ARTICLE pubs.acs.org/jpr
Comprehensive Proteome Analysis of Malignant Pleural Effusion for Lung Cancer Biomarker Discovery by Using Multidimensional Protein Identification Technology Chia-Jung Yu,*,†,‡,§ Chih-Liang Wang,||,^ Chun-I Wang,†,^ Chi-De Chen,† Yu-Min Dan,† Chih-Ching Wu,§,# Yi-Cheng Wu,z I-Neng Lee,§ Ying-Huang Tsai,|| Yu-Sun Chang,†,§ and Jau-Song Yu†,‡,§ †
Graduate Institute of Biomedical Sciences, ‡Department of Cell and Molecular Biology, §Molecular Medicine Research Center and Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Tao-Yuan, Taiwan Division of Pulmonary Oncology and Interventional Bronchoscopy, Department of Thoracic Medicine, and z Division of Thoracic & Cardiovascular Surgery, Chang Gung Memorial Hospital, Tao-Yuan, Taiwan
)
#
bS Supporting Information ABSTRACT: Malignant pleural effusion (MPE) obtained from lung adenocarcinoma may contain potentially useful biomarkers for detection of lung cancer. In this study, we used a removal system for high-abundance proteins followed by onedimensional SDS-PAGE combined with nano-LC MS/MS to generate a comprehensive MPE proteome data set with 482 nonredundant proteins. Next, we integrated the MPE proteome and secretome data sets from three adenocarcinoma cell lines, with a view to identifying potential PE biomarkers originating from malignant cells. Four potential candidates, alpha-2-HSglycoprotein (AHSG), angiogenin, cystatin-C, and insulin-like growth factor-binding protein 2, (IGFBP2), were isolated for preclinical validation using ELISA. Both AHSG and IGFBP2 levels were increased in lung patients with MPE (n = 68), compared to those with nonmalignant pleural effusion (n = 119). Notably, the IGFBP2 level was higher in MPE, compared with that in benign diseases (bacteria pneumonia and tuberculosis pleuritis), and significantly associated with malignancy, regardless of the cancer type. Our data additionally support an extracellular function of IGFBP2 in migration in lung cancer cells. These findings collectively suggest that the adenocarcinoma MPE proteome provides a useful data set for malignancy biomarker research. KEYWORDS: malignant pleural effusion, lung cancer, biomarker, proteome, IGFBP2
’ INTRODUCTION Lung cancer is one of the most common human malignancies and the leading cause of cancer-related deaths worldwide.1 Persistent poor survival of lung cancer patients is largely attributable to late diagnosis, with a recorded 75% of lung cancer patients presenting with advanced stages (III and IV) at the time of diagnosis.2 Nonsmall cell lung cancer (NSCLC), including squamous cell carcinoma, adenocarcinoma, large-cell carcinoma, and other rare subtypes, constitute the most common type of lung cancer, representing about 80% of all cases.3,4 Despite major advances in cancer therapy over the past two decades, prognosis of patients with advanced NSCLC has improved only minimally. The overall 5-year survival for NSCLC remains at 15%. However, if the cancer is detected at stage IA, the 5-year survival rate often exceeds 80%,5 indicating that the stage of lung cancer is related to prognosis. Thus, it is essential to establish successful methods for early cancer detection and distinguish between disease stages to improve survival and patient care. Pleural effusion (PE), a term used to represent accumulation of pleural fluid, contains proteins originating from plasma filtrate r 2011 American Chemical Society
released by inflammatory, epithelial and malignant cells. PE is triggered by a variety of etiologies, including malignancy, pneumonia, tuberculosis infection, autoimmune disease, pulmonary embolism, heart failure, and liver or renal disease.6 Malignant pleural effusion (MPE) is usually diagnosed when exfoliated malignant cells in effusion fluid or cancer cells implanting in parietal pleura are detected via cytological examination or transcutaneous pleural biopsy. A high percentage of MPE (>75%) arises from lung cancer, breast cancer, lymphoma/leukemia or ovarian cancer. Lung cancer is the major etiology of malignant effusion.7 Approximately 50% of lung cancer patients develop PE in their disease course,8 while only 15% of patients have been reported with MPE.9 Paramalignant pleural effusion (PMPE) is defined as malignancy with no etiologic evidence of tumor invasion, which may result from airway obstruction with lung collapse, lymphatic obstruction and systemic effects of cancer treatment.10 Differentiation of MPE from PMPE has profound implications in the therapy and prognosis of cancer. MPE is Received: May 20, 2011 Published: August 02, 2011 4671
dx.doi.org/10.1021/pr2004743 | J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
ARTICLE
Table 1. Laboratory Characteristics of 13 Malignant Pleural Effusions Used for Generation of MPE Proteome Data Seta gender
a
protein (g/dL)
LDH (IU/L)
WBCs (μL-1)
Lym cells (μL-1)
Neut cells (μL-1)
age
glucose (mg/dL)
M = 7; F = 6
67.65 ( 11.48
115.75 ( 23.21
3.97 ( 0.95
78.54 ( 40.90
1246.54 ( 1271.29
838.01 ( 773.04
394.62 ( 1230.91
Range
45 80
87 161
2.1 5.7
35 141
135 4675
80 2148
0 4488
Data are presented as mean ( SD. M, male; F, female; LDH, lactic dehydrogenase; WBCs, white blood cells; Lym, lymphocyte; Neut, neutrophil.
