation of tyrosine-phosphorylated proteins in
the controls and HCV or ethanol-related HCC and surrounding tissues. Panels P
(upper) show the pellets containing tyrosine-phosphorylated form of gamma-
tubulin; while panels S (lower) show the beta-actin levels in supernatants as
loading controls. The images shown are representative of at least three
independent experiments.
JOURNAL OF HEPATOLOGY
Journal of Hepatology 2011 vol. 54 j 956–963 959

5. fective on ethanol-induced mitotic biological effects. On the con-
trary, JNK and p38MAPK inhibitors were able to equally revert
HCV-dependent mitotic alterations in untreated (Fig. 5A and B,
lanes 4–6) and in ethanol-treated HCV polyclones (data not
shown).
In the light of the difficulties encountered in searching possi-
ble intracellular pathways responsible for molecular changes in
ethanol-treated cells, we decided to avoid selective inhibitors
preferring a compound with a broad spectrum of biological activ-
ities. We checked the effects of a new active bio-molecule derived
by the rhizome of Rheum palmatum L., known as Emodin. Emodin
was chosen because it is a drug described to interfere with multi-
ple signaling pathways, including NF-kB (nuclear factor-kB), FAK
(focal adhesion kinase), and PI3K (phosphatydil-inositol-3
kinase) intracellular signaling [30–32]. Interestingly, Emodin
(40 lM) was able to completely revert alcohol effects on tyro-
sine-phosphorylation of gamma-tubulin (Fig. 5C); while cyclin
B1 and Aurora kinase A expression remained unchanged
(Fig. 5D). These last data, confirmed by densitometric analysis
(Fig. 5E), together with the analysis of the mitotic index in etha-
nol treated control polyclones in the presence of Emodin (see
Supplementary Fig. 5), demonstrated that Emodin only partially
reverts the ethanol-dependent mitotic deregulation.
Discussion
HCCs are associated with high incidence of genetic alterations,
which increases during the carcinogenic process. Persistent infec-
tion with HCV has been considered a major risk for the develop-
ment of HCC, as well as heavy alcohol abuse, which has been
linked with earlier progression to HCC in chronic hepatitis C
patients [25]. However, molecular mechanisms inducing this syn-
ergism of action are still controversial. Oxidative stress and
deregulation of cellular gene expression, controlling cell-cycle
progression, seem to be dominant mechanisms for the synergic
action of alcohol and HCV [33].
HCV infection, as well as ethanol, impairs cell-cycle progres-
sion leading to a G2/M arrest in liver cells [18,20]. Interest-
ingly, over-expression of mitotic molecules, such as cyclin B1
and Aurora kinase A, was found in human HCCs; however, no
studies analyzed differences in the expression of these proteins
in relation to possible different etiologies of HCCs [34,16,35].
We decided to study the expression and the activity of three
molecules involved in the control of mitosis: two with func-
tional roles (i.e. cyclin B1 and Aurora kinase A), and another
with structural importance (i.e. gamma-tubulin). Cyclin B1
and gamma-tubulin play different roles during mitosis: the first
regulates entry/exit from M phase while the second is impor-
tant for centrosome maturation. Aurora kinase A is a common
partner for these two proteins: it physically interacts with
cyclin B1 enhancing its stability and disrupting cytokinesis, fur-
thermore it recruits gamma-tubulin, and other centrosomal
proteins to promote centrosome maturation and microtubule
nucleation ability [36,37].
Our results provide novel evidence that the expression of all
HCV proteins, alone and even more in association with ethanol,
may induce mitotic defects in HCC cells. Accordingly, both
HCV- and ethanol-related HCCs, analyzed by us, are characterized
by a deregulation of cyclin B1, Aurora kinase A, and tyrosine-
phosphorylation of gamma-tubulin. The finding that either
HCV or ethanol may enhance the expression of some mitotic
molecules is interesting, but even more fascinating is their effect
on tyrosine-phosphorylation of gamma-tubulin.
Table 2. Clinicopathological findings and quantitative values obtained from densitometric analysis of proteins analyzed in 30 liver samples.
⁄p <0.001, ⁄⁄p <0.01, ⁄⁄⁄p <0.05 vs. control samples without tumor. HCC: HCC tissue; sHCC surrounding HCC tissue.
