Dynamic alteration of protein expression profiles
during neoplastic transformation of rat hepatic
Xuefei Li,1,3Yan Li,1,* Xiaonan Kang,2Kun Guo,1Haiyu Li,2Dongmei Gao,1Lu Sun2and Yinkun Liu1,2,4
1Liver Cancer Institute, Affiliated Zhongshan Hospital of Fudan University, Shanghai;2Research Center for Cancer, Institute of Biomedical Science of Fudan
University, Shanghai;3Central Laboratory, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai, China
(Received October 9, 2009 ⁄Revised January 17, 2010 ⁄ Accepted January 19, 2010 ⁄ Online publication March 15, 2010)
To explore the molecular basis of neoplastic transformation of
hepatic oval cells, a proteomic strategy was utilized to examine
the global protein expression alterations during neoplastic trans-
formation of rat hepatic oval-like cells. N-methyl-N¢-nitro-N-nitros-
oguanidine (MNNG)-initiated WB-F344 cells were treated with
H2O2for neoplastic transformation. The transformed cells were
identified by soft agar assay and MTT assay. The subsequent pro-
teomic separation and identification were performed with 2-DE
followed by MALDI-TOF-MS⁄⁄MS analysis. Of the 148 differentially
displayed protein spots analyzed, 121 spots representing 79
distinct proteins were finally identified. The expression levels of
interested proteins were validated by western blotting including
40 S ribosomal protein A (RPSA) and cytokeratin 8. Bioinformatics
annotations indicated that these identified proteins were enriched
with oxidoreduction and stress response; transcription, transla-
tion, and protein processing; and energy⁄⁄metabolism functions.
Interestingly, 17 of the identified proteins were also found to be
involved in early hepatic differentiation of mouse embryonic stem
(ES) cells in our previous study. Twenty-six proteins had been
reported as being dysregulated in hepatocellular carcinoma and
other cancers. It suggested that these changed proteins may be
implicated in neoplastic transformation of WB-F344 cells. The
results may provide some clues for understanding the molecular
mechanisms of hepatocarcinogenesis as viewed from dysregula-
tion of differentiation. (Cancer Sci 2010; 101: 1099–1107)
lular origin of HCC remains unclear. Recently it is well
accepted that HCC is a disease of the stem cells.(2–4)Hepatic
oval cells⁄hepatic progenitor cells (HOCs⁄HPCs), which are
hepatic stem cells located in the smaller interlobular bile ducts
and canals of Hering, are bipotential and able to differentiate
into at least hepatocytes and biliary epithelial cells. They
express not only hepatocytic markers but also biliary markers.
They have been found to be activated and proliferated in patho-
logical processes of many liver diseases, including chronic viral
hepatitis, hepatic cirrhosis, and liver regeneration, as well as
HCC.(5)Moreover, the magnitude of HOCs⁄HPCs activation
corresponds to the severity of liver fibrosis and inflammation.(6)
Many animal models of hepatocarcinogenesis have also been
characterized by a striking proliferation of oval cells.(3)It is
known that cell proliferation during carcinogen exposure is a
prerequisite to ‘‘fix’’ any genotypic injury into a heritable
form.(7)Therefore, HOCs⁄HPCs are probably one of the cellular
origins of hepatocarcinoma. This raises some questions: how do
cancers arise from HOCs? What molecules are responsible for
To elucidate the molecular mechanisms underlying the above
carcinogenic event, it is helpful to investigate the molecular
epatocellular carcinoma (HCC) is one of the most preva-
lent life-threatening human cancers.(1)However, the cel-
changes during the process. Most of the relevant studies gain
information by comparatively analyzing the cancer tissues⁄cells
with normal ones, which have different genetic backgrounds. To
bypass the disadvantage, hepatic oval-like cells were trans-
formed by continuous carcinogen exposure in this study.
The global molecular changes during this process were then
examined using powerful proteomic techniques, which can study
thousands of proteins simultaneously. Moreover, proteome
alterations were analyzed at two different time points. This
dynamic analysis may contribute to gaining more insight into
the molecular changes, thereby furthering our understanding
of hepatocarcinogenesis. Herein, we used a typical proteomic
approach (2-DE followed by MALDI-TOF-MS⁄MS). Seventy-
nine distinct proteins were identified. Some of them were
previously reported to be involved in both early hepatic differen-
tiation of embryonic stem cells(8)and HCC development.(9,10)
These identified proteins may be early hepatocarcinogenesis-
associated proteins. The data provided some clues for further
study of the molecular mechanisms of hepatocarcinogenesis as
viewed from dysregulation of differentiation.
Materials and Methods
Cellculture andidentification. Diploid WB-F344 cells
(abbreviated to WB cells) are a rat non-tumorigenic epithelial
cell line gifted generously by Dr. W.B. Coleman (University of
North Carolina, Chapel Hill, NC, USA). The cells were cultured
in DMEM⁄F12 medium (Gibco, Grand Island, NY, USA) con-
taining 10% FBS (Gibco) and 100 U⁄mL penicillin (Gibco) and
100 U⁄mL streptomycin (Gibco). WB cells were passaged at
1:3 dilution every 3 days.
