| OS is the most common primary malignant tumor of bone in children and adolescents. Due to the advances in diagnostic modalities and treatment methods, the survival rate (5-year disease-specific survival rates,67%-70%) and postoperative limb function of patients with OS has been improved. However, for patients with metastatic disease at diagnosis or with tumors showing a poor response to chemotherapy are still unsatisfactory (5-year disease-specific survival rates,20%-40%), even with dose-intensive or high-dose chemotherapy. So it is important to develop new targeted therapeutic strategies based on OS-specific proteins and find more biomarkers for diagnosis of this cancer.At present, comparative proteomics provide a powerful approach in screening for alterations in protein levels and post-translational modifications that are associated with tumors and has culminated in the identification of many potential new therapeutic targets and an abundance of cancer-related biomarkers. However, global proteomic profiling of human OS are developed very late and slowly. To our knowledge, only a few papers have reported comparative proteome research in OS. In these researches, tissue and cell lines were used. But due to the complexity and difference of proteome, low copy proteins and membrane proteins were usually undectected in whole cell or tissue, so recently many proteomic investigations have focused on subcellular compartments. The plasma membrane (PM) is an organized system serving as a structural and communication interface for exchanges of information and substances with the extracellular environment. The proteins on the PM act as’doorbells’and’doorways’playing crucial roles in cell function including intercellular communication, cellular development, cell migration, and drug resistance. So it is important to systematically study the PM proteins involved in OS.Material and MethodsCell culture Human osteosarcoma cell lines MG-63, and the normal osteoblastic cell line hFOB1.19(expressing SV40large T antigen) were originally obtained from the American Type Culture Collection (Manassas, VA, USA). The cell lines were cultured according to our reported with a little modification. Briefly, the cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM) from GIBCO with10%FBS. For hFOB1.19, a1:1Ham’s F12medium was added to DMEM without phenol red and with2.5mmol/L L-glutamine and0.3mg/mL G418. All cultures were maintained in10cm diameter plates in a humidified atmosphere of5%CO2at37℃. MG-63cells were passaged every2to3d, and hFOB1.19cells were passaged every4to5d. Furthermore, the same Ham’s F12medium and DMEM with10%FBS without G418was used to incubate the two cell lines for the last48hours before harvesting them for membrane extraction to exclude any unexpected affect that the difference between culture medium will cause. About108cells were collected and used for PM separation.Patient tissue samplesAll patient tissue and clinical information was collected with patient consent after permission by Ethics Committee of SMMU. Eleven archival sections of formalin-fixed, paraffin-enbedded primary osteosarcoma and their respective adjacent nontumorous tissue were collected.Preparation of OS PMsThe PM was isolated as previously described. All steps were carried out at4℃. Briefly, adherent cells (108) were washed three times with PBS, scraped using a plastic cell lifter, and broken in1mL solution containing0.2mM EDTA, lmM NaHCO3using a glass homogenizer. The nuclear and unbroken cells were removed through200g, the supernatant was collected, and centrifuged for0.5h at25000rpm. The cell pellets were resuspended in1mM NaHCO3in an approximate ratio of1ml per5×108cells and used for PM separation by two-phase systems.2g of suspended cell pellets was added to the top of14g of the dextran-poly (ethylene glycol) mixture (6.6%Dextran T500,6.6%PEG3350.0.2M K3PO4, pH7.2). After mixed for40times, the tube was centrifuged for5min at750g. The PM-enriched upper phase was collected and purified again as before. The upper phase was diluted5-fold with1mM sodium bicarbonate, and centrifugated at100000g for2h in a SW32rotor. The pellets were collected and used for purification check and proteomics. Protein Digest, iTRAQ Labeling, and Strong Cation Exchange Fractionation iTRAQ labeling was done according to the kit protocol (Applied Biosystems Inc., Foster City, CA) and the previously reported. Protein (100μg) from the PM of MG-3and hFOB1.19cell lines was acetone precipitated overnight at-20℃and resuspended in30μL iTRAQTM Dissolution Buffer (ABI, Foster City, USA). After reduction and alkylation, proteins solutions were digested overnight at37℃with