Ovarian Cancer Stem Cell-specific Surface Marker Candidate Protein Cdc50a Screening And Characterization Of Stem Cells

BackgroundOvarian cancer is the leading cause of gynecological cancer mortality, and the5-year survival rate of advanced stage(III or IV) patients ranged from20%to30%. Currently, primary cytoreductive surgery followed by chemotherapy based on platinum is considered to be the standard treatment for advanced ovarian carcinoma. However, the chemoresistance, recurrence and metastasis are obstacles in the treatment of ovarian cancer, especially result in the low five-year survival rate of advanced patients. So, deep research in the root causes of tumorigenesis, metastasis and chemoresistance is the key to improve the prognosis. Current studies suggest that cancer stem cells (CSCs) are the putative mediators of chemotherapy resistance and tumor progression. CSCs possess properties of self-renewal, high tumorigenesis and differentiation, besides, it can weaken the effect of anti-tumor drugs by highly expressing ATP-binding cassette (ABC) transporters family membrane proteins, which can transport and excrete metabolites, drugs, endogenous esters, polypeptides and so on. Therefore, research in ovarian cancer stem cells to improve the5-year survival rate of ovarian cancer patients with drug resistance is of great significance.Up to now, there are still no specific surface markers to screen ovarian CSCs. At present, the CSCs are screened by the common process of separating CSCs is first screening the positive cells with the special surface markers of CSCs by fluorescence activated cell sorting (FACS) or magnetic activated cell sorting (MACS), then identifing the characteristics of the cells by testing the cytoactive, transplant, clone formation, cell proliferation and differentiation. To identify and determine the specific molecular markers is one of the key points in CSCs research.In this study, we focused on looking for possible CSCs-specific surface molecular markers, and further verifying whether the positive cell subsets have CSCs potency. As is known to all, side population (SP) cell subsets are rich of stem cells, so we screen the candidate proteins which differentially expressed on the surface of ovarian SP cell subsets by high-throughput proteomics method, and then verity possible surface markers of CSCs.Methods1. SP cell subsets were sorted from epithelial ovarian cancer (EOC) cell line SKOV3and A2780by Moflo fluorescence activated cell sorter, and be enriched to maintain the proportion stabilized at more than5%for passage and freeze. We labeled protein on A2780and SKOV3cells by stable isotope labeling with amino acids in cell culture (SLIAC), then separated the SP and NSP cells membrane composition by gradient centrifugation. After extracting the membrane proteins, we detected the peptide by mass spectrometry, and quantitatively analyzed proteomics. At last, differentially expressed membrane proteins were screened as the ovarian CSCs specific surface marker candidate membrane proteins.2. Membrane proteins were extracted from the SP and NSP cells, then the expression of candidate proteins in the two cell subsets were verified by immunoblotting. Observing the intracellular localization of CDC50A with immunofluorescence confocal laser, and semi-quantitative analysing the expression level of it in SP and NSP cell subsets,we labeled the epithelial ovarian cancer cell line SKOV3and A2780with specific fluorescent antibody, and selected a candidate protein-positive and negative cell subsets with Moflo fluorescent activated cell stream sorter. Then the cell subsets were cultured and recruited to analyze candidate protein-positive and-negative cell subsets by flow cytometry.3. We labeled the epithelial ovarian cancer cell line SKOV3and A2780with specific fluorescent antibody, and selected an alternative protein-positive and negative cell subsets with Moflo fluorescent activated cell stream sorter. We verified its capacity to differentiating into other subsets after cultivated in DMEM (HG) medium in vitro. We tested the difference of paclitaxel and cisplatin resistance to candidate protein positive and negative cell subsets by MTT assay in vitro experiments; while using the same method to determine cell growth curve of the two group cell subsets; We analyzed two subsets of the cell cycle in the SKOV3cell line with BD fluorescence activated flow cytometry, and observed the proportion of each cell cycle to detect the active proliferation of cells. We collected the EOC cell line candidate protein positive and negative cell subsets, vaccinated single cell to96-well plates to observe the single-cell colony formation rate. Efficiency of different subsets of single-cell clone was compared, the same amount of candidate protein positive and negative cells were inoculated into NOD/SCID mouse subcutaneously to observe the speed of the tumor formation. We extracted cell RNA, reverse transcripted it to DNA, then amplified DNA by PT-PCR technology to validate the expression of several other stem cell-related gene in the positive cell subsets.Results1, We successfully sorted the SP cells from the A2780and SKOV3cell lines. The SP ratio was less than0.5%in the first sorting, while the proportion of SP cells in A2780cell lines is up to5%in the secondary sorting, which reached15%in the third sorting. Similarly, the