Shi-Yong Li, Ping An, Hui-Yun Cai, Xue Bai, Ying-Nan Zhang, Bo Yu, Fu-Yi Zuo and Gang Chen

Beijing, China

Proteomic analysis of differentially expressed proteins involving in liver metastasis of human colorectal carcinoma

Shi-Yong Li, Ping An, Hui-Yun Cai, Xue Bai, Ying-Nan Zhang, Bo Yu, Fu-Yi Zuo and Gang Chen

Beijing, China

(Hepatobiliary Pancreat Dis Int 2010; 9: 149-153)

colorectal carcinoma;liver metastasis;proteomic analysis;human arginase

Introduction

The incidence of colorectal cancer is increasing year by year. Although surgical procedures have been improved, the fi ve-year survival rate of such patients after surgery is only 60% and the recurrence rate is 40%-70%. Upon diagnosis, 25% of the patients have already had liver metastases, and another 25%have had liver metastases after surgery for colorectal carcinoma.[1]This is the major cause of death in these patients. To the present, there has been no effective approach for early detection and diagnosis of liver metastasis of colorectal carcinoma.

Therefore, it is extremely important to identify molecular tumor markers for liver metastasis of colorectal carcinoma so as to fi nd out a rationale for early detection and diagnosis of liver metastasis of this tumor, which could improve the fi ve-year survival rate after surgery and the prevention of colorectal carcinoma.Since the change and regulation of protein expression in the development of colorectal cancer have been the targets of investigation,[2,3]we used proteomic analysis to study the differentially regulated proteins in normal mucosa, primary focus, and liver metastases. The proteins involved in liver metastases from colorectal carcinoma were isolated and used to screen for molecular tumor markers.

MethodsSample collection

Sixteen patients were diagnosed with colorectal carcinoma associated with liver metastases by clinical images. At surgery, fresh primary foci of colorectal carcinoma and liver metastatic tissues were immediately frozen in liquid nitrogen. The samples of specimens were further con fi rmed to be colorectal adenocarcinoma with modest or poor differentiation along with liver metastasis. The patients were 9 men and 7 women, aged from 41 to 75 years. Among them, 8 patients suffered from colon cancer, and the other 8 from rectal cancer.According to the TNM system, their tumors were determined to be at stage Ⅳ (T2-4N1-2M1).

Preparation of protein samples and fl uorescence two-dimensional differential in-gel electrophoresis(2-D DIGE)

2-D DIGE was performed following the instructions provided by the Ettan DIGE handbook (GE Healthcare).Protein was extracted with lysis buffer from the tissue samples (7 mol/L urea, 2 mol/L thiourea, 4%3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate (CHAPS), and 30 mmol/L Tris-HCL 9.0, pH 8.4).The concentration of extracted protein was determined using the Bradford assay. Two aliquots of equal amounts of protein from each patient were mixed as an internal standard (Cy2 labeled). Fifty micrograms of protein and 400 pmol of CyDye DIGE Fluor minimal dye were mixed as a sample to be loaded on a gel. Cy3 was used to label the tissues from the primary focus of colorectal carcinoma,and Cy5 was used to label liver metastases. The samples to be mixed were vortexed. After centrifugation, the labeled mixture was collected at the bottom of each vial.The reaction mixture was kept on ice for 30 minutes in dark, followed by the addition of one aliquot of buffer (7 mol/L urea, 2 mol/L thiourea, 2% CHAPS [w/v] 2% DTT[w/v], and 2% IPG buffer). After 10 minutes of reaction on ice, it was mixed with rehydration buffer (7 mol/L urea, 2 mol/L thiourea, 2% CHAPS [w/v], 2% DTT [w/v],1% IPG buffer and a trace amount of bromophenol blue).The samples were loaded into each lane of the gel, and the 24 cm IPG gel was run. PROTEAN IFF CELL (Bio-Rad) was used to perform isoelectric focusing (IEF). The conditions for IEF included 50 V, 13 hours; 200 V, 1 hour;500 V, 1 hour; 1000 V, 1 hour; and 10 000 V, 5 hours. An Ettan DALT Twelve electrophoresis system (Amersham)was used to perform two-dimensional separation of the protein samples. The times for two-dimensional separation were: 0.5 W/piece, 1 hour; 1 W/piece, 1 hour;5 W/piece, until the trace of bromophenol blue migrated towards the bottom of the gel. After electrophoresis, the gel was placed in sodium dodecyl sulfate electrophoresis buffer in a dark container (room temperature) and then immediately scanned.

