Serum Proteomics Analysis Of Rat Models With Epithelial Ovarian Carcinoma

Ovarian carcinoma, which has different pathologic types, is one of the most common gynecologic malignancies in women.90%of ovarian cancer is thought to be derived from ovarian surface epithelium (OSE), that is to say epithelial ovarian carcinoma (EOC), which is the leading cause of gynecologic malignancies deaths. In United States, it is estimated that21,650new cases will be diagnosed, the mortality rate achieves71.7%. Due to the hidden start and a lack of effective screening methods at early stages, over70%of patients with ovarian cancer are diagnosed at an advanced stage. Though, in recent years, greater development has taken place in the aspect of operation and medication, the5year survival rate still fluctuates at the level of25%-30%. Otherwise, if diagnosed earlier, the survival rate can be as high as94%using nowadays measures. Therefore, it is hugely important to seek convenient and non-invasive or minimally invasive serum tumor biomarker acted as significant supplementary mean for early-diagnosis and therapeutic target.The occurrence and development of tumor is always accompanied by changeable protein levels. Serum samples are easy to obtain and monitor at any time, starting from the protein considered as the biological performer, it is excellent to investigate the dynamic protein levels in different stages of EOC as well as the relationship and regularity dynamically, entirely and quantitatively utilizing serum proteomic technology, thus EOC specific serum protein expression profile (PEP) and then specific serum biomarker can be easily obtained, given this, extremely valuable information will be provided for early diagnosis, sub-typing and prognosis, etc. of patients with EOC. Serum proteome pattern for early tumor diagnosis by proteomics is a hot research area in biological medical field. Recently, isobaric tags for relative and absolute quantitation (iTRAQ) technology, which is the new emerging quantitative proteome analysis based on tandem mass spectrometry (MS/MS), can exactly examine differential protein expression from multiple samples, thus making proteome genuinely a differential display technique. EOC is divided into four major histopathological subtypes, including ovarian serous cancer (OSC), ovarian mucinous cancer (OMC), ovarian endometrioid adenocarcinoma (OEA) and ovarian clear cell adenocarcinoma (OCCA). Molecular events and expression pattens of tumorigenesis are different for each EOC subtype, in addition, clinical signs and biological behaviors as well as responses to chemotherapy are also distinct. Therefore, the diagnostic and prognostic value of identical biomarker for different EOC subtypes is diverse. Furthermore, regular expression of a single biomarker is changeable during occurrence and development of tumor, but limited to a single EOC subtype. Otherwise, so far, majority of ovarian cancer biomarkers are not subtype specific and not valuable in late individualized therapies for different ovarian cancers. It is clear that identify and estimate subtype specific biomarkers for early-stage EOC to improve early diagnosis and survival rate for these EOC subtypes only if research different EOC subtypes separately, thus individual therapies for different EOC subtypes will come true.To solve the problems listed above, we constructed an appropriate rat model to mimic the dynamic changes during development of human EOC. The first part of this paper constructed DMBA induced Wistar rat EOC models through cloth strip which was embedded in situ, then estimated the successful inductivity of the models using H.E. staining. To greatest extent, we can obtain rat EOC models as more as possible, whose histopathologial changes were exactly the same as human ovarian cancers, and the alteration of cell characteristics was accordant with tumorigenesis. The inductivity of ovarian cancer that similar to the research results of domestic and foreign literatures, could reach as high as64%in the32nd week after operation. Given the well homogeneity of the induced ovarian cancer, EOC accounted for the majority (60%), especially serous cancer, otherwise, the inductivity of mucinous cancer and clear cell adenocarcinoma with the worst prognosis was lower. Furthermore, the inductivity of ovarian sarcoma was just4%, less than published results. Using the serum collection from different stages of OSC-burdened rat models before and after operation and diverse EOC-subtypes-burdened rats, thus made EOC-subtypes tumorigenesis dynamical analyses possible, and further intensively studied early-stage diagnosis of EOC and sub-typing, then, strong theoretical foundation for targeting therapy