Identification Of Urinary Gc-globulin As A Novel Biomarker For Bladder Cancer By Two-dimensional Fluorescent Differential Gel Electrophoresis (2D-DIGE)

Bladder cancer is one of the most common malignant tumors of the urinary system threating to human health, and with its morbidity and mortality rising. All over the world, bladder cancer is the fourth most common cancer in men and the seventh most common among women, and there are more than350,000bladder cancer patients diagnosised each year. As is estimated by American Cancer Society statistics, In2010, around70,530new cases and14,680deaths from bladder cancer occurred in the United States. Bladder cancer is still the most common malignant tumors of the urinary system in China, and presents the trend of steadily morbidity. In general, it is clinically characterized by high recurrent rates and poor prognosis once tumors progress to muscularis propria invasive disease. At diagnosis,～90%of bladder cancers are urothelial cancer,55%to60%of them are low grade urothelial carcinoma. Some patients with the low grade urothelial carcinoma will recur even after appropriate treatment of the cavity or open. Typically, the patients recurred have a better differentiation of low grade urothelial carcinoma, but16%to25%of the patients will have the tumor grade increased. Furthermore, around10%of low grade urothelial carcinoma will invade deeper layers or distant transfer. On the whole, the development of bladder low gradel urothelial carcinoma progressing is a multifactor, multi-gene complex process and multi-channel change.Currently, cystoscopy and bladder biopsy are considered as the most reliable methods for diagnosis and surveillance of bladder cancer. Repeated cystoscopy enable patients to receive effective treatment in the early stages of disease, potentially slowing disease progress myometrial invasion. However, cystoscopy is an invasive, time-consuming and expensive examination, which can not be accepted well for patients. Urine cytology, a non-invasive means, is still assisted with cystoscopy, especially in the highly invasive bladder cancer with a high sensitivity. But its sensitivity is lower in the low-grade lesions, and its accuracy is dependent on the experience of the pathologists. Improving the predictive ability would greatly benefit treatment of patients and monitoring of their condition. Thus, the development of a reliable non-invasive biomarker would be highly valuable for increasing the early detection rate of bladder cancer and predicting the progression of superficial tumors in time. To date, during the clinical routine process, we still do not have a good sensitivity and specificity of serum or urine tumor diagnostic markers for the detection and tracking bladder cancer patients.The basic research based on the genetic level has provided a strong basis for the occurrence of bladder cancer diseases and the development of the interpretation of the molecular mechanisms, but these studies emphasized the role of a single or a few molecules, lacking of integrity of the protein level and systemthe nature of the study. Ultimately, the protein is a direct execution of the life activities, and the expression of certain genes in the product may be more than one. Moreover, the expression for gene is complicated. The same gene with different surrounding environment of the body and different physiological states of the body, will produce different the protein and play an entirely different role. There is a great contrast on gene structure with stability compared to the complexity and vari of the protein function. In addition, for the functions of proteins, it is also related to the post-processing of the protein, modification and transfer the process of positioning, which these processes can not be decided by the gene, once, any one of these process steps with any little error can result in the body’s disease. At last, discovery of miRNA allows us to realize differentially expressed miRNA post-transcriptional regulation of gene expression, is also possible in the corresponding protein levels. Therefore, studying bladder cancer disease on the protein level is an important issue that should not be overlooked.The Proteome that is a dynamic whole changing with time and space is aimed at analyzing the composition, expression levels and modification state, interaction and association of the proteins in cells from a holistic a holistic perspective, and revealing the functions of proteins as well as laws of cellular activity. The models of humoral proteomics research provide new opportunities to find new, highly sensitive diagnostic tools for the early detection of cancer. The main purpose of the clinical proteomics research is looking for a marker