Before neoplastic transformation, some molecular markers of
oval cells(11,12)were detected by immunocytochemistry and RT-
PCR to confirm that WB cells maintain their original phenotype
of oval cells.
Immunocytochemistry. WB cells were washed twice with
PBS and fixed in 100% methanol at )20?C for 5 min. The fixed
cells were incubated with blocking buffer (5% BSA⁄PBS) for
30 min at room temperature, then with the primary antibodies
(mouse anti- cytokeratin 7 [CK7] antibody, mouse anti-connexin
43 [Cx43] antibody; Abcam, Cambridge, UK) overnight at 4?C,
and then with the secondary antibodies for 1 h at room tempera-
ture. The cells were counterstained with DAPI (Invitrogen,
Carlsbad, CA, USA) for nuclear staining. Cells were visualized
by fluorescence microscopy (BX-60; Olympus, Tatsuno-machi,
(between passages 8–10) were passaged at a density of
4To whom correspondence should be addressed.
ª ª 2010 Japanese Cancer Association
|vol. 101|no. 5|
1.5 · 105cell⁄well on six-well plates. After the cells increased
to 80% confluence, they were treated for 24 h with the initiating
concentration 5 lg⁄mL; Tokyo Kasei Kogyo, Tokyo, Japan)(13)
or with solvent control, PBS (pH 7.4). Twenty-four hours later,
the cells were washed twice with serum-free medium and then
exposed to H2O2(final concentration 100 lM⁄mL; Sigma, St.
Louis, MO, USA) once a week.(14,15)Cells were harvested from
1.5 · 105⁄well every week. This process was repeated up to 21
RT-PCR. To ensure that WB cells maintained phenotypic
properties of oval cells, we detected some markers of oval cells
before neoplastic transformation, including alpha-fetoprotein
(AFP), albumin (ALB), glucose-6-phosphatase (G6P), gamma-
(TAT), and Yp. Total RNA was extracted using Trizol Reagent
(Invitrogen) and was reverse-transcribed into complementary
DNA with oligo (dT)18 primer using a RevertAid first strand
cDNA synthesis kit (Fermentas China Co., Shenzhen, China)
according to the manufacturer’s instructions. PCR was per-
formed using Taq polymerase (MBI Fermentas). The primer
sequences, annealing temperature, number of cycles used for
PCR, and length of the amplified products can be seen Table 1.
In addition, we also detected some hepatocytic markers during
neoplastic transformation, including AFP, CYP1a1, G6P, TAT,
and ALB. The PCR products were separated by electrophoresis
on 1.0–1.5% agarose gels and stained with ethidium bromide.
GAPDH was used as the endogenous control. Each sample was
carried out in triplicate and run in a DNA Thermal Cycler (Per-
kin Elmer, MA, USA).
Soft agar assay. Cells exposed to H2O2at weeks 3, 5, 7, 11,
and 21 (abbreviated to WB3, 5, 7, 11, 21 cells, respectively)
were used to test anchoring-independent growth ability. The
assay was performed according to the method described by
Qi.(16)Herein, 2 · 104cell density was used. Growth of cells
was maintained by weekly overlaying 250 lL of DMEM⁄F12
with 10% FBS. Colony formation and enumeration were evalu-
ated under a microscope (Leica, NY, USA) after 14–20 days’
incubation. The colonies containing 50 cells were scored as
positive, and colony forming efficiency was counted. (Colony
forming efficiency = the number of colony⁄the number of inoc-
MTT assay. MTT assay was performed to determine cell
proliferation ability of WB, WB5, and WB21. Cells were inocu-
lated in a 96-well plate at a density of 5 · 103⁄well. Five double
wells and blank wells were arranged. These cells were cultured
for 48 h. Then, pre-prepared 10 lL MTT (final concentration,
5 mg⁄mL) was added and cells were placed in an incubator.
After 4 h, 100 lL DMSO was added and cells were vibrated for
30 min. Finally, the optical absorptions of each well in 570 nm
were obtained by using an enzyme labeling instrument (Multis-
kan Spectrum; Thermo Electron, MA, USA).
Limiting dilution assay. Limiting dilution assay was per-
formed to select the colonies with high proliferative capacity.
WB5 and WB21 cells were inoculated in 96-well plates. Each
well contained a single cell. The 96-well plates were placed in
an incubator. Every three days, 50% of the old medium was
replaced with fresh medium. When colonies formed and prolif-
erated, each of these cell populations was analyzed by soft agar
assay, to gain the colonies with the strongest anchoring-indepen-
dent growth ability.
