【作者】 韩丽萍；

【导师】 董子明；

【作者基本信息】 郑州大学 ， 病理学与病理生理学， 2005， 博士

【摘要】 Eph(erythropoitin producing hepatocellular carcinoma)受体是已知最大的受体酪氨酸激酶(receptor tyrosine kinase,RTK)家族,至今已有14个成员,Eph受体的配体Ephrin(Eph family receptor interacting proteins)已克隆有8个成员。Eph受体及其配体间的相互作用能够诱发双向的信号传导过程,引起一系列生物效应,参与细胞的生存、生长、迁徙和分化活动,在胚胎发育、血管发生、肿瘤形成等方面具有重要功能。 EphA2是该家族第三个被发现并克隆全长cDNA的成员,在成人上皮细胞中低水平表达,存在于细胞-细胞粘附部位。EphA2在上皮细胞中的作用尚不清楚,但研究表明Eph受体及其配体间的相互作用负性调节细胞生存、生长、迁徙和浸润活动。在肿瘤细胞中EphA2经常高表达并出现功能改变。在乳腺癌、恶性黑色素瘤、前列腺癌、结肠癌及肺癌等多种实体瘤中存在EphA2高表达,其表达水平与疾病的进展及细胞的恶性程度有关。EphA2在肿瘤组织及肿瘤细胞中高表达,尤其高表达于侵袭性更强的肿瘤细胞,提示Eph信号转导促进肿瘤细胞脱离粘附、降低细胞间的接触抑制,导致肿瘤细胞的游离、播散及侵袭。EphA2作为强有力的癌蛋白参与转移性肿瘤的发生发展过程,应用EphA2蛋白的抗体靶向抑制试验,下调EphA2蛋白表达的同时,也抑制了肿瘤细胞的生长,进而降低肿瘤的恶性程度。 EphrinA1是该亚族第一个被发现的配体,其在肿瘤中表达模式的研究刚刚开始,EphrinA1配体的结合有利于EphA2受体的自动磷酸化。EphA2与更多还原

【Abstract】 The Eph (erythropoitin producing hepatocellular carcinoma) receptor family is the largest subfamily of receptor protein tyrosine kinases (RTK), consisting of at least fourteen distinct receptors and eight membrane-bound ligands, known as Ephrins(Eph family receptor interacting proteins) . Signal conduction mediated by Eph receptors and their ligands Ephrins is involved in many processes of the life, including signal conduction of nervous system, angiogenesis and the regulation of cell-cell attachment. Recently, more and more investigators have focused on the roles of Eph family in tumours, espercially the mechanisms of signal conduction and their clinicopathological associations in carcinoma.EphA2 is a transmembrane receptor tyrosine kinase that is found at low levels on adult epithelial cells and enriched within sites of cell-cell adhesion. The cellular functions of EphA2 in epithelial cells remain largely unknown, but much recent evidence indicates that the cellular consequences of ligand binding in normal epithelial cells include negative regulation of cell growth and migration. EphA2 overexpression, however, causes tumourigenesis of mammary epithelial cells. EphA2 is frequently overexpressed and functionally altered in many type of carcinoma. Elevated EphA2 protein levels are more frequently observed in invasive cancers, and the highest levels of EphA2 protein are consistently found in the mostaggressive tumour cells. Overexpression of EphA2 is consistently associated with disease progression and highly malignant cellular phenotypes. The cell line research also implicates its powerful oncogene feature in metastatic cancer. Antibody-based targeting of EphA2 down-regulates expression of EphA2 and inhibits malignant cell behavior.EphrinAl (B16) is the major ligand for EphA2. Ligand binding contributes to EphA2 autophosphorylation. However, little is known about its expression pattern in tumours. EphA2 and EphrinAl are found to be co-expressed in human malignant tumour cells as well as in blood vessels, where Ephrin Al is believed to function as an autocrine growth factor. EphA2 and EphrinAl are both overexpressed in malignant melanomas and EphrinAl is reported to stimulate the growth of EphA2-expressing melanoma cell lines. Eph/ephrin signaling may be involved in tumorigenesis, metastasis, invasion and angiogenesis. All the evidence suggests that Eph receptors and ephrin ligands play critical roles in tumour angiogenesis and tumour growth.The ER, a nuclear hormone receptor, is a member of a family of activated transcription factors that can initiate or enhance the transcription of genes containing specific hormone response elements. Two ERs have been identified, ER a and ER P , which are distinct proteins encoded by separate genes located on different chromosomes. They exhibit considerable homology in the DNA- and ligand-binding domains and have similar relative binding affinities. However, the N- terminal regions of ER a and ERB are poorly conserved, with only 20% amino acid homology. The effects of estrogen on its target cells are mediated through ER a and ER P .The different structure of the two receptors suggests that they may have different functions. Recent studies have documented an inverse relationship between EphA2 and estrogen receptor expression in breast cancer cell lines. EphA2 acts as a downstream target of estrogen. Overexpression of EphA2 decreases estrogen dependence as defined using both in vitro and in vivo criteria. EphA2 overexpression also decreases the ability of tamoxifen to inhibit breast cancer cell growth and tumorigenesis.Ovarian carcinoma is the leading cause of