RESEARCH ARTICLE

Comparative proteomic analysis of fibrosarcoma and skinfibroblast cell lines

Ogunc Meral & Hamdi Uysal

Received: 25 August 2014 /Accepted: 23 September 2014# International Society of Oncology and BioMarkers (ISOBM) 2014

Abstract Comparative proteomic analysis of normal andcancer cell lines provides for a better understanding ofthe molecular mechanism of cancer development and isessential for developing more effective strategies fornew biomarker or drug target discovery. The purposeof this study is to compare protein expression levelsbetween fibrosarcoma and fibroblast cell lines. In ourstudy, two-dimensional polyacrylamide gel electrophore-sis (2-D PAGE) and liquid chromatography coupledwith tandem mass spectrometry (LC-MS/MS) techniqueswere carried out to compare the protein profile betweenfibrosarcoma and fibroblast cell lines. We prepared celllysate samples to analyze intracellular proteins andsecretome samples to analyze extracellular proteins inboth cell lines. Our results revealed 13 upregulatedproteins and 1 downregulated protein of which all ofthem identified in fibrosarcoma cell line after the com-parison with fibroblast cell line cell lysates. When com-paring secretome profiles of both cell lines, we foundand identified 13 proteins only expressed in fibrosarco-ma cell line. These identified proteins have commonfunctions such as cell proliferation, cell differentiation,invasion, metastasis, and apoptosis in cancer. The dataobtained from this study indicates that these proteinshave importance on understanding the molecular mech-anism of fibrosarcoma. These proteins may serve ascandidate biomarkers and drug targets for future clinicalstudies.

Keywords 2-D PAGE . Fibrosarcoma . Fibroblast . LC-MS/MS . Proteomics

Introduction

Adult fibrosarcoma was defined by the World Health Organi-zation (WHO) in 2002 as a “malignant tumor, composed offibroblasts with variable collagen production and, in classicalcases, a herringbone architecture” [1]. Fibrosarcomas are raremalignant mesenchymal tumors originating from fibroblasts.The characteristic appearance of these tumors is immatureproliferating fibroblasts or undifferentiated anaplastic spindlecells in a storiform pattern [2]. In fibrosarcoma, fibroblasts arespindle-shaped with elongated pointed nuclei and cytoplasm,and intercellular collagen fibers and fibroblasts are commonlyarranged in interlaced bands [3].

Fibrosarcoma diagnosis is difficult to determine, and likemost other types of sarcomas, fibrosarcoma causes no char-acteristic symptoms [4]. These tumors usually occur in softtissues like muscles, connective tissue, blood vessels, or lipidtissues [2], but mostly, the tumor can occur in the deep softtissues of the lower extremities, followed by the upper ex-tremities and trunk. In fibrosarcoma, metastasis most com-monly occurs by way of the bloodstream, and the lung is themain metastatic site, followed by the skeletal system [4].Fibrosarcoma which is the least differentiated type of mesen-chymal malignancy and defined as a spindle cell malignantneoplasm has the least heterogeneous characteristics amongthe sarcomas [5]. To date, no recognizable cytogenetic ormolecular characterization has been done in fibrosarcoma [6].

Proteins secreted from cancer cells are essential in cellproliferation, cell differentiation, invasion, metastasis, andangiogenesis of cancer by regulating cell-cell and cell-extracellular matrix interactions. These proteins can also bemeasured in body fluids such as blood or urine [7]. Their rolein cancer makes them a good target and source fortherapeutical and drug-based intervention and potential bio-marker which is important for diagnosis and prognosis ofcancer. In recent years, most studies focused on the

O. Meral :H. Uysal (*)Faculty of Veterinary Medicine, Department of Biochemistry,Ankara University, 06110, Ankara, Turkeye-mail: huysal@veterinary.ankara.edu.tr

Tumor Biol.DOI 10.1007/s13277-014-2672-8

characterization of secreted proteins from neoplastic tissues orcancer cell lines in order to discover new novel biomarkers[8].