to identify disease in body fluids, which can be relatively inexpensive detection and early diagnosis of disease. Currently, serum or plasma proteomics have attracted widespread attention. While, the urine can directly contact with the tumor samples, so it is readily available in the early detection of disease and disease surveillance. Therefore, urine is an attractive option to bladder cancer finding potential markers compared to plasma. Although there are some known protein markers of bladder cancer identified in the blood and urine such as bladder cancer antigen, nuclear matrix proteins and fibrinogen metabolites; they were unsatisfactory specificity and sensitivity. To sum up, studies using proteomics technology for urine analysis diagnosis and monitori prognosis bladder cancer is a current research trendTo date, many studies use two-dimensional gel electrophoresis and mass spectrometry techniques to identify differentially expressed proteins in the urine of bladder cancer, using Maas staining, silver staining or radiolabeled staining marked urinary protein extracts. In2009, Feldman et al conducted a proteomics strategy analysis of the urine of bladder cancer patients and normal control subjects to find differently expressed proteins, and the protein of interest Cystatin B was further tested by immunohistochemical techniques in the organization and semi-quantitative Western blot techniquein urine usin. They found the expression level of urine Cystatin B is positively associated with tumor grade (P=0.062), stage (P=0.0047) and tumor recurrence (P=0.0104) and the progress of time (P=0.0007). In2011, Li et al performed a study by using two-dimensional electrophoresis screening healthy volunteers, low-grade and invasive bladder cancer urine samples in order to identify capable of early detection of bladder cancer biomarkers. They found that the fibrinogen, lactate dehydrogenase B, apolipoprotein Al, profusion protein and haptoglobin5protein expression increased in the urine of patients with bladder cancer of low malignant or aggressive. After further analysis of the urine samples, apolipoprotein A1expression was significantly higher in invasive bladder cancer compared to the low grade bladder cancer. Apo-A1level was measured quantitatively using ELISA and was suggested to provide diagnostic utility to distinguish patients with bladder cancer from controls at18.22ng/ml with a sensitivity and specificity of91.6%and85.7%respectively, and at29.86ng/ml for distinguish patients with low malignant bladder cancer from patients with aggressive bladder cancer with a sensitivity and specificity of83.7%and89.7%respectively.However, with the improvement of proteomics technology, fluorescent differential two-dimensional gel electrophoresis-based proteomics research is a new trend. Two-dimensional fluorescent differential gel electrophoresis (2D-DIGE) is an advanced sensitive gel-based separation and quantification approach. Protein samples are prelabeled with different fluorescent dyes, mixed and run simultaneously on the same gel. Different excitation wavelengths can be recorded by different filters the Noninterference gum diagram results due to the different fluorescent labeled sample.This method could effectively avoid systematic errors between different gels, especially suitable for the differences between the comparison of different samples. In1997, Unlu etc. firstly published papers about the fluorescence difference gel electrophoresis, which the different samples with different fluorescent dyes labeled and mixed in the same gel and then were conducted a two-dimensional electrophoresis, greatly improving the experimental results of repeated and quantitative accuracy. As far as we know, the study for bladder cancer by labeling urinary protein extracts and using the fluorescent two-dimensional gel electrophoresis is rare. Only in2007, Orenes-Pinero used2D-DIGE proteomics strategy to explore urine markers for diagnosis of bladder cancer. In our present study, from the view of the formal level of urine proteomics, based on the establishment of norms and standards for the collection of clinical specimens workflow, developing the scientific and rational inclusion and exclusion criteria, then, we collected the enough the urine samples form patients with bladder cancer and healthy control subjects. The diagnosis was conducted by two different pathologists in Nanfang hospital of Southern Medical University according to the criteria of World Health Organization classification of tumour. The bladder cancer patients were divided into the low-grade urothelial carcinoma group, the high-level urothelial carcinoma group and the invasive urothelial carcinoma group. We used the method of repeated