Study Of Early Detection Of Obarian Carcinoma By Magneatic Beads Integrated With MALDI-TOF-MS And The Biological Effects Of GRP SiRNA On Ovarian Carcinoma Cell Line ES2

1. BackgroundObviously, OC is the most lethal gynecological cancer. It often eludes the clinician, due to its insidious localization in the pelvis and lacking of symptoms and signs during early stages (stageⅠ/Ⅱ). This is the reason for that nearly 70% patients are not diagnosed with OC until the advanced stages (stageⅢ/Ⅳ). The five-year overall survival of advanced stage is no more than 20%, compared to the 90% of early stage cases. Therefore, increasing the number of women diagnosed in early stage should have a direct effect on the mortality and economics of this cancer.Lacking of highly specific biomarker remains the main bottleneck problem in the study of early detection of OC. Proteomic can afford a full view of disease and has been used abroadly in research of many diseases. Although 2-dimensional gel electrophoresis and multi-dimensional liquid chromatography after mass spectrometry (MS) analysis seemed a somewhat economical, these techniques were also laborious and time-consuming. Recently, the discovery of biomarkers in body fluids has been advanced by the introduction of MS based screening methods such as Matrix-assisted Laser Desorption/ionization Time of Flight Mass Spectrometry (MALDI-TOF-MS) and a protein chip coupled with Surface enhanced Laser Desorption/Inionation Time of Flight Mass Spectrometry (SELDI-TOF-MS). The first clinical investigations using SELDI-TOF-MS in different cancer types revealed high diagnostic sensitivities and specificities.SELDI-TOF-MS has been developed to facilitate protein profiling of complex biological mixtures, with high efficacy discovery of cancer biomarkers in serum or plasma. However, different patterns for the same type of cancers have been identified by different individual groups using the same types of biological specimens and analytical platforms of SELDI-TOF-MS. These discrepancies could be attributed to poorly define analytical protocols, which were not established reproducibility. Based on magnetic beads integrated with MALDI-TOF-MS analyzing, serum peptides were fractionated and concentrated on surface-modified targets with specific protein-capture properties by some scholars and they attributed the high reproducibility to the hydrophobic surface of magneatic beads.Following with the development of the technique of proteomic, many kinds of body fluid such as bile, saliva, ascites have been used for the screening of tumor biomarkers. Many studies focus on secreening the biomarker in serum of patients with OC by proteomic. However, the proteins and peptides in the serum came from all over the body, which made the specificity and sensitivity of these serum biomarkers insufficient to capture most of the patients in early stage from the large scale of women without any symptoms. Furthermore, taking patients with OC in different stages as a whole group would undoubtly leaving out some biomarker of early stage.Evidently, combining the serum, ascite and cyst fluid of OC and devide the OC group into early stage and advanced stage groups will enhance the specificity of the biomarker of OC. In the present study, MALDI-TOF-MS integrating with the megneatic beads was used to screening the biomarker of early detection of OC. Immunohistochemistry and enzyme linked immuno adsorbing were used to validated the suspected peptide. Small interfering RNA technique was used to understand the biological effects of suspected peptides on ovarian carcinoma cell line.2. Materials and methods Proteomic: training set:A total of 30 consecutive patients with OC (11 in early stages and 19 in advanced stage; 23 of serous and 6 of mucinous and one of endometrioid cancinoma; nine with age<50years old and 21 with>50 years old; 10 with positive familial history and 20 with negative familial history.),10 with benign ovarian tumors (10 of serous and 2 of mucinous) and 13 age-matching patients without cancer and tumor collected from the Department of Gynecology, the First and Second Affiliated Hospitals of Zhengzhou University, from July 2006 to January 2009, were recruited in this study. Testing set:8 cases of OC (6 serous and 2 mucinous),5 cases of benign ovarian tumor (4 serous and 1 mucinous) and 5 cases of normal ovarian. The ascites and cyst fluid samples were also collected by the same method during the operation.IHC:The matching tissues of these 53 cases of training set, testing set and another 10 cases of benign tissues were selected for IHC.ELISA:All of the cases of training set and testing set were identical to the cases of proteomic except for additional 10 cases of benign patients without matching tissues were also used for training set.Small interfere RNA:ES2 ovarian cancer cell line was used to peform the study of gene silencing of the target gene.All patients were collected the venous blood for 5ml in the morning the day before the operation and stored at -80℃after centrifuging. Ascites and cyst fluid samples further obtained from five patients were dealed with the same method. No treatment was implemented on all patients before surgery. All of the participants were identified as normal renal function. All of the tissues matching with the serum were collected at the beginning of the operation and were disposed by paraffin after being fixed with 10% formalin.2μl Suspention of Weak cation beads (WCX) SPE-CM,95μl of SPE-CB and 100μL of SPE-CW was added into 5μl of serum sample. The prepared samples were obtained after several binging and elution. The ascites and cyst fluid samples were 10μL and 2.5μL respectively for the same preparing method.1μl of eluent with peptides was mixed with 10μl of matrix (0.3%HCCA) and then 1μl of mixture was applicated onto the Anchorchip. Put the buck sracper into Microflex mass spectrometer (Bruker Daltonics) after calibrated by the standard