Identification And Study Of Cell Characteristics Associated CDC50A Positive Cells Dryness In Ovarian Cancer Cell Lines And Primary Cells

BackgroundEpithelial ovarian cancer (EOC) is the leading causes of gynecological cancer mortality, among which75%patients were diagnosed as advanced ovarian cancer. Under the treatment of primary cytoreductive surgery followed by chemotherapy based on platinum, about60～80%patients relapsed. The median survival time of advanced ovarian cancer patients was only16～22months, and the5-year survival rate is only30%. The chemotherapy resistance of the tumor cells is considered to be one of the main reasons for this phenomenon. A recent study of targeted drugs and small molecular compounds in clinical trials failed to improve the prognosis of patients with ovarian cancer. With the emergence of cancer genome map, many scholars are mainly engaged in the effects of the genome and epigenome changes on the clinical outcomes of ovarian cancer. But the phenotype mechanism of ovarian cancer chemotherapy resistance and rumor recurrence is still not clear, so it become the biggest obstacle for ovarian cancer therapy.Stem cells are known as a kind of special cells which has the ability of self-renewal and multi-directional differentiation. According to its origin and differentiation ability, they are mainly divides into embryonic stem cells, mesenchymal stem cells and adult stem cells. With the development of tumor biology, scholars put forward the theory of cancer stem cells (CSCs) in the last century. Lots of studies have found that there were a very small number of CSCs existed in tumors at quiescent state, which not only have the similar characteristics of stem cells, such as self-renewal and differentiation, but also maintained the tumorigenicity and cell heterogeneity of tumor cells. So it can escape from the toxic damage of chemotherapy drugs, and result in the drug resistance, progression and recurrence of tumor. Early in1997, Bonnet and his companies isolated the phenotype of CD34+CD38" leukemia stem cells, and directly confirmed the existence of cancer stem cells for the first time at the cellular level. After that, the theory of CSCs was identified in many solid tumors such as breast cancer, prostate cancer, colon cancer, lung cancer, pancreatic cancer, and established a series of separation, enrichment, identification and molecular mechanism research methods of CSCs. At present, the separation and enrichment of CSCs method mainly includes:the sorting of side population cells (side population, SP) by flow cytometry, sorting according to the specific CSCs surface marker by flow cytometry, and the cultivation of the tumor sphere.Recently, many scholars begin to research in the cancer stem cells to looking for the new treatment of ovarian cancer. They will study on the basis of the outcomes of other solid cancer stem cells, and may open a new chapter of ovarian cancer stem cell research. Sorting according to the specific CSCs surface marker by flow cytometry is still the main way to isolate ovarian cancer stem cells, and the present study mainly include the following surface markers:CD133, CD117, CD44, ALDH1, EPCAM, CD24, CD90and so on. Although some research progress has been made, because these markers are draw lessons from other solid tumor stem cells, the separation and enrichment of ovarian cancer stem cells is not very effective, and there is still a debate. Therefore, looking for a specific surface marker to successfully separate and identifiy the ovarian cancer stem cells, may has great value to targeted therapy of ovarian cancer.In the early stage of our work, we used quantitative proteomic techniques based on the stable isotope labeling and mass spectrometry to compare the membrane protein of SP and SP cells, and successfully screened the ovarian cancer stem cell surface candidate protein-transmembrane protein30A (TMEM30A, also called CDC50A). So, on the basis of the previous study, in this research, we mainly identified the CSCs properties of CDC50A surface markers positive (CDC50A+) cells in vivo and in vitro by flow cytometry and serum-free suspension culture technology, which mainly included the ability of sphere formation, self-renewal, differentiation and tumorigenesis of CDC50A positive ovarian cancer cells. Besides, we also validated the cisplatin resistance of CDC50A positive cells, and analyzed the relationship between the expression of CDC50A protein and the clinical outcomes of EOC patients. At last, we demonstrated the role of estrogen played in the microenvironment of CSCs and discussed the effect of HRT on the prognosis of EOC patients.Methods1. The proportion of CDC50A positive cells in the ovarian cancer cell lines such as SKOV3, A2780, IGROV1, COC1,OVCAR3, ES2were analyzed by flow cytometry; CDC50A positive and CDC50A negative cells were sorted by Moflo fluorescent activated cell stream sorter and cultured in HG-DMEM/10%FBS medium, then we verified the differentiation capability of the two populations in vitro; The sphere formation and serial passage ability of the CDC50A positive and CDC50A negative cells were tested by serum-free suspension culture technology; The enrichment of CDC50A positive cells in SKOV3spheres were tested by immunofluorescence and flow cytometry; SKOV3cells were sorted into CDC50A positive and negative population, then serial dilutions of the two population cells were resuspended (1:1) in PBS:Matrigel and implanted subcutaneously into the scapular region of NOD/SCID mice respectively to identify the tumorigenesis capacity of two populations. When the transplanted tumors formed, sorting out of the two populations, inoculated into mice again to test the self renewal capacity in vivo; The proportion of CDC50A positive cells in transplanted tumors was analyzed