death from gynecologicalmalignancies and surface epithelial carcinomas of the ovary comprise the majority (80-90%) of ovarian carcinomas. The overall 5-year survival rate of ovarian carcinoma patients is <50%. The high mortality rate is usually ascribed to late diagnosis of this tumour, which lacks early symptoms. But even in late stages of the disease, the courses are highly variable. Despite considerable efforts aimed at elucidating the tumourigenesis of ovarian carcinoma, its molecular mechanism is still largely unknown. There has been tremendous interest in identifying markers of prognostic and therapeutic significance. Recently, Thaker er al reported that EphA2 expression was associated with aggressive features in ovarian carcinoma. To date, this is the only study regarding EphA2 in ovarian carcinoma. No study about EphA2 and EphrinAl in this type tumour is available in literature yet. Ovarian carcinoma is steroid hormone-dependent tumour. Estrogen plays an important role during development of ovarian carcinoma. The research on estrogen receptor has been recently highlighted. A large number of reports focusing on the roles of ER a in ovarian carcinoma are available in literature, but few is about ERp\ For better understanding of ovarian carcinogenesis and searching for alternatively effective therapy of ovarian carcinoma, we systematically analysed the expression patterns of EphA2 and ephriAl in a series of Norwegian ovarian carcinoma by immunohistochemistry , LCM (laser capture microdissection)-assisted RT-PCR and Western blot, explored the relationship between protein expressions of EphA2, EphrinAl and clinicopathological features and survival of ovarian carcinoma patients. We detect the expression patterns of estrogen receptor subtypes ER a and ERB in ovarian carcinoma by immunohistochemistry , LCM -assisted Western blot. The relationships between the expressions of estrogen receptor subtypes and EphA2 were analysed. Immunocytochemistry and RT-PCR were used to measure the expressions of both EphA2 and EphrinAl mRNA and protein in Human epithelial ovarian carcinoma cell lines SKOV3 and OVCAR3. In addition, suppression the expression of EphA2 in SKOV3 and OVCAR3 by using liposome-mediated RNAi method was also conducted, in order to study the possibility of gene therapy in vitro. This study is divided into three parts as listed below:The first part: Protein and mRNA expressions of EphA2 and EphrinAl in ovarian carcinoma and their corelations to clinicopathological features and survivalMethods1. Immunohistochemistry was used to assess the protein expressions of EphA2 and EphrinAl in 118 patients with ovarian carcinoma, FIGO stage lib -IVb and 12 patients with normal ovary and the relationships between protein expressions and clinicopathological features and survival were statistically analyzed accordingly.2. Frozen tissues from 30 cases in which the tumours showed different grades of EphA2 and EphrinAl protein expressions were used for laser capture microdissection(LCM). About 50 cancer cells in each frozen section were captured with the LCM method for extraction of mRNA.3. mRNAs of EphA2 and EphrinAl were examined by one-step RT-PCR and compared with their protein expressions.4. Immunocytochemistry and RT-PCR were used to measure the expressions of both EphA2 and EphrinAl mRNA and protein in Human epithelial ovarian carcinoma cell lines SKOV3 and OVCAR3.5. Statistical analysis: All the dates were analyzed by SPSS 11.0 statistical package. Statistical analyse were carried out by using x2 test, / test, ANOVA, LSD and Spearman rank correlation. Survival analyse were done by Kaplan-meier and Cox regression methods. Overall survival was the period from the first diagnose to the latest clinical observation or death time. P values of <0.05 were considered statistically significant.Results1. Among the 12 normal ovary tissue, 8(66.7%) were scored as grade O(-) and 4 (33.3%) were scored as grade 1(+) for EphA2 protein expression. 11 (9.3%)cases were scored as grade O(-), 67 (56.8%) scored as grade 1(+), 26(22%) scored as grade 2(++) and 14 (11.9%) scored as grade 3(+++) in 118 ovarian carcinoma. Thepositive rates of expression of EphA2 protein in normal ovary tissue and ovarian carcinoma were 33.3%(4/12) and 90.7%(113/118), respectively. Expressions of EphA2 were lower in normal ovary tissue than that in ovarian carcinoma (PO.01).2. Among the 12 normal ovary tissues, 5(41.7%) were scored as grade O(-), 6(50%) were scored as grade 1(+) and l(8.3%)scored as grade 2(++) for EphrinAl protein expression. 