Lung Cancer (2006) 54, 8794

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/lungcan

Proteomics-based identication of secreted protein dihydrodiol dehydrogenase as a novel serum markers of non-small cell lung cancer
Ling-Jin Huang a,, Sheng-Xi Chen a, Yan Huang b, Wan-Jun Luo a, Hai-He Jiang a, Qing-Hua Hu a, Peng-Fei Zhang c, Hong Yi c
a

Department of Cardiothoracic Surgery, Xiangya Hospital, Central South University, Changsha 410008, Hunan, PR China Department of Infectious Disease, Xiangya Hospital, Central South University, Changsha 410008, Hunan, PR China c Department of Central Laboratory, Xiangya Hospital, Central South University, Changsha 410008, Hunan, PR China
b

Received 21 February 2006; received in revised form 29 May 2006; accepted 11 June 2006

KEYWORDS
Dihydrodiol dehydrogenase; Secreted protein; Non-small cell lung cancer; Tumor marker

Summary Identication of secreted proteins of lung cancer could provide new candidates of serum biomarkers for cancer diagnosis and prognosis evaluation. In this study, non-small cell lung cancer (NSCLC) cell line A549 was cultured. Proteins in the conditioned medium of A549 were recovered and the proteome analysis was subsequently performed. Secreted proteins of A549 were identied using mass spectrometry and database search. Fourteen human proteins were identied, including peptidylprolyl cistrans isomerase A, manganese superoxide dismutase, peroxiredoxin 1, phosphatidylethanolamine-binding protein, glutathione S-transferase P, PGP9.5, alpha enolase, phosphoglycerate mutase 1, galectin-1 and dihydrodiol dehydrogenase (DDH). DDH was selected for further analysis using RT-PCR, immunoblotting, immunohistochemical staining and ELISA in NSCLC patients. Compared with normal lung tissues, higher DDH mRNA and protein expression level were found in 15 NSCLC cancer tissues (p < 0.05). DDH overexpression was identied to be located in cytoplasm and cell membrane by immunohistochemical staining in NSCLC tissue. The serum level of DDH was signicantly higher in NSCLC patients (n = 64) than nonmalignant lung tumor (n = 20) and healthy controls (n = 20) (p < 0.05). The results show that DDH was one of the secreted proteins in NSCLC. It can serve as a tissue marker and a novel serological marker of NSCLC. Identication of secreted proteins could be a feasible and effective strategy to search potential serum biomarkers of cancer. 2006 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Corresponding author. Fax: +86 731 4327332. E-mail address: drhuanglj@yahoo.com.cn (L.-J. Huang).

Secreted proteins of cancer cells can be shed into blood. Their serum levels may increase in the early stage of cancer, and correlate with cancer cell proliferation and/or

0169-5002/$ see front matter 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2006.06.011

88 secreted protein overexpression. As a result, secreted proteins may be used as potential serum biomarkers of cancer [1,2]. Well-known secreted protein serving as serum biomarkers include CEA for colon cancer, prostasin and human kallikrein 10 for ovarian cancer [3,4], cysteine-rich secretory protein-3 and prostate-specic antigen (PSA) for prostate cancer [5,6]. Secreted protein analysis, together with genomic approaches have facilitated identication of a number of serum biomarkers [7] for malignant tumors, such as macrophage inhibitory cytokine 1 in metastatic prostate, breast, and colorectal carcinomas [2], osteopontin and prostasin in ovarian carcinomas [4,8], tissue inhibitor of metalloproteinase-1 (TIMP-1) in pancreatic adenocarcinoma [9]. Another good example is showed in the investigation of nasopharyngeal carcinoma (NPC) cellsecreted proteomes using SDS-PAGE and matrix-assisted laser desorption/ionization-time of ight mass (MALDI-TOF MS), three serum biomarkers of NPC were identied which included bronectin, Mac-2-binding protein (Mac-2 BP), and plasminogen activator inhibitor 1 (PAI-1) [1]. We postulated that analysis of the secreted proteins of lung cancer cells may be an effective approach for delineating potential serum biomarkers. Lung cancer is the most common cancer and the leading cause of cancer-related death worldwide. The early diagnosis of lung cancer is critical for successful therapy. Serological biomarkers can play important roles in cancer screening, monitoring of cancer progression, treatment response and surveillance for recurrence [10]. Serum protein biomarkers, such as carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA-125), cytokeratin 19 fragment marker (CYFRA 21-1) are not ideal tools in detection of lung cancer due to their low specicity and/or sensitivity [11,12]. For more effective lung cancer diagnosis, we need to nd new biomarkers, which have higher specicity and sensitivity. Lung cancer is divided clinically into small-cell lung cancer (SCLC) and non-SCLC (NSCLC). NSCLC comprises more than 80% of lung cancers. In this study, human NSCLC cell line A549 was cultured and the secreted proteins in conditioned medium were analyzed by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and MALDI-TOF MS. Secreted protein dihydrodiol dehydrogenase (DDH) was identied and conrmed. High DDH expression was detected in NSCLC cancer tissues. The serum level of DDH was evaluated in NSCLC patients, benign lung tumor patients and healthy individuals using enzyme-linked immunosorbent assay (ELISA) which indicated high serum DDH level in NSCLC.

