Proteomic Identification of Paclitaxel-Resistance Associated hnRNP A2 and GDI 2 Proteins in Human Ovarian Cancer Cells Dong Hyeon Lee,†,¶ Kwanghoe Chung,‡,¶ Ji-Ae Song,§ Tae-heon Kim,§,| Haeyoun Kang,| Jin Hyong Huh,| Sang-geun Jung,⊥ Jung Jae Ko,∇ and Hee Jung An*,§,| Department of Physiology, Department of Biochemistry, Institute for Clinical Research, Department of Pathology, Department of Gynecologic Oncology, College of Medicine, and College of Life Science, CHA University, Sungnam, South Korea Received May 15, 2010
Ovarian cancer is a gynecological malignancy with the highest mortality. Chemoresistance is an important subject for the treatment of ovarian cancer, because obtaining significant drug resistance to the first line chemotherapy, paclitaxel, causes major therapeutic obstacles. It is essential to improve the survival rate of ovarian cancer patients by mining the biomarkers indicating the drug resistance and prognosis, and by further understanding underlying mechanisms of drug resistance. In the present study, we established paclitaxel-resistant subline (SKpac) from human epithelial ovarian cancer cell line, SKOV3, and performed comparative analysis of whole proteomes between paclitaxel-resistant SKpac sublines and paclitaxel-sensitive parental SKOV3 cells to identify differentially expressed proteins and useful biomarkers indicating chemoresistance. Proteins related to chemoresistant process were identified by two-dimensional gel electrophoresis (2DE) with mass spectrometry (MALDI-TOF and LC-MS/MS). Eighteen spots were differentially expressed and were identified in SKpac chemoresistant cells compared to SKOV3. The expressions of ALDH 1A1, annexin A1, hnRNP A2, and GDI 2 proteins were validated by Western blot, which was consistent with proteomic analysis. Among the selected proteins, downregulation of hnRNP A2 and GDI 2 was found to be the most significant finding in SKpac cells and chemoresistant ovarian cancer tissues. Our results suggest that hnRNP A2 and GDI 2 may represent potential biomarkers of the paclitaxel-resistant ovarian cancers for tailored cancer therapy. Keywords: paclitaxel • drug resistance • ovarian cancer • proteomics • biomarker • hnRNP • GDI2
Introduction Ovarian cancer is the leading cause of death from all gynecological malignancies because of their late diagnosis and high recurrence rate. Ovarian cancer locates deep in the pelvis, grows rapidly, presents nonspecific major symptoms in the early stage, and lacks form of effective screening; therefore, 70% of the patients with ovarian cancer are diagnosed at an advanced disease stage and have low survival rate.1 The cytoreductive surgery followed by adjuvant taxane-based chemotherapy, such as paclitaxel, is the current standard and effective treatment for the advanced ovarian cancer.2 Even with this multimodality treatment strategy, many patients will have recurrent ovarian cancer and ultimately die from their diseases, * Corresponding author: Hee Jung An, M.D., Ph.D., Department of Pathology, College of Medicine, CHA University, 351 Yatap-dong, Seongnam Si Bundang-gu, Gyeonggi-Do, Republic of Korea, 463-712. Tel: 82-31-7805439. Fax: 82-31-780-5476. E-mail:
[email protected]. † Department of Physiology. ‡ Department of Biochemistry. ¶ These two authors have equally contributed to this work. § Institute for Clinical Research. | Department of Pathology. ⊥ Department of Gynecologic Oncology. ∇ College of Life Science.