Comparative proteomic analysis of cancer cells and normalcells are widely determined by differential protein expression,and studying the mechanism of tumorigenesis at a proteomelevel is powerful a approach to understanding the underlyingmechanism of tumorigenesis [9]. Comparative proteomicstudies using 2-D gel electrophoresis coupled with mass spec-trometry has been applied in various cancers, including lungcancer [10], gastric cancer [11], breast cancer [12], and pros-tate cancer [13].

In this study, different protein expressions between humanskin fibroblast cell line (CCD-1135Sk) and human fibrosar-coma cell line (HT-1080) were studied by two-dimensionalpolyacrylamide gel electrophoresis (2-D PAGE) and liquidchromatography coupled with tandem mass spectrometry(LC-MS/MS) techniques. We compared intracellular proteinsby preparing lysate samples and extracellular proteins bypreparing secretome samples between fibrosarcoma and fibro-blast cell lines. The aim of the present study was to understandthe molecular mechanism of fibrosarcoma and investigatepotential biomarkers for fibrosarcoma.

Materials and methods

Cell culture

Human skin fibroblast cell line CCD-1135Sk and humanfibrosarcoma cell line HT-1080 were purchased from Ameri-can Type Culture Collection (ATCC). Cell lines were culturedin DMEM-F12 medium containing 10 % fetal bovine serum(FBS) and 50 mg/l gentamicin. Both cell lines were incubatedat 37 °C in 5 % CO2 atmosphere.

Preparation of samples

Preparation of secretome samples

To obtain secretome samples, cells were grown to 70 %confluence and washed with phosphate-buffered saline(PBS) three times before incubation in serum-free mediumfor 24 h. After incubation, the medium was harvested andfiltered through a membrane filter (0.45 μm) to remove celldebris or dead cells. Afterwards, the medium was enrichedand concentrated by trichloroacetic acid (TCA) precipitationmethod. For this, TCAwas added to serum-free medium to afinal concentration of 20 %, and the solution was incubatedovernight at 4 °C. The precipitate was collected by centrifu-gation (15,000 rpm, 15 min, 4 °C), and then the supernatantwas carefully removed. The pellet was washed with ethanolthree times. After washing with ethanol, the pellet was dried at

room temperature, resuspended in lysis buffer (7 M urea, 2 Mthiourea, 4 % 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic (CHAPS), 2 % dithiothreitol (DTT), 2 %IPG buffer; pH 3–10), and then centrifuged (15,000g,30 min, 4 °C), and the supernatant including secreted proteinswas collected.

Preparation of lysate samples

To obtain lysate samples, cells were grown to 70% confluenceand harvested with PBS-EDTA solution. Cells were collectedby centrifugation (3000 rpm, 10 min) and washed with PBS(x2) and sucrose solution. Afterwards, the supernatant wascarefully removed and cells were lysed in the lysis buffer (7Murea, 2 M thiourea, 4 % CHAPS, 2 % DTT, 2 % IPG buffer;pH 3–10) and then centrifuged (15,000g, 30 min, 4 °C), andthe supernatant was collected.

Cell viability assay

After incubating the cells in complete media or serum-freemedia for 24 h, viable cells and dead cells were determined bytrypan blue dye exclusion method. The percentage of cellviability was expressed as the ratio of the total viable cells tothe sum of the total viable and dead cells.

Determination of protein concentration

The protein concentration was determined by a dye-metal-based colorimetric protein assay [14]. Commercially availablePierce 660 nm protein assay reagent (Pierce/Thermo Scientif-ic, Rockford, IL) which is not affected by the levels of thereducing agents was used, and results are reported in micro-grams per microliter.