sample (Clinprot standard). Massspectrograms of peptides peaks with different m/z were obtained after detecting the samples. Then the common elevated peaks in the serum, ascites and cyst fluid were selected for the candidated peptide peaks. At last, the matching peptides were searching in the http://us.expasy.org/tools/tagident. html protein data bank. After retrievaling from prvious data, the deducted cnadidated peptides were selected for next validation.Immunostaining was performed with the antibodies of the GRP (rabbit polyclonal, MAB35841 R&D Co, Ameraican) and IgG fluorescence Secondary Antibodies (surpported by Yueyan Biotech Co. Shanghai). According to the manufacture's protocol, a standard streptavidin-biotin-peroxidase method was used.Serum level of GRP was detected by gastrin-releasing peptide ELISA kit (QC01-1, R & D Co, American). The reactions were done by following the manufacture's protocol.A GRPsiRNA targetting the human GRPmRNA common sequence was constructed and inserted onto the Bam Hi-HindⅡlinearized p-genesil-1 vector. After transfected the plasmid into the ES2 cell, level of GRPmRNA and protein were detected by RT-PCR and Western bloting after the transfection. The proliferation, apoptosis and capability of migration and invasiveness were evaluated by cell dounting, flow cytometry and transwell.The build-in ClinProtTM software (Bruker Daltonics Company) and FlexAnalysis 3.0 were used to compile the peaks across the spectra and analyze the intensity of single peak. All signals with a signal-to-noise (S/N) ratio>5 in a mass range of 1000-10000 Da were recorded with the ClinProtTM software (Ver.2.2; Bruker Daltonics). A linear SVM classifier was used to distinguish between the different groups of data. The capability of each peak in dstinguishing different groups of data was estimated by the P value of Kruskal-Wallis test. The combinations of the selected peaks was analyzed by the leave-one-out cross-validation. The results of IHC and ELISA were analyzed byχ2 test, q-test and Kruskal-Wallis. Spearman correlations and Pearson correlation coefficient analysis was used to get the relevance of the IHC and ELISA and the GRP and pro-GRP level. Two sample t-test was used to analysize the different level of proliferation, apoptosis, capability of migration, mRNA and protein of GRP before and after transfection in the cell. These data analysis were completed by analysis software SPSS 13.0.3.ResultsMALDI-TOF-MS was used to profile the mass spectral patterns of serum samples from three groups and found that there were no different peaks between benign and normal groups. The common elevated peaks in early and advanced stage of OC were 2881 and 2897Da compared with the normal group (P<0.05). The common elevated peaks in early and advanced stage of OC were 2881,2897 and 4466Da (P<0.05). It was further found that the 2881 and 2897Da peaks were also elevated in the ascite and cyst fluid samples. The pattern of two combined peaks had a specificity of 100% and a sensitivity of 100%, as evaluated by leave-one-out crossvalidation. The remaining 18 serum samples were analyzed as a blind test set. The specificity and sensitivity of the blind test were 100% and 87.5%, respectively.Compared with normal ovarian group, there were no significant defferential peaks in below and above 50 years old groups (P all>0.05). Compared with normal ovarian group, only 2953 and1466Da peaks were found downregulated in the positive and negative familial history groups (P<0.05) while other differential peaks were thoroughly different.The candidate peptides matching with 2881 and 2897Da were searching in the http://us.expasy.org/tools/tagident. html peptide data bank. According to the statistic and the reported data, only Gastrin-releasing peptide (GRP) were found closely related to many kinds of cancers among the nine kinds of candidate peptide of OC searching by us.The positive expression of GRP in tissues of OC were higher than the benign tumor and normal ovary. The expressions of GRP in the normal group were all low to negative expression (χ2= 24.599, P= 0.00). The higher serum level of GRP in OC than benign and normal groups was verified by ELISA (P<0.05, by q-test). There was a high correlation between level of Gastrin-releasing peptide in tissue and serum (r=0.809, P=0.000). The correlations of the serum level of 2881,2897Da peaks and serum level of GRP in three groups were all high. There was an obviously coefficient between GRP and pro-GRP in the training set (r=0.90, P<0.001). Pro-GRP level in serum of training set significantly higher than the benign and normal groups(χ2= 41.81, P=0.00). The area below ROC curve was 0.974 when the proGRP was used for diagnostic marker (z=27.88, P<0.001). The cut-off-value was 89.2851, which had a sensitivity of 100% and a specificity of 90.91%(Youden index was 0.9091) in training set and a sensitivity of 93.33% and a specificity of 80%(Youden index was 0. 7333) in testing set.Cells stably expressing GRPsiRNA were got by G418 screening after transfection. The proliferation and capability of migration and invasiveness of ES2 cell were significantly decreased (P<0.05) after the transfection accompanied by the down-regulation of the mRNA and protein level of GRP (P<0.05). As a result, the apoptosis of ES2 cell was also activated by the transfection (P<0.05).4. conclusion4.1 The 2881 and 2897 Da peaks are highly specific and sensitive serum biomarkers for early detection of OC. The mechanism of the carcinogenesis of hereditary and nonhereditary OC might be partly different4.2 Gastrin-releasing peptide might be the candidated peptide matching with the 2881 and 2897Da peaks. Pro-GRP had a high diagnostic value in detecting OC with early stage.4.3. ES2 cell transfected by the GRPsiRNA show a decreased capability of proliferation, migration and invasiveness as well as the activation of the apoptosis. GRP might be a potential target for treatment of OC.