by flow cytometry to test the differentiation ability of CDC50A positive and negative cells in vivo.2. Tissues and ascites of the EOC patients were collected and the ovarian cancer cells were isolated; The proportion of CDC50A positive cells in the primary ovarian cancer cells were analyzed by flow cytometry; We established the primary tumor sphere culture system, and test the sphere formation capacity and serial passage ability of the primary CDC50A positive and CDC50A negative tumor cells; The enrichment of CDC50A positive cells in primary ovarian tumor spheres were tested by immunofluorescence and flow cytometry; Then, ovarian tumor cells were sorted into CDC50A positive and negative population, and serial dilutions of the two population cells were resuspended (1:1) in PBS:Matrigel and implanted subcutaneously into the scapular region of NSG mice respectively to identify the tumorigenesis capacity of two populations. When the transplanted tumors formed, sorting out of the two populations, inoculated into mice again to test the self renewal capacity in vivo; The proportion of CDC50A positive cells in transplanted tumors was analyzed by flow cytometry to test the differentiation ability of CDC50A positive and negative cells in vivo.3. We tested the difference of cisplatin resistance to CDC50A positive and negative cells by MTT assay in vitro experiments; Then the flow cytometry analysis were used to detect the enrichment of SKOV3CDC50A positive cells with different concentrations of cisplatin; we used the same method to test the enrichment of primary CDC50A positive tumor cells sorted from EOC patient with cisplatin chemotherapy; Besides, we established the transplanted tumor model with NOD/SCID mice, and test the enrichment of CDC50A positive tumor cells with cisplatin chemotherapy.4. We analyzed the proportion of CDC50A positive cells in primary tumors and metastasis of EOC patients respectively by flow cytometry; Then we analyzed the proportion of CDC50A positive cells in different EOC patients; Immunohistochemistry were used to analyzed CDC50A protein expression of76EOC patients and its correlation with clinical prognosis.5. Different concentration of estrogen were added to culture of CDC50A positive cells, and demenstrated the role of it in maintaining the growth and the self-renewal of ovarian cancer stem cells. Then, in combination with clinical data, we analyzed the impact of postoperative hormone replacement therapy on the patients with ovarian cancer. Results1. We found that CDC50A positive population were exist in six ovarian cancer cell lines, and the proportion of CDC50A positive cells were ranged from0.4%to2%; We cultured CDC50A positive cells in vitro after sorting from SKOV3cell line, which could again differentiate to CDC50A negative cells and maintain its positive rate, but the CDC50A negative cells couldn’t differentiate to CDC50A positive cell subsets, which indicated that CDC50A positive cells had the potential to differentiate to CDC50A negative cells; The SKOV3sphere were formed by serum-free suspension culture after one week, and compared to the CDC50A negative cells, CDC50A positive cells were more capable of generating spheres. Besides, CDC50A positive spheres can passage consecutively, while the CDC50A-spheres can’t. Besides, we examined the expression of CDC50A protein in SKOV3spheres and attached cells by immunofluorescence, which showed that the majority spheroid cells stained for CDC50A, while the attached cells can’t. Similarily, we found that the expression of CDC50A in spheroid cells was much higher than the attached cells by flow cytometry. The tumor formation experiment showed that1/3mice injected with1×104CDC50A negative cells can formed tumors, but no tumor formation detected following injection of1×103cells, While CDC50A positive cells were tumorigenic in mice even with1×102cells. We also found that only CDC50A positive cells were able to generate secondary tumors in NOD/SCID mice. The H&E staining and flow cytometry analysis showed that not only the histology of the first and second passage xenografts was similar, but also the frequency of CDC50A positive cells was maintained in xenografts obtained by serial in vivo passaging of CDC50A positive cells,which indicated that CDC50A positive cells also have the capacity of differentiation m vivo.2. To further demonstrated the properties of CDC50A+cells in primary tumor cells, we analyzed primary EOC specimens to find that CDC50A was present at variable expression in all primary tumors and ascite samples. Similar to cell lines, we isolated and inoculated the human ovarian tumor and ascite cells at a density of104cells/ml to low-attachment6-well plates in serum-free medium with cytokines. Spheres were observed about2weeks. In order to further demonstrate the sphere formation capacity of CDC50A+/Lin-cells, CDC50A+/Lin-and CDC50A7Lin-populations were sorted from three human ovarian tumors to inoculate in suspension culture at a density of10000cells/well. Compared to CDC50A-/Lin-cells, CDC50A+/Lin-cells were more capable of generating spheres and even with serial passage. Then, we found that CDC50A protein was expressed at a high level in spheres by immunofluorescence. Similarily, compared to the isolated tumor cells from human tissues, the proportion of CDC50A+/Lin-cells in spheres was much higher. To test if the CDC50A+/Lin-cells isolated from human EOC to generate tumors in mice, sorted cells were suspended in Matrigel and subcutaneously injected into NSG mice. We found that104CDC50A-/Lin-cells did not grow tumors, While104CDC50A+/Lin-cells generated tumors from3/8. Tumor generation required approximately～5months. Similarly, We found that only CDC50A+/Lin-cells were able to generate secondary tumors in NSG mice. Histological analysis of the