2-DE and image analysis. After washing cells three times with
cold PBS, 1 mL lysis buffer (40 mM Tris, 7 M urea, 2 M thio-
urea, 4% CHAPS, 65 mM DTT, 1 mM PMSF, and 2% pharma-
lyte [pH 3–10]) was added to 1.5 · 107cells to extract the total
proteins by homogenization. The supernatant was collected by
centrifugation (20 000g for 30 min at 4?C and stored at )80?C
or further use. The protein concentration was determined by 2-D
Quant Kit (Amersham Biosciences, Piscataway, NJ, USA). 2-
DE was performed using the Amersham Biosciences 2-DE sys-
tem according to the manufacturer’s instructions.(17)Briefly,
400 lg of total protein was firstly isoelectrically focused on 13-
cm pH 3–10 NL IPG strips (Amersham Pharmacia, Uppsala,
Sweden) by an Ettan IPGphor system (Amersham Pharmacia)
for a total of 50 000 Vh. After equilibration twice, the second-
dimension separation was carried out using 12.5% SDS-PAGE
gels at a constant current of 7.5 mA⁄gel for 30 min and then
15 mA⁄gel till the Bromphenol blue front reached the bottom of
the gels. For each cell sample, triplicate 2-DE gels were run
under the same conditions. Proteins on gels were stained with
coomassie brilliant blue (CBB) G-250. The stained gels were
imaged using an ImageScanner (Amersham Biosciences) and
then analyzed with ImageMaster 2D Platinum 5.0 software
(Amersham Biosciences). Protein spots that were determined to
be differentially displayed (difference fold more than 2.0 or less
than 0.5, P < 0.05 with Student’s t-test) automatically by the
software were evaluated manually by local comparison to elimi-
Protein identification by mass spectrum analysis. The differ-
entially displayed spots were excised from the gels, destained
with 50 mM NH4HCO3and 50% acetonitrile (ACN) twice, and
Table 1. RT-PCR primer sequences and reaction conditions
Gene Primer sequences (5¢ fi 3¢)Annealing (ºC)CyclesSize (bp)
Albumin (ALB)58 35442
Glucose-6-phosphatase (G6Pase)58 35 489
Tyrosine-aminotransferase (TAT)5635 471
Cytochrome P450, family 1, subfamily A,
polypeptide 1 (CYP1a1)
ª ª 2010 Japanese Cancer Association
then dried with ACN twice. The dried gel pieces were digested
overnight at37?C with12.5 ng⁄mL
NH4HCO3. The peptides were extracted twice with 50%
ACN⁄0.1% TFA for 20 min each, and dried with N2. The dried
peptides were mixed with matrix of 5 mg⁄mL a-cyano-4-
hydroxycinnamic acid (Sigma) in 50% ACN⁄0.1% TFA, spotted
onto a MALDI target, then analyzed by a 4700 Proteomics
Analyzer (Applied Biosystems, Foster City, CA, USA). Trypsin-
digested peptides of horse myoglobin were used as a mass stan-
dard to calibrate the instrument, and then default calibration was
applied on the sample peptides. After MS acquisition, four
strongest peptides per spot were selected automatically for the
MS⁄MS analysis. Spectra were searched against the National
Center for Biotechnology Information non-redundant 20070627
database and the International Protein Index rat-3.35 database
using the search engine Mascot (Matrix Science, London, UK).
The mass tolerance was set as 0.3 Da, and MS⁄MS tolerance
was 0.4 Da. The automatic data analysis and database searching
were performed with the GPS Explore software (Applied Bio-
Western blotting. Equal amounts of total proteins (30 lg)
were separated by 10% SDS-PAGE and transferred to a 0.45-
mm PVDF membrane (Millipore, Bedford, MA, USA) using a
Bio-Rad SemiDry apparatus. The membrane was incubated with
trypsin in 25 mM
the blocking buffer (5% non-fat milk in TBS⁄0.5% Tween-20)
for 1 h at room temperature, then with mouse anti-cytokeratin 8
(1:500 dilution; BD Biosciences), rabbit anti-40 S ribosomal
protein A (anti-RPSA) (1:200 dilution; Santa Cruz Biotechnol-
ogy, Santa Cruz, CA, USA), or mouse anti-GAPDH (1:5000
dilution; Kang-Chen, Shanghai, China) overnight at 4?C, fol-
lowed by HRP-conjugated secondary antibodies for 1 h at room
temperature. After washing three times in TBST, the membrane
was incubated with Chemiluminescent Detection Reagent (Tian-
Gen, Beijing, China) for 2 min and exposed to X-ray film.
Characteristics of transformed cells. Rat WB-F344 cells were
isolated from the liver of an adult male Fisher-344 rat. Their
phenotypic properties in culture most resembled the ‘‘oval’’
cells.(18)Before neoplastic transformation, several biomarkers of
oval cells were detected including AFP, ALB, G6P, GGT IV,
TAT, Yp, CK7, and Cx43 (Fig. 1). The result displayed that
WB cells still maintained the phenotypic properties of oval cells.
At the second week after H2O2 treatment, transforming foci
were observed via phase-contrast microscopy in a multilayer,
overlapping, and confused arrangement (Fig. 2a-1). The level of
AFP expression did not clearly alter and CYC1a1 expression
(a2) (c1) (c2)
The transforming foci at the second week after
microscopy; (a2) untreated WB cells. (b) RT-PCR
analysis of RNA samples extracted from untreated
WB cells, transformed cells at weeks 3, 7, 11, and 21
(G6P), and albumin (ALB). (c1–4) Soft agar assay.