ultrafiltration and the salt removing to purify the total urine protein. Two-dimensional fluorescent differential gel electrophoresis was used for screening and separation. After electrophoresis, the gel was scaned by Typhoon9400imaging system and analyzed using DeCyder software to find the differently expressed proteins. The proteins were identified using matrix-assisted laser desorption ionization time-of-flight technology. Then, we applied bioinformatics methods to select the interested candidate proteins (GC). After the verification of the result of the candidate protein reliability, a large number of clinical bladder cancer urine of patients and the control group urine samples was used to measure GC protein concentration by ELISA. Statistical analysis was performed by ELISA results combined with the clinical data of clinical specimens to establish the relationship between the candidate protein molecular markers and bladder cancer disease so as to further determine the effective diagnosis and monitor of bladder cancer disease, study the pathogenesis of bladder cancer. These will be a basis for the study to diagnosis, disease monitor and treatment of bladder cancer.However, up to date, studies using two-dimensional fluorescent differential gel electrophoresis for urine analysis related to bladder cancer are scarce. Therefore, to our present research, Identification of urinary Gc-globulin as a novel biomarker for bladder cancer by two-dimensional fluorescent differential gel electrophoresis (2D-DIGE) is to acquire the ideal biomarker for the early detection, development and prognosis of bladder cancer on the level of urine samples. Undoubtedly, this project has important theoretical and practical significance on the in-depth understanding of the occurrence, development and prognosis of bladder cancer.Material and methods1. Specimen collection and preparationStudies were done with the approval of the bioethics committee of Nanfang hospital. All subjects were informed about the purpose of the study and gave their written consent. All patients had their upper tracts cleared via examinations, healthy volunteers with no evidence of disease were used as control group. All subjects were recruited from the Chinese Han population at Nanfang hospital from January2011to April2012. Pre-cystoscopy voided urine specimens were collected from patients presenting positive findings under suspicion of bladder cancer. Bladder cancer tissue and normal urothelial tissue were harvested form the cystectomy specimens of individual patients. The bladder cancer was confirmed by cystoscopy combined with histopathological information after subsequent surgical operations. The diagnosis was conducted by two pathologists in Nanfang hospital of Southern Medical University according to the criteria of World Health Organization classification of tumours. Each urine sample (20ml) was collected into a sterile plastic tube and then immediately centrifuged at1500x g for5min at4℃to remove cell debris and particulate matter. The supernatant was stored at-80℃for further analysis.2. The screening of the candidate protein marker(1) Equal volume urine specimens from12bladder cancer patients and12controls were pooled respectively for2D-DIGE analysis. Of these bladder cancer patients, there were8cases of non-invasive papillary urothelial carcinoma (4were low grade and4were high) and4cases of infiltrating urothelial carcinoma. For processing, samples were first thawed on ice, adding protease inhibitors,1mmol/L phenylmethylsulfonyl fluoride,5mmol/L phenanthroline, and5mmol/L benzamidine, and then centrifuged using Centricon Plus-20,10,000MWCO devices. Then proteins in the concentrated urine were precipitated using a ReadyPrep2-D clean up kit to remove other interfering components according to the manufacturer’s instructions. Protein concentration was measured by using the Bradford method. (2) To establish the good stability, high repeatability urine proteomic profiles by two-dimensional gel electrophoresis(2.1) The Immobiline Dry strip (pH3-10, length24cm) was rehydrated with100pg protein in450ml was rehydrated buffer for13h at room temperature. The first dimensional electrophoresis was performed on Protean IEF cell with a total of60kVh, The soelectric focusing conditions were500V1h,1,000V1h,5,000V3h,8,000V60KVh,500V3h.Then the strip was subjected to two-step equilibration for each step15min. The second dimensional electrophoresis was carried out in a homogeneous SDS-PAGE (12%) using a Protean II xi2D cell until the bromophenol blue front reached the bottom of the gel.