Tumor Biol. (2012) 33:1785–1802 DOI 10.1007/s13277-012-0438-8

RESEARCH ARTICLE

Platelet-derived growth factor may be a potential diagnostic and prognostic marker for cholangiocarcinoma Sirintip Boonjaraspinyo & Thidarut Boonmars & Zhiliang Wu & Watcharin Loilome & Paiboon Sithithaworn & Isao Nagano & Somchai Pinlaor & Puangrat Yongvanit & Phuangphaka Sadee Nielsen & Chawalit Pairojkul & Narong Khuntikeo

Received: 21 April 2012 / Accepted: 30 May 2012 / Published online: 26 June 2012 # International Society of Oncology and BioMarkers (ISOBM) 2012

Abstract Our previous report showed that platelet-derived growth factor (PDGF) and related genes were upregulated in a Syrian hamster model and could be detected in all human cholangiocarcinoma (CCA) tissues. We therefore hoped that PDGF could be used as a diagnostic and prognostic marker. We analyzed 78 samples of human CCA and adjacent tissues for PDGF and related gene expression, and localized PDGF protein expression. The mechanism of anti-cancer drugs on PDGF and related genes or proteins in CCA cell lines (OCA17, M156, and KKU100) was studied through MTT cell viability assay, quantitative real-time PCR, and immunoblotting. Mutagenesis of the PDGFRA coding region was analyzed. Moreover, the PDGFRA in sera of CCA patients and healthy controls was investigated. PDGFA was found to be upregulated in CCA tissue (84.6 %). Positive PDGFA immunohistochemical staining was significantly

correlated with status (P00.000), stage of CCA (P00.013), metastasis (P00.017), and short survival rate (P00.005), and the multivariate analysis confirmed that PDGFA positive immunostaining had a higher likelihood of the risk of death (HR 02.907, P00.016). For DNA point mutation of the PDGFRA sequence, silent mutations were found at tyrosine kinase 2 V824V (exon 18) and A603A (exon 13), and a missense mutation in S478P (exon 10); there was only a missense mutation in S478P (29 %) that has significant correlation with the histopathological grading (P00.037) and positive immunoreactive PDGFA (P00.021). In vitro cell line study by immunowestern blotting found that sunitinib malate had an inhibitory effect on the PDGFA pathway by decreasing p-PDGFRA, AKT, and p-AKT expression. The serum level of PDGFA in CCA patients was significantly higher than those of healthy control by

S. Boonjaraspinyo : T. Boonmars (*) : P. Sithithaworn : S. Pinlaor Department of Parasitology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand e-mail: [email protected]

W. Loilome : P. Yongvanit Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand

T. Boonmars e-mail: [email protected] S. Boonjaraspinyo : T. Boonmars : W. Loilome : P. Sithithaworn : S. Pinlaor : P. Yongvanit : C. Pairojkul : N. Khuntikeo Liver Fluke and Cholangiocarcinoma Research Center, Khon Kaen University, Khon Kaen 40002, Thailand S. Boonjaraspinyo : Z. Wu (*) : I. Nagano Department of Parasitology, Graduate School of Medicine, Gifu University, Gifu 501-1194, Japan e-mail: [email protected]

P. S. Nielsen Division of Clinical Immunology, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand C. Pairojkul Department of Pathology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand N. Khuntikeo Department of Surgery, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand

1786

1.4-fold (P 00.014). The present results suggest that PDGFA and PDGFRA may be used for CCA prognosis and/or as diagnostic candidate markers. Keywords Cholangiocarcinoma . Opisthorchiasis . PDGFA . PDGFRA . Carcinogenesis

Introduction Platelet-derived growth factor (PDGF) is an important growth factor that is involved in the regulation of cell growth and division, including cell proliferation, differentiation, survival, transformation, and chemotaxis; moreover, it plays a significant role in angiogenesis [1]. To date, several reports have found that PDGFs are frequently produced by tumor cells and may affect tumor growth and dissemination in several different ways [2, 3]; therefore, PDGFs may further be involved in the recruitment of tumor fibroblasts and pericytes [4, 5]. Tumor fibroblasts may function, in turn, to produce factors that directly act on tumor cells to promote their proliferation and migration. In addition, tumor fibroblasts may also secrete angiogenic factors that help to sustain tumor angiogenesis and promote metastasis of tumor cells [6]. The PDGF/PDGFR (platelet-derived growth factor receptor) complex engages several signaling pathways, such as RAS-MAPK, PI3K, and PLC, which are known to be involved in multiple cellular and developmental responses, including carcinogenesis [7]. AKT/PKB is one of the serine/ threonine kinase effectors of PI3K signaling [8, 9]. Activation of the PI3K pathway by PDGFRs can promote actin reorganization, directed cell movements, stimulation of cell growth, and inhibition of apoptosis [10]. PDGF can activate the STAT proteins [11, 12] which can be observed with high frequency in many blood malignancies and solid tumors [13, 14]. Cholangiocarcinoma (CCA), a malignant tumor of the bile duct, is a rapidly lethal disease. The highest incidence of this cancer is found in northeast Thailand and is caused by liver fluke infection in combination with many other factors including nitrosamine in fermented food [15, 16]. The relationship between O. viverrini infection and the prevalence of CCA has been demonstrated [17]; however, the mechanisms by which genes or proteins are regulated are still largely unknown. This has led to a lack of effective diagnostic and prognostic markers, even though this cancer has been known for more than 100 years. A better understanding of its molecular pathogenesis is warranted in order to define new biomarkers that could be used for potential diagnosis as well as prediction of treatment response and prognosis. According to our previous report on an animal model of opisthorchiasis-associated CCA, the highest expression of PDGFA (platelet-derived growth factor alpha)

Tumor Biol. (2012) 33:1785–1802

occurred at the early stage (2 months) of cholangiocarcinogenesis. Thus, we expanded our work by increasing the number of human CCA cases and studied the functional analysis. Seventy-eight cases of CCA tumors collected from patients in an O. viverrini-endemic region were analyzed for the expression of PDGFA and its receptor by real-time RT-PCR (reverse transcription–polymerase chain reaction) and immunostaining. The frequency of PDGFRA point mutations was evaluated by DNA sequencing; correlation with clinicopathological parameters was also analyzed. Moreover, we investigated the different mechanisms of action of sunitinib malate and imatinib mesylate in vitro in CCA cell lines by MTT cell viability, migration assay, real-time RT-PCR, and Western blot. The overexpression and/or mutation of target genes may play a critical role in cholangiocarcinogenesis and tumor progression of CCA. This may be used as a new marker in the diagnosis and prognosis of CCA.

Material and methods Collection of human CCA and adjacent tissues Seventy-eight pairs of liver samples from CCA patients, including CCA and adjacent tissues, were provided by the Liver Fluke and Cholangiocarcinoma Research Center, Faculty of Medicine, Khon Kaen University, Thailand. The utilization of the specimens in the present study was approved by the Human Ethics Committee of Khon Kaen University (Ethical Clearance No. HEKKU501153). These CCA samples were histologically confirmed by a medical pathologist. Total RNA was isolated from the CCA and adjacent tissues. The expression levels of PDGFA and PDGFRA (platelet-derived growth factor receptor alpha) were determined by quantitative real-time RT-PCR. The altered expression was calculated (expression level in CCA tissue/expression level in adjacent liver tissue); more than threefold or less than threefold was considered to be significant. The clinicopathological features of CCA patients were represented by status, gross type, histological type, histological grading, staging, and metastasis. Intrahepatic CCA is classified as mass-forming, periductal-infiltrating, or intraductal-growing, based on its growth characteristics as ascertained by the Liver Cancer Study Group of Japan [18]. The histological grading classifications are well differentiated, moderately differentiated, and poorly differentiated [19]. The most common histology of intrahepatic CCA is an adenocarcinoma, which shows tubular and/or papillary structures. This has been classified using the International Union Against Cancer (UICC)/American Joint Committee

Tumor Biol. (2012) 33:1785–1802

on Cancer (AJCC) TNM (tumor-node-metastasis) system [20] as follows: stage I—solitary tumor without vascular invasiona (T1); stage II—solitary tumor with vascular invasiona (T2a) or multiple tumors with or without vascular invasiona (T2b); stage III—tumor perforating the visceral peritoneum or involving local extrahepatic structures by direct invasion (T3); stage IVA—tumor with periductal invasionb (T4) or any T and regional lymph node metastasesc (N1); stage IVB—any T, any N, and distant metastases (M1). **Note: a includes major vascular (portal or hepatic vein) and microvascular invasion; bincludes tumors with periductal-infiltrating or mixed mass-forming and periductal-infiltrating growth pattern; cnodal involvement of the celiac, periaortic, or caval lymph nodes is considered to be distant metastasis (M1). Collection of CCA cell lines Three CCA cell lines were established from different histological grading of primary CCA tumors: OCA17 represented a well-differentiated type, M156 represented a moderately differentiated type, and KKU100 represented a poorly differentiated type, as described by Sripa et al. [21] CCA cells were cultured in Ham’s F-12 (Gibco®; Invitrogen, Carlsbad CA, USA) supplemented with 10 % fetal calf serum, 100-U⁄ml penicillin, and 100-μg ⁄ml streptomycin at 37 °C and 5 % CO2. CCA cell lines were seeded in a sixwell plate (1×105 per well) and incubated for 24 h. The media were replaced, and the cells were incubated in either a control medium or sunitinib malate; total RNA from each sample was isolated using the protocol described in the following.