3 cases (2.5%) were scored as grade O(-), 67 (56.8%) scored as grade 1(+), 32(27.1%) scored as grade 2(++) and 16(13.6%) scored as grade 3(+++) in 118 ovarian carcinoma. The positive rates of expression of EphrinAl protein in the normal ovary and ovarian carcinoma were 58.3%(5/12) and 97.5%(113/118), respectively. Expressions of EphrinAl were lower in normal ovary tissue than that in ovarian carcinoma (P<0.01).3. EphA2 and EphrinAl protein expressions often co-localized in the same tumour areas and vascular endothelial cells, and expression patterns of EphA2 and EphrinAl were basically consistent. There was a positive correlation between positive expression of EphA2 and EphrinAl (PO.001) .4. There was no significant correlation between the expression of EphA2 protein and age, the size of residual tumour, histological type and efficacy of primary chemotherapy(P > 0.05). The expression of EphA2 protein was significant associated with FIGO stage and histological grade(P<0.05).5. There was no significant correlation between the expression of EphrinAl protein and age, FIGO stage ,the size of residual tumour and histological type (P>0.05). The expression of EphrinAl protein was significant associated with histological grade and efficacy of primary chemotherapy (PO.05).6. Survival analysis revealed that FIGO stage, histological type and EphA2 immunostaining were significantly associated with poorer overall survival (PO.05). They were of independent significance for disease specific survival .No significant association was observed between Ephrin Al expression and survival(P>0.05).7. Variable amount of mRNA expressions of EphA2 was observed in the 30 tumoursanalysed. There was difference in mRNA expressions between protein expression gradeO-1 and grade2-3(P<0.05). Our results indicate that the level of EphA2 mRNA expression correspond to its proteins.8. Variable amount of mRNA expressions of EphrinAl was observed in the 30 tumours analysed. There was difference in mRNA expressions between protein expression gradeO-1 and grade2-3(P<0.05). Our results indicate that the level of EphrinAl mRNA expression correspond to its proteins.9. Both immunocytochemistry and RT-PCR showed strong EphA2 and Ephrin A-l mRNA and protein expressions in Human epithelial ovarian carcinoma cell lines SKOV3 and OVCAR3.The second part: Protein expressions of ER a and ER ￡ in ovarian carcinoma and their corelations to EphA2 protein expressionMethods1. Immunohistochemistry was used to assess the protein expressions of ER a and ERP in 118 patients with ovarian carcinoma, FIGO stage lib -IVb and the relationships between protein expressions and clinicopathological features and survival were analyzed accordingly.2. Frozen tissues from 20 cases were used for laser capture microdissection(LCM). About 2000 ~ 5000 cancer cells in each frozen section were captured with the LCM method for extraction of protein.3. Protein expressions of EphA2 , EphrinAl, ER a and ERPwere detected by Western blot and the protein expressions of ER a and ER P were compared with those of EphA2.4. Statistical analysis: All the dates were analyzed by SPSS 11.0 statistical package. Statistical analyse were carried out by using °￡ test, t test, ANOVA, LSD and Spearman rank correlation. Survival analyse were done by Kaplan-meier and Cox regression methods. Overall survival was the period from the first diagnose to the latest clinical observation or death time. P values of <0.05 were considered statistically significant.Results1. The positive immunostainings of ER a and ER P were located in the nuclear area of the tumor cells. 32 (27. 1%) ,20 (16.9%) , 17 (14.4%) ,49 (41.6%) of the 118 tumours demonstrated negative, weak, moderate and strong for ER a protein expression, respectively. 