5668 Journal of Proteome Research 2010, 9, 5668–5676 Published on Web 09/21/2010
and one of major reasons is the development of chemoresistance of cancer cells to anticancer drugs. Paclitaxel is a kind of antimicrotubule agent, stabilizing polymerized microtubules, and is currently used as first-line chemotherapy for the treatment of ovarian, breast, and nonsmall cell lung cancers.3 The drug-resistance against paclitaxel is one of the important subjects in the chemoresistance of ovarian cancer. Paclitaxel-resistance can be mediated by the overexpression of P-glycoprotein and altered expression or post-translational modification of β-tubulin or other microtubule regulatory proteins.4 However, overall molecular mechanisms and markers for paclitaxel-resistance remain to be investigated. Biomarkers indicating drug-resistance is very important to guide the most appropriate chemotherapy and improve the survival rate as well as to understand the underlying mechanism developing the chemoresistance. The biomarkers for detecting tumor resistance to chemotherapy have been investigated using cDNA microarrays in ovarian cancers;5,6 however, the expression of the genes may be inconsistent with protein due to post-translational modification. Therefore, the high-throughput studies using proteomics have recently been performed for identifying the biomarkers for early detection7 and chemoresistance.8-10 2DE coupled with MS is a widely used efficient method for large-scale protein 10.1021/pr100478u
2010 American Chemical Society
Paclitaxel-Resistance Associated Proteins in Ovarian Cancer expression analysis and quantifying various proteins in different samples. The development of protein biomarker for chemoresistance of ovarian cancer is necessary to validate with ovarian cancer tissues in vivo. However, most proteomic studies searching for chemoresistant biomarkers employed several cultured cell lines only. In terms of this issue, further validation is necessary for candidate protein biomarkers in the chemoresistant ovarian cancer tissues as well as in ovarian cancer cell lines. The aim of the present study was to identify proteins associated with paclitaxel resistance as valuable biomarkers for chemoresistance of ovarian cancers. In this study, we produced 4 different paclitaxel-resistant sublines (SKpacs) from a human epithelial ovarian cancer cell line, SKOV3, and performed the comparative analysis of whole proteomes between parental SKOV3 cells and paclitaxel-resistant SKpac sublines using 2DE coupled with MS analysis. Proteins differentially expressed in paclitaxel-sensitive and paclitaxel-resistant ovarian cancer cells are considered to be involved in the development of chemoresistance, and the candidate proteins were further confirmed not only in chemoresistant cell lines but also in chemoresistant ovarian cancer tissues to identify potential biomarkers for chemoresistance of ovarian cancer. In the present study, we utilized serous type carcinoma cell lines and tissues, because serous carcinoma is the most common type of ovarian cancer and accounts for approximately 50% of ovarian carcinomas.
Material and Methods Generation of Paclitaxel-Resistant Cell Line and Cell Cultures. Human epithelial ovarian cancer cell line SKOV3 was purchased from American Type Culture Collection (ATCC, Manassas, VA). Paclitaxel-resistant sublines (SKpacs) were generated in our laboratory by exposing cells to stepwise concentrations of paclitaxel from IC50 10% to 1000% for 12 months, and 4 different sublines (Skpac-1,2,3,4) were produced. All cell lines were cultured in MaCoy’s 5A media containing fetal bovine serum (10%) and antibiotics-antimycotics (Invitrogen, Carlsbad, CA) in a humidified incubator (5% CO2, 37 °C). These cell lines grew in a monolayer and were passaged when cultures were 70-80% confluent. Drug Sensitivity Assay. The sensitivity to paclitaxel was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-dyphenyltetrazolium bromide (MTT) assay. Briefly, cells were diluted with culture medium to the seeding density of 1 × 104 cells/well, plated on 96-well tissue culture plates, and incubated at 37 °C overnight. The next day, the cells were incubated