Two-dimensional gel electrophoresis

Equal amount of proteins (250 μg proteins for lysate samples,75 μg proteins for secretome samples) diluted with rehydra-tion buffer (7 M urea, 2 M thiourea, 4 % CHAPS, 1 % DTT,0.5 % IPG buffer; pH 3–10, 10 % glycerol) and applied to a13-cm pH 3–10 IPG (GE Healthcare, USA) strip. Passiverehydration was performed overnight at room temperature.After passive rehydration, first dimensional separation of pro-teins was performed using 50 V for 4 h, 500 V for 6 h, 1000 Vfor 1 h, and 6000 V for 5 h. After isoelectric focusing, stripswere first equilibrated with 1 % DTT in equilibration buffer(6 M urea, 30 % glycerol, 2 % sodium dodecyl sulfate (SDS),50 mM Tris-HCl, pH 8.8) and then with 2.5 % iodoacetamidein the same buffer. The equilibrated strips were placed on topof the prepared 12 % SDS polyacrylamide gels and sealed inplace with 0.5 % low-melt agarose. Second dimensional sep-aration of proteins was performed at a current of 10 mA/gel

Tumor Biol.

for 30 min and 20 mA/gel for about 6 h. After 2-D electro-phoresis, gels were stained with colloidal Coomassie G-250.Triplicate gels were made for each sample.

Analysis of gel images

Stained gels were scanned using a densitometer (Bio-Rad,GS-800) and analyzed by PDQuest 2-DE analysis softwarev7.4.0 (Bio-Rad) as described in the manufacturer’s manual.

In-gel trypsin digestion

Spots corresponding to the proteins of interest were excisedfrom the 2-DE gels and washed with acetonitrile until all tracesof colloidal Coomassie G-250 were removed. Reduction andalkylation of proteins were performed by 10mMDTT (30min,80 °C) and 55 mM iodoacetamide (20 min, in the dark),respectively, both in 50 mM NH4HCO3. Gel particles werewashed with 50 mM NH4HCO3 and then incubated with thedigestion solution (20 ng/μl of trypsin in 50 mM NH4HCO3).

LC-MS/MS analysis

The trypsinized proteins were identified by liquidchromatography-electrospray ionization-quadrupole time-of-fl ight mass spectrometry (LC-ESI-qTOF) system(nanoACQUITY ultra-pressure liquid chromatography(UPLC) and SYNAPT high-definition mass spectrometerwith NanoLockSpray ion source, Waters). Firstly, peptideswere trapped on a nanoACQUITY UPLC Symmetry C18Trap column at 5 μl/min of flow rate; afterwards, peptideswere eluted from the trap column by gradient elution onto ananalytical column (nanoACQUITYUPLC BEH C18,Waters)at 300 nl/min of flow rate with a linear gradient from 5 to 35%acetonitrile. Parallel collision-induced dissociation mass spec-trometry (MSE) was done at positive ion V mode, applyingthe MS and MS/MS functions over 1.5-s intervals with 6 Vlow-energy and 15–40 V high-energy collusions. Mass driftwas corrected by infusing glu-fibrinopeptide (500 pmol/μl)every 45 s through the NanoLockSpray ion source at 300 nl/min of flow rate. Peptide signal data between 50 and 1600m/zwere collected.

LC-MS/MS data processing

Low collision energy was performed for peptide scan andmassmeasurement, and higher collision energywas employedto retrieve the data for the peptide sequence. ProteinLynxGlobal Server v2.4 software (PLGS) (Waters Corp., Milford,MA) was used to extract tandem mass spectra, deisotoping,and charge state deconvolution as well. The Apex3D dataacquisition parameters were configured with the minimumrequirements for a 0.2-min chromatographic peak width,

10,000MS TOF resolution, 150 counts for low energy thresh-old, 50 counts for elevated energy threshold, and 1200 countsfor the intensity threshold. Protein sequence database fromUniprot was used. Databank search query was set and limitedto the following: minimum three fragment ion matches perpeptide, minimum seven fragment ion matches per protein,minimum one peptide match per protein, and one missedcleavage. Carbamidomethyl cysteine was selected as fixedmodification, and for the variable modifications, acetyl N-terminal, deamidation of asparagine and glutamine, and oxi-dation of methionine were chosen. The false-positive rate ofthe IdentityE algorithm was found to be around 3–4 % using arandomized database which is five times larger when com-pared to the original one.