1787

Quantitative real-time PCR The primer pairs for the PDGF pathway in this experiment were PDGFA, PDGFRA, PDGFB, PDGFRB, PI3KCA, AKT1, STAT3, PDPK1, PTEN, CCL5, G3PDH, and GUSB, designed based on the published sequence as summarized in Table 1. Before use, the specific primers were checked for specificity by conventional polymerase chain reaction. A single band of each gene was observed. Real-time PCR was performed to determine gene expression levels using a SYBR Premix Ex Taq Kit and Thermal Cycler Dice™ Real Time System (Takara Bio) according to manufacturer’s instructions. In brief, 20 μl of reaction mixture was composed of 2 μl of the template (appropriate dilution was determined by gene), 10 μl of SYBR Premix Ex Taq, 0.8 μl of 5 μM of each primer, and 6.4 μl of distilled water. PCR amplification was performed with 1 cycle of predenaturation at 95 °C for 10 s and 40 cycles of amplification at 95 °C for 5 s and 60 °C for 30 s. Melting curve analysis was carried out from 60 °C to 95 °C with 0.1°C/s temperature transition. Specific external controls were constructed for all target genes. Tenfold serial dilutions (101 to 107 copies/2 μl) of the DNA were used to generate standard curves for each gene. Differences in the amounts of cDNA from different samples were normalized by quantification of the housekeeping genes G3PDH and GUSB. The expression levels were represented as the copy numbers of the target gene/106 housekeeping gene copies. Values were expressed as mean±SD; P-values <0.05 were considered significant. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE)

Total RNA isolation and complementary DNA synthesis Total RNA was extracted from the liver specimen or CCA cell line using TRIzol® reagent (Invitrogen) according to the manufacturer’s instructions. The isolated RNA was treated with DNase I (RQ1 RNase-Free DNase; Promega, Madison WI, USA) in the presence of ribonuclease inhibitor (Takara Bio, Shiga, Japan). The treated RNA was extracted with phenol/chloroform, precipitated with ethanol, and dissolved in RNase-free water. Reverse transcription was performed with Prime-Script™ Reverse Transcriptase (Takara Bio) according to the manufacturer’s instructions. In brief, each reaction consisted of 3 μg of the sample RNA, 1 μl of 0.5-μg/μl oligo(dT)12–18 primer, 1 μl of 10mM deoxyribonucleotide triphosphate mix, 4-μl firststrand buffer, 1-μl RNase inhibitor, and 1-μl reverse transcriptase. The reaction was incubated at 42 °C for 60 min and then inactivated by heating at 75 °C for 15 min [22].

The protein components of the CCA cell lines were separately resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) using a Mini-Protean II cell (Bio-Rad, Hercules CA, USA) under reducing conditions on 8 % gel prepared using the method of Laemmli [23]. Before loading, the samples were prepared by mixing with SDS-PAGE sample buffer and heating for 5 min at 95 ° C; the samples were then loaded onto the gel. After electrophoresis, the resolved polypeptides were electrophoretically transferred to a nitrocellulose membrane for immunoblotting [24] using Trans-Blot Semi-Dry (Bio-Rad). The electrophoretic transfer was performed in Towbin buffer for 90 min at 90 mA. Immunoblot analysis The effect of sunitinib malate on PDGFRA downstream signaling pathways in cells of the M156 CCA cell line

1788 Table 1 Summary of the realtime RT-PCR primer pairs

a

Upper row forward primer, lower row reverse primer

Tumor Biol. (2012) 33:1785–1802

Gene

Product length (bp)

PDGFA

116

PDGFRA

176

PDGFB

163

PDGFRB

210

PIK3CA

180

AKT1

207

STAT3

184

PDPK1

207

PTEN

131

CCL5

154

G3PDH

182

GUSB

131

Sequencea

GenBank accession number/reference

5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′5′-

Boonjaraspinyo et al. [25]

was assessed by immunoblotting, using the specific antibodies GAPDH, PDGFA, p-PDGFRA (Tyr 754), AKT, p-AKT (Ser 473), PDPK1, and p-PDPK1 (Ser 241) (Santa Cruz Biotechnology, Santa Cruz CA, USA). Then, membranes were washed in TBS (tris-buffered saline) containing 0.1 % Tween 20 and incubated for 1 h with peroxidase-conjugated secondary antibody. After that, the slides were washed three times with TBS for 10 min each time. The immunoreactive proteins were visualized by Western Lightning® Chemiluminescence Reagent (PerkinElmer Life Sciences, Boston MA, USA). Quantitative analysis of protein expression was performed using a luminescent image analyzer (Fuji Xerox, Tokyo, Japan). cDNA sequencing of coding region of PDGFRA Total RNA was extracted from human tumor tissue with TRIzol® reagent, as previously described [25]. Primers were designed based on the published sequence in GenBank, accession no: NM_006206, and included the coding sequencing (CDS) region from exon 10 to exon 22 of the human PDGFRA gene: PDGFRA1 (forward AAAT CAAACCCACCTTCAGC and reverse TCAAC CACCTTCCCAAACG), PDGFRA2 (forward T G T C C T G G T T G T C AT T T G G A a n d r e v e r s e

CCCATTCGGAGGAAGAGAAG -3′ ATCAGGAAGTTGGCGGACG -3′ GAAGGTGGTTGAAGGAACAGC -3′ AGGCTCCCAGCAAGTTTACAA -3′ GCCTCATAGACCGCACCAA -3′ GGCTTCTTCCGCACAATCTC -3′ ACAGCCTCTACACCACCCTGA -3′ TCTCAAACTTCTCTTCCCAGCA -3′ CGTGTGCCATTTGTTTTGAC-3′ CATTCCAGAGCCAAGCATCA -3′ GCCCAACACCTTCATCATCC -3′ AGGGACACCTCCATCTCTTCAG -3′ GCTTCTCCTTCTGGGTCTGG -3′ ACGCCTCCTTCTTTGCTG -3′ GTGAGGAAATGGAAGGATACGG -3′ AGTGTGAGGAGGAGGACGAAGA -3′ TGAAGACCATAACCCACCACAG -3′ TTACACCAGTTCGTCCCTTTCC -3′ TGCTGCTTTGCCTACATTGC -3′ ACTCCCGAACCCATTTCTTCTC -3′ GAACATCATCCCTGCCTCTACT -3′ CCTGCTTCACCACCTTCTTG -3′ ATGGAAGAAGTGGTGCGTAGG -3′ AAGGATTTGGTGTGAGCGA -3′

Boonjaraspinyo et al. [25] NM_002608 NM_002609 NM_006218 NM_ 005163 NM_139276 NM_002613 NM_000314 NM_002985 Boonmars et al. [22] Boonjaraspinyo et al. [25]

GCTCCGTGTGCTTTCATCA), and PDGFRA3 (forward T G A A C C C T G C T G AT G A A A G C a n d r e v e r s e GCATTGTCTGAGTCCACACG). RT-PCR was performed using the Thermal Cycler Dice Real Time System. The PCR reaction mixture was comprised of 2 μl cDNA, 2 μl of 10× Hot Start Taq buffer, 1.6 μl of 25-mM dNTP, 2 μl of 5-μM primer pairs, 0.12 μl of Hot Start Taq DNA polymerase (Takara Bio), and 10.4 μl of distilled water to give a final volume of 20 μl. The condition was 1 cycle at 98 °C for 10s and 38 cycles at 98 °C for 10 s, 58 °C for 30 s, and 72 °C for 60 s. PCR-amplified cDNA fragments were a pure single band and corresponded to the predicted size upon analysis by agarose gel electrophoresis. PCR products were purified using a GENECLEAN® II Kit (Qbiogene, Carlsbad CA, USA). Both strands of each fragment were sequenced by the dideoxynucleotide chain termination method using an ABI PRISM® dye terminator cycle sequencing ready reaction kit and automated DNA sequencing system (Applied Biosystems, Foster City CA, USA). The reaction mixture consisted of 0.5 μl of 2.5× Ready Reaction Premix, 4 μl of 5× BigDye® sequencing buffer (Applied Biosystems), 2 μl of 5-μM primer pairs, and distilled water to give a final volume of 20 μl. The PCR sequencing reaction was performed using the Thermal Cycler Dice Real Time System. The conditions were 1 cycle at 96 °C for 3 min; 25 cycles at