Image capture and analysis

A Micotech K8 scanner was used to scan the stained gel at a resolution of 600 dpi. Images were analyzed with Melanie 3.0 software (GeneBio, Geneva). After background subtraction, spot detection, and matching,standard gels were generated. After comparison and matching, differential expression of proteins was obtained.

Proteomic analysis and database searching

After being digested from the gel, the differentially regulated proteins appearing were analyzed with liquid chromatography coupled to tandem mass spectrometry with electrospray ionization (LC-ESI-MS/MS) and mass spectrometry (MALDI-TOF-TOF). The information about peptide fragmentation sequence (post-source decay) and peptide mass fi ngerprint was obtained,followed by data searching in the SWISSPORT database with Mascot software for protein identi fi cation.

Con fi rmation by immunohistochemistry

The expression of human arginase in the primary focus of colorectal carcinoma and liver metastasis tissues was con fi rmed by routine immunohistochemical staining. Thirty specimens were taken from colorectal carcinoma with liver metastasis. Immunostaining was performed on paraf fi n-embedded samples with an SP kit (ZYMED Co., USA). The general procedure was performed with the following steps: dehydration of paraf fi n-embedded tissues, quenching with 3% H2O2,soaking in phosphate buffer saline (PBS), antigen retrieval in the microwave, blocking with 5%-10%normal goat serum, storage at 4 ℃ overnight with the primary antibody, washing with PBS, dilution with biotinylated secondary antibody, DAB staining, nuclear staining with hematoxylin, and mounting on slides.

Statistical analysis

SPSS10.0 software was used for statistical analysis.Student's t test was performed to analyze differentially regulated proteins. The Chi-square analysis was made of the results of immunohistochemical staining. P≤0.05 was considered statistically signi fi cant.

ResultsTwo-dimensional electrophoresis

Fig. 1. Cy3 staining for colorectal carcinoma (green), and Cy5 staining (red) for liver metastasis on DIGE images.

In 2-D DIGE of tissue specimens from the 16 patients with colorectal cancer and liver metastases, protein spots were clearly visualized, with around 900 protein spots on each gel (Fig. 1). A signi fi cant difference was found between the proteins from the primary foci of colorectal carcinoma and the liver metastasis. DeCyder v.5.02 image analysis software was used to determine the differences between the images. After background subtraction, calibration within the gel, and erasing artifact spots, the matching rate of each gel and the master gel reached over 90%. In comparison of liver metastases and colorectal cancer, 46 protein spots were found to be signi fi cantly different, with a 1.5-fold change (Fig. 2). Within the ranges of molecular weight 10-40 kD and pH 6.0-9.0, three protein spots were found to have an over 5-fold difference in colorectal cancer tissues as compared with liver metastases and normal mucosa. Within the ranges of molecular weight 20-40 kD and pH 5.0-9.0, 17 protein spots were found to have an over 5-fold difference in liver metastases compared with colorectal cancer tissues and normal mucosa.These results all reproducibly appeared on parallel gels.Protein spots with signi fi cant differences were collected for mass spectrometry analysis.

Fig. 3. Expression of human arginase in carcinoma tissues (original magnification ×200). A: liver metastasis; B: primary focus of colon cancer.