Research Article
960 Journal of Hepatology 2011 vol. 54 j 956–963

6. Taken together, our findings may imply that not only pro-
found changes in gene expression but also early phosphorylation
events may contribute to hepatocyte transformation opening a
new attractive research field in molecular hepatocarcinogenesis,
which surely requires further investigations. Furthermore, the
study of a larger sample of tissues, including HCC and other
sources, may not only strengthen our results but also identify
some possible correlations between the expression of these pro-
teins and the degree of tumor transformation.
Moreover, here, we reinforce our previous findings demon-
strating that HCV proteins modulate mitotic molecules via PKR
[18]. In particular, here we find that the HCV-dependent altera-
tions of cyclin B1, Aurora kinase A, and tyrosine-phosphorylation
of gamma-tubulin are mediated by a mechanism strongly
dependent on PKR, as well as on p38MAPK and JNK pathways.
Interestingly, PKR is a relevant mediator of p38MAPK and JNK
activity in several conditions. Goh et al. demonstrated that PKR
mediates the activation of p38MAPK and JNK by specific proin-
flammatory stress stimuli, such as: interleukin-1beta, lypopoly-
saccharide, TNF (tumor necrosis factor)-alpha, etc. [38]. In
mouse fibroblasts, PKR influences TNF-alpha signaling positively
C
C + siPKR
HCV
EtOH
HCV + siPKR
EtOH + siPKR
**
*
*****
*
PKR
Aurora A
β-Actin
Cyclin B1 Aurora A
Foldchanges
0.0
1.0
0.5
2.5
2.0
1.5
3.0
3.5
4.0
4.5
A
B
C
Cyclin B1
C HCV Et-OH
C HCV Et-OH
+
+
-
- -
- +
+
-
- -
-+
+
-
- -
-GFP siRNA
PKR siRNA
+
+
-
- -
- +
+
-
- -
-+
+
-
- -
-GFP siRNA
PKR siRNA
IP: P-Tyr
P
S
P-γ-Tubulin
P-γ-Tubulin
β-Actin
Fig. 4. Role of PKR in the HCV-related mitotic effects. (A) Protein expression
levels of PKR, cyclin B1 and Aurora kinase A, analyzed 24 h after release from G2/
M blocking, in control and HCV polyclones exposed or not to the ethanol,
transfected or not transfected with siPKR (10 nM) or with siGFP (10 nM) as the
control. Beta-actin is present as the control of equal protein loading. (B) Tyrosine-
phosphorylation levels of gamma-tubulin 24 h after release from G2/M blocking,
in the control and HCV polyclones exposed or not to ethanol, and transfected or
not with siPKR (10 nM) or siGFP (10 nM) as the control. Panels P (upper) show the
pellets containing tyrosine-phosphorylated form of gamma-tubulin; while panels
S (lower) show the beta-actin levels in supernatants as loading controls.
Immunoblots are representative of at least three independent experiments. (C)
Densitometric analysis reported as fold changes in protein levels ± SD (bars)
respect to the control considered as 1 after normalization against beta-actin.
⁄p <0.001 versus control polyclones.
** *
*
*
Cyclin B1 Aurora A
Foldchanges
0
1
2
3 C
C + Emodin
EtOH
EtOH + Emodin
Aurora A
β-Actin
A
B
C
E
D
Cyclin B1
Aurora A
β-Actin
Cyclin B1
C HCV Et-OH
C HCV Et-OH
+
+
-
- -
- +
+
-
- -
-+
+
-
- -
-SP
SB
+
+
-
- -
- +
+
-
- -
-+
+
-
- -
-SP
SB
IP: P-Tyr
P-γ-Tubulin
β-Actin
P-γ-Tubulin
P-γ-Tubulin
β-Actin
P
S
P
S
C EtOH
C EtOH
- + - +Emodin
- + - +Emodin
Fig. 5. Role of Emodin in ethanol-related mitotic effects. (A) Protein expression
levels of cyclin B1 and Aurora kinase A observed 24 h after release from G2/M
blocking, in HCV polyclones and in control polyclones exposed or not to ethanol,
treated or not with SP600125 (10 M) and SB203580 (10 M). (B) Protein expression
levels of tyrosine-phosphorylated gamma-tubulin, observed 24 h after release
from G2/M blocking, in HCV polyclones and in control polyclones exposed or not
to ethanol, treated or not with SP600125 (10 M) and SB203580 (10 M). (C) Protein
expression levels of tyrosine-phosphorylated gamma-tubulin, 24 h after release
from G2/M blocking, in control polyclones exposed or not to ethanol, and treated
or not with Emodin (40 M). (D) Protein expression levels of Aurora A and cyclin
B1, in the control polyclones exposed or not to ethanol and treated or not with
Emodin. Beta-actin is reported as the control of equal loading. Immunoblots are
representative of at least three independent experiments. (E) Densitometric
analysis reported as fold changes in protein levels compared to the control
considered as 1 after normalization against beta-actin. ⁄p <0.001 versus control
polyclones.