Tumor Biol. (2012) 33:1785–1802

96 °C for 10 s, 50 °C for 5 s, and 60 °C for 4 min; and 1 cycle at 60 °C for 2 min. Electrophoresis of the purified samples was performed on a model 3100 Genetic Analyzer automated DNA sequencing system (Applied Biosystems). Computer analysis of sequence data was performed using Hitachi DNASIS® sequence analysis software (Helixx Technologies, Toronto, Canada). Immunohistochemical staining for PDGFA proteins Immunohistochemistry procedure was performed according to the streptavidin–biotin–peroxidase complex principle, using rabbit anti-PDGFA polyclonal antibody (dilution 1:50; Santa Cruz Biotechnology). Briefly, the paraffin sections of human tissues were deparaffinized and rehydrated. The sections were autoclaved at 121 °C for 10 min in 10-mM citrate buffer (pH 6.0) for epitope antigen retrieval. The sections were incubated with 3 % H2O2 in methanol for 10 min for the inactivation of endogenous peroxidase and then blocked with 5 % skim milk for 30 min at room temperature. After incubation with PDGFA primary antibody at 37 °C for 60 min, the sections were incubated with the secondary biotinylated anti-rabbit antibody (dilution 1:100; Zymed, San Francisco CA, USA) for 60 min, followed by incubation with streptavidin–peroxidase (dilution 1:800; Zymed) for 30 min. The immune reaction was visualized by AEC (aminoethylcarbazole) as a chromogen. All sections were counterstained with hematoxylin. Immunohistochemical reactions were analyzed; tumor cells with intracytoplasmic immunoreactivity were evaluated as positive. Sections were semi-quantitatively scored by Reis et al. [26] independently, with the observers blinded to the clinical information and results of the other molecular tests, as follows: (−), 0 % of immunoreactive cells; (+), <5 % of immunoreactive cells; (++), 5–50 % of immunoreactive cells; and (+++), >50 % of immunoreactive cells. Samples with scores (−) and (+) were considered negative, and those with scores (++) and (+++) were considered positive [32]. MTT cell viability assay The 3-(4,5-dimethyl-2-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (Sigma-Aldrich, St. Louis MO, USA) assay is a colorimetric measurement of MTT reduction to a blue formazan product by mitochondrial dehydrogenases of viable cells. The three CCA cell lines were seeded in a 96-well plate (1–2 × 104 per well) and incubated for 24 h. The medium was replaced, and the cells were incubated for 3 day either with complete medium or with sunitinib malate or imatinib mesylate in various concentrations at 37 °C under 5 % CO2 atmosphere. Then, the media were replaced

1789

with 100-μM MTT working solution in each well at a final concentration of 0.5 mg/ml. Cells were incubated for 4 h at 37 °C. Finally, the media were replaced with 100 μl of DMSO to dissolve the formazan. Optical density (A) was measured at 570 nm using a microplate spectrophotometer. The number of viable cells was calculated using the following formula: cell viability rate (%)0(optical density A of experimental group/optical density A of control group)× 100 %. In vitro migration assay Cell migration was performed using a 24-well transwell chamber with 8-μm pore membrane (Falcon™ HTS; BD Biosciences, Bedford MA, USA). Cells were treated with sunitinib malate or imatinib mesylate for 24 h and then seeded to the upper compartment of the transwell at a density of 5 × 10 4 /well in 1 % fetal bovine serum (FBS). The lower chamber was filled with 10 % FBS after 24 h of incubation at 37 °C in a humidified CO2 incubator, cells in the transwell were fixed with 4 % paraformaldehyde and stained using 0.5 % toluidine blue in 2 % Na2CO3. Cells from the upper surface were removed with a cotton swab. Cells that migrated to the under surface were counted under a microscope with a 10× objective in five random fields. Enzyme link immunosorbant assay for PDGFA The PDGFA in serum was quantified using PDGF-AA (R&D systems; Wiesbaden-Norderstedt, Germany) done according to the manufacturer’s instructions. Briefly, 100 μl samples (1:20 PDGF-AA), standard, and controls were dispensed into microplate wells precoated with antiPDGFAA monoclonal antibody and then incubated for 120 min at room temperature. After incubation, the wells were washed for 4 times, and then 150 μl of HRPconjugated PDGF-AA was added and incubated for 120 min at room temperature. After washed time, 200 μl of chromogenic enzyme substrate solution was added and incubated for 30 min at room temperature; 50 μl stop solution was added and measured spectrophotometrically using microplate reader (machine) at 450 nm. The sample concentration was calculated based on standard concentrations. Statistical analysis Statistical correlation between PDGFA and PDGFRA genes, protein expressions, mutations, and various clinicopathological variables (status, gross type, histological type, histological grading, staging, and metastasis) was performed by chi-square and Student’s t-tests using SPSS version 19.0.

1790

Tumor Biol. (2012) 33:1785–1802

Survival curves were constructed by the Kaplan–Meier method and compared with the log-rank test using SigmaPlot version 11.0. The effect of prognostic factors on survival has been summarized using Cox proportional hazard regression analysis with 95 % CI by Stata version 11. PDGFA in healthy and CCA patient sera was analyzed by Student’s t-tests using SPSS version 19.0. The results were considered statistically significant when the P-value was less than 0.05.

Results Expression of PDGFA and PDGFRA in human CCA tissues To observe whether the expression of PDGFA and PDGFRA was involved in CCA, total RNA was isolated from each sample of CCA and adjacent liver tissues. The expression levels were determined by quantitative real-time PCR, and folds of expression alteration were Fig. 1 Expression levels of PDGFA and PDGFRA mRNA in 78 CCA patients. Total RNA was isolated from the CCA and adjacent tissues. The expression level was determined with realtime PCR and was presented as copy numbers within 106 GUSB copies. a PDGFA mRNA expression, b PDGFRA mRNA expression, c fold of PDGFA expression alteration, d fold of PDGFRA expression alteration

calculated (Fig. 1). The expression of PDGFA was upregulated in 66 cases (84.6 %; 3.0- to 92.4-fold). The expression of PDGFRA was upregulated in 36 cases (46.2 %; 3.1- to 32.3-fold).

Immunohistochemical staining for PDGFA in human opisthorchiasis-associated CCA To observe whether there is a change in the expression of PDGFA in human opisthorchiasis-associated CCA, immunohistochemical staining for PDGFA was performed. In 47 cases (60.3 %), tumor cells demonstrated a positive reaction, with strong and diffuse expression of PDGFA in the cytoplasm of epithelial and endothelial cells. As shown in Fig. 2c–f, strong staining for PDGFA protein was observed in malignant bile duct tissue. Staining specificity was confirmed by the absence of staining when using normal rabbit serum (Fig. 2a) and in adjacent normal tissue (Fig. 2b).