detected in lung cancers classified as incurable metastatic disease, and the median survival time of these patients is limited to 8 months, with a recorded 5-year survival rate of only 2%.11 Therefore, MPE implies advanced disease and grave prognosis in other malignancies.12 However, a limited number (only 40 70%) of patients with MPE can be diagnosed via cytological examination or pleural biopsy.13 15 Nondetection of MPE (31.5% of the false negative rate) leads to underestimation of disease status and inadequate therapy and, consequently, poor survival.16 Thus, it is important to identify potential cancer biomarkers to aid in distinguishing MPE from benign diseases. Owing to organ proximity and a smaller dilution effect, detection of biomarkers from PE may enhance the probability of discovering potential biomarkers for lung cancer. To date, a number of potential biomarkers, including lung surfactant protein A,17 carcinoembryonic antigen (CEA),17 cystatin-C,18 vascular endothelial growth factor,19 pigment epithelium-derived factor,20 22 and epididymal secretory protein E1 precursor,23 have been evaluated in PE. However, these markers are not currently recommended or encouraged in routine clinical practice, due to limitations in sensitivity and specificity. Proteomic approaches have been widely applied to identify biomarkers of malignant diseases, particularly in human body fluids corresponding to plasma/serum, cerebrospinal fluid, saliva and urine.24,25 The biggest challenge in the discovery of biomarkers from body fluids is the high complexity and dynamic range of proteins. The high-abundance proteins mask the low-abundance proteins, and thus detection of proteins with low concentrations in body fluid requires the application of additional prefractionation and advanced mass spectrometry techniques. In this study, we employed a high-abundance protein removal system and onedimensional SDS-PAGE combined with nano-LC MS/MS (GeLC MS/MS) approaches to circumvent the analytical challenges mainly arising from the high dynamic range of protein concentrations in human body fluid and LTQ-OrbiTrap mass spectrometry to optimize the identification of proteins expressed in malignant PE. With the aid of these comprehensive proteomic approaches, we generated a PE proteome data set containing 482 nonredundant proteins and established the clinical relevance of PE biomarkers in NSCLC.
’ MATERIALS AND METHODS Patient Populations and Clinical Specimens
MPE diagnosis is based on the detection of malignant cells via cytological examination of effusion fluid or pathologic examination of pleural biopsy. PMPE is defined as cancer cases with PE, but with no evidence of pleural involvement, based on negative results in biopsy and cytological examination. Patients with PMPE are radiologically monitored regularly over 6 months to exclude the possibility of occult malignancy within the effusion. Thirteen malignant pleural effusion specimens from lung adenocarcinoma patients were used to generate the MPE proteome data set. The laboratory characteristics of these 13 MPE specimens are summarized in Table 1. Sixty PE (37 MPE and 23 PMPE)
obtained from lung cancer patients (including 34 adenocarcinomas, 13 squamous cell carcinomas, 10 NSCLCs, and three small cell lung cancers) were used for validation in the first batch to select potential candidates for distinguishing malignancy from paramalignancy (Supplementary Table S1, Supporting Information). To investigate the specificity and accuracy of potential biomarkers in distinguishing malignancy from non-malignancy (PMPE, bacteria pneumonia and tuberculosis pleuritis), we extended the PE sample size to 187 specimens obtained from 68 patients with MPE (38 lung cancers, 11 breast cancers, 6 gastric cancers, 5 colon cancers, 3 lymphomas, 1 bladder cancer, 1 ovarian cancer, 1 melanoma, 1 pancreatic cancer and 1 thyroid cancer; 32 men and 36 women; age range 22 84 years), 59 patients with PMPE (30 lung cancers, 6 hepatomas, 4 lymphomas, 3 breast cancers, 3 esophageal