行政院國家科學委員會專題研究計畫 成果報告
以螢光二次元差異性電泳及質譜儀來分析肝癌與其相對應
的非癌肝組織，蛋白質表現的差異
計畫類別： 個別型計畫 計畫編號： NSC92-2314-B-002-143- 執行期間： 92 年 08 月 01 日至 93 年 07 月 31 日 執行單位： 國立臺灣大學醫學院內科 計畫主持人： 許金川 報告類型： 精簡報告 處理方式： 本計畫可公開查詢中 華 民 國 93 年 11 月 2 日
! " # $ % & ' ( ) * + , - . / 0 1 234 5 6 7 8 9 : ; < = > ? @- A B C D = E F G H H H I3H H H J K ? L M N O P Q R S T U V W X Y Z [ \ ] ^ _ `Q a [ b ^ c N d - e f g h < i -j k lm n h o - p @q r s t u v . w x s y ' z { | - p @l@ } ~ k - e f - U - a [ g ' - l$ D L > - @ ' ¡ ¢ £ i ¤ Q U ¥ ¦ § i ¤ @¨ t lM N ©ª« ' i ¤ @¨ t { - ¬ - ® ¯ 7 ' ° i ¤ c § - ± ² ' ³ ´ µ ¶ d ³ ´ µ ¶ - ¨ t h Z · - ¨ t ¸ ¹ lº » ¼ ½ ¾ ¿ À ¹ Á  à - { Ä Å h ' Æ $Ç µ ¶ $È$´ µ ¶ $i ¤ 7 `É Ê ~ D § Ë Ì - Í Î Ï Ð Ñ U - Ò Ó Ô Õ Ö × Ø i ¤ @} - Ù Ú i ¤ Û¼ ½ Ü FÝ - Õ l `Q Þ ' Ûn @ ß à - á â ã ¬ ä l ! i ¤ ` Q Þ ' å æ D - ç l0 z è é ê ë ì í c § Q ! i ¤ . Z c ° - O î ¯ ïð c } î ¯ n `× Ø Ûñ Q ò ó ô õ . ö Q $÷ ø ù ú $Èû ü ! i ý c þ } ` " # È Ó Z # i ¤ Ó ¦ ° É Ó ¦ Æ ` p - i lM N ! i ¤ - Û N d F - ñ ' ( è - ©É h ñ ·h Ù Ú - { ¿ ! - l$ " # ! `Q $ % Ü & - ® ¯ l©z ' ( ' ) L ú * + · ú * , © W h - " # » § - - W N n . © W Û ñ Ü / ! N n 0 1 § i ¤ © W " # - 2 3 } § 4 ¨ t 5 - l. F - ¯ ð c Q ò ó ô õ . 0 " # È Ó Z # i ¤ Ó ¦ ` p - i lÆ L © z ' Û ñ Ü / ! ° Q i ¤ ß `Q 6 7 ñ ·ñ . - É i ¤ - 8 `Ö ß y `Q 9 ñ . - l Û} : ; < 3 · = - > ? @ § " # ! i ¤ l; < s t F 2 z AB ~ $C D E F $G C H I J $K L H F D M N OÛ ? @ - ¨ t = 8 1 1 z Û ? @ - ¨ t = Pl; < O P ¿ Û ! Z © ª ' Q ¾ ¿ `Q ä  - R S l$ $ T
Keywordshepatocellular carcinoma, proteomics, differential gel electrophoresis, DIGE, MALDI, mass spectrometry
Hepatocellular carcinoma (HCC) has been the leading cause of cancer death in Taiwan. About 6000-8000 people died of this cancer every year in Taiwan. Though regular sonographic examination can early detect small HCC and there are many therapeutic modalities for HCC, the therapeutic results remains unsatisfactory. To improve the survival, further investigation of the early diagnostic markers and the mechanisms of hepatocarcinogenesis is very important.
In the recent years, investigating the genome-wide expression profiles of cancers has been the predominant method to identify cancer-related genes. Though using cDNA microarray for genome-wide expression profiling is a very powerful tool to clarify the genetic changes in cancers, the major pitfall of these methods is that the mRNA expression does not parallel protein expression in many cases.