JOURNAL OF HEPATOLOGY
Journal of Hepatology 2011 vol. 54 j 956–963 961

7. regulating JNK and negatively regulating p38MAPK [39]. More
recently, it has been reported that the depletion of PKR impairs
p38 and JNK phosphorylation induced by either the E3L deletion
mutant of vaccinia virus or double-stranded RNA [40]. Based on
our previous data demonstrating that HCV core expression leads
to deregulation of the mitotic checkpoint via a p38/PKR-depen-
dent pathway [19], we hypothesize that also JNK might be a
downstream effector of PKR in the HCV-dependent mitotic
effects. On the other hand, our results demonstrated that ethanol
treatment modifies the expression of the same mitotic molecules
targeted by HCV virus, but in a PKR, JNK, and p38MAPK-indepen-
dent way. These data suggest that other signaling molecules may
be involved in the ethanol-dependent mitotic effect. Interest-
ingly, Emodin a novel anticancer drug that interferes with the
activity of multiple signaling pathways including NF-kB, FAK,
and PI3K, completely reverts the ethanol-associated over-expres-
sion and up-regulation of the tyrosine-phosphorylated form of
gamma-tubulin [30–32]. Further investigations are required to
analyze molecular pathways involved in the ethanol-dependent
effects on cyclin B1 and Aurora kinase A. In addition, Emodin,
which has a well documented hepatoprotective effect [41,42],
has been recently reported as capable to inhibit hepatoma cell
growth affecting genes potentially associated with liver tumor
progression, including cyclins [43]. These findings make this nat-
ural agent a potential candidate to improve hepatocarcinoma
treatment.
In conclusion, our study demonstrates, for the first time, that
HCV proteins and alcohol synergistically alter the mitotic appara-
tus, using different intracellular pathways; furthermore, we have
identified new molecular mechanisms associated with HCV- and
alcohol-dependent mitotic abnormalities. Our findings provide
important new insights into HCV- and alcohol-associated hepato-
carcinogenesis furnishing a good starting point to develop inno-
vative combined therapeutic strategies [44,45]. However,
specific multiple effectors and downstream signal cascades have
to be deeply investigated and the possible correlations among
these signaling molecules and the stage and grade of HCC are still
unclear.
Financial support
For this work Dr. Anna Alisi was supported by a fellowship from
Italian Association for the Study of the Liver: AISF.
Conflict of interest
The authors who have taken part in this study declared that they
do not have anything to disclose regarding funding or conflict of
interest with respect to this manuscript.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jhep.2010.08.016.
References
[1] Motola-Kuba D, Zamora-Valdes D, Uribe M, Mendez-Sanchez N. Hepatocel-
lular carcinoma. An overview. Ann Hepatol 2006;5:16–24.
[2] Morgan TR, Mandayam S, Jamal MM. Alcohol and hepatocellular carcinoma.
Gastroenterology 2004;127:87–96.
[3] Starley BQ, Calcagno CJ, Harrison SA. Nonalcoholic fatty liver disease and
hepatocellular carcinoma: a weighty connection. Hepatology 2010;51:
1820–1832.
[4] Tan A, Yeh SH, Liu CJ, Cheung C, Chen PJ. Viral hepatocarcinogenesis: from
infection to cancer. Liver Int 2008;28:175–188.
[5] Yu MC, Yuan JM, Lu SC. Alcohol, cofactors and the genetics of hepatocellular
carcinoma. J Gastroenterol Hepatol 2008;23:92–97.
[6] Safdar K, Schiff ER. Alcohol and hepatitis C. Semin Liver Dis 2004;24:
305–315.
[7] Bellentani S, Pozzato G, Saccoccio G, Crovatto M, Crocè LS, Mazzoran L, et al.
Clinical course and risk factors of hepatitis C virus related liver disease in the
general population report from the Dionysos study. Gut 1999;44:874–880.
[8] Hassan MM, Hwang LY, Hatten CJ, Swaim M, Li D, Abbruzzese JL, et al. Risk
factors for hepatocellular carcinoma synergism of alcohol with viral hepatitis
and diabetes mellitus. Hepatology 2002;36:1206–1213.