cancers, 3 headand-neck cancers, 2 colon cancers, 1 gastric cancer and 1 pancreatic cancer, 1 cholangiocarcinoma, 1 renal cell carcinoma, 1 cervical cancer, 1 malignant thymoma, 1 malignant neurofibroma and 1 leukemia; 41 men and 18 women; age range 22 92 years) and 60 patients with benign pulmonary disease (30 bacteria pneumonia and 30 tuberculosis pleuritis; 44 men and 16 women; age range 26 93 years) (Supplementary Table S2, Supporting Information). All the PE samples were obtained from patients subjected to PE aspiration at Chang Gung Memorial Hospital. Pleural effusion samples were stored at 80 C until further analysis. Written informed consent was received from all patients, prior to collection. The study was approved by the Institutional Review Board of Chang Gung Memorial Hospital. Medical records of patients were reviewed, and all patient identities were protected. Removal of High-Abundance Proteins of Pleural Effusion
PE from 13 lung adenocarcinoma patients were pooled, and 80 μL of the pooled samples were subjected to depletion of six high-abundance proteins (albumin, IgG, IgA, transferrin, antitrypsin and haptoglobin) using the Multiple Affinity Removal System affinity column (Hu-6HC, 4.6 100 mm, Agilent Tech€ KTA purifier-10 fast perfornologies, Wilmington, DE) via A mance liquid chromatography (GE Healthcare, U.K.). The buffers used for binding and elution were purchased from Agilent Technologies, and removal procedures were performed according to the manufacturer’s instructions (Agilent Technologies). Unbound fractions (PE depleted of the six proteins) were concentrated and desalted via centrifugation in Amicon Ultra-4 tubes (molecular weight cutoff 3000 Da; Millipore, Billerica, MA). Protein concentrations were determined with the BCA protein assay kit (Pierce, Rockford, IL). One-Dimensional SDS-PAGE and In-Gel Digest of Proteins
PE proteins with high abundant protein depletion (50 μg) were resolved on 10% SDS-PAGE and stained by Coomassie Brillant Blue G-250. The whole gel lane was cut into 66 pieces and subjected to in-gel tryptic digestion as described in Supporting Information and Methods.
4672
dx.doi.org/10.1021/pr2004743 |J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
ARTICLE
Reverse Phase Liquid Chromatography-Tandem Mass Spectrometry
Each peptide mixture was reconstituted in HPLC buffer A (0.1% formic acid, Sigma, St. Louis, MO), loaded into a trap column (Zorbax 300SB-C18, 0.3 5 mm, Agilent Technologies, Wilmington, DE) at a flow rate of 0.2 μL/min in HPLC buffer A, and the salts were washed with buffer A at a flow rate of 20 μL/min for 10 min. The desalted peptides were then separated on a resolving 10-cm analytical C18 column (inner diameter, 75 μm) with a 15-μm tip (New Objective, Woburn, MA). The peptides were eluted by a linear gradient of 0 10% HPLC buffer B (99.9% ACN containing 0.1% formic acid) for 3 min, 10 30% buffer B for 35 min, 30 35% buffer B for 4 min, 35 50% buffer B for 1 min, 50 95% buffer B for 1 min, and 95% buffer B for 8 min at a flow rate of 0.25 μL/min across the analytical column. The LC setup was coupled on line to a two-dimensional linear ion trap mass spectrometer LTQ-Orbitrap (Thermo Fisher, San Jose, CA) operated using the Xcalibur 2.0.7 software (Thermo Fisher, San Jose, CA). The MS full-scan was performed in the LTQ-Orbitrap with a MS range from 350 Da to 2,000 Da and the intact peptides were detected at a resolution of 30 000. Internal calibration was performed using the ion signal of (Si(CH3)2O)6H+ at m/z 445.120025 as a lock mass.26 The data-dependent procedure that alternated between one MS scan followed by six MS/MS scans for the six most abundant precursor ions in the MS survey scan was applied. The m/z values selected for MS/MS were dynamically excluded for 180 s. The electrospray voltage applied was 1.8 kV. Both MS and MS/MS spectra were acquired using the one microscan with a maximum fill-time of 1000 and 100 ms for MS and MS/MS analysis, respectively. Automatic gain control was used to prevent overfilling of the ion trap, 5 104 ions were accumulated in the ion trap for generation of MS/MS spectra. Database Searching and Bioinformatics