The introduction of fluorescent 2D differential gel electrophoresis (DIGE) has now made it possible to detect and quantitate differences between experimental pairs of samples resolved on the same 2D gel. The basis of this technique is to use two fluorescent dyes (Cy3 and Cy5) to differentially label lysine residues of two protein samples for comparative analysis on a single gel. The ability to directly compare two samples on the same gel not only avoids the complications of gel-to-gel variation but also enables a more accurate and rapid analysis of differences and reduces the number of gels that need to be run. Following automated image analysis, using the novel and innovative software, spots of interest are selected for gel excision, subjected to in-gel enzymatic digestion, and mass spectrometry identification.
In this current project, we enrolled eight paired of HCC and the corresponding non-tumor liver tissues, and subjected to DIGE analysis. We found that 9 proteins (eg. heat shock protein) were upregulated in the HCC tissues, while 11 proteins were downregulated in the HCC tissues. We are currently to investigate whether these proteins could be used as the new HCC diagnostic markers.
U V W :
Introduction
Hepatocellular carcinoma (HCC) is one of the most common malignant tumors worldwide, mainly in South Africa and Southeast Asia. Major risk factors are chronic hepatitis resulting from infection with hepatitis B virus (HBV) or hepatitis C virus (HCV) and exposure to various carcinogens including aflatoxin B1 (1). However, the molecular mechanisms underlying HCC in most patients remain unclear.
The global analysis of cellular proteins has recently been termed proteomics and is a key area of research that is developing in the postgenomic era. With respect to cancer, proteomics has the potential to identify novel targets for therapy or markers for diagnosis. Methods of proteomics at present involve combination of serial sophisticated techniques including two-dimensional gel electrophoresis (2-DE), image analysis, mass spectrometry, amino acid sequencing, and bioinformatics to resolve comprehensively, to quantify, and to characterize proteins. However, no two gel images are directly superimposable. This limitation makes image comparison complex and difficult to determine what are the true differences between these two gels.
The introduction of fluorescent 2D differential gel electrophoresis (DIGE) by Unlu et al. (17), which is further modified by Amersham Biosciences, Inc, has now made it possible to detect and quantitate differences between experimental pairs of samples resolved on the same 2D gel. The basis of this technique is to use two fluorescent dyes (Cy3 and Cy5) to differentially label lysine residues of two protein samples for comparative analysis on a single gel. A third fluorescent dye (Cy2) has also been introduced to label the pooling of aliquots of all biological samples in the experiment.
This Cy2-labeled proteins are used as the internal standard. The ability to directly compare two samples on the same gel not only avoids the complications of gel-to-gel variation but also enables a more accurate and rapid analysis of differences and reduces the number of gels that need to be run.
Materials and Methods
Protein extraction from liver tissues
A total of eight HCC and the corresponding non-tumor liver tissues were enrolled. The liver tissues were ground in a mortar filled with liquid nitrogen,then the protein were extracted. Adjust the pH to 8-9 to fit the condition for the subsequent dye labeling. The protein concentrations of the homogenates are determined using the Coomassie protein assay reagent (Pierce).
Protein labeling
Cell lysates are labeled with N-hydroxy succinimidyl ester-derivatives of the cyanine dyes Cy2, Cy3, and Cy5 (Amersham Biosciences, Inc.) according to the manufacture’s protocol. Typically, 100 µg of lysate is minimally labeled with 400 pmol of either Cy3 or Cy5 for comparison on the same 2D gel. Labeling reactions are performed on ice in the dark for 30 min and then quenched with a 50-fold molar excess of free lysine to dye for 10 min on ice. Differentially labeled samples are mixed and reduced with 65 mM dithiothreitol for 15 min. Ampholines/pharmalytes, pH 3–10 (1% (v/v) each; Amersham Biosciences, Inc.), and bromphenol blue are added, and the final volume was adjusted to 350 µl with lysis buffer. A pool of all samples is also prepared and labeled with Cy2 to be used as a standard on all gels to aid image matching and cross-gel statistical analysis. The Cy3 and Cy5 labeling reactions (100 µg of each) from each time point are mixed and run on the same gels with an equal amount (100 µg) of Cy2-labeled standard.