Result

Cell viability (trypan blue exclusion)

The effect of serum-free medium incubation (24 h) on theviability of the human skin fibroblast (CCD-1135Sk) andhuman fibrosarcoma (HT-1080) cell lines was tested by trypanblue exclusion method, and the results showed that this treat-ment has no significant effect on cell viability (Fig. 1).

2-D PAGE

Two-dimensional PAGE was performed three times for eachsample. Two-dimensional gels from lysate samples displayed

Fig. 1 Viability of human skin fibroblast and human fibrosarcoma cellscultured in complete media or serum-free media for 24 h. Cell viabilitywas determined by trypan blue exclusion method. Data are expressed asmean±SD of from three independent experiments

Tumor Biol.

221±18 and 455±30 protein spots in human skin fibroblastand human fibrosarcoma lysate groups, respectively. In theanalysis of gels between human fibrosarcoma and human skinfibroblast cell line according to the PDQuest 2-DE analysissoftware (Bio-Rad), 13 proteins were over-expressed and 1protein was under-expressed in human fibrosarcoma cell line(P>0.05) (see Figs. 2 and 3).

Two-dimensional gels from secretome samples displayed94±7 and 118±5 protein spots in human skin fibroblast andhuman fibrosarcoma secretome groups, respectively. In theanalysis of gels between human fibrosarcoma and human skinfibroblast cell line according to the PDQuest 2-DE analysissoftware (Bio-Rad), 13 proteins were only expressed in hu-man fibrosarcoma cell line. All these proteins have at least 30-fold higher density than the background (see Figs. 4 and 5).The differentially expressed proteins were selected and num-bered by the PDQuest 2-DE analysis software.

Protein identification

Fourteen significant protein spots from lysate samples and 13significant protein spots from secretome samples were excisedand subjected to in-gel tryptic digestion. The extracted pep-tides were analyzed by LC-ESI-qTOF system usingProteinLynx Global Server v2.4 software. Compared withthe expressed proteins in human fibroblast lysate samples,human fibrosarcoma lysate samples exhibited 13 over-expressed proteins including protein disulfide isomerase A6,endoplasmin, 60-kDa heat shock protein, protein NDRG1, t-complex protein one subunit epsilon, stress-70 protein, t-complex protein 1 subunit beta, protein disulfide isomeraseA3, stress-induced phosphoprotein 1, peptidyl prolyl cis transisomerase A, transgelin-2, heterogeneous nuclear ribonucleo-proteins A2/B1, and one under-expressed protein major vaultprotein (MVP) (see Table 1); compared with the expressed

proteins in human fibroblast secretome samples, 13 proteinsincluding secreted protein acidic and rich in cysteine(SPARC), calreticulin, isoform APP751 of amyloid beta A4,endoplasmin, galectin-1, 78-kDa glucose-regulated protein(GRP-78), alpha-enolase, peptidyl prolyl cis trans isomeraseA, peptidyl prolyl cis trans isomerase B, glyceraldehyde-3-phosphate dehydrogenase, putative elongation factor 1 alpha-like 3, urokinase-type plasminogen activator only expressedin human fibrosarcoma secretome samples (see Table 2).