The resulting MS/MS spectra were searched using Mascot algorithm (version 2.2.06, Matrix Science) against the SwissProt database (version 56.0, selected for Homo sapiens, 20401 entries). Search parameters included differential amino acid mass shifts for oxidized methionine (16 Da) and fixed modification for carbamidomethyl cysteine. The maximum mass tolerance was set to 10 ppm for precursor ions and 0.5 Da for fragment ions. The validation of MS/Ms based peptides and protein identifications were performed by the Scaffold proteome software (Version _ 2_06_02, Proteome Software Inc., Portland, OR), in which the cutoffs for peptide thresholds were 95.0% minimum, and protein thresholds were 95.0% minimum and minimum of 2 peptides. The identified proteins were further analyzed used ProteinCenter (Proxeon Bioinformatics, Odense, Denmark), a proteomics data mining and management software, to compare cell line secretomes with PE proteome, in which the identified proteins were functionally classified based on universal Gene Ontology (GO) annotation terms, and linked to at least one annotation term within the GO molecular function and biological process categories, respectively. Enzyme-Linked Immunosorbent Assay
Angiogenin (ANG) and cystatin-C (CST3) protein levels in human PE were determined using sandwich enzyme-linked immunosorbent assay (ELISA) kits purchased from R&D systems (Minneapolis, MN). Protein levels of IGFBP2, AHSG and CEA were determined using sandwich ELISA kits obtained from BioVendor R&D (BioVender Laboratory Medicine, Modrice,
Figure 1. Generation of a malignant pleural effusion data set using multidimensional protein identification approaches. (A) Workflow used in this study to generate a potential MPE biomarker data set using depletion of high-abidance proteins from pooled MPE followed by GeLC MS/MS analysis. The strategy consists of the MPE proteome profiling, combined with the secretome data sets from three adenocarcinoma (AD) cancer cell lines, and subsequent validation in clinical specimens. (B) Efficacy of high-abundance protein depletion from malignant pleural effusions. MPE from 13 lung adenocarcinoma patients were pooled and subjected to a six high-abundance plasma protein depletion system, as described in Materials and Methods. Aliquots of proteins (50 μg) prior to (lane 1) and after (lane 2) depletion were prepared and separated using one-dimensional SDS-PAGE, followed by staining with Coomassie Brilliant Blue R-250, to examine the efficacy of MPE depletion.
Czech Republic) and Bio-Quant (Bio-Quant Inc., San Diego, CA), respectively. Immunodetection of IGFBPs, IGFs and IGF Receptors in Cell Extracts or Conditioned Media of Lung Cancer Cell Line
Protein samples prepared from CL1 0 cells were subjected to SDS-PAGE and Western blot analysis using primary antibodies against candidates of interest for 16 h at 4 C as described previously.27 The detail procedures for conditioned media preparation and antibodies information are described in Supporting Information and Methods. Transwell Migration Assay
Transwell cell migration was performed using cells treated with recombinant IGFBP2 (R&D Systems, Inc.) as described in Supporting Information and Methods. Statistical Analysis
All data were processed using the statistical package, SPSS 13.0 (SPSS Inc., Chicago, IL). All continuous variables were expressed as means ( SD. The nonparametric Mann Whitney U test was employed to analyze the variations in ELISA results for different clinical parameters. Two-tailed p values of 0.05 or less 4673
dx.doi.org/10.1021/pr2004743 |J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
Figure 2. Functional classification of MPE proteins using ProteinCenter software based on universal GO annotation terms. Proteins were linked to at least one annotation term within the (A) GO molecular function and (B) biological process categories.