L.-J. Huang et al. treated with DOCTCA (deoxycholatetricarboxylic acid) sediment method to recover the total protein. Added 0.1 mL 0.15% DOC (w/v) into per milliliter supernatant and mixed well, incubated on ice for 10 min, then added 50 L 100% TCA (v/v) per milliliter to the mixture, centrifuged at 20,000 g for 30 min, carefully discarded the supernatant, and air dry at last. Protein in medium of RPMI-1640 with 5% FBS was collected in the same way as the conditioned medium and used as negative control. Precipitated protein samples were resuspended in cell lysis buffer (8 mol/L urea, 4% CHAPS, 40 mmol/L Tris, 65 mmol/L DTT) and the supernatant was used for proteomic analysis. Cells left on the dishes were washed twice with PBS, and disrupted in cell lysis buffer for 30 min. The protein extracts were centrifuged at 20,000 g for 30 min and the supernatants were used for immunoblotting. Protein concentrations of samples were determined by the BCA protein assay reagent from Pierce (Rockford, IL, USA).

2.2. Patient population and clinical specimens

Fifteen newly pathological identied NSCLC tumors samples and distant (distance to the edge of tumor >5 cm) nonmalignant lung samples of the same patient were obtained at the time of surgery at the Department of cardiothoracic surgery, Xiangya Hospital, China, and stored at 80 C. These were used for reverse transcription polymerase chain reaction (RT-PCR) analysis and immunoblotting. Samples used for immunohistochemical staining were xed in 10% buffered formalin and embedded in parafn. There were 5 adenocarcinomas and 10 squamous carcinomas. The median age was 55 years (range 4572 years); 5 were female and 10 were male. Eighty-four serum samples were collected which included 34 adenocarcinomas, 30 squamous carcinomas and 20 benign lung tumor patients. The clinical data was listed on Table 1. Sera were obtained from the Department of cardiothoracic surgery, Xiangya Hospital. The control group for ELISA consisted of 20 apparently healthy blood donors (ages 4566 years; mean 51 years; 10 females and 10 males). Blood samples (5 mL) were collected, clotted at 4 C, centrifuged at 3000 rpm for 10 min then stored at 80 C. The study was approved by the Medical Ethics and Human Clinical Trial Committee at Xiangya Hospital.

2.3. Protein separation by two-dimensional electrophoresis (2-DE)

Total protein (1200 g) of conditioned medium and control were separated by 2-DE which separated proteins according to their isoelectric points and molecular weights. Isoelectric focusing was performed using IPGstrip (pH 3-10L, 24 cm) on IPGphor isoelectric focusing cell (Amersham Biosciences) and second-dimension SDS-PAGE using Ettan Dalt II system (Amersham Biosciences) was conducted as described by manufacturer and Gorg [13]. After electrophoresis, the protein spots were visualized by blue silver staining technique described by Candiano et al. [14]. Stained with 0.12% Coomassie blue G-250 (Sigma, USA), 10% ammonium sulfate, 10% phosphoric acid, and 20% methanol for 24 h, and the background was destained with 10% methanol and 10% acetic acid. The stained 2-DE gels were scanned using the