A) PDGFA mRNA expression p < 0.005

C) Fold of PDGFA expression alteration

D) Fold of PDGFRA expression alteration

B) PDGFRA mRNA expression p < 0.005

Tumor Biol. (2012) 33:1785–1802

1791

Fig. 2 Immunohistochemical expression of PDGFA in human opisthorchiasis-associated CCA was stained with anti-PDGFA antibody and counterstained with hematoxylin. The brown color indicates the positive reaction. a Negative control, b adjacent normal control, c well differentiation (tubular type), d well differentiation (papillary type), e moderate differentiation (tubular type), and f poor differentiation (tubular type). Arrow: hepatic bile duct; magnification (×10)

Table 2 Frequency of single base pair substitutions among PDGFRA exon 10–22 mutations from CCA patients PDGFRA

Exon 10 Exon 13 Exon 18

Nucleotide change

Amino-acid substitution

1763 T>C 2140 G>A 2803 C>T

S478P A603A V824V

No. of cases (n078)

dbSNP

23 (29%) 36 (46%) 47 (60%)

rs35597368 rs10028020 rs10015469

dbSNP single nucleotide polymorphism database (http:// www.ncbi.nlm.nih.gov/SNP/)

Gene-activating mutation in PDGFRA in human opisthorchiasis-associated CCA To further characterize the potential participation of PDGFRA in CCA cancer development and to determine the presence of gene-activating mutations, we performed a mutational analysis of PDGFRA (exons 10 to 22 in the cDNA from frozen CCA tumors). The results (Table 2 and Fig. 3) indicated a silent mutation in the enzymatic domain at tyrosine kinase 2 codon 824 C>T, leading to a Val>Val change (SNP rs10015469, previously reported) in exon 18

1792

Tumor Biol. (2012) 33:1785–1802

Fig. 3 Electropherogram of part of PDGFRA sequence. DNA sequencing of base substitution: a 1763 T>C in PDGFRA exon 10 (reverse primer), b 2140 G>A) in PDGFRA exon 13 (forward primer), c 2803 C>T in PDGFRA exon 18 (reverse primer). Arrow indicates base substitution

(60 %), and at tyrosine kinase 1 codon 603 G>A, leading to an Ala>Ala change (SNP rs10028020, previously reported) in exon 13 (46 %). On the other hand, in the receptor regular domain at the dimerization domain encoded by exon 10 presented a missense mutation in codon 478 T>C, leading to a Ser>Pro substitution (SNP rs 35597368, previously reported) (Table 2). Correlation between PDGFA and PDGFRA expression and clinicopathological features of CCA patients The correlations of clinicopathological characteristics of the 78 CCA patients (Table 3) with PDGFA and PDGFRA expressions are shown in Table 4. Positive immunostaining of tumor cells was significantly correlated with tumor status (P00.000), stage (P00.013), and metastasis (P00.017). For DNA point mutation, there was only a missense mutation in S478P that had significant correlation with the positive immunostaining of PDGFA (P 00.021) and histological grading (P00.037), as shown in Table 5. There were no correlations of positive immunoreactive expression, mRNA

expression, and point mutation with various clinicopathological parameters (such as age, gender of patient, and histological type). Overall survival rate of CCA patients Among the 78 CCA patients studied, the number of deaths was 50 (68 %), and the mean survival time was 1.01 years after resection. For patients with positive immunostaining for PDGFA in tumor cells, the survival rates at 1, 2, and 3 years were 36.6 %, 19.5 %, and 12.2 %, respectively; for the negative immunostaining group, survival rates were 71.4 %, 38.0 %, and 28.5 %, respectively. Patients in the PDGFA-positive immunostaining group had a poorer prognosis than those in the negative immunostaining group (P00.005), a statistically significant difference (Fig. 4). This result correlated with high mRNA expression of PDGFA: These patients had a poorer prognosis than those with low expression of PDGFA. For patients with high expression of PDGFA, the survival rates at 1, 2, and 3 years were 42.9 %, 21.0 %, and 15.3 %, respectively; for those with low

Tumor Biol. (2012) 33:1785–1802

1793

Table 3 Clinicopathological characteristics of 78 CCA patients Clinicopathological factors

Number

Age (n078) Gender (n078) Male Female Survival (year) (n078) Status (n078) Dead Alive

56.64±8.76

CCA gross type (n048) Mass forming Mixed type Histological type (n074) Papillary type Tubular type Mixed type Histological grading (n077) Well differentiation Moderate to poor differentiation Staging of CCA (n078) I & II III & IV Metastasis (n078) No metastasis Metastasis

50 28 1.44±1.48

(64.10%) (35.90%)

65 13

(83.33%) (16.67%)

35 13

(72.92%) (27.08%)

19 47 8

(25.68%) (63.51%) (10.81%)

33 44

(42.86%) (57.14%)

31 47

(39.74%) (60.26%)

42 36

(53.85%) (46.15%)

expression, the rates were 55.6 %, 32.7 %, and 20.8 %, respectively—however, this difference was not statistically significant (data not shown). Additionally, the survival rates of patients with point mutations of PDGFRA in exons 10, 13, and 18 were similar. There was no difference in the survival rates between the point mutation group and non-mutation groups for all point mutations (data not shown). To determine which of these CCA prognostic factors such as age, gender, gross type, histological type, histological grading, and metastasis including clinical stage were independent predictors of mortality, the Cox proportional hazards regression was used. Table 6 shows the results of the univariable and multivariable analysis of overall survival analysis. In a univariate analysis of potential CCA prognostic factors, the hazard ratio for CCA gross of mixed type compared with mass forming type was 0.376 (95 % CI 0.171–0.823, P00.014). The hazard ratio for histological grading of moderate to poor differentiation compared with those with well differentiation was 2.041 (95 % CI 1.214– 3.431, P00.007). The prognosis profile of histological type was associated with a significantly higher hazard ratio in mixed type versus papillary (HR03.210; 95 % CI 1.268– 8.126, P00.014) than tubular type versus papillary (HR0

2.141; 95 % CI 1.098–4.172, P00.025). The difference between stages I, II, and III, IV in the survival demonstrated a significantly increased risk for patients with stages III, IV, with a hazard ratio of 2.230 (95 % CI 1.241–4.005, P00.007). The prognostic profile was also associated with the metastasis; the hazard ratio for metastases was 2.066 (95 % CI 1.235–3.454, P00.006). However, the effect of age, gender, PDGFA, and PDGFRA gene expression on survival was not significant. In a multivariate analysis of prognostic factors, only positive immunostaining of PDGFA (HR02.907; 95 % CI 1.218–6.937, P00.016) was a significant and independent unfavorable prognostic factor. Effects of PDGFRA inhibitor on CCA cell line proliferation and migration As PDGFA has been detected in CCA and other related cancers and as its expression level has been shown to correlate with the stage of CCA and metastasis, the role of PDGFA in the proliferation and motility of CCA cell lines was investigated by examining the effects of sunitinib malate and imatinib mesylate on these properties using migration and MTT assay. CCA cell lines were treated with various concentrations of sunitinib malate (1, 1.25, 2.5, 5, and 10 μM). The percentage of viable cells was determined by MTT assay. Incubation with sunitinib malate for 72 h resulted in inhibition of cell proliferation, with M156 showing the most sensitivity and KKU100 the least sensitivity. Sunitinib malate significantly inhibited the growth of CCA cell lines in a dose-dependent manner in all three cell lines (IC50 05–6 μM). Treatment with imatinib mesylate in various concentrations (20, 25, 30, 35, and 40 μM) resulted in inhibition of cell proliferation, with OCA17 showing the most sensitivity and M156 the least sensitivity. Imatinib mesylate inhibited the growth of CCA cell lines in a dose-dependent manner in all three cell lines (IC50 031–41 μM) (Fig. 5). Treatment with sunitinib malate for 24 h suppressed cell migration in all three CCA cell lines, with M156 showing the most sensitivity (62 %) and KKU100 the least (81.6 %). For imatinib mesylate treatment, the most sensitive was KKU100 (54.87 % suppression), and the least sensitive was OCA17 (69.48 %) (Fig. 6). Involvement of PDGFR inactivation of downstream effectors The role of PDGFA in CCA was investigated in the M156 cell line established from tumor tissues of CCA patients. The level of PDGFA mRNA was determined by quantitative real-time RT-PCR and normalized to G3PDH mRNA, as shown in Fig. 7.

1794

Tumor Biol. (2012) 33:1785–1802

Table 4 The correlations of clinicopathological characteristics of the CCA patients with the PDGFA and PDGFRA expression Parameter

PDGFA immunoreactive expression (n 0 78)

PDGFA mRNA expression (n 0 78)

Negative (%)

Positive (%)

Pvalue

Lower expression (%)

Over expression (%)

Pvalue

50

19 (38)

31 (62)

0.468

8 (16)

42 (84)

0.840

≥ 60 28 Gender (n 0 78) Male 50 Female 28 Status (n 0 78) Dead 65 Alive 13 CCA gross type (n 0 48) Mass forming 35 Mixed type 13 Histological type (n 0 74) Papillary type 19 Tubular type 47 Mixed type 8 Histological grading (n 0 77) Well differentiation 33 Moderate to poor 44 differentiation Staging of CCA (n 0 78) I & II 31 III & IV 47 Metastasis (n 0 78) No metastasis 41 Metastasis 37

13 (46)

15 (54)

4 (14)

24 (86)

21 (42) 11 (39)

29 (58) 17 (61)

0.815

7 (14) 5 (18)

43 (86) 23 (82)

21 (32) 11 (85)

44 (68) 2 (15)

0.000*

11 (17) 1 (8)

15 (43) 5 (38)

20 (57) 8 (62)

0.784

9 (47) 18 (38) 3 (37.5)

10 (53) 29 (62) 5 (62.5)

17 (52) 15 (34)