(c1) WB cells could not grow in soft agar; (c2,3)
transformed cells at weeks 5 (WB5) and 21 (WB21);
purified-WB21. (d) MTT assay. The proliferation
ability of transformed cells significantly increased
compared with WB cells (*P < 0.001), and WB21
cells also proliferated
(4P < 0.001). Magnifications (a) ·100, (c) ·200.
Neoplastic transformation of WB cells. (a1)
faster thanWB5 cells
hepatic oval cells. (a) RT-PCR analysis of RNA
analysis of the expression of cytokeratin 7 (CK7)
and connexin 43 (Cx43). (b1) Green fluorescence
denotes CK 7 expression; (b2) negative control
withoutaddition of primary
fluorescence displays Cx43 expression; (b4) negative
control without addition of primary antibodies,
blue denotes nuclear
Comparison of biomarkers in WB-F344 and oval
cells according to references. Magnification (b),
Identificationof severalbiomarkers of
WB cells. (b1–b4)ICC
by (b3) Green
Li et al. Cancer Sci|
| vol. 101| no. 5|
ª ª 2010 Japanese Cancer Association
increased at a different time point of neoplastic transformation.
G6P was expressed in WB21 and TAT was expressed in WB,
WB7, WB11 and WB21, but ALB was not always expressed
during the progression of transformation (Fig. 2b). Remarkably,
anchoring-independent growth was the important characteristic
of transformed cells. The transformed cells could form colonies
in soft agar, and the frequencies of colony formation were 0,
0.1%, 0.3%, 0.8%, and 2% in WB3, WB5, WB7, WB11, and
WB21 cells, respectively, but WB cells could not grow in soft
agar (Fig. 2c-1–3). MTT assay displayed that transformed cells
proliferated faster than WB cells (Fig. 2d). These results
strongly indicated that by MNNG initiation and induction with
H2O2, WB cells gained the characteristics of transformed cells.
Purification of transformed cells. For improving the ratio of
transformed cells, WB5 and WB21 cells were firstly cloned by
limiting dilution assay to select the colonies with high prolifera-
tive capacity. Eight and 36 colonies were gained in WB5 and
WB21 cells, respectively. Each of these cell populations was
further analyzed by soft agar assay. The two of colonies of WB5
and WB21 with the strongest anchoring-independent growth
ability were chosen for further culturing for subsequent prote-
ome analysis (their frequency of colony formation in soft agar
as 1.5% and 7%, respectively). Figure 2(c-4) shows the colony
of purified-WB21 cells.
formation. Proteome expression alteration during neoplastic
transformation of rat hepatic oval-like cells was investigated.
Total proteins, extracted from WB cells (non-treated), WB5
(treated with H2O2for 5 weeks, earliest anchoring-independent
growth, purified by limited dilution and soft agar assay) and
WB21 (treated with H2O2for 21 weeks, strongest anchoring-
independent growth ability, purified by limited dilution and soft
agar assay) were separated by 2-DE. About 1100–1500 protein
spots were resolved in one gel by CBB staining. For each cell
sample, we integrated triplicate 2-DE gels for image analysis.
Figure 3 exhibits the representative 2-DE maps of the three cell
The subsequent difference analysis, spot cut and MS⁄MS
identification were all carried out in three subdivided compara-
tive groups: group A compared WB5 with WB (WB–WB5), B
compared WB21 with WB (WB–WB21), and C compared
WB21 with WB5 (WB5–WB21). We searched spectra data
against both the National Center for Biotechnology Information
and the International Protein Index database, and got the same
positive protein identifications with high confidence (protein
scores greater than 60 and at least two matched peptides
sequenced with good MS⁄MS spectra) (Table S1). In group A,
of the 41 picked spots, 30 positive identifications representing
28 proteins were identified. In group B, of the 73 selected spots,
61 identifications representing 43 proteins were determined. In
group C, of the 34 analyzed spots, 30 identifications represent-
ing 28 proteins were obtained. Taken together, of the 148 differ-
representing 79 distinct proteins. Among the identified proteins,
cytokeratin-8 and RPSA were randomly selected for western
blot analysis to validate the reliability of proteomic analysis.
Their expression patterns were shown to be consistent with
those in 2-DE gels (Fig. 4).
In all identified proteins, 17 proteins were found in more than
one of the comparative pairs (Table 2). Their expression
displayed eight different expression patterns. For example,
heterogeneous nuclear ribonucleo-protein (hnRNP) A3 down-
regulated firstly, then up-regulated; the expression of cytoplasmic
cells at weeks 5 (WB5) and weeks 21 (WB21). Four hundred
microgram of total protein was subjected to the 2-DE system (13 cm,
pH 3–10, NL IPG strip; 12.5% SDS-PAGE). Proteins were visualized by
coomassie brilliant blue (CBB) staining. About 1300 protein spots were
resolved in one gel.
Reference gels of untreated WB cells (WB) and transformed
(RPSA) and cytokeratin 8 in untreated WB cells
(WB) and transformed cells at weeks 5 (WB5) and
weeks 21 (WB21). The expression patterns analyzed
by western blotting (b) were consistent with those
in 2-DE gels (a).