Mass spectrometry analysis

Mass spectrometry was performed for the identifi cation of 20 protein spots. Eighteen proteins were found to be signi fi cantly relevant. Two proteins, activator protein 2B and Homo sapiens S-adenosylmethionine transgelin variant, were increased in the primary focus of colorectal carcinoma but decreased in liver metastasis.Sixteen proteins were increased in liver metastatic tissues, including zinc fi nger protein 64 homolog,guanine nucleotide exchange factor 4, Homo sapiens arginase, Homo sapiens glutathione S-transferase A3,tumor necrosis factor, alpha-induced protein 9, chain A, human glutamate dehydrogenase-apo form, tolllike receptor adaptor molecule 2, and ribonucleotide reductase M2 polypeptide.

Con fi rmation with immunohistochemical staining

Human arginase was localized in the cytosol of colorectal carcinoma and metastatic liver tissues. In the cytosol, yellow and brown pixel staining was visualized. Of 30 samples of liver metastasis, 93.3%showed a positive arginase staining, whereas 66.7% of the primary foci of colorectal carcinoma tissues showed a positive staining (P＜0.01). The positive rate and signal were signi fi cantly higher in liver metastatic tissues than in the primary focus of colorectal carcinoma (Fig. 3).This result is consistent with that obtained from the proteomic analysis.

Discussion

With the development of technology, proteomic analysis may become an effective approach for detecting the initiation and invasion of colorectal cancer and for screening the hallmarks of tumors.[4-8]2-D DIGE, which is a new and widely-used proteomic technology, has been used to study colorectal cancer.[9-11]Its accuracy and reproducibility exceed those of other routine approaches.In our study, 2-D DIGE and mass spectrometry were performed to compare and analyze the protein expression in normal mucosa,[12-14]the primary focus of colorectal cancer, and liver metastases. Proteomic technology has been used in bioinformatic analysis of protein spots with a signi fi cant differential regulation.Some differentially regulated proteins associated with metastasis were identi fi ed.

Among the screened proteins, two were highly expressed in the primary focus of colorectal carcinoma,but had a lower expression in liver metastases;these proteins were the activator protein 2B and S-adenosylmethionine transgelin variant. Proteomic analysis has shown that S-adenosylmethionine is highly expressed in stomach cancer tissues.[15]The mechanism for the high expression of these two proteins in the primary focus is poorly understood, and their role during metastasis remains to be elucidated.

Among the 16 proteins that were highly expressed in liver metastases, some may be related to resistance to drugs, such as glutathione S-transferase (GST) A3 and chain A and human glutamate dehydrogenaseapo form. GST can bind to hydrophobic cytotoxic drugs, which may increase their water solubility and accelerate their metabolism. This effect can decrease the cytotoxicity of anti-cancer drugs. GSTA3 is also a highly ef fi cient catalytic isomerase in the biosynthesis of steroid hormone.[16]But the mechanisms of its drug resistance need to be further studied, and it may be related to metastasis of colorectal cancer to the liver.Human arginase is a hydrolase expressed in the liver and catalyzes the fi nal step of the ornithine cycle to generate urea and ornithine.[17,18]It was reported that in patients with elevated plasma arginase activity, the incidence of recurrence and metastasis of colorectal cancer was signi fi cantly higher.[19]In our study, arginase activity was signi fi cantly higher in liver metastases than in the primary focus, which was further con fi rmed by immunohistochemistry. Human arginase is directly associated with liver metastases of colorectal cancer, and it may be a molecular biomarker for the detection and diagnosis of liver metastasis of colorectal cancer.