Age (n 0 78) < 60

*

Number

PDGFRA mRNA expression (n 0 78) Lower expression (%)

Over expression (%)

Pvalue

0.610

28 (56)

22 (44)

14 (50)

14 (50)

0.651

29 (58) 13 (46)

21 (42) 15 (54)

0.325

54 (83) 12 (92)

0.400

35 (54) 6 (46)

30 (46) 7 (54)

0.612

3 (9) 3 (23)

32 (91) 10 (77)

0.177

20 (57) 7 (54)

15 (43) 6 (46)

0.838

0.780

1 (5) 9 (19) 2 (25)

18 (95) 38 (81) 6 (75)

0.297

9 (47) 28 (60) 4 (50)

10 (53) 19 (40) 4 (50)

0.631

16 (48) 29 (66)

0.125

5 (15) 7 (16)

28 (85) 37 (84)

0.928

16 (48) 25 (57)

17 (52) 19 (43)

0.468

18 (58) 14 (30)

13 (42) 33 (70)

0.013*

2 (6) 10 (21)

29 (94) 37 (79)

0.076

16 (52) 26 (55)

15 (48) 21 (45)

0.748

22 (54) 10 (27)

19 (47) 27 (73)

0.017*

4 (10) 8 (22)

37 (90) 29 (78)

0.147

21 (51) 21 (57)

20 (49) 16 (43)

0.624

P-value is significant

PI3K/AKT and STAT3 are the two major signaling pathways that regulate cell proliferation, motility, and invasion in response to a variety of growth factors/receptor tyrosine kinases. To investigate the pathways by which PDGF promotes invasion and proliferation of CCA cells, we examined the effects of a receptor tyrosine kinase inhibitor (sunitinib malate) on these two main signaling pathways. Suppression of PDGFRA gene expression by sunitinib malate (24 h) reduced PDGFB, AKT, PDPK1, and STAT3 expression levels in the M156 CCA cell line. The expression levels of PDGFA were slightly decreased in the treated group, while PDGFRB was significantly increased. For PI3KCA, PTEN, and CCL5, the level of gene expression was not different. Because sunitinib malate is a tyrosine kinase inhibitor, we examined several major downstream signaling molecules of tyrosine kinases, including PDGFRA, AKT, and PDPK1, by western blot analysis. No inhibitory effects were

observed on PDGFA, p-PDGFRA, and PDPK1 levels in the M156 cell line following 24-h sunitinib malate treatment. Only AKT and p-AKT protein expressions were decreased at 24 h (Fig. 8). However, increased treatment time (48 h) resulted in decreased p-PDGFRA, AKT, and p-AKT (Fig. 8). Western blot data indicated that sunitinib malate had an inhibitory effect on the PDGFA pathway in these cells.

PDGFA in sera of CCA patients The serum PDGFA level obtained from 14 healthy controls was 3,821±272 pg/ml and ranged from 1,993 to 5,128 pg/ ml (Fig. 9). Sera from 51 CCA patients were 5,336±305 pg/ ml and ranged from 2,772 to 14,539 ng/ml. The PDGFA level in CCA patient sera was significantly higher than the

Tumor Biol. (2012) 33:1785–1802

1795

Table 5 The correlations of clinicopathological characteristics of the CCA patients with the PDGFRA point mutation Clinicopathological parameter

Number

SNP 478 S>P

SNP 603 A>A

SNP 824 V>V

No mutation (%)

Mutation (%)

pvalue

No mutation (%)

Mutation (%)

pvalue

No mutation (%)

Mutation (%)

pvalue

50 28

36 (72) 19 (68)

14 (28) 9 (32)

0.700

25 (50) 17 (61)

25 (50) 11 (39)

0.363

16 (32) 15 (54)

34 (68) 13 (46)

0.062

50 28

35 (70) 20 (71)

15 (30) 8 (29)

0.894

27 (54) 15 (54)

23 (46) 13 (46)

0.971

18 (36) 13 (46)

32 (64) 15 (54)

0.367

65 13

46 (71) 9 (69)

19 (29) 4 (31)

0.912

34 (52) 8 (62)

31 (48) 5 (38)

0.542

25 (38) 6 (46)

40 (62) 7 (54)

0.605

35 13

26 (74) 9 (69)

9 (26) 4 (31)

0.726

20 (57) 6 (46)

15 (43) 7 (54)

0.497

15 (43) 6 (46)

20 (57) 7 (54)

0.838

19 47

12 (63) 36 (77)

7 (37) 11(23)

0.072

11 (58) 26 (55)

8 (42) 21 (45)

0.599

10 (58) 17 (36)

8 (42) 30 (64)

0.170

8

3 (38)

5 (63)

3 (38)

5 (63)

2 (25)

6 (75)

33

19 (58)

14 (42)

0.037*

9 (20)

12 (36) 19 (43)

21 (64) 25 (57)

0.546

35 (80)

14 (42) 21 (48)

0.644

44

19 (58) 23 (52)

31 47

18 (58) 37 (79)

7 (42) 16 (21)

0.843

15 (48) 27 (57)

10 (52) 26 (43)

0.454

10 (32) 21 (45)

15 (68) 32 (55)

0.975

41 37

30 (73) 25 (68)

11 (27) 12 (32)

0.588

24 (59) 18 (49)

17 (41) 19 (51)

0.382

16 (39) 15 (41)

25 (61) 22 (59)

0.891

Downregulation 12 Upregulation 66 PDGFRA Downregulation 42 Upregulation 36 PDGFA immunoreactive expression Negative 32

9 (75) 46 (70)

3 (25) 20 (30)

0.711

6 (50) 36 (55)

6 (50) 30 (45)

0.771

5 (42) 26 (39)

7 (58) 40 (61)

0.882

30 (71) 25 (69)

12 (29) 11 (31)

0.848

21 (50) 21 (58)

21 (50) 15 (42)

0.462

17 (40) 14 (39)

25 (60) 22 (61)

0.886

18 (56)

14 (44)

0.021*

14 (44) 22 (48)

11 (34) 20 (43)

21 (66) 26 (57)

0.419

9 (20)

18 (56) 24 (52)

0.722

37 (80)

Age <60 ≥60 Gender Male Female Status Dead Alive CCA gross type Mass forming type Mix type Histological type Papillary Tubular Mix type Histological grading Well differentiation Moderate to poor differentiation Staging of CCA I & II III & IV Metastasis No metastasis Metastasis PDGFA

Positive *

46

P-value is significant

PDGFA level in healthy control sera by 1.4-fold (P00.014) as shown in Fig. 9.

Discussion Our present study found that PDGFs are involved in cholangiocarcinogenesis by overexpression of PDGFA in human CCA tissues and low expression of p-PDGFRA,

AKT, and p-AKT after sunitinib malate treatment. Moreover, positive PDGFA staining was located at the cytoplasm of cancer cell, fibroblast, and endothelial cell. These results suggest that PDGFA may be used for CCA prognosis and/or as a diagnostic candidate marker. It is well established that PDGF exists in epithelial cells, endothelial cells, and fibroblasts, and that it enhances cell proliferation, migration, and survival [27]. This conforms with our present results, which showed different levels of

1796

Tumor Biol. (2012) 33:1785–1802

PDGFA Immunoreactive Survival Analysis

1.0

Positive Negative

Survival rate

0.8

0.6

0.4

0.2

0.0 0

2

4

6

8

Time (year) Fig. 4 Kaplan–Meier curves for survival analysis according to positive immunoreactive of PDGFA in CCA patients (P00.005)

expression in both cancer tissues and adjacent tissues (Fig. 1a, b). Moreover, PDGFs are also present within the extracellular matrix (ECM) and in tissues during the acutephase response to injury [28, 29]. Several reports have demonstrated that the PDGF/PDGFR signaling pathway plays a crucial role in embryonic development, especially in organogenesis, including alveologenesis, glomerulogenesis, angiogenesis, oligodendrogenesis, and spermatogenesis [27, 30, 31]. PDGFA/PDGFRA signaling is necessary for building normal epithelium of the gastrointestinal (GI) tract. PDGFRA is also involved in the regulation of proliferation and differentiation of some mesenchymal cells by paracrine PDGFs [32]. The present results showed that the expressions of PDGFA and PDGFRA mRNA in tumor tissues were 84.6 % and 46.2 %, respectively. This is in agreement with our previous study, in which we identified PDGFA and PDGFRA upregulation in CCA tissue (80 % and 40 %, respectively) [25]. Similar to these observations, Martinho et al. [33] reported that PDGFA and PDGFRA expressions were detected in 81.2 % and 29.6 % of gliomas, respectively. In colon carcinoma, PDGFA and PDGFRA were expressed in