2. Material and methods

2.1. Cell culturing and sample preparation
A549 cells (ATCC no. CCL-185) were grown in RPMI 1640 supplemented with 15% fetal bovine serum (FBS) at 37 C under a humidied atmosphere of 95% air and 5% CO2 (v/v). To obtain culture supernatants, cells were grown to conuence in 15 cm tissue culture dishes, then washed with serumfree medium and incubated in 5% FBS medium for 48 h. Five milliliters conditioned medium were collected and centrifuged to eliminate the intact cells. The supernatant was

Proteomics-based identication of secreted protein dihydrodiol dehydrogenase

Table 1 Clinical data and results of ELISA analysis Sex (M/F) Squmaous carcinoma Adenocarcinoma Benign lung tumor Healthy control
* #

89

Age (years) 56 50 50 45 12 14 16 10

Number 34 30 20 20

OD value 0.173 0.179 0.139 0.137 0.016*,# 0.015*,# 0.018 0.011

25/9 18/12 12/8 10/10

Compared with healthy control, p < 0.05. Compared with benign lung tumors, p < 0.05. Compared with healthy control, p > 0.05.

Immagescanner (Amersham Biosciences) and the digitalized gel images were normalized and comparatively analyzed using software PDQuest (Bio-Rad, CA).

2.4. Mass spectrometric analysis of secreted protein

The specic protein spots in gels of conditioned medium of A549 were excised from gels, washed with deionized water, and destained in the destaining solution that consisted of 50 mmol/L ammonium bicarbonate and 100% acetonitrile (ACN) (V/V, 1:1), After dehydrated with 100% ACN, the gel pieces were dried thoroughly in a vacuum centrifuge (Savant, USA) for 30 min. The dried gelpieces were incubated in the digestion solution that consisted of 50 mmol/L ammonium bicarbonate and 0.1 g/L TPCKtrypsin (Sigma, USA) for 16 h at 37 C. Digested peptides were extracted and condensed in vacuum centrifuge to about 5 L. Recovered peptides were mixed with matrix cyano-4-hydroxycinnamic acid (CCA, Sigma, USA). Samples (1 L) were loaded on plate and dried for crystallization. Masses of peptides were determined using MALDI-TOF mass spectrometer (PE Biosystem, Framingham, MA). Database was searched by Mascot software with the following parameters: mammalian species and one missed cleavage site. Carbomidomethylation of cysteine and oxidation of methionine were considered as xed modication or possible partial modication. A trypsin fragment peak served as internal standard for mass calibration. A list of the corrected mass peaks was the peptide mass ngerprinting (PMF). The probability scores calculated by software were used as criterion for correct identication.

phosphate dehydrogenase (GAPDH) was employed as internal control and amplied separately. The forward primer sequence of GAPDH was 5 -TGGGTGTGAACCATGAGAAGT-3 and reverse primer 5 -GAGGAGTGGGTGTCGCTGT-3 . All of the PCR reactions used a volume of 25 L. The PCR products were separated on 2% agarose gels and the bands were visualized with the ethidium bromide. The intensity of DDH and GAPDH bands was quantied by densitometric scanning (GIS-2020, Tanon, Shanghai) and the ratio of the DDH/GAPDH gene expression from each sample was calculated.

2.6. Immunoblotting of DDH in conditioned medium

Proteins (60 g/lane) of A549 cells extraction, conditioned medium and control medium were separated in a 10% polyacrylamide gel with 5% stacking gel. The resolved proteins were transferred onto PVDF membrane (Millipore) using BioRad wet transfer unit, and the blots were blocked with 5% (w/v) non-fat dry milk in TBS-T solution (25 mmol/L Tris, pH 7.5, 150 mmol/L NaCl, 0.05% (w/v) Tween 20). After washing in TBS-T, the blots were incubated for 1 h with mouse antihuman DDH antibodies (Cat. no. X1265M, Exalpha Biologicals Inc., USA) at 1:500 dilutions, followed by horseradish peroxidase (HRP)-conjugated goat anti-mouse secondary antibodies diluted at 1:2000. Immunoreactive band was visualized by exposing the membrane to an X-lm (Kodak) with enhanced chemiluminescent reagent (Pierce, USA).