Protein Separation by 2D Gel Electrophoresis
Immobilized non-linear pH gradient (IPG) strips, pH 3–10 (Amersham Biosciences, Inc.), are rehydrated with Cy-labeled samples in the dark at room temperature overnight, according to the manufacturers guidelines. Isoelectric focusing is performed using a IPGphor II apparatus (Amersham Biosciences, Inc.) for a total of 80 kV-h at 20 °C, 10 mA. Strips are equilibrated for 15 min in 50 mM Tris-HCl, pH 8.8, 6 M urea, 30% (v/v) glycerol, 1% (w/v) SDS containing 65 mM dithiothreitol and then for 15 min in the same buffer containing 240 mM iodoacetamide. Equilibrated IPG strips are transferred onto 18X20 cm 12% uniform polyacrylamide gels poured between low fluorescence glass plates. Strips are overlaid with 0.5% (w/v) low melting point agarose in running buffer containing bromphenol blue. Gels are run in Protean II gel tanks (Bio-Rad) at 30 mA per gel at 10 °C until the dye front had run off the bottom of the gels.
Gel Imaging
2D gels are scanned directly between glass plates using the highly sensitive Typhoon 9400 Variable Mode Imager (Amersham Biosciences, Inc.) according to the manufacture’s protocol. Gels are fixed in 30% (v/v) methanol, 7.5% (v/v) acetic acid overnight and washed in water, and total protein is detected by post-staining with Sypro-Ruby dye (Molecular Probes) for 3 h at room temperature. Excess dye was removed by washing twice in water, and gels are imaged at the appropriate excitation and emission wavelengths for the stain.
Images analysis
The images are analyzed with DeCyder software (Amersham Biosciences, Inc.) according to the manufacture’s protocol. The DeCyder software consists of two analysis modules: Differential In-gel Analysis and Biological Variation Analysis. DeCyder automatically detects, matches and analyzes protein spots in multiplexed fluorescent images, and is able to give routine detection of <10% differences with >95% confidence. Statistical analysis is carried out on each and every difference.
Protein Identification by delayed extraction matrix-assisted laser desorption ionization (MALDI) MS
Changes observed by 2D-DIGE analyses are aligned with Sypro-Ruby protein patterns, and spots are selected for picking according to this post-stained image. Spots of interest are excised from 2D gels using an automated spot picker (Amersham Biosciences, Inc.) following the manufacturer’s instructions.
The excised spots are subjected to in-gel tryptic digestion. The digests are mixed with saturated α-cyano-4-hydroxycinnamic acid solution in acetonitrile/H2O and spotted onto a MALDI sample plate, then MALDI MS analysis is performed on a Voyager DE-STR workstation (PerSeptive Biosystems, Framingham, MA) equipped with a 337-nm nitrogen laser. The peptide spectra, acquired in reflectron mode at an accelerating voltage of 20 kV, were the sum of 50 laser shots. The mass spectra are externally calibrated using low mass peptide standards. This procedure typically results in mass accuracies of 50–100 ppm. The peptide mass fingerprint data are compared with those in the National Center for Biotechnology Information nonredundant protein database using the MS-Fit search tool (University of California San Francisco Mass Spectrometry Facility, San Francisco, CA).
Result
HCC and their corresponding nontumor tissue were obtained from eight patients (Table 1) who received surgical resection, and samples were homogenized and then total proteins were extracted as described in Materials and Methods. Protein expression was compared between each tumor/nontumor pair utilizing DIGE technology with a mixed-sample internal standard. As shown in Table 2, protein extracts were labeled with either Cy3 or Cy5 fluorescence dyes, and then each patient’s Cy3/Cy5-labeled sample pair was mixed with a Cy2-labeled sample containing an equal mixture of all 16 samples prior to running all three samples together on each of 8 individual gels. In order to get more reliable data, we interchange the Cy3 or Cy5 for labeling each tumor/nontumor pair. After 2D gel electrophoresis, the Cy2, Cy3, and Cy5 channels were individually imaged from each of the eight gels using dye-specific excitation and emission wavelengths, and the images were analyzed using DeCyder software. Using the DIGE technique with the mixed-sample internal standard, protein spot-features were identified from over 2000 features as changing in abundance across the 8 patient samples. The Cy3 (tumor) and Cy5 (nontumor) spot maps from patient 1 (Fig. 1) are representative
of the spot maps from the other patients (data not shown). Proteins of interest were excised from a preparative gel for subsequent mass spectrometry and database interrogation.