Expression of 40 S ribosomal protein A
ª ª 2010 Japanese Cancer Association
actin decreased continuously, etc. This suggested that the
regulation mechanisms may be complex and elaborate in the
Bioinformatics annotation of the differential proteins. Further
database mining indicated that these identified differential
proteins could be classified into seven groups based on functions
and biological processes described in the UniProt database
(Table S1). They are oxidoreduction and stress response; tran-
scription, translation, and protein processing; cystoskeleton, cell
movement, and adhesion; energy⁄metabolism; cell proliferation
and apoptosis; signal transduction; and others. Figures 5 and 6
shows subcellular distributions and functional categories of the
identified proteins. Overall, the most common functional annota-
tions were energy⁄metabolism, oxidoreduction and stress
response, as well as transcription, translation, and protein pro-
cessing. In the three divided groups, the oxidoreduction and
stress response was an important functional annotation. The
most differential proteins comprised 24% and 26%, respectively,
and were involved in energy⁄metabolism when transformed
cells were compared with WB cells (groups A and B). 23% and
32% of identified proteins participated in transcription, transla-
tion, and protein processing in the progression of transformation
(groups A and C). In comparing WB21 with WB⁄WB5 (groups
B and C), the cystoskeleton, cell movement, and adhesion
accounted for a higher proportion in seven functional groups.
Regarding the proteins identified in group A, the subcellular dis-
tributions were enriched in the cytoplasm and nucleus; neverthe-
less, in group B, most of the identified proteins were localized in
cytoplasm, endoplasmic reticulum (ER), and mitochondria. In
addition, the identified proteins in group C mainly distributed
cytoplasm, nucleus, and cystoskeleton.
To gain more insight into the complex regulatory networks that
govern the neoplastic transformation of WB cells, interactions
between the identified proteins and the other known proteins were
mapped using interaction database (IntAct) mining and computa-
tional analysis (Pajek software, http://vlado.fmf.uni-lj.si/pub/
networks/pajek/) in three divided groups and also overall
(Fig. 7). Interestingly, glucose transporter-4 (GLUT4⁄gtr4) was
shown to interact with many identified proteins. In addition,
hnRNP K and dynein light chain 1 (dyl1) were also found to be
interact with several identified proteins. But GLUT4 and dyl1
were not the proteins identified in this study.
To understand the complicated molecular mechanisms of
hepatocarcinogenesis as viewed from dysregulation of differen-
tiation, results were compared with those from the previous
study about the differentiation of mouse ES cells along hepatic
direction.(8)Overall 17 proteins were commonly identified
(Table S2). Although some of them exhibited different expres-
sion patterns or isoforms, to some extent, they may commonly
be involved in neoplastic transformation and hepatic differentia-
tion. The result suggested that some regulatory mechanisms
may be common to both differentiation and neoplastic transfor-
mation. Meanwhile, the identified proteins were also compared
with the dysregulated proteins found in HCC and other can-
(Table S3). In these proteins as above, the expression changes
of ten proteins have been found in neoplastic transformation and
hepatic differentiation, as well as HCC development. Taken
together, those of proteins or protein families were suggested to
be candidate factors highly associated with neoplastic transfor-
mation of WB cells. They included in ubiquitin-conjugating
enzyme 2L3, fructose-bisphosphate aldolase A, heat shock preo-
tein 8, tubulin, cytokeratin 8, vimentin, ATP synthase, eukary-
otic translation elongation factor, annexin V, and heterogeneous
Studies of aggregation chimeras and X-linked polymorphisms
strongly suggest that liver tumors are derived from single
cells (monoclonal).(3)But which cell originates HCC? Many
studies suggest that bipotential HOCs⁄HPCs in adult liver
are more likely to become target cells of carcinogens than
other adult cells because they are long lived and have long-
term repopulating potential.(19,20)The oval cell activation
precedes the development of HCC in almost all animal mod-
els of hepatocarcinogenesis.(20)Furthermore, HOCs⁄HPCs are
activated and proliferated in chronic inflammation and are
possibly involved in the histogenesis of HCC.(3)Several stud-
ies have suggested that many human HCCs may originate
from HPCs, including some HCCs with features of chronic
(CHC), etc.(21–24)Therefore, HOCs⁄HPCs have been consid-
ered as one cellular source of HCCs. However, the molecular
Table 2. Expression patterns of the proteins identified in more than one of the comparative groups
Protein expression pattern in three groups
Hnrpa3, 34 kDa protein (2)
Rpsa 40S, ribosomal protein SA (3)
Sorting nexin 3 (2)
Shmt2, Serine hydroxymethyltransferase (4,5)
Similar to actin, cytoplasmic 2 (7)
Similar to pyruvate kinase 3 isoform (1–4)
Acat2, acetyl-CoA acetyltransferase cytosolic (4)
Aldh3a1, Aldehyde dehydrogenase (1,4,5)
Serpinh1, Serpin H1 (1)
Nono, non-POU domain-containing octamer-binding protein (2)
Col1a1, Col1a1 protein (1)
Ppic, Peptidyl-prolyl cis-trans isomerase (2)
Cald1, non-muscle caldesmon (3,5)
Cnn3, 36 kDa protein (3)
Hspa8, heat shock cognate 71 kDa protein (1)
Vim, vimentin (3)
Prph1, peripherin (3)
–, unchanged; ƒ, up-regulated; !, down-regulated. The bold numbers in the protein species column designate the functional grouping of the
protein and correspond to those in Table S1.