Discussion

The structure and function of the cancer cells are differentfrom those of the normal cells that they originate; for thisreason, protein expression is altered [12]. These protein

Fig. 2 Representative images of two-dimensional electrophoresis (2-DE) maps of fibroblast lysate (a) and fibrosarcoma lysate (b) samples

Fig. 3 Differentially expressed proteins between human fibrosarcomalysate and human skin fibroblast lysate samples. The numbered spotsrepresent proteins up- and downregulated in fibrosarcoma lysate samples

Tumor Biol.

expression changes can be associated with tumorigenesis, andanalysis of these proteins is important to understand the mo-lecular mechanism of cancer and also essential for biomarkerdiscovery in cancer [11]. In this study, we identified 14 pro-teins in lysate samples and 13 proteins in secretome samples.In lysate samples, 13 identified proteins were over-expressedand 1 protein under-expressed in fibrosarcoma, and insecretome samples, 13 identified proteins were only expressedin fibrosarcoma.

The heat shock proteins (Hsps) are involved in proteinfolding, transport, and helping to reach the final conformationof proteins for the ability of cells to overcome stress [10].Their expression is changed by environmental and pathophys-iological conditions, such as increased temperature or oxida-tive stress [15]. Hsps were reported to be over-expressed invarious cancers such as lung cancer [10], gastric cancer [11],

prostate cancer [13], and adrenocortical carcinomas [16]. Hspssuch as mortalin, endoplasmin, and HSP-60 were found over-expressed in fibrosarcoma lysate samples.

Chaperones are molecular complexes that involved correctfolding of proteins which are important for cell growth andcell survival [17]. TCP-1-beta and epsilon are molecularchaperones and were found over-expressed in fibrosarcomalysate samples. Calreticulin is also one of the molecular chap-erones and was only expressed in fibrosarcoma secretomesamples. Also, co-chaperone STI1 was found over-expressedin fibrosarcoma lysate samples. TCP-1 proteins were reportedto be over-expressed in various cancer types such as colorectalcancer [17], human lung squamous carcinoma [18], and adre-nocortical carcinoma [16]. Besides this, it is reported thatcalreticullin is over-expressed in adrenocortical carcinoma[16], and STI1 is over-expressed in ovarian cancer [19].

Among the upregulated proteins, heterogeneous nuclearribonucleoproteins A2/B1 and NDRG1 increased in fibro-sarcoma lysate samples. It is known that heterogeneousnuclear ribonucleoproteins A2/B1 is strongly associated incell proliferation and cell survival in cancer [20]. NDRG1has also potential functions in processes such as celldifferentiation and cell cycle regulation [21]. Heteroge-neous nuclear ribonucleoproteins A2/B1 and NDRG1 werereported to be over-expressed in various cancer types suchas gastric cancer [11], lung cancer [22], breast cancer [12],and adrenocortical carcinomas [16].

Galectin-1, SPARC, urokinase-type plasminogen activator,and isoform APP751 of amyloid beta A4 proteins were se-creted proteins, and, compared to fibroblast secretome, theseproteins were only detected in fibrosarcoma secretome sam-ples. Recent studies have shown that proteolytic cleavage ofamyloid precursor protein has a role in cancer developmentand is associated with NF-kB pathway activation [23].Galectin-1 has also a role in cancer such as in angiogenesisand metastasis processes [24]. SPARC modulates cell-matrix

Fig. 4 Representative images of two-dimensional electrophoresis (2-DE) maps of fibroblast secretome (a) and fibrosarcoma secretome (b) samples

Fig. 5 Differentially expressed proteins between human fibrosarcomasecretome and human skin fibroblast secretome samples. The numberedspots represent proteins only expressed in fibrosarcoma secretomesamples

Tumor Biol.

interactions and over-expressed in many types of cancer [25].Urokinase-type plasminogen activator is involved in cell pro-liferation and metastasis in cancer with degradation of extra-cellular matrix and basement membrane [26]. These proteinswere reported to be expressed in many cancer types such asnasopharyngeal carcinoma [27] and prostate cancer [28].

We detected a remarkable downexpression of MVP infibrosarcoma cell line. It is known that MVP expressions havebeen showed in various cancer types, and it functions as atransport-associated protein that can be related with multidrugresistance [29].