2.7. Immunoblotting analysis of DDH in NSCLC

About 100 mg liquid-nitrogen-frozen tissue specimens were grounded into powder gen and lysed in 400 L cell lysis buffer. The lysates were incubated at 37 C for 30 min, and centrifugated at 20,000 g for 30 min at 4 C. Protein concentration of the supernatants was determined and the supernatants were used for immunoblotting analysis. The procedure of immunoblotting analysis of DDH was the same as above, and GAPDH (Santa Cruz, USA) was used as the internal control. The intensity of DDH and GAPDH bands was quantied by densitometry and the ratio of the DDH/GAPDH from each sample was examined.

2.5. RT-PCR analysis

Total RNA was isolated from tumors and distant lung tissues, respectively, using TRIzol reagent (MRC, Cincinnati, OH, USA). After measurement of RNA yield, cDNA was synthesized by random hexamers according to the manufacturers instructions with RevertAidTM First Strand cDNA Synthesis Kit (Fermentas life science). The primer sequences for DDH gene were as follows: forward primer 5 -CCTGGGATTTGGCACCTAT-3 and reverse primer 5 TTGGGATCACTTCCTCACC-3 . The cycling conditions were as follows: initial denaturation at 94 C for 2 min, followed by 30 cycles at of 94 C for 30 s, 50.2 C for 30 s, and 72 C for 30 s, and nally 72 C for 10 min. Glyceraldehyde-6-

2.8. Immunohistochemical staining

Immunohistochemical staining for DDH was carried out using the SuperPicTureTM Polymer Detection kit (Zymed,

90 San Francisco, CA) according to the manufacturers instructions. The slides with 4- m sections of parafn tissue blocks, were dried at 60 C for 30 min, treated with xylenes then dehydrated in alcohol. Endogenous peroxidase was blocked with 3% H2 O2 . Microwave treatment of the slides was performed for 15 min in 0.01 mol/L citrate buffer (pH 6.0). After incubating the DDH antibody with the tissue at 1:200 dilutions, the HRP polymer was added, and after a wash step, the DAB chromogen is then added to visualize the antibody binding. Slides were counterstained with hematoxylin.

L.-J. Huang et al.

2.11. Statistical analysis

Mean values (mean S.D.) between the groups were compared using the Students unpaired two-tailed t-test. The Fishers exact test was used to examine the association between the DDH expression status and clinicopathological features. All statistical tests were two-sided, and differences were considered signicant when p 0.05. Statistical analysis was performed using SPSS statistical software (Chicago, IL).

2.9. Slide evaluation

The slides were classied independently by two investigators as positive or negative [15]. A specimen was considered positive if more than 10% of all cancer cells were positively stained at cell membrane or cytoplasmic area, and negative if less than 10% positively stained.

3. Results
3.1. Identication of proteins in conditioned medium
In this study, A549 was cultured in RPMI-1640 medium supplemented with low concentration (5%) of fetal bovine serum. It was better for cells propagation and might be better for cells secreting function. However, it would bring exogenous proteins into the conditioned medium samples. To identify the potential secreted protein spots, gels of conditioned medium and control medium were compared, and the specic protein spots in the conditioned medium gel were picked up. These protein spots were either secreted proteins of A549 or degraded products of bovine proteins. In order to identify A549-secreted proteins, we conned the taxonomy to mammalian species in database searching, so that proteins which belonged to Homo sapiens were secreted by A549 and others were bovine proteins. A total of 45 protein spots were identied and 28 belonged to bovine proteins. Seventeen protein spots were identied as human proteins which included 14 proteins (Table 2). On the gel of conditioned medium, the specic protein spot with an apparent pI 7.1 and molecular mass of 32 kDa was iden-