sequence grade modified trypsin (Promega)(1:10). The peptides were pooled, desalted with Sep-Pak Cartridge (Waters) and fractionated by Strong Cation Exchange (SCX) chromatography on an Ultimate HPLC system (LC Packings) using a Column (5μm,300A,0.5×23mm,Waters). Peptides were eluted with a linear gradient of0-500mM KC1(25%v/v acetonitrile,10mM KH2PO4, pH2.8) for60min at a flow rate of200μl/min.15fractions were collected. The iTRAQ experiments were carried out twice:the first experiment compared MG-63cells (115reporter ions) and hFOB1.19(116reporters), while the second experiment was performed using MG-63(115reporters), hFOB1.19(114reporters)LC-MS AnalysisEach SCX fraction was dried down, dissolved in0.1%formic acid, and analyzed on Qstar PulsarTM (Applied Biosystems-MDS Sciex). Peptides were separated on a reverse-phase column packed with ZORBAX300SB-C18enrichment column (5μm,300A,0.5×23mm, Waters) and separated by a75-μm-internal diameter PepMap RP column from LC Packings packed with3-μm C18beads with100-A pores. Buffer A:5%ACN,95%water,0.1%FA and Buffer B:95%ACN,5%water,0.1%FA. The flow rate used for separation on the RP column was400nl/min with gradient5%-45%during90min. MS data was acquired automatically using Analyst QS1.0software Service Pack8(ABI/MDS SCIEX, Concord, Canada). An analysis survey scans were acquired from400-1800with up to6precursors selected for MS/MS from m/z100-2000. The two most intense peaks over30counts, with a charge state2-4were selected for fragmentation. Curtain gas was set at10, nitrogen was used as the collision gas, and the ionization tip voltage was4000V.Data analysisRatios of the114.1.115.1and116.1amu signature mass tags generated upon MS/MS fragmentation from the iTRAQTM-labeled tryptic peptides were calculated using Protein Pilot (ABI,USA)(version2.0.1)(ABI) in Analyst. The MS and MS/MS tolerances were set to0.2Da. The IPI databases was used for searching iTRAQTM-identified peptides. Methyl methanethiosulphonate modification of cysteines was used as a fixed modification, and one missed tryptic cleavage was allowed. All proteins identified must have≥95%confidence and the protein confidence threshold cutoff was set to1.3(unused) with at least more than one peptide above the95%confidence level. The true value for the average ratio was expressed as an error factor (EF=1095%confidence interval) and calculated according to the reports. An EF<2was set for the quantification quality to be satisfied. In addition, a p-value<0.05was significant for protein quantification. To designate significant changes in protein expression, fold-changes>1.5or<0.66were set as cutoff values. The peptide and proteins were exported, and saved as excel files.BioinformaticsThe theoretical isoelectric point (pI) and molecular weight (MW) and Grand average of hydropathicity (GRAVY) were calculated through inhouse developed software. The subcellular location and function of the identified proteins were elucidated by UniProt knowledgebase (Swiss-Prot/TrEMBL) and Gene Ontology Database. The mapping of putative transmembrane helices (TMHs) in the identified proteins was carried out using the transmembrane hidden Markov model (TMHMM) algorithm, available at http://www.cbs.dtu.dk/services/TMHMM. A protein-protein interaction network was done by STRING software through inputting IPI number (http://string.embl.de)ResultsProduction and characterization of PM derived from cell linesDue to cell lines--MG-63and hFOB1.19with different survival characteristics, the cells were cultured with similar nutrition. In order to decrease the difference caused by culture nutrition, the cells were cultured for at least24hours in the same nutrition before they were collected. About108cells were used for PM separation. In this work, a combination of differential centrifugation and aqueous two-phase partitioning was used to separate plasma membrane. Two phases were obtained after centrifugation, including upper and down-phase. Fractions containing PM were determined based on the enrichment of a PM marker enzyme--Na+/K+ATPase and the decrease of a mitochondrial marker-Prohibitin. PM was enriched in upper phase through comparing the signal strength of Nor U and MG U with that of Nor D and MG D. According to the results analyzed by Image J software (http://rsb.info.nih.gov/ij), PM was enriched for11.2or15.3-fold in upper phase and6.5or3.9-fold in down-phase in hFOB1.19or MG-63cell line compared with homogenizations. While mitochondrial was increased for1.2or1.4-fold in upper phase and6.9-or6.3-fold in the down phase in hFOB1.19or MG-63cell line. Basically the purification was successful in two cell lines.Identification of differentially