Among the 27 identified proteins in lysate and secretomesamples, 6 metabolic enzymes such as protein disulfide isom-erase A3 and A6, alpha-enolase, and peptidyl prolyl cis transisomerase A and B were identified. These enzymes haveinvolved in many biological processes such as protein folding[10, 15] and glucose metabolism [30]. In particular, alpha-enolase and peptidyl prolyl cis trans isomerase Awere report-ed to be over-expressed in lung cancer [10].

Comparative proteomics studies are essential in under-standing the molecular mechanism of cancer and biomarkerdiscovery. Our results will benefit in the identification of

Table 1 Differentially expressed proteins between human fibrosarcoma and human skin fibroblast lysate samples

Spot no. Accession no. Protein name Molecularweight (Da)

Isoelectricpoint (pI)

PLGS score Protein function

1506 Q15084 Protein disulfide isomerase A6 53,867 5.0 27342 Chaperone

1903 P14625 Endoplasmin 92,411 4.6 15993 Chaperone

2605 P10809 60-kDa heat shock protein, mitochondrial 61,016 5.6 30952 Chaperone

3413 Q92597 Protein NDRG1 42,807 5.4 5928 Stress-responsive protein

3602 P48643 t-complex protein 1 subunit epsilon 59,632 5.3 14835 Chaperone

3707 P38646 Stress-70 protein, mitochondrial(GRP-75)

73,634 5.8 14715 Chaperone

3902 Q14764 MVP (major vault protein) 99,266 5.2 5029 Ribonucleoprotein

4505 P78371 t-complex protein 1 subunit beta 57,452 6.0 18799 Chaperone

4601 P30101 Protein disulfide isomerase A3 56,746 5.9 23115 Isomerase

5706 P31948 Stress-induced phosphoprotein 1 62,599 6.4 5942 Co-chaperone

6002 P62937 Peptidyl prolyl cis trans isomerase A 18,000 7.9 10210 Isomerase

7001 P62937 Peptidyl prolyl cis trans isomerase A 18,000 7.9 8605 Isomerase

7006 P37802 Transgelin-2 22,377 8.4 27775 Tumor suppressor

8209 P22626 Heterogeneous nuclear ribonucleoproteinsA2/B1

37,406 9.2 14281 Ribonucleoprotein

Table 2 Differentially expressed proteins between human fibrosarcoma and human skin fibroblast secretome samples

Spot no. Accession no. Protein name Molecularweight (Da)

Isoelectricpoint (pI)

PLGS score Protein function

0203 P09486 SPARC 34,609 4.5 4886 Extracellular matrix protein

0403 P27797 Calreticulin 48,111 4.1 5534 Chaperone

0902 P05067 Isoform APP751 of amyloid beta A4 protein 84,764 4.5 1520 Serine protease inhibitor

1704 P14625 Endoplasmin 92,411 4.6 7611 Chaperone

2002 P09382 Galectin-1 14,706 5.1 2855 Galactoside-binding protein

2602 P11021 78-kDa glucose-regulated protein (GRP-78) 72,288 4.9 6867 Chaperone/stress response

6202 P06733 Alpha-enolase 47,139 7.2 3430 Lyase

7204 P06733 Alpha-enolase 47,139 7.2 3083 Lyase

8001 P62937 Peptidyl prolyl cis trans isomerase A 18,000 7.9 1652 Isomerase

9002 P23284 Peptidyl prolyl cis trans isomerase B 23,727 9.9 7594 Isomerase

9106 P04406 Glyceraldehyde-3-phosphate dehydrogenase 36,030 8.7 13439 Oxidoreductase

9306 Q5VTE0 Putative elongation factor 1 alpha-like 3 50,153 9.4 3961 Elongation factor

9307 P00749 Urokinase-type plasminogen activator 48,475 8.3 5618 Hydrolase

Tumor Biol.

potential tumor markers and understanding of the mechanismsof fibrosarcoma.