were considered significant. Receiver operator characteristic (ROC) curves were constructed by plotting sensitivity versus 1-specificity, and the areas under the curves (AUC) analyzed with the Hanley and McNeil method.28 The optimal cutoff point for establishing an accuracy score in each case was determined using Youden’s index (J) calculated as J = 1 (false positive rate + false negative rate) = 1 ((1 sensitivity) + (1 specificity)) = sensitivity + specificity 1.29
’ RESULTS AND DISCUSSION Generation of a Malignant Pleural Effusion Proteome Data set using Multidimensional Protein Identification Approaches
MPE represent a manifestation of cancer metastasis and frequently indicate poor prognosis.6,30 In this anatomic cavity, we expected to identify serum proteins, secretory proteins from cancer, immune cells or mesothelium, membrane proteins from shed exosomes, and proteins released from dead cells.20,30 Identification of these proteins would be an efficient strategy in the search for cancerous protein markers and provide an opportunity to develop serum/plasma accessible markers for clinical usage. Previously, Tyan et al. identified 40 and 124 PE proteins by using two-dimensional polyacrylamide gel electrophoresis and 2D nano-HPLC ESI MS/MS, respectively.31 Their results showed significant similarities between the proteomes of plasma and PE, particularly in terms of the high- to moderate-abundance protein profile. Therefore, the analytical challenges mainly arising from the high dynamic range of protein concentrations in human serum/plasma are also expected in PE analysis. To circumvent the masking of low-abundance proteins, we applied a prefractionation approach with antibody-based affinity chromatography (immunodepletion) followed by GeLC MS/ MS analysis to generate a potential MPE biomarker data set for lung cancer (Figure 1A). As expected, this prefractionation method allowed the depletion of six highly abundant plasma proteins
ARTICLE
from MPE (Figure 1B). After immunodepletion, unbound protein fractions were subjected to concentration and desalting. To generate a comprehensive MPE proteome data set, prefractionated MPE proteins (50 μg) were subjected to SDSPAGE, sliced into 66 fractions, digested individually with trypsin, and analyzed using LC ESI MS/MS. A spectral search was performed against the Swiss-Prot database using the Mascot algorithm and the results further analyzed using Scaffold software, resulting in the identification of 482 unique proteins with high confidence (95.0% minimum peptide probability, 95.0% minimum protein probability and minimum of 2 peptides). The detail information for 482 MPE proteins, including identified probability, mass score and peptide information are shown in Supplementary Table S3 (Supporting Information). To predict the functions of the MPE proteins, ProteinCenter software based on universal GO annotation terms was used. These proteins were linked to at least one annotation term within the GO molecular function and biological process categories, respectively. The top three most common molecular functions were determined as protein binding (78.8%), catalytic activity (68.0%), and metal ion binding (55.4%) (Figure 2A). The major biological process categories included metabolic processes (82.8%), response to stimulus (63.9%), and regulation of biological processes (57.3%) (Figure 2B). To our knowledge, the current study provides the most comprehensive PE analysis in this field. The MPE profile partly overlaps with the protein data set of Pernemalm et al., who performed a quantitative proteomic study using narrow-range peptide IEF to improve the detection of lung adenocarcinoma markers in plasma and PE.23 The group identified 20 proteins that differed significantly between the lung adenocarcinoma and pleuritis groups and successfully validated 7 proteins (gelsolin, NPC2, VCAM1, A2M, SERPINA1, EFEMP1 and CLEC3B) using Western blot analysis. In comparison analyses, all 7 validated proteins were also identified in our MPE data set, supporting the utility of comprehensive MPE proteomic analysis and useful data sets for the discovery of cancer biomarkers. Selection of Potential Pleural Effusion Biomarkers for NSCLC