(2.2)40.0%ethanol,10.0%acetic acid fixed2times, each time15min.30.0%ethanol,0.2%Na2S2O3,6.8%sodium acetate-sensitized for30min. Following distilled water washed three times, each time5min.2.50‰AgNO3staining20min. Following distilled water washed two times, each time1min.2.5%Na2CO3,0.04‰formaldehyde to develop an image until protein spots clear. Then we immediately added the5.0%acetic acid to terminate for10min, following distilled water washed three times, each time5min. At last, we used30.0%ethanol and4.6%glycerol to preserve.(2.3) the gel was scanned through the UMAX PowerLook1100scanner to obtain2-DE gel image, then using Melanie4image analysis software to analyse the image and produce differentially expressed protein spots data.(3) To screen and separate the differentially expressed proteins between patients with bladder cancer and normal controls by two-dimensional fluorescent differential gel electrophoresis(3.1) Protein isolated from the pooled urine samples were labeled with cyanine dyes according to the manufacturer’s instructions.50μg of urine protein samples from bladder cancer and control group were minimally labeled with400pmol of Cy3and Cy5fluorescent dyes, respectively. To assess the reproducibility and statistical inferences, an internal standard pool was labeled with Cy2. The internal standard pool was generated by combining equal amounts of extracts from urinary samples of all neoplastic and control group subjects. The labeling reactions were carried out on ice for30min protecting from light, and then quenched with1ml of10mM lysine for10min. All three labeled samples were mixed and resolved in one gel.(3.2) The method of the resting isoelectric focusing was similar to2-DE(3.3) The three gels were visualized with a Typhoon9410scanner at the excitation emission of488/520nm (Cy2),532/580nm (Cy3) and633/670nm (Cy5), respectively. The images were analyzed with the DeCyder5.0software. Its differential in-gel analysis (DIA) module was used for pairwise comparisons of each sample with the internal standard within each gel by calculating the normalized spots volumes. The DeCyder biological variation analysis (BVA) module was used to calculate average abundance changes for each spot across the different spot maps. The spots whose ratios of Cy5/Cy2and Cy3/Cy2were up-or down-regulated equal or greater than2-fold were considered for further analysis(3.4) In addition, another strip was performed in parallel as a preparative gel for spots picking as marked in2D-DIGE. Noticeably,1000mg of proteins were loaded onto the IPG strip and the gel was stained with Coomassie brilliant blue. The differentially expressed protein spots were cut from the Coomassie stained gels, and then identified by MALDI-FOF/TOF-MS. Briefly, Protein spots from gels were digested with trypsin solution, and the peptide mixtures was analyzed. A list of the corrected mass peaks was the peptide mass fingerprinting (PMF). The search was restricted to the Homo sapiens subsets of the sequences in the Swiss-Prot and NCBI nonredundant protein sequence databases.3. GC was selected for further investigation after bioinformatics analysisWe conducted the bioinformatics research to study to explore the biological processes participated of thosed identified proteins, protein-protein interactions and protein-specific structures and functions, in order to select the candidate proteins for further study by those tools, such as:SWISS-PROT, String and Pubgene.4. The expression of candidate proteins GC in the urine samples form healthy controls and patients with bladder cancer(1) For western blot analyses, the samples involved two low grade and two high grade non-invasive papillary urothelial carcinoma, four infiltrating urothelial carcinoma and eight controls.30μg prepared proteins from urine and tissue were electrophoresed respectively on a12%SDS polyacrylamide gel and then transferred onto Polyvinylidene Xuoride (PVDF) membranes. The membranes were blocked in a solution of TBS containing5%nonfat milk powder and0.1%Tween-20for1h at room temperature and then incubated overnight at4℃with the primary antibody and rabbit monoclonal primary antibody against human β-actin. After washing with TBS-T for three times, the membranes were incubated with the secondary antibody dilution at room temperature for1h. The proteins were detected using an enhanced chemiluminescence detection system.