Li et al. Cancer Sci|
| vol. 101| no. 5|
ª ª 2010 Japanese Cancer Association
The WB-F344 cell line shares many characteristics with rat
liver progenitor cells(25)and is considered to be an in vitro
model of hepatic oval cells. It has been used in many studies on
of hepatocarcinogenesis from HOCs remain
differentiation, transdifferentiation, as well as HCC develop-
ment.(12,26–29)As known to all, reactive oxygen species (ROS)
plays a central role in carcinogenesis. In this study, H2O2, a
member of the ROS family, was selected to induce the neoplas-
tic transformation of WB cells. To investigate the elaborate
component and functional and biological process annotation, respectively.
Subcellular distributions and functional categories of the identified proteins in three comparative groups and overall. (a,b) Cellular
ª ª 2010 Japanese Cancer Association
molecular regulations during this process, we selected cells from
two time points according to their characteristics: WB5 were
that gained the earliest anchored-independent growth ability;
WB21 cells had the strongest anchored-independent growth
Comparative proteome analysis was accordingly divided into
three groups. For the identified proteins in subdivided groups,
the most common biological process annotation was oxidoreduc-
tion and stress response. Alterations of this sort of protein were
possibly a response to H2O2exposure. Moreover such regula-
tions might further promote the development of transformation.
With the progression of transformation (groups A and C), tran-
scription, translation, and protein processing was the significant
biological process. This kind of protein is directly or indirectly
identified proteins in this study; black dots represent those which interacted with the red ones but were not detected in this study. Lines
between dots indicate interactions.
Interactions between the identified proteins and other known proteins in three comparative groups and overall. Red dots denote the
biological process annotation in three comparative
groups and overall according to Figure 5.
Percentage distribution of functional and
Li et al. Cancer Sci|
| vol. 101|no. 5|
ª ª 2010 Japanese Cancer Association
involved in mRNA processing, translation regulation, and main-
tenance and transportation of proteins, etc. They may contribute
to the progression of transformation. The proteins involved in
energy⁄metabolism biological process changed significantly
when the transformed cells compared with WB cells (groups A
and B), such as pyruvate kinase 3, ATP6v1a1, fructose-bisphos-
phate aldolase A, etc. Transformed⁄tumor cells often display
fundamental changes in pathways of energy metabolism in order
to meet the increased requirements of proliferation.(30)For
example, the embryonic M2 isoform of pyruvate kinase
expressed in tumor cells shifts cell metabolism to aerobic glycol-
ysis and may promote tumorigenesis.(31)However, it is still
debated whether these metabolic changes are a consequence or a
cause of tumorigenesis.(30)Moreover, 22% identified proteins
belonged to the biological process of cytoskeleton, cell move-
ment, and cell adhesion in group C (WB21–WB5). This sort of
protein also comprised a large proportion of group B (WB21–
WB). This kind of protein is highly related to both anchorage-
independent growth and cellular tumorigenicity,(32)suggesting
fundamental roles for cytoskeleton proteins in the progress of
neoplastic transformation. In addition, most of recurring proteins
in the three comparative groups (see Table 2) also belonged to
three biological process annotations discussed above. So these
biological processes may have important regulation roles in the
neoplastic transformation of WB cells; moreover, several of the
identified proteins may be responsible for the transformation.
Protein–protein interactions analysis at a system-wide level is
helpful for understanding the essential mechanisms of various
pathological phenomena, including carcinogenesis. In this study,
GLUT4⁄gtr4 was not one of the identified proteins, but was
shown to interact with many of them. As we know, malignant
cells have characteristics including accelerated metabolism and
increased glucose uptake, while facilitative GLUT proteins are
important mediators of glucose transport. Activation of GLUT
genes represents one of the earliest events in oncogenesis. Fur-
thermore, some studies have demonstrated involvement of
GLUT4 in human astrocytic tumors, gastric cancer, and rhabdo-
myosarcoma, as well as in neoplastic transformation of ovarian
epithelium.(33,34)So, this suggested that GLUT 4 may be a key
regulator of the neoplastic transformation of WB cells. Further
study is needed to unveil its role.
One interesting finding of this study was that some differen-
tial proteins had been found to be involved in the hepatic differ-
entiation of mouse ES cells or the development of HCC⁄other
cancers. The data suggested that some regulatory mechanisms
may be common to normal differentiation and neoplastic trans-
formation, as well as hepatocarcinogenensis. In other words,
dysregulation of some differentiation mechanisms may lead to
hepatocarcinogenesis. The alteration of several central proteins
involved in differentiation may be responsible for transforma-
tion or hepatocarcinogenesis. In this study, 10 kinds of these
commonly identified proteins were implicated in hepatic differ-
entiation, neoplastic transformation, and tumor development.
Furthermore, they participated mostly in the regulation of three
biological processes mentioned above. This suggested that these
proteins may be key factors in the neoplastic transformation of
WB cells. Further, identifications of different isoforms of identi-
cal protein species in these studies also reflected their delicate
and complex regulatory functions in the process of both differ-
entiation and transformation⁄hepatocarcinogenensis.