Table 6 Summary of overall survival analysis by univariate and multivariate Cox regression analysis Parameter

Age <60 versus ≥60 years Gender Male versus female CCA gross type Mass forming versus mixed type Histological type Papillary versus tubular type Papillary versus mixed type Histological grading Well differentiation versus moderate–poor differentiation Metastasis No metastasis versus metastasis Stage of tumor I–II versus III–IV PDGFA gene expression Low versus high expression PDGFRA gene expression Low versus high expression PDGFA immunostaining Negative versus positive CI confidence interval, HR hazard ratio *

P-value is significant

Univariate analysis

Multivariate analysis

HR

P-value

95% CI

HR

P-value

95% CI

1.081

0.764

0.649–1.800

1.377

0.428

0.624–3.033

1.443

0.164

0.860–2.420

1.828

0.140

0.820–4.075

0.376

0.014*

0.171–0.823

0.438

0.084

0.172–1.118

2.141 3.210

0.025* 0.014*

1.098–4.172 1.268–8.126

1.523 3.229

0.412 0.141

0.557–4.162 0.678–15.360

2.041

0.007*

1.214–3.431

1.898

0.225

0.674–5.343

2.066

0.006*

1.235–3.454

0.831

0.685

0.340–2.030

2.230

0.007*

1.241–4.005

1.381

0.603

0.409–4.658

0.857

0.643

0.446–1.646

0.942

0.919

0.292–3.026

0.876

0.599

0.535–1.433

0.881

0.733

0.426–1.821

2.189

0.004*

1.292–3.707

2.907

0.016*

1.218–6.937

Tumor Biol. (2012) 33:1785–1802

1797

Fig. 5 Cytotoxic effects of a sunitinib malate b imatinib mesylate. Human CCA cell lines were incubated with sunitinib malate and imatinib mesylate at various concentrations prior to the determination of cellular viability through MTT assay

100 % and 67 % of cases, respectively [34], and in astrocytoma, PDGFRA was detected in 45 % [35]. This suggests that PDGFA and PDGFRA are involved in tumorigenesis. Their increased expression was confirmed by immunohistochemical analysis of PDGF receptors and ligands, based on the distribution of positive cells in cytoplasmic or membrane reactivity [36]. This is in accordance with the present results, where positive staining was observed in endothelial cells, fibroblasts, and the cellular cytoplasm of cancerous tissues (Fig. 2). The tumor diffusely expressed the ligand PDGFA in the cytoplasm and in the endothelial cells, which agreed with previous reports on Ewing’s sarcoma [37], Merkel cell carcinoma [1], and CCA in a hamster model [25]. All cells strongly and diffusely expressed the ligand PDGFA in the cytoplasm,

suggesting that an autocrine loop might be involved in the pathogenesis of CCA. A mild immunoreactivity for the ligand PDGFA was documented in the stroma surrounding the tumor cells and also in endothelial cells of the tumor’s vascular tree. It is probable that PDGF modulates the stromal differentiation process in coordination with ECM molecules [38], because PDGF cytokine effects are dependent on the origin of the stroma [39]. After release of TGF-1 and PDGF, formation of granulation tissue is facilitated by chemotaxis of neutrophils, monocytes, fibroblasts, and myofibroblasts, as well as by synthesis of new extracellular matrix (ECM) and neoangiogenesis [40]. Our analysis also indicated that 59 % of the tumor cells demonstrated positive expression of PDGFA protein. The

1798

Tumor Biol. (2012) 33:1785–1802

Fig. 6 Sunitinib malate and imatinib mesylate treatment inhibited CCA cell migration. a OCA17 control, b OCA17 with sunitinib malate, c OCA17 with imatinib mesylate, d M156 control, e M156

with sunitinib malate, f M156 with imatinib mesylate, g KKU100 control, h KKU100 with sunitinib malate, i KKU100 with imatinib mesylate

positive immunostaining for PDGFA in the tumor cells was significantly correlated with the status (P00.000), stage of the tumor (P00.013), and metastasis (P00.017). Survival analysis showed that patients in the PDGFA-positive immunostaining group had a significantly poorer prognosis than those in the negative immunostaining group (P00.005). Moreover, the multivariate analysis confirmed that positive PDGFA protein expression had a significantly poorer prognosis

by 2.9-fold of a risk of death than those in the negative immunostaining group. Supporting data for this finding is provided by a recent study which found similar clinicopathological characteristics, angiogenic factors, and significantly higher expression levels of VEGF and PDGFA in advanced-stage tumors (stages 3 and 4) in human neuroblastomas [41]. In a study on gastric carcinoma, the PDGFA-positive group demonstrated a shorter overall survival rate compared with

Tumor Biol. (2012) 33:1785–1802

1799

Fig. 7 Sunitinib malate treatment inhibits PDGFRA transactivation in the M156 cell line. *P-value is significant

the PDGFA-negative group (p <0.0001) [42]. Therefore, our data suggest that PDGFA expression is correlated with the stage of tumor advancement, metastasis, and poor prognosis in CCA. The 78 CCA tissue samples were further evaluated for the presence of activating mutations of PDGFRA (exons 10–22). Mutational analysis of PDGFRA showed the presence of a silent mutation (V824V); this silent mutation was also found in 60 % of CCA tissues studied. Our results indicate that activating mutations of PDGFRA are absent in CCA. Nevertheless, the presence of a PDGFRA/PDGFA autocrine/ paracrine stimulation loop in a number of cases supports the potential role of specific tyrosine kinase inhibitors in CCA treatment [43, 44]. PDGFRA would result in constitutive activation of kinase activity, but only a silent polymorphism at valine 824 (GTC-GTT); this is in agreement with a previous report on malignant solitary fibrous tumors [45]. In addition, a missense mutation in S478P (29 %) has a significant correlation with histopathological grading (P00.037) and positive immunoreactivity of PDGFA (P00.021). Moreover, mutant receptors strongly associated with cancer and fibrotic diseases

have been shown to remain catalytically active on the plasma membrane rather than internalizing upon ligand binding [46]. This study suggests that the PDGFR gene has several exonic changes, but only S478P that was involved in histopathological change. Sunitinib malate and imatinib mesylate are inhibitors of receptor tyrosine kinases (RTK) such as PDGFR, KIT, etc., proto-oncogenes expressed by tumor cells. In the present study, the multityrosine kinase inhibitor was shown to effectively inhibit CCA cell line proliferation, with an IC50 (the concentration that inhibits the proliferation by 50 %) of approximately 5 μM for sunitinib malate and 30 μM for imatinib mesylate after 72 h of culture. Similarly, a recent study found that the IC50 of sunitinib malate for treatment of nasopharyngeal carcinoma ranged from 2.0–7.5 μM [47]. PDGF plays a critical role in angiogenesis during tumor growth and metastasis. Tumor angiogenesis requires chemotactic and mitogenic activities [48] from the PDGF/ PDGFR paracrine signaling loop for the formation, branching, and maintenance of blood vessels [49]. In the present study, inhibition of the receptor tyrosine kinases appeared to

1800

Tumor Biol. (2012) 33:1785–1802

Fig. 8 Western blot analyses of the M156 cell line after sunitinib malate treatment at 24 and 48 h using indicated antibodies. *P-value is significant

suppress CCA cell migration and inhibit metastasis, similar to previous reports on medulloblastoma cells [50] and transitional cell carcinoma [51]. Moreover, we found that PDGFA positive immunostaining located in fibroblasts at the tumor stroma may be caused by the PDGF that plays an important role in wound healing by stimulating smooth cell migration, angiogenesis [52], secretion of other growth factors involved in the healing process, and the production of matrix components such as collagen [53, 54]. RTK activation occurs when, by ligand binding, the receptor dimerizes and undergoes conformational transformations, which induce activation of the kinase domains. These, in turn, lead to activation of important intracellular signaling pathways, such as RAS/mitogen activated protein kinase (RAS/MAPK), phosphoinositide-3 kinase (PI3K), and signal transducers and p = 0.014

PDGFA in serum (pg/ml)