expressed proteinsDue to2D-LC-MS/MS method providing a powerful alternative to gels especially in hydrophobic proteins, newly developed iTRAQ technique was used to compare protein expression between MG-63and hFOB1.19cells. After duplicate LC-MS/MS analysis,342proteins were quantified, which had p-values>95%confidence level(ProtScore>1.3)and at least more than one peptide above the95%confidence level.60of them were detected in both independent experiments.66%(226of342) proteins were identified by more than5peptides,10%(36out of342) by4peptides;11%(38out of342) by3peptides,8%(27out of342) by2peptides and only5%(17out of342) by one peptide.According to the following criteria:(1) cutoff iTRAQ ratios of fold-change for protein expression were>1.5for up-regulation and<0.66for down-regulation;(2) A protein had to be quantified with at least three spectra (allowing generation of a p-value), a p-value<0.05;(3) An EF<2was set for the quantification quality to be satisfied. A total of63differentially expressed proteins were detected in the two experiments. In the first experiment,22proteins were found to be changed for more than1.5-fold, including6proteins up-regulated and16down-regulated with a p-value<0.05. In the second,47proteins were changed,6proteins up-regulated and41down-regulated. Of which,6proteins were identified by the two experiments including2up-regulated proteins and4down-regulated. Representative MS/MS spectra for three peptides identified from ITGβ1antigen are detected. Consistent changes were found in the three peptides. Almost total y or b ions were detected in the sequence.Bioinformatic analysis of differentially expressed proteinsAccording to the annotations from UniProt knowledgebase (Swiss-Prot/TrEMBL) and Gene Ontology Database.69.8%(44of63) differential proteins located in plasma membrane including proteins annotated as membrane, Single-pass type I membrane protein anchored to membrane, intermediate filament, actin cytoskeleton. microtubule, cell-cell adherens junction as well as plasma membrane. There are also proteins located in other subcellular organelles such as nucleus (12.7%), cytoplasm (6.3%), endoplasmic reticulum (ER)-Golgi intermediate (1.6%). There are also6proteins not to be annotated by GO and UniProt database.37%of these differential proteins are transmembrane proteins, including3proteins with more than10TMHs,1protein with4TMHs,4proteins with2, and11proteins with1TMH. Furthermore, we simply analyzed the protein function annotated by GO database, and found proteins involved the following biological processes more frequently changed:binding (30.8%), cell structure (8.8%), and signal transduction (7.8%)As we known, cancer is caused by many proteins, so in this work, bioinformatics analysis was performed to elucidate the network between the identified differential proteins. A wide protein-protein interaction network was affected in OS patients, for example, the network of ITGβ1-ITGA2-ITGβ1-CTNNB1-CTNND1. According to the information in String interaction database (http://string.embl.de), these proteins have important function in signal transportation, cell adhesion, etc. ITGβ1(CD151antigen, up-regulated in OS) is involved in T-cell adhesion processes and spontaneous rosette formation with erythrocytes. ITGA2(Integrin alpha-2precursor, down-regulated in OS) is a receptor for laminin, collagen, collagen C-propeptides, fibronectin and E-cadherin. ITGβ1(Integrin beta-1precursor, down-regulated in OS) is a receptor for collagen. CTNNB1(Catenin beta-1, down-regulated in OS) is involved in the regulation of cell adhesion and in signal transduction through the Wnt pathway. CTNND1(catenin delta-1, down-regulated in OS) binds to and inhibits the transcriptional repressor ZBTB33, which may lead to activation of target genes of the Wnt signaling pathway. In order to further verify our proteomic results and offer some help for OS diagnosis and treatment, ITGβ1antigen was selected for further research for their plasma membrane location, protein binding and signal transportation function, and multiple interactions in the network.ConclusionsIn summary, in this work, we reported a subcellular proteomic research combining PM purification. iTRAQ label. LC-MS separation and identification in OS cell lines with clinical verification in patient’s tissues and plasma. Many PM proteins with several TMs were detected to be differentially expressed. ITGβ1were identified and verified to be a potential biomarker of OS and target for therapy in OS. Our research might offer some clue to understand the mechanism of OS progress and offer novel biomarkers for OS diagnosis and treatment. |
PDF Full Text Request