1(0.8%), 7 (5.9%), 13 (11.1%), 97 (82.2%) of the tumours were negative, weak, moderate and strong for ERB protein expression, respectively. The positive rates of ER a and ERB expression in ovarian carcinoma were 72.9%(86/118) and 99.2%(1/118), respectively. There was no correlation between positive expression of ERa and ERB (P>0.05).2. There was no significant correlation between ERa expression and age, FIGO stage, histological grade, residual tumour size and efficacy of primary chemotherapy (P > 0.05). A positive association trend between ER a protein expression and histological type was, however, observed (P = 0.05).3. There was no significant correlation between ERB expression and age, histological type, FIGO stage, residual tumour size and efficacy of primary chemotherapy(P> 0.05). But ERB expression was significantly associated with histological grade(P<0.05).4. Survival analysis revealed that there was no significant association between ER a expression and survival(P > 0.05). High levels of ERB expression was significantly associated with a shorter overall survival (PO.05).5. Among the 78 samples with lower EphA2 protein expressions (grade 0 and grade 1), 23 were scored as grade 0(-), 16 were scored as grade 1(+), 9 scored as grade 2(++) and 30 were scored as grade 3(+++)for ER a protein expression; 1 were scored as grade 0(-), 7 were scored as grade 1(+), 9 scored as grade 2(++) and 61 were scored as grade 3(+++) for ERB protein expression. Among the 40 samples with higher EphA2 protein expressions (grade 2 and grade3), 9 were scored as grade 0(-), 4 were scored as grade 1(+), 8 scored as grade 2(++) and 19 were scored as grade 3(+++)for ER a protein expression; 4 scored as grade 2(++) and 36 were scored as grade 3(+++) for ERB protein expression. There was nosignificant correlation between ER a expression and EphA2 expression (P>0.05). But ERfl expression was significantly associated with EphA2 expression (PO.05).6. Among the 20 samples with higher EphA2 mRNA expressions, cancer cells were successfully captured with the LCM method. EphA2 protein band (130kD) in each frozen sample was shown clearly by Western blot with optimized operating conditions. EphA2 protein expression was consistent with results of immunohistochemistry and RT-PCR analysis. 4 sampls from 20 samples showed EphrinAl protein band (26kD) , and EphrinAl protein expression was coincident with the results of immunohistochemistry and RT-PCR.7. Protein expressions of ER a and ERJ3 were assayed by Western blot in the 20 samples. ER a protein bands (66kD) were shown in 3 sampls,which was coincident with results of immunohistochemistry . ERJ3 protein bands (56kD) were shown in all samples, which was coincident with results of immunohistochemistry. ERB protein bands and EphA2 protein bands were compared. The results showed that they were coincident.The third part: Influence of EphA2siRNA on the expressions of EphA2 mRNA and protein in human epithelial ovarian carcinoma cell lines SKOV3 and OVCAR3Methods1. 2 pairs of EphA2siRNA and one pair of non-silencing control siRNA were designed.2. Human epithelial ovarian carcinoma cell SKOV3 and OVCAR3 were transfected by EphA2siRNA in three different concentrations.3. Immunocytochemistry and Western blot were used to evaluate the inhibition effects of EphA2siRNA on expressions of EphA2 and EphrinAl protein in ovarian carcinoma cell SKOV3 and OVCAR3.4. RT-PCR was performed to measure the inhibitory effects of EphA2siRNA onexpressions of EphA2 and EphrinAl mRNA in ovarian carcinoma cell SKOV3 andOVCAR3.5. The effects of EphA2 siRNA on proliferation in SKOV3 and 0VCAR3 cells were observed by drawing cell growth curves and measuring cloning efficiency.6. The effects of EphA2 siRNA on adhesion in SKOV3 and 0VCAR3 cells were detected by cell adhesion assay.7. The inhibitory abilities of EphA2 siRNA on tumor cell invasion in SKOV3 and OVCAR3 cells were evaluated by Matrigel-boyden chamber method.8. Statistical analysis: All the dates were analyzed by SPSS 11.0 statistical package. Statistical analyse were carried out by using x2 test, / test, ANOVA, LSD and Spearman rank correlation. P values of <0.05 were considered statistically significant.Results1. Transfection of Alexa Fluor 488 labeled, non-silencing siRNA allowed determination of transfection efficiency by fluorescence microscopy . The transfection efficiency in SKOV3 and OVCAR3 cells were 100%.2. Immunocytochemistry showed the expression level of EphA2 antigen in SKOV3 and OVCAR3 obviously reduced after EphA2siRNA transfection (P<0.05) . No inhibitory effect of EphA2siRNA on expression of EphrinAl antigen was observed(P>0.05).3. The expression of EphA2 mRNA in SKOV3 cells was obviously depressed by EphA2siRNA. There was obvious difference between control group and EphA2siRNA group( PO.05). There was significant difference between transfected groups at different transfecting time (PO.05). The most effective was EphA2siRNA group 96h after