PE proteins are derived from plasma filtrate, inflammatory cells or malignant cells,32 and therefore, efficient strategies to select potential biomarkers from the complicated PE pool are critical. Conditioned media of cancer cell lines contain secreted or shed proteins released through both classical and nonclassical secretion pathways. Proteins identified in conditioned media (secretome) from three lung adenocarcinoma cell lines (H23, CL1 0 and CL1 5) generated recently using nongel based proteomic approaches were comparable. The H23 secretome data set (871 unique proteins) was established by Professor Diamandis and his colleague using a 2D-LC MS/MS strategy.33 The CL1 0 (1157 unique proteins) and CL1 5 (1947 unique proteins) secretome data sets were additionally generated by our group using the GeLC MS/MS approach, as described previously.34 We integrated the MPE proteome data set and the three secretome data sets with the aim of detecting potential PE biomarkers originating from malignant cells. Using this strategy, we identified 93 MPE proteins in all three secretomes and 88 MPE in any two of the three secretomes (Supplementary Table S4, Supporting Information). The 181 MPE proteins detected in at least two of the three secretome data sets were further analyzed using bioinformatics programs designed to predict protein secretion pathways. The SignalP program predicted that 68 of the 4674
dx.doi.org/10.1021/pr2004743 |J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
ARTICLE
Table 2. Identification Information of Four Candidate Proteins Selected for Preclinical Validation by ELISA gene name
protein accession no
biology sample
no. unique peptides
no. unique spectra
AHSG
FETUA_HUMAN
CL1 0a
4
5
peptide sequence CDSSPDSAEDVR
sequence coverage 7.08%
CDSSPDSAEDVRK CNLLAEK QYGFCK CL1 5a
5
7
AHYDLR
8.72%
CDSSPDSAEDVR CDSSPDSAEDVRK CNLLAEK QYGFCK MPEb
10
18
CNLLAEK
21.30%
CNLLAEKQYGFCK EATEAAKCNLLAEK EHAVEGDCDFQLLK EHAVEGDCDFQLLKLDGK FSVVYAK HTFMGWSLGSPSGEVSHPR HTLNQIDEVK LDGKFSWYAK QLKEHAVEGDCDFQLLK ANG
ANGI_HUMAN
CL1 0
5
6
DDRYCESIMR
30.60%
DINTFIHGNK SSFQVTTCK YCESIMR CL1 5
6
9
YTHFLTQHYDAKPQGR DINTFIHGNK
34.00%
DINTFIHGNKR GLTSPCK SSFQVTTCK YCESIMR YTHFLTQHYDAKPQGR H23a
3
3
DINTFIHGNKR
32.00%
NWVACENGLPVHLDQSIFR YTHFLTQHYDAKPQGR MPE
2
2
DDRYCESIMR
12.90%
SSFQVTTCK CST3
CYTC_HUMAN
CL1 0
4
5
ALDFAVGEYNK
30.80%
LVGGPMDASVEEEGVR LVGGPMDASVEEEGVRR TQPNLDNCPFHDQPHLK CL1 5
5
7
ALDFAVGEYNK LVGGPMDASVEEEGVR
30.80%
LVGGPMDASVEEEGVRR RALDFAVGEYNK TQPNLDNCPFHDQPHLK H23
6
8
AFCSFQIYAVPWQGTMTLSK
45.20%
ALDFAVGEYNK LVGGPMDASVEEEGVR LVGGPMDASVEEEGVRR TQPNLDNCPFHDQPHLK TQPNLDNCPFHDQPHLKR MPE
2
3
ALDFAVGEYNK
18.50%
LVGGPMDASVEEEGVR IBFBP2
IBP2_HUMAN
CL1 0
7
8
GDPECHLFYNEQQEAR
26.80%
HGLYNLK 4675
dx.doi.org/10.1021/pr2004743 |J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
ARTICLE
Table 2. Continued gene name
protein accession no
biology sample
no. unique peptides
no. unique spectra
peptide sequence
sequence coverage
LAACGPPPVAPPAAVAAVAGGAR LEGEACGVYTPR LIQGAPTIR MPCAELVR TPCQQELDQVLER CL1 5
2
2
HGLYNLK
4.88%
LIQGAPTIR H23
11
39
CYPHPGSELPLQALVMGEGTCEK
44.20%
GDPECHLFYNEQQEAR GECWCVNPNTGK GPLEHLYSLHIPNCDK GPLEHLYSLHIPNCDKHGLYNLK HHLGLEEPKK LAACGPPPVAPPAAVAAVAGGAR LEGEACGVYTPR LPDERGPLEHLYSLHIPNCDK MPCAELVR TPCQQELDQVLER MPE
6
6
GGKHHLGLEEPKK
17.40%
HHLGLEEPKK LEGEACGVYTPR LIQGAPTIR SGMKELAVFR TPCQQELDQVLER a
CL1 0, CL1 5 and H23 are lung adenocarcinoma cell lines. b MPE, malignant pleural effusion.
Figure 3. Detection of AHSG, AGN, CST3 and IGFBP2 in pleural effusions. The PE levels of AHSG, AGN, CST3 and IGFBP2 from 60 patients with lung cancer were determined with sandwich ELISA. A p value of less than 0.05 using the Mann Whitney U test indicates statistical significance.
proteins were secreted via the classical secretory pathway (i.e., endoplasmic reticulum/Golgi-dependent pathway; SignalP probability g0.90), based on the presence of a specific signal peptide.35,36 The SecretomeP program estimated the release
36 proteins via the nonclassical secretory pathway (SignalP probability e0.90 and SecretomeP score g0.50).37 In addition, TMHMM predicted that 3 integral membrane proteins were not secreted via the classical or nonclassical secretion pathways.38 4676
dx.doi.org/10.1021/pr2004743 |J. Proteome Res. 2011, 10, 4671–4682
Journal of Proteome Research
ARTICLE
Table 3. Clinical Significance of IGFBP2 and AHSG Levels in Malignancy patient no. (%)
IGFBP2 (ng/mL)
pa