Conclusion

The comparative proteomic analysis of human fibrosarcomaand human skin fibroblast cell lines by 2-D electrophoresiscoupled with mass spectrometry was studied, and differential-ly expressed proteins were identified. These proteins havevarious functions in cancer, and our data show that theseidentified proteins have importance in understanding the mo-lecular mechanism of fibrosarcoma. These identified proteinsmay also serve as candidate biomarkers for future clinicalstudies of fibrosarcoma.

Acknowledgments The authors thank Dr. Tarik Baykal from MedipolUniversity for the assistance with LC-MS/MS analysis and Dr. MertPekcan from Ankara University for providing technical assistance andsuggestions. This study was supported by Ankara University ScientificResearch Grant (grant no. 12B3338001).

Conflicts of interest None

References

1. Fletcher CDM, Krishnan Unni K, Mertens F. Pathology and geneticsof tumours of soft tissue and bone. 3rd ed. Lyon: IARC; 2006.

2. Nikitovic D, Kouvidi K, Karamanos NK, Tzanakakis GN. The rolesof hyaluronan/RHAMM/CD44 and their respective interactionsalong the insidious pathways of fibrosarcoma progression. BiomedRes Int. 2013;2013:929531.

3. Hajdu SI. Fibrosarcoma: a historic commentary. Cancer. 1998;82:2081–9.

4. Goldblum JR, Folpe AL,Weiss SW. Enzinger andWeiss’s soft tissuetumors. 6th ed. Philadelphia: Elsevier Saunders; 2013.

5. Song B, Kim B, Choi SH, Song KY, Chung YG, Lee YS, et al.Mesenchymal stromal cells promote tumor progression in fibrosar-coma and gastric cancer cells. Korean J Pathol. 2014;48(3):217–24.

6. Klijanienko J, Lagace R. Soft tissue tumors: a multidisciplinary,decisional diagnostic approach. 1st ed. Hoboken: Wiley; 2011.

7. Xue H, Lu B, Lai M. The cancer secretome: a reservoir of bio-markers. J Transl Med. 2008;6:52.

8. Zwickl H, Traxler E, Staettner S, Parzefall W, Grasl-Kraupp B,Karner J, et al. A novel technique to specifically analyze thesecretome of cells and tissues. Electrophoresis. 2005;26:2779–85.

9. Guoqing L, Zhefeng X, Jianping L, Cui L, Feng L, Zhuchu C.Cancer: a proteomic disease. Sci China Life Sci. 2011;54:403–8.

10. Rubporn A, Srisomsap C, Subhasitanont P, Chokchamnankit D,Chiablaem K, Svasti J, et al. Comparative proteomic analysis of lungcancer cell line and lung fibroblast cell line. Cancer GenomicsProteomics. 2009;6:229–37.

11. Zang J, Wang P, Gao S, Xiao D, Zhang J, Wang K. Differentialproteins expression between gastric cancer and normal cell lines. LifeSci J. 2008;5(4):28–32.

12. Deng SS, Xing TY, Zhou HY, Xiong RH, Lu YG, Wen B, et al.Comparative proteome analysis of breast cancer and adjacent normal

breast tissues in human. Genomics Proteomics Bioinforma.2006;4(3):165–72.

13. Johansson B, Pourian MR, Chuan Y, Byman I, Bergh A, Pang ST,et al. Proteomic comparison of prostate cancer cell lines LNCaP-FGCand LNCaP-r reveals heatshock protein 60 as a marker for prostatemalignancy. Prostate. 2006;66:1235–44.

14. Antharavally BS, Mallia KA, Rangaraj P, Haney P, Bell PA.Quantitation of proteins using a dye-metal-based colorimetric proteinassay. Anal Biochem. 2009;385(2):342–5.