transfection. No significant difference was found in different concentration and other EphA2siRNA groups (P>0.05).4. The expression of EphA2 mRNA in OVCAR3 cells was obviously depressed by EphA2siRNA, and there was significant difference between control groups and EphA2siRNA groups( PO.05).5. No significant difference for EphrinAl mRNA in SKOV3 and OVCAR3 cells was found in different concentration and EphA2siRNA groups (P>0.05).6. The expression of EphA2 protein in SKOV3 cells was obviously depressed by EphA2siRNA, and there was significant difference between control group and EphA2siRNA group( PO.05). There was significant difference between transfected groups at different transfecting time(P<0.05). The most effective was lOOnM EphA2siRNA-l group 96h after transfection. No significant difference was found in different concentration and EphA2siRNA groups(P>0.05).7. The expression of EphA2 protein in 0VCAR3 cells was obviously depressed by EphA2siRNA, and there was significant difference between control groups and EphA2siRNA groups( P<0.05). No significant difference was found in different concentration, transfecting time and EphA2siRNA groups (P>0.05).8. The expressions of EphA2 mRNA and protein in SKOV3 and 0VCAR3 cells were suppressed significantly by different concentrations of EphA2siRNA, and there was a correlation between the suppressive effection for expression of EphA2 mRNA and expression of EphA2 protein (PO.05) in both SKOV3 and 0VCAR3 cells. In SK0V3 cells, the most effective inhibition for EphA2 mRNA and EphA2 protein was lOOnM EphA2siRNA-l group 96h after transfection.9. The cell growth of SKOV3 and 0VCAR3 cells were obviously depressed with different concentration of EphA2siRNA transfecting at least 48h, and there was a significant difference between control groups and EphA2siRNA groups (.P<0.05). The inhibitory effects depended on the EphA2siRNA concentration after 72h transfection.10. No significant difference for the ability of cloning efficiency in SKOV3 and OVCAR3 cells was found in between control groups and EphA2siRNA groups (P>0.05).11. The cell adhesion of SKOV3 and OVCAR3 cells was significantly inhibited by EphA2siRNA transfection, and there was a significant difference between control groups and EphA2siRNA groups( P<0.05).12. SKOV3 and OVCAR3 cell Matrigel invasion was significantly inhibited byEphA2siRNA transfection, and there was a significant difference between control groups and EphA2siRNA groups( P<0.05).Conclusions1. High levels of EphA2 and EphrinAl protein expressions in ovarian carcinoma tissue and ovarian carcinoma cell lines SKOV3 and OVCAR3 indicate that both of them are involved in ovarian carcinogenesis. EphA2 and EphrinAl colocalized in tumour cells in ovarian carcinoma, their protein expressions are associated with histological grade, suggesting that EphA2 and EphrinAl may cooperate in tumour angiogenesis.2. EphA2 positive staining is significantly associated with poorer overall survival. This indicates that EphA2 may be an independent prognostic marker for ovarian carcinoma.3. The higher positive rates of protein expressions of ER a and ERB in ovarian carcinoma indicates that ovarian carcinoma is steroid hormone-dependent disease. A positive association trend between ER a protein expression and histological type has been observed in our study. ERB expression is significantly associated with histological grade. These findings suggest that estrogen receptor subtypes may have different biological functions.4. That high levels of ERB expression was significantly associated with a shorter overall survival indicates that ERB expression may be of predictive value for ovarian carcinoma in clinical management.5. LCM combination of highly sensitive Western blot could specifically and accurately detect target protein in a limited number of ovarian carcinoma cells. The optimized procedure can be used for a particular protein expression study in tissues where the number of target cells is limited.6. Both the expressive levels of mRNA and protein in SKOV3 and 0VCAR3 cells could be successfully suppressed by using liposome-mediated RNAi technique. Down-regulation of EphA2 gene by EphA2siRNA in SKOV3 and OVCAR3 showed retardation of cell growth, proliferation and partial reversion of themalignant phenotype, including the ability of adhesion and invasion. EphA2siRNA may be a new and potent target of gene therapy in ovarian carcinoma.更多还原