with various concentrations (0.15-80 nM) of paclitaxel. After incubation for 72 h, 10 µL of MTT solution (5 mg/mL) (Sigma, St. Louis, MO) was added to each well, and the plates were incubated for another 3 h. Following incubation, 100 µL of dimethylsulfoxide (Sigma, St. Louis, MO) was added to each well to solubilize the MTT formazan product. Absorbance at 550 nm was measured with a microplate reader (Chemila, Labotech, Ochten). Growth inhibition was calculated as the percentage of viable cells compared with untreated cultures. Experiments were done in triplicate. Sample Preparation and 2DE. SKOV3 and Skpac-1,2,3,4 were prepared for 2DE as described by Chung et al.7 In brief, the collected samples were suspended in 50 mM Tris buffer containing 7 M urea, 2 M thiourea, 4% (w/v) CHAPS, and protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). The lysates were homogenized and centrifuged at 12 000g for 15 min. Benzonase (Sigma, St. Louis, MO) was
research articles added to the mixture and stored at -80 °C until use. For 2DE analysis, nonlinear Immobiline DryStrip pH 3-10, 18 cm (GE Healthcare Bio-Sciences Corp., Uppsala, Sweden) were rehydrated in swelling buffer containing 7 M urea, 2 M thiourea, 0.4% (w/v) dithiothreitol (DTT), and 4% (w/v) CHAPS. The protein lysates were loaded into the rehydrated IPG strips using a Multiphor II apparatus (GE Healthcare Bio-Sciences Corp.) for a total of 57 kVh. The 2DE separation was performed on 8-16% (v/v) gradient SDS-polyacrylamide gels. Following fixation of the gels in a solution of 40% (v/v) methanol containing 5% (v/v) phosphoric acid, the gels were stained with Colloidal Coomassie Blue G-250 solution. The gels were destained in 1% (v/v) acetic acid and then imaged using a GS710 imaging calibrated densitometer (Bio-Rad Laboratories, Hercules, CA). Protein spot detection and 2DE pattern matching were carried out using ImageMaster 2D Platinum software (GE Healthcare Bio-Sciences Corp.). To ensure the reproducibility of 2DE experiments, each sample was analyzed four times. In-Gel Digestion with Trypsin and Extraction of Peptides. The procedure for in-gel digestion of protein spots from Coomassie Blue-stained gels was performed. In brief, each protein spot was excised from stained gels and cut into pieces. The gel pieces were washed in 25 mM ammonium bicarbonate buffer, pH 7.8, containing 50% (v/v) acetonitrite (ACN). Following the dehydration of gel pieces in a SpeedVac concentrator for 10 min, gel pieces were rehydrated in sequencing grade trypsin solution (Promega Co., Madison, WI). After incubation in 25 mM ammonium bicarbonate buffer, pH 7.8, at 37 °C overnight, the tryptic peptides were extracted with 0.5% trifluoroacetic acid (TFA) containing 50% (v/v) ACN with sonication. The extracted solution was reduced to 1 µL in a vacuum centrifuge. Prior to mass spectrometric analysis, the resulting peptides solution was subjected to a desalting process using a reversedphase column. A constricted GEloader tip (Eppendorf, Hamburg, Germany) was packed with Poros 20 R2 resin (PerSpective Biosystems, MA). After an equilibration step with 5% (v/v) formic acid, the peptides solution was loaded on the column and washed with 5% (v/v) formic acid. The bound peptides were eluted with R-cyano-4-hydroxycinnamic acid (CHCA) (5 mg/mL in 50% (v/v) ACN/5% (v/v) formic acid). Analysis of Peptides Using MALDI-TOF MS and LC-MS/ MS, and Identification of Proteins. Mass measurement of tryptic peptides was performed with a Voyager-DE STR mass spectrometer (PerSpective Biosystems) in reflectron positive ion mode as described by Bahk et al.11 Close external calibration was performed for every four samples with calibration mixtures of adrenocorticotropic fragment 18-39, Neurotensin, and Angiotensin I as standard calibrants. Mass spectra were acquired for the mass range of 900-3500 Da. The proteins were identified by peptide mass fingerprinting (PMF) searching, against the NCBI databases, using the search program MASCOT (http://www.matrixscience.com). Only significant hits as defined by program were considered initially with at least 4 matching peptide masses. (protein scores >66 and p-value 50 and p-value