2.10. ELISA detection of DDH in serum

Serum levels of human DDH were determined by enzymelinked immunosorbent assay (ELISA). Microtiter plates (96well) were washed with 0.01 mol/L PBS (pH 7.4) and coated with serum (100 L per well) overnight at 4 C. The plates were then washed with PBS and blocked with 200 L of ovalbumin (Sigma) (1 mg/mL in PBS) for 2 h. Subsequently, DDH antibody at 1:1000 dilutions in PBS (100 L/well) were applied and incubated for 1 h. After washing, 100 L of HRP-conjugated goat anti-mouse IgG (diluted 5000-fold in PBS, Sigma) was added and incubated for 1 h. 3,5,3 ,5 tetramethylbenzidine (TMB) was used as the chromogen and examined at 450 nm with a plate reader (Bio-Rad model 680). All samples were assayed in triplicate.

Table 2

Protein spots searched by Mascot software in database Accession codea P06733 P62937 Q6LEN1 Q06830 P30086 P09211 P09936 P63261 P09382 CAD34991 Q96E67 P48546 P52895 P18669 Mw /pI 47350/6.99 18098/7.82 22176/6.86 22324/8.27 20443/7.18 23583/5.43 23840/5.43 29678/5.50 14917/5.34 53597/8.36 40536/5.55 53964/9.09 37111/7.13 28769/6.85 Matched number 8 15 6 7 9 9 12 15 9 9 15 6 8 10 Coverage (%) 21 54 53 53 74 52 61 60 73 20 61 20 41 51 Function Metabolism Signal transduction Antioxidants Antioxidants Signal transduction Antioxidants Thiol protease Cytoskeleton Signal transduction Cytoskeleton Metabolism Metabolism Metabolism

Protein name Alpha enolase (ENO1) Peptidylprolyl cistrans isomerase A (PPIA) MnSOD Peroxiredoxin 1 (PDX1) Phosphatidylethanolamine-binding protein (PEBP) Glutathione S-transferase P (GSTP1-1) Ubiquitin carboxyl-terminal hydrolase isozyme L1 (PGP9.5) Gamma-actin Galectin-1 (GAL1) Sequence 299 from Patent WO0222660 ACTB protein Gastric inhibitory polypeptide receptor (GIPR) Dihydrodiol dehydrogenase 2 (DDH) Phosphoglycerate mutase 1 (PGAM1)
a

Accession codes refer to the Swiss-PROT database.

Proteomics-based identication of secreted protein dihydrodiol dehydrogenase

91

Fig. 1 Identication of DDH by peptide mass ngerprints. (A) Peptide mass spectrum was obtained by MALDI-TOF mass spectrometry. (B) The masses of eight tryptic peptides were matched with DDH by database searching. (C) The sequence of DDH is represented by single-letter code for amino acids. Sequence coverage by eight peptides is indicated by capital letters, which is 41% of full length.

tied as DDH by mass spectrometry (Fig. 1). Since it was known that DDH was located in cytoplasm and augmented expression of DDH was detected in NSCLC [1518], DDH was selected for further analysis.

the control medium were examined at the same time as control. As shown in Fig. 2, DDH were detected in conditioned medium and A549 cells, and no positive band was detected in the control medium.

3.2. Verication of secreted protein DDH by immunoblotting

To verify the results obtained from MALDI-TOF MS, protein samples of conditioned medium and protein extract of A549 cells were subjected to immunoblotting analysis. Proteins in

3.3. Highly expressed DDH in NSCLC tissues

Differentially expression of DDH between cancer tissues and distant nonmalignant lung tissue in NSCLC patients was studied using RT-PCR and immunoblotting. As shown in Fig. 3, DDH mRNA overexpression was detected in cancer tissues

92

L.-J. Huang et al.

Fig. 2 Identication of DDH by immunoblotting: DDH band were positive in conditioned medium of A549 (A) and A549 cell lysate (A549), and negative in control medium (C). Compared with paired distant nonmalignant lung tissues (L), DDH expression were higher in NSCLC cancer tissues (T).

Fig. 4 Highly expressed DDH in NSCLC. The expression levels of DDH were compared by immunoblotting between lung (L) and cancer (C) tissues. GAPDH were served as internal control.

(p < 0.05). Same as DDH gene, it was found that DDH protein was highly expressed in cancer tissues identied by immunoblotting (p < 0.05) (Figs. 2 and 4).

cated that DDH might be one of potential serum biomarker for NSCLC.