Of the 278 features with statistical significance (paired t-test; p<0.05) in a DeCyder BVA analysis, 30 were targeted for protein identification. The proteins subsequently
identified from these features using MALDI-TOF mass spectrometry are listed in Table 3. Up-regulation protein spots include cGMP-specific 3’, 5’-cyclic nucleotide phosphodiesterase, T cell receptor α chain, Tropomyosin, Heat shock protein 71 kD, Triosephosphate isomerase, DNA polymerase gamma, Lactate dehydrogenase-C, Inorganic pyrophosphatase , Actin capping protein -subunit. Down-regulated protein spots include similar to transforming acidic-coil containing protein 2, carbonic anhydrase II, ferritin light chain, MHC class I antigen, high mobility group box 1 and fructose-1,6-biphosphatase. There are some proteins that can not be identified by MALDI-TOF will be subjected to LC MS/MS analysis.
Discussion
We used 2-D DIGE technology to simultaneously screen paired liver cancers and corresponding nontumor tissues in eight patients for protein abundance changes, followed by protein identification by mass spectrometry and database interrogation. Compared with the traditional 2-D electrophoresis, DIGE technology allows for each patient comparison to be performed on proteins resolved in the same 2-D gel separation, thereby removing error caused by gel-to-gel variation, and also allows for the quantification of the abundance change for each protein-pair. In this case, however, a direct Cy3/Cy5 comparison limits the detection of significant abundance changes to the 95th percentile confidence level based on the variation present between the samples for each tumor/nontumor pair-wise analysis. Thus the greater the difference between the two samples, the higher the abundance change must be to fall within the 95th percent confidence level. Using the mixed-sample internal standard in this experimental design allowed for the detection of significant abundance changes based on the variance of the mean change within the cohort. It allowed for the detection of 278 statistically significant protein abundance changes across multiple paired samples that were not apparent from individual comparisons due to the large variation between samples. When applied to an expanded patient cohort, this approach may identify useful biomarkers for liver cancer, as well as correlate specific changes in protein expression with different disease stages. Future studies may benefit from the use of western blots and immunohistochemistry staining to confirm interest protein expression in HCC.
In this study, we found heat shock protein 71(HSP71) was up-regulated in HCC. HSP71 is a constitutively expressed chaperone and its function was related to protein stabilization and folding. Heat shock protein 70 family members showed a tendency toward overexpression in HCV-related HCC (1). The expression of ferritin light chain decreased in our HCC tissues, and this decrease has been reported in previously proteomic study of HCC (2). Ferritin plays an important role in iron homeostasis, and excess of iron was associated with increased risk for HCC.
Using 2D-DIGE made us to identify same protein targets with previously HCC proteomic studies(3,4,5), suggesting this is a reliable technique. The identification of differentially expressed proteins is still in progress. We hope this study can help to understand HCC carcinogenesis, and find biomarkers related to HCC.
Table 1 Clinical and pathologic data of patient tumor samples
Pt# Age Sex AFP(ng/ml) HBsAg Anti-HCV Tumor Size ___Grade ____Cirrhosis____________ 1 52 M 10785 + - 12*12*6 2-3 -
3 66 M <20 - + 7*6*5 2 + 4 74 F 340 + - 7*6*8 2 + 5 67 F <20 + + 4.5*4*4 2 + 6 81 M 2486.6 + - 14*12*7.5 1-2 - 7 71 M <20 + + 15*13*12 3-4 + 8 65 M <20 - - 11*8.5*8 2 -
AFP, -fetoprotein; HBsAg, hepatitis B virus surface antigen; Anti-HCV, anti-Hepatitis C virus antibody Table 2 Experimental design for 2D-DIGE comparison of 9 HCCs and their corresponding nontumor tissues
GelNo. Cy2_____________Cy3__________Cy5__________________________________ 1 Pooled Std Tumor Nontumor
2 Pooled Std Tumor Nontumor 3 Pooled Std Tumor Nontumor 4 Pooled Std Nontumor Tumor 5 Pooled Std Nontumor Tumor 6 Pooled Std Tumor Nontumor 7 Pooled Std Nontumor Tumor 8 Pooled Std Nontumor Tumor Std, standard pI 3 10 MW(kD) -174 -124 -79 -48 -38 -25 -18 -13
Figure 1 2-D DIGE analysis using the mixed-sample internal standard. Cy3 (green) and Cy5(red) pseudocolor DIGE spot map image from patient #1 was shown for comparison. pI range is from 3 to 10 (left to right), and molecular mass separation 174 kDa to 13 kDa (top to bottom).