Among the commonly identified proteins, hnRNPs were a
group of interesting proteins. During the neoplastic transforma-
tion of WB cells, a set of hnRNPs were identified, including
hnRNP K, A2⁄B1, A3, A1, and D. This kind of protein has cen-
tral roles in DNA repair, telomere biogenesis, cell signaling, and
in regulating gene expression at both the transcriptional and
translational levels.(35)Some of this protein family could be
classed as oncogenes. For example, the expression of hnRNP
A2⁄B1 was modulated at different stages of the cell cycle in
tumor cells, peaked at the S phase, declined in the G2 and M
stages, but was restored in late G1. Therefore, it is considered as
having oncogene characteristics.(35)In addition, p37 isoforms of
hnRNP D have been shown to be involved in tumor develop-
ment. In transgenic mouse overexpressed p37, c-myc, c-fos, and
c-jun were up-regulated and in rodents tumors were observed in
the esophagus, lung, urethra, and soft tissue of the head and
neck.(36)So, hnRNP members are possibly involved in tumori-
genesis. Whether they also play central roles in the neoplastic
transformation of rat oval-like cells remains to be investigated.
In summary, as an initial step toward elucidating the mecha-
nisms underlying the neoplastic transformation of rat hepatic
oval-like cells, this study analyzed the global protein expression
changes during this process. Our results may provide some clues
for understanding hepatocarcinogenesis as viewed from dysre-
gulation of differentiation. The proteins we have identified may
play important roles in the neoplastic transformation of WB-
F344 cells. Further detailed investigations are needed to define
their concrete roles.
This work was supported by a grant from The State 863 High Technol-
ogy R&D project of China (2006AA02A308) and China National Key
Projects for Infectious Disease (2008ZX10002-021 and 2008ZX10002-
017). We are grateful to Dr. W.B. Coleman (School of Medicine, Uni-
versity of North Carolina, NC, USA) for kindly providing the WB-F344
cell line. We thank Dr. Z. Yu (Fudan University, Shanghai, China) for
the electron microscopic analysis. We also thank Mr. X.W. Zhou and
Miss X.H. Liu (Fudan University, Shanghai, China) for MALDI-TOF
The authors have no conflict of interest.
1 Seeff LB, Hoofnagle JH. Epidemiology of hepatocellular carcinoma in areas
of low hepatitis B and hepatitis C endemicity. Oncogene 2006; 25: 3771–7.
2 Sell S. Cellular origin of cancer: dedifferentiation or stem cell maturation
arrest? Environ Health Perspect 1993; 101: 15–26.
3 Alison MR, Lovell MJ. Liver cancer: the role of stem cells. Cell Prolif 2005;
4 Wu XZ, Chen D. Origin of hepatocellular carcinoma: role of stem cells.
J Gastroenterol Hepatol 2006; 21: 1093–8.
5 Falkowski O, An HJ, Ianus IA et al. Regeneration of hepatocyte ‘buds’ in
cirrhosis from intrabiliary stem cells. J Hepatol 2003; 39: 357–64.
6 Bird TG, Lorenzini S, Forbes S. Activation of stem cells in hepatic diseases.
Cell Tissue Res 2008; 331: 283–300.
7 Roskams T. Liver stem cells and their implication in hepatocellular and
cholangiocarcinoma. Oncogene 2006; 25: 3818–22.
8 Li Y, Kang X, Guo K et al. Proteome alteration of early-stage differentiation
of mouse embryonic stem cells into hepatocyte-like cells. Electrophoresis
2009; 30: 1431–40.
9 Sun S, Lee NP, Poon RT et al. Oncoproteomics of hepatocellular carcinoma:
from cancer markers’ discovery to functional pathways. Liver Int 2007; 27:
10 Kuramitsu Y, Nakamura K. Proteomic analysis of cancer tissues: shedding
light on carcinogenesis and possible biomarkers. Proteomics 2006; 6: 5650–
11 Alison MR, Islam S, Lim S. Stem cells in liver regeneration, fibrosis and
cancer: the good, the bad and the ugly. J Pathol 2009; 217: 282–98.
12 Couchie D, Holic N, Chobert MN, Corlu A, Laperche Y. In vitro
differentiation of WB-F344 rat liver epithelial cells into the biliary lineage.
Differentiation 2002; 69: 209–15.
13 Tsao MS, Grisham JW, Nelson KG, Smith JD. Phenotypic and karyotypic
changes induced in cultured rat hepatic epithelial cells that express the ‘‘oval’’
ª ª 2010 Japanese Cancer Association
cell phenotype by exposure to N-methyl-N’-nitro-N-nitrosoguanidine. Am J
Pathol 1985; 118: 306–15.
14 Huang RP, Peng A, Hossain MZ et al. Tumor promotion by hydrogen
peroxide in rat liver epithelial cells. Carcinogenesis 1999; 20: 485–92.
15 Okamoto M, Kawai K, Reznikoff CA, Oyasu R. Transformation in vitro of a
nontumorigenic rat urothelial cell line by hydrogen peroxide. Cancer Res
1996; 56: 4649–53.