4. Discussion
Currently, invasive techniques, such as bronchoscopy, and percutaneous ne-needle aspiration are still required to make the nal diagnosis of NSCLC [16]. Noninvasive methods, such as testing serum biomarker level of NSCLC showed variable sensitivity and specicity [1012]. However, compared to invasive methods, serum biomarker tests have great potential to facilitate the early detection, monitoring of lung cancer in a non-painful way. The analysis of the secreted proteins of lung cancer cells could provide new clues to identify potential serological biomarkers of lung cancer. In this study, a total of 14 human secreted proteins were identied. Some of these identied proteins were conrmed as secreted proteins in previous studies, it included PPIA [17], PGP9.5 [18], PEBP [19] and GAL1 [20], which were consistent with our study. Some of the identied secreted proteins in this investigation had been conrmed as tumor markers of NSCLC, which included PPIA [20,21], PDX1 [1,22], PGP9.5 [18,23,24], PGAM1 [25] and DDH [15,26]. In the previous study, PGP9.5 was identied in serum and found autoantibodies against PGP9.5 in the circulation from patients with lung cancer which suggested that PGP9.5 may have utility in lung cancer screening and diagnosis [18]. Serum concentration of secreted proteins were related with the

3.4. Immunohistochemical study of DDH

The tissue samples were immunostained to examine the differential expression of DDH in cancer tissues and lung tissues, and to evaluate the cellular distributions of DDH. Positive immunostainings were found in all cancer tissues (n = 15, Fig. 5), and no immunoreactivity was found in the lung tissues. The immunoreactivity against DDH was strong in the cytoplasm and cell membrane (Fig. 5C).

3.5. ELISA analysis for serum DDH

The relative levels of DDH in sera collected from adenocarcinoma patients (n = 34), squamous carcinoma patients (n = 30), benign lung tumor patients (n = 20) and healthy controls (n = 20) were examined using the ELISA system. As shown in Table 1, the serum levels of DDH were signicantly higher in NSCLC patients than in healthy controls (p < 0.05) and benign lung tumor patients (p < 0.05). There was no signicantly difference between lung adenocarcinoma patients and squamous carcinoma patients. And no signicantly differences were detected between healthy controls and benign lung tumor patients. These results indi-

Fig. 3 RT-PCR analysis of DDH expression in NSCLC. (A) DDH expression in samples of NSCLC tissue and distant nonmalignant lung tissue of the same patient; (1) DNA marker, (2) GAPDH of cancer tissue, (3) GAPDH of lung tissue, (4) DDH of cancer, (5) DDH of lung. (B) Overexpression of DDH in NSCLC was detected (n = 15, p < 0.05).

Proteomics-based identication of secreted protein dihydrodiol dehydrogenase

93

cell proliferation and/or its expression, so the secreted proteins which overexpressed would have more possibility to be serum biomarkers of lung cancer. Further investigation would be undertaken to seek new serum biomarkers from these identied proteins. DDH was a member of aldo-keto reductase superfamily; it was considered as cytoplasmic protein in lung, liver, kidney, prostate and mammary gland [27]. Overexpression of DDH in NSCLC has been shown in several previous study [1518], but those studies focused on detection of cytoplamic DDH [15,26,28,29], our study is the rst in the world to investigate DDH as a secreted protein. In this study, we revealed that DDH not only existed in conditioned medium of A549 cell, on the membrane of cancer cells of NSCLC, but also in the sera of NSCLC patients. DDH overexpression was conrmed in NSCLC in our investigation using RT-PCR, immunoblotting and immunohistochemistry. The serum DDH level was checked using indirect ELISA in this study to validate the probability of using DDH as serum biomarkers. The serum level of DDH was signicantly higher in NSCLC patients than nonmalignant lung tumor and healthy controls; thus, it might represent a new potential serum biomarker for NSCLC. However, the specicity and sensitivity of DDH in diagnosis of NSCLC need further investigation. DDH could divert polycyclic aromatic hydrocarbons (PAHs) trans-dihydrodiols to the deleterious o-quinones, which had the potential to cause covalent and oxidative DNA lesions, increasing the mutation of PAH-exposed lung cells. PAHs, as ubiquitous environmental pollutants and tobacco carcinogens, are implicated in the causation of lung cancer [28]. In human, four isoforms of DDH (DDH1DDH4) have been identied with monomeric mass of 36 kDa [27]. There is >97% identity in amino acid sequence. DDH1 and DDH2 was the major isoforms in NSCLC detected by RT-PCR and nucleotide sequencing [15,26]. Correlation between clinicopathological parameters and DDH expression in patients with NSCLC was evaluated, which indicated overexpression of DDH was a prognostic marker [15] and might serve as an early biomarker for patients with resectable stage I NSCLC [26]. In summary, identication of secreted proteins could be a feasible and effective strategy to search potential serum biomarkers of cancer. DDH represents a new diagnostic and prognostic factor in NSCLC, and can be used as a serum marker. Further investigation of other secreted proteins detected in this study, may reveal interesting results too.