Zinc fi nger protein 64 homolog, guanine nucleotide exchange factor 4 (ASEF4), and ribonucleotide reductase M2 polypeptide are associated with the transcription and translation of genes. Zinc fi nger protein is the most abundant motif in mammalian cells, and it is the largest category in all the speci fi c DNA-binding proteins. About 2% of human genes encode zinc fi nger proteins. These proteins are closely associated with the expression and regulation of eukaryocytic genes. ASEF may be an anticancer gene during the initiation and development of colorectal cancer.[20]The role of other proteins during liver metastasis is still not clear, and they include hCG2039834, PRP19/PSO4 pre-mRNA processing factor 19 homolog and Chain A, and the crystal structure of human Fcari bound to Iga1-Fc. Further investigation of these proteins may shed some lights on the mechanism of liver metastasis.

In this study proteomic analysis was made of proteins involving in the metastasis of colorectal cancer to the liver, and important differentially-regulated proteins were found. These proteins may play a role during the initiation and metastasis of colorectal cancer.Some proteins are directly associated with metastasis and could be molecular biomarkers for the detection and diagnosis of liver metastases of colorectal cancer.With the progress in proteomic research, the molecular mechanism of liver metastases of colorectal cancer can be further elucidated. Finally molecular tumor biomarkers with a high speci fi city and sensitivity can be identi fi ed.

Funding: This study was supported by grants from the National Natural Science Foundation of China (No. 30471700) and the"Tenth one fi ve" Science and Technique Foundation of the PLA,China (No. 06G027).

Ethical approval: Not needed.

Contributors: LSY proposed the study. AP wrote the fi rst draft.CHY analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. LSY is the guarantor.

Competing interest: No bene fi ts in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Sadahiro S, Suzuki T, Ishikawa K, Nakamura T, Tanaka Y,Masuda T, et al. Recurrence patterns after curative resection of colorectal cancer in patients followed for a minimum of ten years. Hepatogastroenterology 2003;50:1362-1366.

2 Greco C, Alvino S, Buglioni S, Assisi D, Lapenta R, Grassi A, et al. Activation of c-MYC and c-MYB proto-oncogenes is associated with decreased apoptosis in tumor colon progression. Anticancer Res 2001;21:3185-3192.

3 Thorstensen L, Qvist H, Nesland JM, Giercksky KE, Lothe RA. Identi fi cation of two potential suppressor gene regions on chromosome arm 14q that are commonly lost in advanced colorectal carcinomas. Scand J Gastroenterol 2001;36:1327-1331.

4 Bichsel VE, Liotta LA, Petricoin EF 3rd. Cancer proteomics:from biomarker discovery to signal pathway pro fi ling. Cancer J 2001;7:69-78.

5 Srinivas PR, Srivastava S, Hanash S, Wright GL Jr. Proteomics in early detection of cancer. Clin Chem 2001;47:1901-1011.

6 Alfonso P, Núnez A, Madoz-Gurpide J, Lombardia L, Sánchez L, Casal JI. Proteomic expression analysis of colorectal cancer by two-dimensional differential gel electrophoresis.Proteomics 2005;5:2602-2611.

7 Mazzanti R, Solazzo M, Fantappié O, Elfering S, Pantaleo P,Bechi P, et al. Differential expression proteomics of human colon cancer. Am J Physiol Gastrointest Liver Physiol 2006;290:G1329-1338.

8 Friedman DB, Hill S, Keller JW, Merchant NB, Levy SE,Coffey RJ, et al. Proteome analysis of human colon cancer by two-dimensional difference gel electrophoresis and mass spectrometry. Proteomics 2004;4:793-811.

9 Lawrie LC, Fothergill JE, Murray GI. Spot the differences:proteomics in cancer research. Lancet Oncol 2001;2:270-277.

10 Simpson RJ, Dorow DS. Cancer proteomics: from signaling networks to tumor markers. Trends Biotechnol 2001;19:S40-48.

11 Ong SE, Mann M. Mass spectrometry-based proteomics turns quantitative. Nat Chem Biol 2005;1:252-262.

12 Unlü M, Morgan ME, Minden JS. Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis 1997;18:2071-2077.

13 Colangelo CM, Williams KR. Isotope-coded af fi nity tags for protein quanti fi cation. Methods Mol Biol 2006;328:151-158.