[9] Wilkens L, Flemming P, Gebel M, Bleck J, Terkamp C, Wingen L, et al.
Induction of aneuploidy by increasing chromosomal instability during
dedifferentiation of hepatocellular carcinoma. Proc Natl Acad Sci USA
2004;101:1309–1314.
[10] Plentz RR, Schlegelberger B, Flemming P, Gebel M, Kreipe H, Manns MP, et al.
Telomere shortening correlates with increasing aneuploidy of chromosome
8 in human hepatocellular carcinoma. Hepatology 2005;42:522–526.
[11] Teoh NC, Dan YY, Swisshelm K, Lehman S, Wright JH, Haque J, et al. Defective
DNA strand break repair causes chromosomal instability and accelerates
liver carcinogenesis in mice. Hepatology 2008;47:2078–2088.
[12] Bharadwaj R, Yu H. The spindle checkpoint, aneuploidy, and cancer.
Oncogene 2004;23:2016–2027.
[13] Löbrich M, Jeggo PA. The impact of a negligent G2/M checkpoint on genomic
instability and cancer induction. Nat Rev Cancer 2007;7:861–869.
[14] Fourest-Lieuvin A, Peris L, Gache V, Garcia-Saez I, Juillan-Binard C, Lantez V,
et al. Microtubule regulation in mitosis: tubulin phosphorylation by the
cyclin-dependent kinase Cdk1. Mol Biol Cell 2006;17:1041–1050.
[15] Ito Y, Takeda T, Sakon M, Monden M, Tsujimoto M, Matsuura N. Expression
and prognostic role of cyclin-dependent kinase 1 (cdc2) in hepatocellular
carcinoma. Oncology 2000;59 (1):68–74.
[16] Jeng YM, Peng SY, Lin CY, Hsu HC. Overexpression and amplification of
Aurora-A in hepatocellular carcinoma. Clin Cancer Res 2004;10:2065–2071.
[17] Miller LM, Menthena A, Chatterjee C, Verdier-Pinard P, Novikoff PM, Horwitz
SB, et al. Increased levels of a unique post-translationally modified betaIVb-
tubulin isotype in liver cancer. Biochemistry 2008;47:7572–7582.
[18] Alisi A, Mele R, Spaziani A, Tavolaro S, Palescandolo E, Balsano C. Thr 446
phosphorylation of PKR by HCV core protein deregulates G2/M phase in HCC
cells. J Cell Physiol 2005;205:25–30.
[19] Spaziani A, Alisi A, Sanna D, Balsano C. Role of p38 MAPK and RNA-
dependent protein kinase (PKR) in Hepatitis C virus core-dependent nuclear
delocalization of cyclin B1. J Biol Chem 2006;281:10983–10989.
[20] Clemens DL, Calisto LE, Sorrell MF, Tuma DJ. Ethanol metabolism results in a
G2/M cell-cycle arrest in recombinant Hep G2 cells. Hepatology
2003;38:385–393.
[21] Stickel F, Schuppan D, Hahn EG, Seitz HK. Cocarcinogenic effects of alcohol in
hepatocarcinogenesis. Gut 2002;51:132–139.
[22] Dey A, Cederbaum AI. Alcohol and oxidative liver injury. Hepatology
2006;43:63–64.
[23] Clemens DL. Effects of ethanol on hepatic cellular replication and cell cycle
progression. World J Gastroenterol 2007;13:4955–4959.
[24] Balsano C, Alisi A. HCV-related transformation and new therapeutic
strategies: an update. Curr Cancer Ther Rev 2006;2:41–56.
[25] Bedogni G, Miglioli L, Masutti F, Ferri S, Castiglione A, Lenzi M, et al. Natural
course of chronic HCV and HBV infection and role of alcohol in the general
population: the Dionysos Study. Am J Gastroenterol 2008;103:2248–2253.
[26] Barr AR, Gergely G. Aurora-A: the maker and breaker of spindle poles. J Cell
Sci 2007;120:2987–2996.
[27] Hirota T, Kunitoku N, Sasayama T, Marumoto T, Zhang D, Nitta M, et al.
Aurora-A and an interacting activator, the LIM protein Ajuba, are required
for mitotic commitment in human cells. Cell 2003;114:585–598.
[28] Blindenbacher A, Duong FH, Hunziker L, Stutvoet ST, Wang X, Terracciano L,
et al. Expression of hepatitis c virus proteins inhibits interferon alpha
signaling in the liver of transgenic mice. Gastroenterology 2003;124:
1465–1475.