proportion of SP cells in SKOV3cell lines stabilized was around6%. We Labeled intracellular lysine of SP and NSP by stable culture isotope labeling method (SILAC), and analyzed2730proteins by mass spectrometry, then screened1561differentially expressed proteins, in which920up-regulation proteins and627down-regulation proteins were found, and also no variation in the other14proteins. Most of the differentially expressed proteins were membrane proteins, but there were also some cytoplasmic and nuclear proteins, which may due to the proteins pollution during the extracted process. We comparatively analyzed2000kinds differentially expressed proteins of SKOV3and A2780cell line and filtered out2membrane proteins CD59and CDC50A (also known as TMEM30A) as candidate proteins of the CSCs specific surface markers.2,we sorted candidate proteins CDC50A and CD59-positive and-negative cell subsets with Moflo fluorescent activated cell flow sorter, cultured and enriched them in vitro. The first separation ratio of CDC50A was less than0.2%, and the second sorting enrichment ratio was about0.2%to0.4%, while the third sorting enrichment ratio was more than1%. However, the positive rate of CD59in the first sorting was nearly100%, so we gave up CD59as an ovarian CSCs marker protein for further study. The following study was the further identification of CDC50A. We used the immunoblot technique to detecting the expression of CDC50A from the membrane proteins which were extracted from the SP and NSP cell subsets and found the expression level of CDC50A in the SP cells was significantly up-regulated. We stained SP and NSP cells by immunofluorescence and observed the intracellular localization of CDC50A by immunofluorescence confocal laser; we successfully detected the expression of CDC50A in cells which located on the membrane.3,Compared to NSP cells, expression of CDC50A in SP cells was significantly upregulated. We cultured CDC50A+cells in vitro after sorting, which could again differentiate to CDC50A-cells, but the CDC50A-cell couldn't differentiate to CDC50A+cell subsets, which indicated that CDC50A+cells had the potential to differentiate to CDC50A-cells. There was a significant difference between growth curves of CDC50A+and CDC50A-cells by MTT assay, and the growth speed of CDC50A+cell was significantly faster than the CDC50A-cells. Both in the SKOV3and A2780cell lines, there were significant differences in the drug resistance of CDC50A+and CDC50A-cells to paclitaxel and cisplatin, especially the resistance to cisplatin, the drug resistance index of CDC50A+cells to cisplatin was significantly higher than the CDC50A-cells. The cell cycle suggested that most CDC50A+cells stayed at quiescent stage compared to the cells didn't be sorted. We found that the monoclonal forming efficiency of CDC50A+SKOV3cell lines was12.5%, while the CDC50A-cells was4.86%, and there was a significant difference, P=0.0142<0.05; The monoclonal forming efficiency of CDC50A+A2780cell lines was18.75%, while CDC50A-cells was5.56%, which also has a significant difference, P=0.0075<0.05. The tumor formation experiment showed that3NOD/SCID mice of the CDC50A+cells group all had tumor and the average time was about20days, while the CDC50A-cells group and SKOV3cells group have not yet found the tumor formation at the end of observation. We detected the expression of three kinds of stem cell-related genes Oct-4, beta-catenin and ABCG2, except for beta-catenin in A2780cell strains has no significant change, the expression of three genes in CDC50A+cell sub-group of two cell lines were significant up-regulated.Conclusion1. The ratio of SP cell subsets in ovarian carcinoma cell line is low, but after continuous culturing and sorting SP cells, which can be enriched. By proteomic analysis of membrane proteins extracted with grandient centrifugation, we screened the differentially expressed protein CDC50A and CD59as candidate proteins of stem cell surface specific markers, and further to verity them.2. The ratio of CDC50A+subsets of ovarian cancer cell lines cultured in vitro was less than1%, while the CD59+was close to100%, so we select CDC50A as a candidate protein in further study. As a transmembrane protein, the expression level of CDC50A+in the SP cell subsets was significantly upregulated. The proportion of CDC50A+subsets can be increased by flow sorting, culturing and enrichment in vitro.3. CDC50A+cells had the potential to differentiate to CDC50A-cells, while the CDC50A-cell doesn't. We found that CDC50+cells had a faster growth rate and more resistance to chemotherapy drugs by testing the growth curve and the drug resistance with MTT assay. The cell cycle detection suggested that most CDC50A+cells keep quiescent state compared to the unsorted cells. Above all, we proved that CDC50A was a significantly up-regulated membrane protein in the SP cell subsets, and the CDC50A+cell subsets had a stronger growth capacity, drug resistance and differentiation potential than the normal cells. Human ovarian cancer cell CDC50A+cell sub-group had CSCs properties, including self-renewal, multi-differentiation and unlimited proliferation and so on; Compared to CDC50A-cells, the expression of Oct-4, beta-catenin, ABCG2in CDC50A+cells were significant up-regulated. And CDC50A+cells had a highly tumorigenic tendency, suggesting that CDC50A+cells may be the ovarian CSCs.