14 Stulík J, Hernychová L, Porkertová S, Knízek J, Macela A,Bures J, et al. Proteome study of colorectal carcinogenesis.Electrophoresis 2001;22:3019-3025.

15 Ryu JW, Kim HJ, Lee YS, Myong NH, Hwang CH, Lee GS, et al. The proteomics approach to fi nd biomarkers in gastric cancer. J Korean Med Sci 2003;18:505-509.

16 Johansson AS, Mannervik B. Human glutathione transferase A3-3, a highly ef fi cient catalyst of double-bond isomerization in the biosynthetic pathway of steroid hormones. J Biol Chem 2001;276:33061-33065.

17 del Ara RM, González-Polo RA, Caro A, del Amo E, Palomo L, Hernández E, et al. Diagnostic performance of arginase activity in colorectal cancer. Clin Exp Med 2002;2:53-57.

18 Polat MF, Taysi S, Polat S, Boyük A, Bakan E. Elevated serum arginase activity levels in patients with breast cancer. Surg Today 2003;33:655-661.

19 Porembska Z, Mielczarek M, Nyckowski P, Baranczyk-Kuzma A. Arginase as a marker of cancerogenesis. I. Monitoring patients after resection of colorectal cancer. Pol Merkur Lekarski 2002;13:284-285.

20 Mitin N, Betts L, Yohe ME, Der CJ, Sondek J, Rossman KL.Release of autoinhibition of ASEF by APC leads to CDC42 activation and tumor suppression. Nat Struct Mol Biol 2007;14:814-823.

BACKGROUND: Early diagnosis of liver metastasis of colorectal carcinoma is very important for the appropriate treatment of such patients. However, there has been no effective approach available for clinical application. The present study aimed to investigate the differential expression of proteins in patients with liver metastasis of colorectal carcinomas using proteomic analysis and evaluate its potentiality in clinical diagnosis.

METHODS: Fluorescence two-dimensional differential in-gel electrophoresis (2-D DIGE) was used to analyze and compare the protein expression between normal mucosa, the primary focus,and liver metastases. Proteomic analysis was made to identify the differentially expressed proteins. Immunohistological staining was used to con fi rm the expression of differentially expressed proteins in colorectal carcinomas and areas of liver metastasis.

RESULTS: A 1.5-fold difference was found with 46 differentially expressed proteins. In 20 differentially expressed proteins, 3 were down-regulated and 17 up-regulated in liver metastases.Proteomic analysis showed that the S-adenosylmethionine transgelin variant was down-regulated in liver metastasis tissues. Zinc fi nger protein 64 homolog (Zfp64), guanine nucleotide exchange factor 4 (GEF4), human arginase,glutathione S-transferases (GSTs) A3, and tumor necrosis factor α (TNF-α)-induced protein 9 were up-regulated in liver metastasis tissues. Immunohistochemical staining con fi rmed that human arginase expression was higher in liver metastases than in the primary focus.CONCLUSIONS: There was a signi fi cant difference in protein expression between the primary focus of colorectal carcinoma and liver metastases. The differentially regulated proteins were closely related to liver metastasis of colorectal carcinoma.Elevated human arginase may be an important molecular marker for liver metastasis from colorectal carcinoma.

Author Af fi liations: Department of General Surgery, General Hospital of Beijing Military Command, Beijing 100700, China (Li SY, An P, Cai HY, Bai X, Zhang YN, Yu B, Zuo FY and Chen G)

Shi-Yong Li, MD, PhD, Department of General Surgery, General Hospital of Beijing Military Command, Beijing 100700,China (Tel: 86-10-66721188; Fax: 86-10-64001388; Email: lsybz@126.com)

? 2010, Hepatobiliary Pancreat Dis Int. All rights reserved.

September 21, 2009

Accepted after revision March 14, 2010