20000 15000 10000 5000 0 Healthy

CCA

Fig. 9 Serum PDGFA level in CCA patients compared with healthy controls. The PDGFA serum level was presented as the mean±SEM

activators of transcription (STAT), which regulate many physiological functions such as cell survival, proliferation, differentiation, adhesion, and apoptosis. Several reports have found that most tumor cell lines and fibroblasts express and secrete PDGFA [55], to stimulate its proliferation and provide a microenvironment for producing extracellular matrix, some growth factors, and tumor growth which are necessary nutritional factors to directly or indirectly promote tumor growth, blood supply, and metastasis [56, 57]. Our results show that sunitinib malate inhibited the tyrosine phosphorylation of PDGFRA transactivation by decreasing the mRNA expression level of PDGFRA, AKT, PDPK1, and STAT3 at 24 h post-treatment. This result was confirmed by immunoblotting, which demonstrated that AKT protein expression was significantly decreased at 24 h and that p-PDGFRA, AKT, and p-AKT expressions were significantly decreased at 48 h posttreatment. Similar to our observations, recent studies have found that sunitinib malate induces tumor cell apoptosis and growth arrest in renal cell carcinoma, which correlates with STAT3 and AKT activity inhibition [58, 59]. Moreover, serum PDGFA level of CCA patients was significantly higher than healthy control (P00.014), supported by [28, 60] that the secreted PDGFA was a response to inflammation, injury during the infection or bile duct obstruction, and tumorigenesis because PDGFA plays a role in angiogenesis, cell proliferation, and survival [61]. Our present study is in agreement with several reports which show that PDGFA is expressed in other tumors such as skin, breast,

Tumor Biol. (2012) 33:1785–1802

colorectal, lung, pancreatic, prostate, and ovarian cancers [62], and might also be involved in epithelial growth, survival, and proliferation. In animal models, PDGF expression in the stromal cells of tumors was found to contribute to both the recruitment of fibroblasts and angiogenesis, thus supporting the growth of tumor cells [63, 64]. In summary, the present study for the first time demonstrated the expression of PDGFA in CCA and its correlation with clinicopathology. The results suggest that PDGFA is likely involved in the tumorigenesis of opisthorchiasisassociated CCA; hence, this factor can be a promising candidate biomarker for diagnosis and therapy of CCA. Acknowledgements This work was supported by a grant from the Thailand Research Fund through the Royal Golden Jubilee PhD program (Grant No. PHD/0139/2551) to Miss Sirintip Boonjaraspinyo and Associate Professor Thidarut Boonmars; a Grant-in-Aid for Scientific Research (24590504) from the Ministry of Education, Culture, Sports, Science and Technology of Japan; and a grant from the Office of the Higher Education Commission and the National Research University Project, Faculty of Medicine (Grant No. I54325), Khon Kaen University, Thailand. We also are grateful to the Department of Parasitology, the Liver Fluke and Cholangiocarcinoma Research Center, the Animal Experimental Unit, the Department of Research Affairs (AS54301), and the Faculty of Medicine, Khon Kaen University. We would also like to thank the Department of Parasitology, Graduate School of Medicine, Gifu University, for their support.

References 1. Kartha RV, Sundram UN. Silent mutations in KIT and PDGFRA and coexpression of receptors with SCF and PDGFA in Merkel cell carcinoma: implications for tyrosine kinase-based tumorigenesis. Mod Pathol. 2008;21:96–104. 2. Heldin CH, Westermark B. Platelet-derived growth factor: three isoforms and two receptor types. Trends Genet. 1989;5:108–11. 3. Langley RR, Fidler IJ. Tumor cell-organ microenvironment interactions in the pathogenesis of cancer metastasis. Endocr Rev. 2007;28:297–321. 4. Ostman A. PDGF receptors-mediators of autocrine tumor growth and regulators of tumor vasculature and stroma. Cytokine Growth Factor Rev. 2004;15:275–86. 5. Pietras K, Sjoblom T, Rubin K, Heldin CH, Ostman A. PDGF receptors as cancer drug targets. Cancer Cell. 2003;3:439–43. 6. John AE, Berlin AA, Lukacs NW. Respiratory syncytial virusinduced CCL5/RANTES contributes to exacerbation of allergic airway inflammation. Eur J Immunol. 2003;33:1677–85. 7. Shih AH, Holland EC. Platelet-derived growth factor (PDGF) and glial tumorigenesis. Cancer Lett. 2006;232:139–47. 8. Burgering BM, Coffer PJ. Protein kinase B (c-Akt) in phosphatidylinositol-3-OH kinase signal transduction. Nature. 1995;376:599–602. 9. Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR, Tsichlis PN. The protein kinase encoded by the Akt proto-oncogene is a target of the PDGF-activated phosphatidylinositol 3-kinase. Cell. 1995;81:727–36. 10. Hu JC, Zhang C, Slavkin HC. The role of platelet-derived growth factor in the development of mouse molars. Int J Dev Biol. 1995;39:939–45.

1801 11. Leaman DW, Leung S, Li X, Stark GR. Regulation of STATdependent pathways by growth factors and cytokines. FASEB J. 1996;10:1578–88. 12. Schindler C, Darnell Jr JE. Transcriptional responses to polypeptide ligands: the JAK-STAT pathway. Annu Rev Biochem. 1995;64:621–51. 13. Bowman T, Garcia R, Turkson J, Jove R. STATs in oncogenesis. Oncogene. 2000;19:2474–88. 14. Catlett-Falcone R, Landowski TH, Oshiro MM, Turkson J, Levitzki A, Savino R, Ciliberto G, Moscinski L, Fernandez-Luna JL, Nunez G, Dalton WS, Jove R. Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity. 1999;10:105–15. 15. IARC. Infection with liver flukes (Opisthorchis viverrini, Opisthorchis felineus and Clonorchis sinensis). IARC Monogr Eval Carcinog Risks Hum. 1994;61:121–75. 16. Sithithaworn P, Haswell-Elkins M. Epidemiology of Opisthorchis viverrini. Acta Trop. 2003;88:187–94. 17. Sripa B, Bethony JM, Sithithaworn P, Kaewkes S, Mairiang E, Loukas A, Mulvenna J, Laha T, Hotez PJ, Brindley PJ. Opisthorchiasis and opisthorchis-associated cholangiocarcinoma in Thailand and Laos. Acta Trop. 2011;120 Suppl 1:S158–68. 18. Liver Cancer Study Group of Japan. The general rules for the clinical and pathological study of primary liver cancer. 4th ed. Tokyo: Kanehara; 2000. 19. Nakanuma Y, Sripa B, Batanasapt V, Leong ASY, Ponchon T, Ishak KG. Intrahepatic cholangiocarcinoma. In: Hamilton SR, Aaltonen LA, editors. World Health Organization classification of tumours: pathology and genetics—tumours of the digestive system. Lyon, France: IARC Press; 2000. 20. Blechacz B, Gores GJ. Cholangiocarcinoma: advances in pathogenesis, diagnosis, and treatment. Hepatology. 2008;48:308–21. 21. Sripa B, Leungwattanawanit S, Nitta T, Wongkham C, Bhudhisawasdi V, Puapairoj A, Sripa C, Miwa M. Establishment and characterization of an opisthorchiasis-associated cholangiocarcinoma cell line (KKU100). World J Gastroenterol. 2005;11:3392–97. 22. Boonmars T, Wu Z, Boonjaruspinyo S, Pinlaor S, Nagano I, Takahashi Y, Kaewsamut B, Yongvanit P. Alterations of gene expression of RB pathway in Opisthorchis viverrini infectioninduced cholangiocarcinoma. Parasitol Res. 2009;105:1273–81. 23. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–85. 24. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A. 1979;76:4350–4. 25. Boonjaraspinyo S, Wu Z, Boonmars T, Kaewkes S, Loilome W, Sithithaworn P, Nagano I, Takahashi Y, Yongvanit P, Bhudhisawasdi V. Overexpression of PDGFA and its receptor during carcinogenesis of Opisthorchis viverrini-associated cholangiocarcinoma. Parasitol Int. 2012;61:145–50. 26. Reis RM, Martins A, Ribeiro SA, Basto D, Longatto-Filho A, Schmitt FC, et al. Molecular characterization of PDGFR-alpha/ PDGF-A and c-KIT/SCF in gliosarcomas. Cell Oncol. 2005;27:319– 26. 27. Andrae J, Gallini R, Betsholtz C. Role of platelet-derived growth factors in physiology and medicine. Genes Dev. 2008;22:1276–312. 28. Badylak SF. Xenogeneic extracellular matrix as a scaffold for tissue reconstruction. Transpl Immunol. 2004;12:367–77. 29. Soma Y, Dvonch V, Grotendorst GR. Platelet-derived growth factor AA homodimer is the predominant isoform in human platelets and acute human wound fluid. FASEB J. 1992;6:2996–3001. 30. Betsholtz C, Lindblom P, Bjarnegard M, Enge M, Gerhardt H, Lindahl P. Role of platelet-derived growth factor in mesangium development and vasculopathies: lessons from platelet-derived growth factor and platelet-derived growth factor receptor mutations in mice. Curr Opin Nephrol Hypertens. 2004;13:45–52.