Table 3 (A) Up-regulated and (B) Down-regulated proteins in human hepatocellular carcinoma
(A) Up-regulated proteins Spot
no. Name MW/pI Function
3 cGMP-specific 3’,
5’-cyclic
15.92 kD/6.14 Catalysis of the reaction: guanosine 3',5'-cyclic nucleotide phosphodiesterase phosphate + H2O = guanosine 5'-phospha
4 T cell receptor α chain 15.78 kD/6.4 Immune response
9, 10 Tropomyosin 32.99 kD/4.63 Cytoskeletal protein binding
11-14 Heat shock protein 71 kD 70.90 kD/5.38 Molecular chaperones
21 Triosephosphate isomerase 26.64 kD/6.45 Catalysis of the reaction: D-glyceraldehyde 3- phosphate = glycerone phosphate
23 DNA polymerase gamma 22.22 kD/6.21 Replication of mitochondrial DNA
25 Lactate dehydrogenase-C 36.31 KD/7.08 Catalysis of the oxidation of lactate to produce pyruvate
27 Inorganic pyrophosphatase 31.84 kD/5.42 Catalysis of the hydrolysis of a pyrophosphate bond between two phosphoric groups
28 Actin capping protein α-subunit 32.75 kD/5.58 Actin filaments assembly
(B) Down-regulated proteins Spot
no. Name MW/pI Function
5 Similar to transforming, acidic-coil containing
protein 2 22.52 kD/6.12
7 Carbonic anhydrase II 29.25 kD/6.87 Catalysis of the reaction: H2CO3 = CO2 + H2O
8 Hypothetical protein 26.28 kD/5.7
15 Ferritin light chain 20.02 kD/5.51 Iron-storing protein complex
16 MHC class I antigen 20.91 kD/5.53 Lymphocyte recognition and antigen
17 Unnamed protein 15.42 kD/5.73
18 Unnamed protein 20.21 kD/4.09
20 High mobility group box 1 24.99 kD/5.76
24 Hypothetical protein 32.34 kD/8.17
26 Fructose-1,6-biphosphatase 36.81 kD/6.54 Catalysis of the reaction: D-fructose 1,6- bisphosphate + H2O
30 Unnamed protein 34.73 kD/7.19
(X) Y Z [
1. Takashima M, Kuramitsu Y, Yokoyama Y, Iizuka N, Toda T, Sakaida I, Okita K, Oka M, Nakamura K. Proteomic profiling of heat shock protein 70 family members as biomarkers for hepatitis C virus-related hepatocellular carcinoma. Proteomics. 2003 Dec;3(12):2487-93.
2. Park KS, Kim H, Kim NG, Cho SY, Choi KH, Seong JK, Paik YK. Proteomic analysis and molecular characterization of tissue ferritin light chain in hepatocellular carcinoma. Hepatology. 2002 Jun;35(6):1459-66.
3. Yokoyama Y, Kuramitsu Y, Takashima M, Iizuka N, Toda T, Terai S, Sakaida I, Oka M, Nakamura K, Okita K. Proteomic profiling of proteins decreased in hepatocellular carcinoma from patients infected with hepatitis C virus. Proteomics. 2004 Jul;4(7):2111-6.
4. Park KS, Cho SY, Kim H, Paik YK. Proteomic alterations of the variants of human aldehyde dehydrogenase isozymes correlate with hepatocellular carcinoma. Int J Cancer. 2002 Jan 10;97(2):261-5.
5. Lim SO, Park SJ, Kim W, Park SG, Kim HJ, Kim YI, Sohn TS, Noh JH, Jung G. Proteome analysis of hepatocellular carcinoma. Biochem Biophys Res Commun. 2002 Mar 8;291(4):1031-7.
(\) / ]
In this study, we have shown using of 2D-DIGE technology to search candidate proteins that are differentially expressed in HCC. Due to the advantages of pre-labeling technology and reduced inter-gel variance, it is suitable to apply this technology in comparing two proteomes from different physiological conditions. Because the identification protein step is important to decode DIGE data, we strongly recommend combination MALDI-TOF and other mass spectrometry (ex, LC MS/MS) to help this process.