xenograft tumor generated by CDC50A+/Lin-cells was identical to that seen in the primary tumor; Flow cytometry analysis showed that the frequency of CDC50A+/Lin-cells was maintained in CDC50A positive xenografts obtained by serial in vivo passaging.3. Cell survival assay showed greater cisplatin resistance of CDC50A positive cells from SKOV3and ES2cell lines. Then we assayed the proportion of CDC50A positive cells after treatment with increasing doses of cisplatin. A clear, dose dependent decrease in the total number of viable cells were observed, While there was a significant increase in the proportion of CDC50A+cells, the most interesting thing is that the absolute number of CDC50A positive cells were maintained during the treatment of cisplatin, which suggested CDC50A positive cells played an important role in ovarian cancer cisplatin chemoresistance. Besides, we compared the proportion of CDC50A+/Lin-cells in the same patient with ovarian cancer before and after cisplatin chemotherapy, the result indicated that the chemotherepy can enrich the CDC50A+/Lin-cells in human ovarian tumors. To further demonstrated CDC50A positive cells are resistant to cisplatin in vivo, NOD/SCID mice were inoculated with SKOV3cells, and treated by saline or cisplatin for a month. Then we observed that cisplatin significantly inhibited the increase of the mean volumes of tumors in the treatment group. Besides, we identified that the proportion of CDC50A+/Lin-cells from the cisplatin treated tumors were much higher than the control tumors, which further demonstrated that the cisplatin can enrich the CDC50A+/Lin-cells.4. Flow cytometry analysis showed that a higher proportion of CDC50A positive cells were observed in metastasis. During the primary ovarian tumors we collected, the patients with advanced stage serous adenocarcinoma were classified to three group: initial diagnosed patients, patients with neoadjuvant chemotherapy based on cisplatin and the relapsed patients. According to the analysis by flow cytometry, the average proportion of CDC50A positive cells was1.47%in tumors of initial diagnosed patients,9.76%in tumors of neoadjuvant chemotherapy patients and13.35%in tumors of relapsed patients. It is suggested that the expression of CDC50A may correlated with the cisplatin resistance and relapse. To further assess whether our experimental findings on CDC50A are relevant to human ovarian cancer patients in the clinic, we therefore scored a panel of76ovarian tumors for the presence of CDC50A expressing cells using immunohistochemistry. There were22patients with CDC50A high expression and54with low or negative expression. The Logistic regression analysis showed that CDC50A expression and resection status were the independent factor that impact on the platinum resistance of patients. Besides, Kaplan-Meier analysis demonstrated that patients with CDC50A high expression tumor cells have worse progression free survival (PFS) compared to the CDC50A low expression group. And the median survival time for PFS were10months for CDC50A high expression patients and17months for CDC50A low expression patients. Then we performed a Cox multivariate analysis of PFS, which showed that CDC50A expression was an independent prognostic factor, as well as tumor stage. The relative risk of relapse due to ovarian cancer was2.85for patients with CDC50A positive tumors compared to patients with CDC50A negative tumors (95%CI1.32-6.16, p<0.01), which indicated that the expression of CDC50A may predict poor clinical outcome of ovarian cancer patients.5. We found that the growth and self-renewal ability of sphere had no significant influence under the different concentration of estrogen. According to the multivariate analysis, HRT did not significantly influence progression-free or overall survival. Similarly, different types of HRT had no significant effect on the prognosis of epithelial ovarian cancer patients. And the strongest independent variable in predicting both progression-free survival and overall survival was the FIGO stage of the disease.Conclusion1. The ratio of CDC50A positive cell subsets in six ovarian carcinoma cell lines is low, which is similar to the ratio of stem cells; The experiments in vitro and in vivo showed that CDC50A positive cells sorted from SKOV3cell lines have the ovarian cancer stem cell properties, such as serial sphere generation, self-renewal, differentiation and tumorigenesis capability.2. we successful isolated the ovarian tumor cells from EOC patients, and identified the CDC50A positive cell subsets exist in a low proportion in all samples we collected. Similarly, we found the primary CDC50A positive tumor cells also have the ovarian cancer stem cell properties, such as serial sphere generation, self-renewal, differentiation and tumorigenesis capability.3. The expriments of SKOV3, primary ovarian tumor cells and NOD/SCID mice all demonstrated that CDC50A positive cells are highly resistant to cisplatin chemotherapy in vitro and in vivo.4. In the clinic, compared to primary EOC, higer proportion of CDC50A positive cells located in metastasis. Besides, compared to the diagnosed patients, the CDC50A positive cells are rich in the tumors of neoadjuvant chemotherapy patients and the relapsed patients. What’ more, Logistic regression analysis and Cox multivariate analysis showed that CDC50A expression were the independent factors of cisplatin resistance as well as progression free survival in EOC patients, patients with CDC50A high expression have the high risk of tumor relapse.5. This study found estrogen was not the important factor which played a role in the growth and self-renewal of the CSCs. Similarily, the postoperative HRT does not have a negative effect on the progression free and overall survival of epithelial ovarian cancer patients.