1802 31. Hoch RV, Soriano P. Roles of PDGF in animal development. Development. 2003;130:4769–84. 32. Karlsson L, Lindahl P, Heath JK, Betsholtz C. Abnormal gastrointestinal development in PDGF-A and PDGFR-(alpha) deficient mice implicates a novel mesenchymal structure with putative instructive properties in villus morphogenesis. Development. 2000;127:3457–66. 33. Martinho O, Longatto-Filho A, Lambros MB, Martins A, Pinheiro C, Silva A, Pardal F, Amorim J, Mackay A, Milanezi F, Tamber N, Fenwick K, Ashworth A, Reis-Filho JS, Lopes JM, Reis RM. Expression, mutation and copy number analysis of plateletderived growth factor receptor A (PDGFRA) and its ligand PDGFA in gliomas. Br J Cancer. 2009;101:973–82. 34. Ito M, Yoshida K, Kyo E, Ayhan A, Nakayama H, Yasui W, Ito H, Tahara E. Expression of several growth factors and their receptor genes in human colon carcinomas. Virchows Arch B Cell Pathol Incl Mol Pathol. 1990;59:173–8. 35. Liang ML, Ma J, Ho M, Solomon L, Bouffet E, Rutka JT, Hawkins C. Tyrosine kinase expression in pediatric high grade astrocytoma. J Neurooncol. 2008;87:247–53. 36. Wilczynski SP, Chen YY, Chen W, Howell SB, Shively JE, Alberts DS. Expression and mutational analysis of tyrosine kinase receptors c-kit, PDGFRalpha, and PDGFRbeta in ovarian cancers. Hum Pathol. 2005;36:242–9. 37. Dalal S, Berry AM, Cullinane CJ, Mangham DC, Grimer R, Lewis IJ, Johnston C, Laurence V, Burchill SA. Vascular endothelial growth factor: a therapeutic target for tumors of the Ewing’s sarcoma family. Clin Cancer Res. 2005;11:2364–78. 38. Dennis JE, Charbord P. Origin and differentiation of human and murine stroma. Stem Cells. 2002;20:205–14. 39. Satomura K, Derubeis AR, Fedarko NS, Ibaraki-O’Connor K, Kuznetsov SA, Rowe DW, Young MF, Gehron Robey P. Receptor tyrosine kinase expression in human bone marrow stromal cells. J Cell Physiol. 1998;177:426–38. 40. Coussens LM, Werb Z. Inflammation and cancer. Nature. 2002;420:860–7. 41. Eggert A, Ikegaki N, Kwiatkowski J, Zhao H, Brodeur GM, Himelstein BP. High-level expression of angiogenic factors is associated with advanced tumor stage in human neuroblastomas. Clin Cancer Res. 2000;6:1900–8. 42. Katano M, Nakamura M, Fujimoto K, Miyazaki K, Morisaki T. Prognostic value of platelet-derived growth factor-A (PDGF-A) in gastric carcinoma. Ann Surg. 1998;227:365–71. 43. Roberts WG, Whalen PM, Soderstrom E, Moraski G, Lyssikatos JP, Wang HF, Cooper B, Baker DA, Savage D, Dalvie D, Atherton JA, Ralston S, Szewc R, Kath JC, Lin J, Soderstrom C, Tkalcevic G, Cohen BD, Pollack V, Barth W, Hungerford W, Ung E. Antiangiogenic and antitumor activity of a selective PDGFR tyrosine kinase inhibitor, CP-673,451. Cancer Res. 2005;65:957–66. 44. George D. Targeting PDGF receptors in cancer—rationales and proof of concept clinical trials. Adv Exp Med Biol. 2003;532:141–51. 45. De Pas T, Toffalorio F, Colombo P, Trifiro G, Pelosi G, Vigna PD, Manzotti M, Agostini M, de Braud F. Brief report: activity of imatinib in a patient with platelet-derived-growth-factor receptor positive malignant solitary fibrous tumor of the pleura. J Thorac Oncol. 2008;3:938–41. 46. Grandal MV, Zandi R, Pedersen MW, Willumsen BM, van Deurs B, Poulsen HS. EGFRvIII escapes down-regulation due to impaired internalization and sorting to lysosomes. Carcinogenesis. 2007;28:1408–17.

Tumor Biol. (2012) 33:1785–1802 47. Hui EP, Lui VW, Wong CS, Ma BB, Lau CP, Cheung CS, Ho K, Cheng SH, Ng MH, Chan AT. Preclinical evaluation of sunitinib as single agent or in combination with chemotherapy in nasopharyngeal carcinoma. Invest New Drugs. 2010;29:1123–31. 48. Dong J, Grunstein J, Tejada M, Peale F, Frantz G, Liang WC, Bai W, Yu L, Kowalski J, Liang X, Fuh G, Gerber HP, Ferrara N. VEGF-null cells require PDGFR alpha signaling-mediated stromal fibroblast recruitment for tumorigenesis. EMBO J. 2004;23:2800–10. 49. Yu J, Ustach C, Kim HR. Platelet-derived growth factor signaling and human cancer. J Biochem Mol Biol. 2003;36:49–59. 50. Abouantoun TJ, Castellino RC, MacDonald TJ. Sunitinib induces PTEN expression and inhibits PDGFR signaling and migration of medulloblastoma cells. J Neurooncol. 2011;101:215–26. 51. Ping SY, Wu CL, Yu DS. Sunitinib can enhance BCG mediated cytotoxicity to transitional cell carcinoma through apoptosis pathway. Urol Oncol. 2010;in press. 52. Risau W, Drexler H, Mironov V, Smits A, Siegbahn A, Funa K, Heldin CH. Platelet-derived growth factor is angiogenic in vivo. Growth Factors. 1992;7:261–6. 53. Heldin P, Pertoft H, Nordlinder H, Heldin CH, Laurent TC. Differential expression of platelet-derived growth factor alpha- and beta-receptors on fat-storing cells and endothelial cells of rat liver. Exp Cell Res. 1991;193:364–9. 54. Zagai U, Fredriksson K, Rennard SI, Lundahl J, Skold CM. Platelets stimulate fibroblast-mediated contraction of collagen gels. Respir Res. 2003;4:13. 55. Heldin CH, Johnsson A, Wennergren S, Wernstedt C, Betsholtz C, Westermark B. A human osteosarcoma cell line secretes a growth factor structurally related to a homodimer of PDGF A-chains. Nature. 1986;319:511–4. 56. Corless CL, Schroeder A, Griffith D, Town A, McGreevey L, Harrell P, Shiraga S, Bainbridge T, Morich J, Heinrich MC. PDGFRA mutations in gastrointestinal stromal tumors: frequency, spectrum and in vitro sensitivity to imatinib. J Clin Oncol. 2005;23:5357–64. 57. Heinrich MC, Corless CL, Duensing A, McGreevey L, Chen CJ, Joseph N, Singer S, Griffith DJ, Haley A, Town A, Demetri GD, Fletcher CD, Fletcher JA. PDGFRA activating mutations in gastrointestinal stromal tumors. Science. 2003;299:708–10. 58. Xin H, Zhang C, Herrmann A, Du Y, Figlin R, Yu H. Sunitinib inhibition of Stat3 induces renal cell carcinoma tumor cell apoptosis and reduces immunosuppressive cells. Cancer Res. 2009;69:2506–13. 59. Yang F, Jove V, Xin H, Hedvat M, Van Meter TE, Yu H. Sunitinib induces apoptosis and growth arrest of medulloblastoma tumor cells by inhibiting STAT3 and AKT signaling pathways. Mol Cancer Res. 2010;8:35–45. 60. Shimizu I. Recent therapeutic developments in hepatic fibrosis— fibrogenesis: cellular and molecular basis. Georgetown: Landes Bioscienc; 2005. 61. Farooqi AA, Waseem S, Riaz AM, Dilawar BA, Mukhtar S, Minhaj S, et al. PDGF: the nuts and bolts of signalling toolbox. Tumour Biol. 2011;32:1057–70. 62. Fleming TP, Matsui T, Aaronson SA. Platelet-derived growth factor (PDGF) receptor activation in cell transformation and human malignancy. Exp Gerontol. 1992;27:523–32. 63. Shao ZM, Nguyen M, Barsky SH. Human breast carcinoma desmoplasia is PDGF initiated. Oncogene. 2000;19:4337–45. 64. Skobe M, Fusenig NE. Tumorigenic conversion of immortal human keratinocytes through stromal cell activation. Proc Natl Acad Sci U S A. 1998;95:1050–5.