about EOC ultimately could be provided.In order to estimate the possibility that rat EOC model mimicked human protein expression profile of carcinogenesis, the second part of this paper collected serum from rat OSC models before and after experiment, coordinated with serum from the same period of Blank and Control rat groups, then combined into pools respectively. After enriching low abundant serum proteins using Proteominer kit, coupled with iTRAQ-labeled and two-dimensional liquid chromatography-tandem mass spectrometry (2DLC-MS/MS) to comparatively analyze rat OSC model specific serum PEP, aiming at finding out OSC-related early-stage-specific serum biomarkers. Compared within an OSC model group, a total of22differentially expressed proteins were identified in serum of OSC-burdened rats, of which,14proteins were up-regulated and8proteins were down-regulated; while relative to Blank and Control group, a total of27significant discrepantly expressed proteins were identified in serum of OSC-burdened rat,13out of27were up-regulated whereas another14proteins were down-regulated. Compared to Blank rat group, Clu was increased significantly in serum of OSC-burdened rat while Gsn decreased significantly according to western blot validations which were not only identical to the results of MS/MS, but also coincide with the published literature elsewhere, thus confirmed the feasibility of screening OSC-related proteins by proteomics to some extent. Besides, some differential proteins such as Apoal, Apob, Hpx and Itihl, etc. might be associated with OSC tumorigenesis, then could be considered as potential OSC-specific early biomarkers and promising targets for serous cancer treatment.The aim of the third part was to find out different EOC subtypes specific serum PEP and then subtype-specific early-stage serum biomarkers. The experiment was performed as follows:①collected serum from different rat EOC subtypes models, coordinated with serum from the same period of Blank and Control rat models, combined into pools respectively,②removing the most high abundant serum proteins (albumin, transferrin and immunoglobulin G) by Agilent’s multiple affinity removal system (MARS) column,③coupled with quantitative proteome based on iTRAQ-labeled and2DLC-MS/MS. Compared to Blank and Control rat group, a total of14differentially expressed proteins were identified in the four EOC subtypes,8out of14proteins were continuously high expressed and another6were continuously low expressed. In addition to the maintaining proteins, relative to Blank and Control rat group, a total of21proteins discrepantly expressed in serum from rat OSC-burdened model, of which11proteins were significantly up-regulated and another10were significantly down-regulated. A total of20proteins were identified differentially expressed in serum from rat OMC-burdened model, of which 13proteins were statistically up-regulated while another7were statistically down-regulated. A total of18proteins were identified differentially expressed in rat OEA-burdened model,5out of18were up-regulated while another13were down-regulated significantly. A total of7proteins were identified differentially expressed in rat OCCA-burdened model, only2were statistically high level and5were statistically low level. The above results suggested that F2and Saa4were OMC-specific differentially expressed proteins, however, not a specific protein could be seen in any other EOC subtypes. According to western blot analyses, when compared to Blank, the level of Ponl was decreased in all the four subtypes, and the level of Clu was increased in rat OMC, OEA and OCCA model, but no significant difference could be seen in rat OSC model, besides, the level of Plg was decreased in rat OSC, OMC and OEA model, but no significant difference could be seen in rat OCCA model. Due to the coincidence with the results of MS/MS, it is probability that diverse subtype-specific proteins could be screen by proteomics to some extent. Then, in the future, extremely valuable information will be provided for early-stage diagnosis of different EOC subtypes, sub-typing and prognosis, then, making individual therapies true for EOC subtypes and ultimately laying favourable experimental foundations for improving clinical therapies and survival rates of patients with EOC.In conclusion, the Wistar rat EOC models, constructed in our project, were capable of demonstrating serum PEP of OSC tumorigenesis and different EOC subtypes entirely and dynamically, thus early-stage-OSC and EOC subtypes specific biomarkers could be found out easily. It is clear that our research established a strong theoretical and experimental foundation for the diagnosis, treatment and sub-typing against EOC.