[29] Rakotomalala L, Studach L, Wang WH, Gregori G, Hullinger RL, Andrisani O.
Hepatitis B virus X protein increases the Cdt1-to-geminin ratio inducing
DNA re-replication and polyploidy. J Biol Chem 2008;283:28729–28740.
Research Article
962 Journal of Hepatology 2011 vol. 54 j 956–963

8. [30] Li HL, Chen HL, Li H, Zhang KL, Chen XY, Wang XW, et al. Regulatory effects
of emodin on NF-kappaB activation and inflammatory cytokine expression
in RAW 264.7 macrophages. Int J Mol Med 2005;16:41–47.
[31] Huang Q, Shen HM, Shui G, Wenk MR, Ong CN. Emodin inhibits tumor cell
adhesion through disruption of the membrane lipid Raft-associated integrin
signaling pathway. Cancer Res 2006;66:5807–5815.
[32] Olsen BB, Bjørling-Poulsen M, Guerra B. Emodin negatively affects the
phosphoinositide 3-kinase/AKT signalling pathway: a study on its mecha-
nism of action. Int J Biochem Cell Biol 2007;39:227–237.
[33] Koike K, Tsutsumi T, Miyoshi H, Shinzawa S, Shintani Y, Fujie H, et al.
Molecular basis for the synergy between alcohol and hepatitis C virus in
hepatocarcinogenesis. J Gastroenterol Hepatol 2008;23:87–91.
[34] Ito Y, Takeda T, Sakon M, Monden M, Tsujimoto M, Matsuura N. Expression
and prognostic role of cyclin-dependent kinase 1 (cdc2) in hepatocellular
carcinoma. Oncology 2000;59:68–74.
[35] Llovet JM, Chen Y, Wurmbach E, Roayaie S, Fiel MI, Schwartz M, et al. A
molecular signature to discriminate dysplastic nodules from early hepato-
cellular carcinoma in HCV cirrhosis. Gastroenterology 2006;131:1758–1767.
[36] Qin L, Tong T, Song Y, Xue L, Fan F, Zhan Q. Aurora-A interacts with cyclin B1
and enhances its stability. Cancer Lett 2009;275:77–85.
[37] Fu J, Bian M, Jiang Q, Zhang C. Roles of aurora kinases in mitosis and
tumorigenesis. Mol Cancer Res 2007;5:1–10.
[38] Goh KC, deVeer MJ, Williams BR. The protein kinase PKR is required for p38
MAPK activation and the innate immune response to bacterial endotoxin.
EMBO J 2000;19:4292–4297.
[39] Takada Y, Ichikawa H, Pataer A, Swisher S, Aggarwal BB. Genetic deletion of
PKR abrogates TNF-induced activation of IkappaBalpha kinase, JNK, Akt and
cell proliferation but potentiates p44/p42 MAPK and p38 MAPK activation.
Oncogene 2007;26:1201–1212.
[40] Zhang P, Langland JO, Jacobs BL, Samuel CE. Protein kinase PKR-dependent
activation of mitogen-activated protein kinases occurs through mitochon-
drial adapter IPS-1 and is antagonized by vaccinia virus E3L. J Virol
2009;83:5718–5725.
[41] Bhadauria M. Dose-dependent hepatoprotective effect of emodin against
acetaminophen-induced acute damage in rats. Exp Toxicol Pathol 2010;62:
627–635.
[42] Dong MX, Jia Y, Zhang YB, Li CC, Geng YT, Zhou L, et al. Emodin protects rat
liver from CCl(4)-induced fibrogenesis via inhibition of hepatic stellate cells
activation. World J Gastroenterol 2009;15:4753–4762.
[43] Hsu CM, Hsu YA, Tsai Y, Shieh FK, Huang SH, Wan L, et al. Emodin inhibits
the growth of hepatoma cells: finding the common anti-cancer pathway
using Huh7, Hep3B, and HepG2 cells. Biochem Biophys Res Commun
2010;392:473–478.
[44] Tanaka S, Arii S. Molecularly targeted therapy for hepatocellular carcinoma.
Cancer Sci 2009;100:1–8.
[45] Warner SL, Gray PJ, Von Hoff DD. Tubulin-associated drug targets: Aurora
kinases, Polo-like kinases, and others. Semin Oncol 2006;33:436–448.
JOURNAL OF HEPATOLOGY
Journal of Hepatology 2011 vol. 54 j 956–963 963