15. Shen H, Huang J, Pei H, Zeng S, Tao Y, Shen L, et al. Comparativeproteomic study for profiling differentially expressed proteins be-tween chinese left and right sided colon cancers. Cancer Sci.2013;104(1):135–41.

16. Yang MS, Wang HS, Wang BS, Li WH, Pang ZF, Zou BK,et al. A comparative proteomic study identified calreticulin andprohibitin up-regulated in adenocortical carcinomas. Diagn Pathol.2013;8:58.

17. Coghlin C, Carpenter B, Dundas SR, Lawrie LC, Telfer C, MurrayGI. Characterization and over-expression of chaperonin t-complexproteins in colorectal cancer. J Pathol. 2006;210:351–57.

18. Li C, Chen Z, Xiao Z, Wu X, Zhan X, Zhang X, et al. Comparativeproteomic analysis of human lung squamous carcinoma. BiochemBiophys Res Commun. 2003;309:253–60.

19. Wang TH, Chao A, Tsai CL, Chang CL, Chen SH, Lee YS, et al.Stress-induced phosphoprotein 1 as a secreted biomarker for humanovarian cancer promotes cancer cell proliferation. Mol CellProteomics. 2010;9(9):1873–84.

20. Chang HY, Hor SY, Lim KP, Zain RB, Cheong SC, Rahman MA,et al. Oral cancer secretome: identification of cancer associatedproteins. Electrophoresis. 2013;34(15):2199–208.

21. Dos Santos M, Da Cunha Mercante AM, Nunes FD, LeopoldinoAM, De Carvalho MB, Gazito D, et al. Prognostic significance ofndrg1 expression in oral and oropharyngeal squamous cell carcino-ma. Mol Biol Rep. 2012;39(12):10157–65.

22. Pino I, Pio R, Toledo G, Zabalegui N, Vicent S, Rey N, et al. Alteredpatterns of expression of members of the heterogeneous nuclearribonucleoprotein (hnRNP) family in lung cancer. Lung Cancer.2003;41(2):131–43.

23. Chaker S, Kashat L, Voisin S, Kaur J, Kak I, Macmillan C, et al.Secretome proteins as candidate biomarkers for aggressive thyroidcarcinomas. Proteomics. 2013;13(5):771–87.

24. Ito K, Stannard K, Gabutero E, Clark AM, Neo SY, Onturk S, et al.Galectin-1 as a potent target for cancer therapy: role in the tumormicroenvironment. Cancer Metastasis Rev. 2012;31:763–78.

25. Yin J, Chen G, Liu Y, Liu S,Wang P,Wan Y, et al. Downregulation ofSPARC expression decreases gastric cancer cellular invasion andsurvival. J Exp Clin Cancer Res. 2010;29:59.

26. Kaneko T, Konno H, Baba M, Tanaka T, Nakamura S. Urokinase-type plasminogen activator expression correlates with tumor angio-genesis and poor outcome in gastric cancer. Cancer Sci. 2003;94(1):43–9.

27. Ge S, Mao Y, Yi Y, Xie D, Chen Z, Xiao Z. Comparative proteomicanalysis of secreted proteins from nasopharyngeal carcinoma-associated stromal fibroblasts and normal fibroblast. Exp Ther Med.2012;3:857–60.

28. Sardana G, Marshall J, Diamandis EP. Discovery of candidate tumormarkers for prostate cancer via proteomic analysis of cell cultureconditioned medium. Clin Chem. 2007;53(3):429–37.

29. Sasaki T, Hankins GR, Helm GA. Major vault protein/lungresistance-related protein (MVP/LRP) expression in nervous systemtumors. Brain Tumor Pathol. 2002;19(2):59–62.

30. Diaz-RamosA, Roig-Borrellas A, Garcia-Melero A, Lopez-AlemanyR. α-Enolase, a multifunctional protein: its role on pathophysiolog-ical situations. J Biomed Biotechnol. 2012;2012:156795.

Tumor Biol.