Acknowledgements
The authors thank Zhi-Ling Huang for excellent technical assistance and Chang-Qing Xie and Ming-Qing Li for their critical reading of this article.

References
Fig. 5 Immunohistochemical staining of DDH in NSCLC. (A) Strong cytoplasmic staining of DDH in lung squmaous cancer and (B) strong cytoplasmic immunoreactivity of DDH in adenocarcinoma. (C) DDH shows strong cell membrane positivity.
[1] Wu CC, Chien KY, Tsang NM, Chang KP, Hao SP, Tsao CH, et al. Cancer cell-secreted proteomes as a basis for searching potential tumor markers: nasopharyngeal carcinoma as a model. Proteomics 2005;5(12):317382. [2] Welsh JB, Sapinoso LM, Kern SG, Brown DA, Liu T, Bauskin AR, et al. Large-scale delineation of secreted protein biomarkers

94
overexpressed in cancer tissue and serum. Proc Natl Acad Sci USA 2003;100(6):34105. Luo LY, Katsaros D, Scorilas A, Fracchioli S, Bellino R, van Gramberen M, et al. The serum concentration of human kallikrein 10 represents a novel biomarker for ovarian cancer diagnosis and prognosis. Cancer Res 2003;63(4):80711. Mok SC, Chao J, Skates S, Wong K, Yiu GK, Muto MG, et al. Prostasin, a potential serum marker for ovarian cancer: identication through microarray technology. J Natl Cancer Inst 2001;93(19):145864. Balk SP, Ko YJ, Bubley GJ. Biology of prostate-specic antigen. J Clin Oncol 2003;21(2):38391. Kosari F, Asmann YW, Cheville JC, Vasmatzis G. Cysteine-rich secretory protein-3: a potential biomarker for prostate cancer. Cancer Epidemiol Biomarkers Prevent 2002;11(11):141926. Buckhaults P, Rago C, St Croix B, Romans KE, Saha S, Zhang L, et al. Secreted and cell surface genes expressed in benign and malignant colorectal tumors. Cancer Res 2001;61(19):69967001. Kim JH, Skates SJ, Uede T, Wong KK, Schorge JO, Feltmate CM, et al. Osteopontin as a potential diagnostic biomarker for ovarian cancer. JAMA 2002;287(13):16719. Zhou W, Sokoll LJ, Bruzek DJ, Zhang L, Velculescu VE, Goldin SB, et al. Identifying markers for pancreatic cancer by gene expression analysis. Cancer Epidemiol Biomarkers Prevent 1998;7(2):10912. Seemann MD, Beinert T, Furst H, Fink U. An evaluation of the tumour markers, carcinoembryonic antigen (CEA), cytokeratin marker (CYFRA 21-1) and neuron-specic enolase (NSE) in the differentiation of malignant from benign solitary pulmonary lesions. Lung Cancer 1999;26(3):14955. Schneider J, Velcovsky HG, Morr H, Katz N, Neu K, Eigenbrodt E. Comparison of the tumor markers tumor M2-PK, CEA, CYFRA 21-1, NSE and SCC in the diagnosis of lung cancer. Anticancer Res 2000;20(6D):50538. Schneider J, Bitterlich N, Velcovsky HG, Morr H, Katz N, Eigenbrodt E. Fuzzy logic-based tumor-marker proles improved sensitivity in the diagnosis of lung cancer. Int J Clin Oncol 2002;7(3):14551. Gorg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R, et al. The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 2000;21(6):103753. Candiano G, Bruschi M, Musante L, Santucci L, Ghiggeri GM, Carnemolla B, et al. Blue silver: a very sensitive colloidal Coomassie G-250 staining for proteome analysis. Electrophoresis 2004;25(9):132733. Hsu NY, Ho HC, Chow KC, Lin TY, Shih CS, Wang LS, et al. Overexpression of dihydrodiol dehydrogenase as a prognostic marker of non-small cell lung cancer. Cancer Res 2001;61(6): 272731. Hoffman PC, Mauer AM, Vokes EE. Lung cancer. Lancet 2000;355:47985.