16 Qi YJ, He QY, Ma YF et al. Proteomic identification of malignant
transformation-related proteins in esophageal squamous cell carcinoma. J Cell
Biochem 2008; 104: 1625–35.
17 Song HY, Liu YK, Feng JT et al. Proteomic analysis on metastasis-associated
proteins of human hepatocellular carcinoma tissues. J Cancer Res Clin Oncol
2006; 132: 92–8.
18 Tsao MS, Smith JD, Nelson KG, Grisham JW. A diploid epithelial cell line
from normal adult rat liver with phenotypic properties of ‘oval’ cells. Exp Cell
Res 1984; 154: 38–52.
19 Chiba T, Zheng YW, Kita K et al. Enhanced self-renewal capability in hepatic
stem⁄progenitor cells drives cancer initiation. Gastroenterology 2007; 133:
20 Alison MR. Liver cancer: a disease of stem cells? Panminerva Med 2006; 48:
21 Lee JS, Heo J, Libbrecht L et al. A novel prognostic subtype of human
hepatocellular carcinoma derived from hepatic progenitor cells. Nat Med
2006; 12: 410–6.
22 Yamamoto T, Uenishi T, Ogawa M et al. Immunohistologic attempt to find
carcinogenesis from hepatic progenitor cell in hepatocellular carcinoma. Dig
Surg 2005; 22: 364–70.
23 Theise ND, Yao JL, Harada K et al. Hepatic ‘stem cell’ malignancies in
adults: four cases. Histopathology 2003; 43: 263–71.
24 Zhang F, Chen XP,Zhang
immunohistochemical and double-fluorescence immunostaining evidence.
Histopathology 2008; 52: 224–32.
25 Vondra ´cek J, Krcma ´r P, Procha ´zkova ´ J et al. The role of aryl hydrocarbon
receptor in regulation of enzymes involved in metabolic activation of
polycyclic aromatic hydrocarbons in a model of rat liver progenitor cells.
Chem Biol Interact 2009; 180: 226–37.
26 Fan J, Shen H, Dai Q, Minuk GY, Burzynski FJ, Gong Y. Bone
morphogenetic protein-4 induced rat hepatic progenitor cell (WB-F344 cell)
differentiation toward hepatocyte lineage. J Cell Physiol 2009; 220: 72–81.
27 Coleman WB, McCullough KD, Esch GL et al. Evaluation of the
differentiation potential of WB-F344 rat liver epithelial stem-like cells in vivo.
Differentiation to hepatocytes after transplantation into dipeptidylpeptidase-
IV-deficient rat liver. Am J Pathol 1997; 151: 353–9.
28 Muller-Borer BJ, Cascio WE, Anderson PA et al. Adult-derived liver stem
cells acquire a cardiomyocyte structural and functional phenotype ex vivo.
Am J Pathol 2004; 165: 135–45.
29 Hooth MJ, Coleman WB, Presnell SC, Borchert KM, Grisham JW, Smith GJ.
Spontaneous neoplastic transformation of WB-F344 rat liver epithelial cells.
Am J Pathol 1998; 153: 1913–21.
30 Garber K. Energy deregulation: licensing tumors to grow. Science 2006; 312:
31 Christofk HR, Vander Heiden MG, Harris MH et al. The M2 splice isoform of
pyruvate kinase is important for cancer metabolism and tumour growth.
Nature 2008; 452: 230–3.
32 Pawlak G, Helfman DM. Cytoskeletal changes in cell transformation and
tumorigenesis. Curr Opin Genet Dev 2001; 11: 41–7.
33 Shibata K, Kajiyama H, Mizokami Y et al. Placental leucine aminopeptidase
(P-LAP) and glucose transporter 4 (GLUT4) expression in benign, borderline,
and malignant ovarian epithelia. Gynecol Oncol 2005; 98: 11–8.
34 Macheda ML, Rogers S, Best JD. Molecular and cellular regulation of glucose
transporter (GLUT) proteins in cancer. J Cell Physiol 2005; 202: 654–62.
35 Carpenter B, MacKay C, Alnabulsi A et al. The roles of heterogeneous
nuclear ribonucleoproteins in tumour development and progression. Biochim
Biophys Acta 2006; 1765: 85–100.
36 Gouble A, Grazide S, Meggetto F, Mercier P, Delsol G, Morello D. A new
player in oncogenesis: AUF1⁄hnRNPD overexpression leads to tumorigenesis
in transgenic mice. Cancer Res 2002; 62: 1489–95.
Additional Supporting Information may be found in the online version of this article:
Table S1. Differentially expressed proteins identified by MALDI-TOF-MS⁄MS during neoplastic transformation of WB-F344 cells.
Table S2. Expression patterns of the proteins also identified in study of early hepatic differentiation of mouse embryonic stem (ES) cells.
Table S3. Expression patterns of the proteins also identified to be dysregulated in hepatocellular carcinoma and other cancers.
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries
(other than missing material) should be directed to the corresponding author for the article.
Li et al.Cancer Sci|
|vol. 101|no. 5|
ª ª 2010 Japanese Cancer Association