(2) ELISA was used to quantify urinary GC levels in91patients with bladder cancer,11benign bladder damage cases and42healthy controls. According to the criteria of World Health Organization classification of tumours,68cases were non-invasive papillary urothelial carcinoma (38were low grade and30were high) and23were infiltrating urothelial carcinoma.5. The analysis of GC level with clinical dataTo explore whether there was any correlation between the GC expression and features of bladder cancer such as age, gender, hematuria, recurrence and metastasis and classification of pathology. Receiving operating curve (ROC) analyses were used to define the optimal diagnostic cut-off value by estimating the sensitivity versus its false-positive rate at optimal cutoffs6. Expression of GC in normal and tumour bladder tissue(1) For western blot analyses, Tissue specimens were immediately frozen in liquid nitrogen, then ground to powder and homogenized in lysis buffer. The mixture was placed on a shaker at4℃for1h, and then followed by centrifugation. The supernatant was used for further analysis after measuring the protein concentrations. Then, the protein was used for western blotting.(2) For Immunohistochemical analysis, There were eight low grade and eight high grade non-invasive papillary urothelial carcinoma, ten infiltrating urothelial carcinoma and ten normal controls in immunohistochemistry. The samples of normal controls were the normal-adjacent to-cancer tissues. Paraffin sections was dropped the wax then put into the water. The sections were incubated in3%hydrogen peroxide to quench the endogenous peroxidase activity and washed in PBS for three times. Plus the primary antibody conjugated with biotin (Biotin) at37℃for30minutes, then put in the wet box. Following PBS washed three times, then plus enzyme labeled avidin (Avidin) at37℃for30. PBS washed three times, following DAB chromogenic and washed in the running water. The sections was stained, dehydration and mounted at last.Results1. With the improvement of optimization steps of repeated ultrafiltration and desalination to sample, we could obtain high-resolution, reproducible two-dimensional electrophoresis maps from human urine samples.2. A representative DIGE image from bladder cancer versus control group is acquired., a total of24differential protein spots whose volumes changed by or over2-fold were selected for further identification. Sixteen differentially expressed proteins were identified. The up-regulated proteins in bladder cancer were ALB, GC, HP, FGB, APOA1, RBP4and SECTM1, and the other nine proteins were down-regulated. They were UMOD,KNG1, AMY1A, AMY2A, ITIH4, AMBP, HSPG2, CST5and MASP2.3. The analysis of Pubgene and String indicated that GC was predicted to play important roles in the bioprocesses of growth, secretion, induction, pathogenesis, signal transduction, digestion, translation, apoptosis and death. Furthermore, GC may tightly correlate with the other identified proteins.4. Western blotting results demonstrated that GC was significantly upregulated in bladder cancer cases in comparison to controls, consistent with our DIGE results.5. The expression level of GC was significantly higher in bladder cancer tissue in comparison to normal control bladder tissue. GC protein was detected, apparently in the cytoplasmic compartments of normal bladder transitional cells and cancerour cells by immunohisochemical staning of bladder sections.6. GC-Cr was significantly elevated in patients with bladder cancer compared to controls and benign cases (1013.70±851.25versus99.34±55.87,105.32±47.81ng/mg, respectively, P<0.05). The urinary GC concentration was significantly higher in infiltrating urothelial carcinoma cases than in cases with low grade and high grade non-invasive papillary urothelial carcinoma (1906.69±840.86versus472.92±348.02and1014.06±753.16ng/mg, P<0.05).7. The optimum cut-off value was161.086ng/mg with92.31%sensitivity and83.02%specificity for diagnosis of bladder cancer. ROC analyses rendered a cut-off value with1407.481ng/mg corresponding to82.61%sensitivity and88.24%specificity d to distinguish infiltrating urothelial carcinoma form bladder cancer patients.Conclusion1. Establishing stability and good reproducibility human urine2-DE techniques and methods in our study, which laid a foundation for further bladder cancer research on urine proteme.2. Comparing the urine proteome map of bladder cancer compared with normal controls by2D-DIGE.3. Sixteen differentially expressed proteins were identified. The up-regulated proteins in bladder cancer were ALB, GC, HP, FGB, APOA1, RBP4and SECTM1, and the other nine proteins were down-regulated. They were UMOD, KNG1, AMY1A, AMY2A, ITIH4, AMBP, HSPG2, CST5and MASP2.4. GC was selected for further investigation after bioinformatics analysis.5. The expression level of GC was significantly higher in bladder cancer tissue in comparison to normal control bladder tissue. GC protein was detected, apparently in the cytoplasmic compartments of cells.6. GC is significantly elevated in bladder cancer, and is positively associated with the pathological classification of bladder cancer7. Urinary GC may be a potential biomarker for the early diagnosis and effective surveillance of bladder cancer.