L.-J. Huang et al.

[17] Andersen H, Jensen ON, Eriksen EF. A proteome study of secreted prostatic factors affecting osteoblastic activity: identication and characterisation of cyclophilin A. Eur J Cancer 2003;39(7):98995. [18] Brichory F, Beer D, Le Naour F, Giordano T, Hanash S. Proteomics-based identication of protein gene product 9.5 as a tumor antigen that induces a humoral immune response in lung cancer. Cancer Res 2001;61(21):790812. [19] Hengst U, Albrecht H, Hess D, Monard D. The phosphatidylethanolamine-binding protein is the prototype of a novel family of serine protease inhibitors. J Biol Chem 2001;276(1):53540. [20] Bao L, Kimzey A, Sauter G, Sowadski JM, Lu KP, Wang DG. Prevalent overexpression of prolyl isomerase Pin1 in human cancers. Am J Pathol 2004;164(5):172737. [21] Campa MJ, Wang MZ, Howard B, Fitzgerald MC, Patz Jr EF. Protein expression proling identies macrophage migration inhibitory factor and cyclophilin a as potential molecular targets in non-small cell lung cancer. Cancer Res 2003;63(7):16526. [22] Lehtonen ST, Svensk AM, Soini Y, Paakko P, Hirvikoski P, Kang SW, et al. Peroxiredoxins, a novel protein family in lung cancer. Int J Cancer 2004;111(4):51421. [23] Hibi K, Westra WH, Borges M, Goodman S, Sidransky D, Jen J. PGP9.5 as a candidate tumor marker for non-small-cell lung cancer. Am J Pathol 1999;155(3):7115. [24] Sasaki H, Yukiue H, Moriyama S, Kobayashi Y, Nakashima Y, Kaji M, et al. Expression of the protein gene product 9.5, PGP9.5, is correlated with T-status in non-small cell lung cancer. Jpn J Clin Oncol 2001;31(11):5325. [25] Chen G, Gharib TG, Wang H, Huang CC, Kuick R, Thomas DG, et al. Protein proles associated with survival in lung adenocarcinoma. Proc Natl Acad Sci USA 2003;100(23):1353742. [26] Chen CY, Hsu CP, Hsu NY, Shih CS, Lin TY, Chow KC. Expression of dihydrodiol dehydrogenase in the resected stage I non-small cell lung cancer. Oncol Rep 2002;9(3):5159. [27] Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, et al. Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Biochem J 2000;351(Pt 1):6777. [28] Palackal NT, Lee SH, Harvey RG, Blair IA, Penning TM. Activation of polycyclic aromatic hydrocarbon trans-dihydrodiol proximate carcinogens by human aldo-keto reductase (AKR1C) enzymes and their functional overexpression in human lung carcinoma (A549) cells. J Biol Chem 2002;277(27):24799808. [29] Provost PR, Blomquist CH, Godin C, Huang XF, Flamand N, Luu-The V, et al. Androgen formation and metabolism in the pulmonary epithelial cell line A549: expression of 17beta-hydroxysteroid dehydrogenase type 5 and 3alpha-hydroxysteroid dehydrogenase type 3. Endocrinology 2000;141(8):278694.

[3]

[4]

[5] [6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]