| Among all subtypes of primary brain gliomas, astrogliomas are most frequent and lethal. According to 2000 World Health Organization (WHO) classification criteria, all astrogliomas are classified into four pathologic grades (from grade I to grade IV). Although a little portion of low-grade astrogliomas display benign features, most of these tumors are malignant and usually exhibit a poor prognosis. For example, patients with the most malignant histopathologic subtype, glioblastoma, carry the worst prognosis, with median survival period less than 12 months, despite surgical treatment combined with radiotherapy and/or chemotherapy. To further improve current treatments for brain astrogliomas, investigators have been attempting to identify new biomarkers of prognostic and/or therapeutic significance, but few promising molecules have been identified.Because of significantly elevated expression in a series of cancers, several receptor tyrosine kinases (RTK) are regarded as valuable candidate biomarkers for these cancers. Eph receptor family, the largest one of the RTK family, is therefore of great interest to be fully investigated for its role and significance in the biologic behavior or the treatment of certain cancers.EphA2, a member of A- type Eph receptor, is a widely expressed transmembrane RTK in epithelial tissues, and has been implicated in neuronal development, as well as regulating cell migration and adhesion. In many non-CNS cancers, upregulation of EphA2 was observed and suggested to play a role in malignant progression. As to brain gliomas, overexpression of EphA2 was only detected in advanced tumors, such as glioblastoma multiforme (GBM) and anaplastic astrocytomas (AA). However, EphA2 expression in low-grade astrogliomas has not been clearly characterized and its possibility as a new candidate biomarker for astrogliomas remains to be substantially studied.Here, we reported our detailed study on the expression of EphA2 in 90 cases of human brain astrogliomas and its correlation to the pathologic grade, proliferation, and apoptosis of astrogliomas, revealing a new functional dimension of the glioma expressed EphA2. Furthermore, we also investigated the effects of induced EphA2 overexpression or inhibiting EphA2 expression by RNAi on the proliferation, invasion, apoptosis and tumorigenesis capacity in nude mice of astroglioma cells, which would contribute to indicating the role of EphA2 in the malignant progression of brain astrogliomas.1. Significance of EphA2 expression in human brain astrogliomasFour human brain astroglioma cell lines (U251, U87, BT325 and SHG44), 90 cases of human brain astroglioma tissue and 10 cases of normal human brain tissue were investigated by reverse transcription-polymerase chain reaction (RT-PCR) for expression of EphA2 mRNA. The results showed that, EphA2 mRNA was expressed in U251, U87, and SHG44 cells, whereas no EphA2 mRNA was expressed in BT325 cells. 43 (47.8%) cases of human brain astroglioma tissue expressed EphA2 mRNA, while none was expressed in normal human brain tissues. There was significant difference (P<0.01) between the positive expression rate of EphA2 mRNA in brain astroglioma tissues and that in normal brain tissues. With the ascending of pathologic grade of tumor specimen, the positive expression rate of EphA2 mRNA increased (P<0.05). There were no significant correlation of positive expression rate of EphA2 mRNA with sex and age of patients, as well as size and site of tumors, respectively (P>0.05 for all).EphA2 protein expression in four human brain astrocytoma cell lines (U251,U87,BT325 and SHG44) was investigated by immunofluorscence staining. The results showed that EphA2 protein was highly expressed in U251,U87 and SHG-44 cells, whereas no EphA2 protein expression was detected in BT325 cells. Positive staining of EphA2 protein was mainly located in cytoplasm of tumor cells.The EphA2 and Ki67 protein expression, as well as apoptosis in 90 cases of human brain astroglioma tissue and 10 cases of normal human brain tissue were investigated by immunohistochemistry SABC method and in situ TUNEL assay, respectively. The results showed that, 39 (43.3%) cases of tumor specimens expressed EphA2 protein, whereas no EphA2 protein expression was detected in normal human brain tissues. Positive immunoreactivity of EphA2 protein was located in cytoplasm and/or membrane of tumor cells. The immunoreactivity scoring (IRS) of EphA2 protein in 90 cases of tumor specimens was 2.90±3.89, which is significantly higher than that of normal human brain tissues (P<0.01). With the ascending of pathologic grade of tumor specimens, positive expression rate of EphA2 protein increased markedly (P<0.05); however, there was no significant correlation of positive expression rate of EphA2 protein with sex and age of patients, as well as size and site of tumors (P>0.05 for all). Ki67 expression could be detected in each tumor specimen, and the proliferative index (PI) of tumor specimens ranged from 2.8% to 90.3%, averaging (30.42±24.91)%. With the ascending of pathologic grade, the PI value of tumor specimens increased markedly (P<0.01); in addition, there was significant difference (P<0.01) between PI in EphA2-positive group (53.71±20.39)% and that in EphA2-negative group (12.61±6.45)%. PI of tumor specimens was positively correlated with EphA2 IRS (P<0.01). Apoptotic tumor cells could be detected in all tumor specimens, and apoptotic index (AI) ranged from 0.2% to 4.9%, averaging (1.42±0.99)%. With the ascending of pathologic grade, AI increased markedly (P<0.01), and there was not significant difference (P<0.01) between AI in EphA2-positive group (1.26±0.71)% and that in EphA2-negative group (1.55±1.16)%. AI was significantly negatively correlated with EphA2 IRS (P<0.05).The results suggested that EphA2 mRNA and protein were highly expressed in human brain astroglioma tissues and cell lines; in addition, the expression level of EphA2 mRNA and protein increased markedly with the ascending of pathologic grade of tumor specimens. The results also suggested that EphA2 protein expression was positively correlated with tumor proliferation, as well as negatively correlated with apoptosis of tumor cells, which indicated EphA2 may be a valuable biomarker of brain astrogliomas.2. EphA2 RNAi vector and eukaryotic expression vector transfecting astroglioma cells in vitroAccording to designd cDNA sequence, the specific RNAi fragment targeting EphA2 and relative fragment with single-nucleotided mutatation were synthesized. The two fragments were cloned into pSilencer vector to produce two new vectors which were named as pSilencer-EphA2-SR and pSilencer-EphA2-mSR, respectively. The two vectors were identified by enzyme digestion and sequencing assay. After that, this two vectors and pSilencer vector used as control were transfected into U251 cells by lipofectin, respectively. Based on transfected vectors, the successfully transfected cells were named as U251-SR,U251-mSR, as well as 251-P. Subsequently, RT-PCR and Western blot were performed to investigate the expression of EphA2 mRNA and protein in U251,U251-P,U251-SR and U251-mSR cells. The results showed that the two vectors were successfully constructed. RT-PCR demonstrated EphA2 mRNA expression in U251-SR cells was obviously inhibited, whereas no obvious changes of EphA2 mRNA expression in U251-P and U251-mSR cells were detected. Western blot showed EphA2 protein expression in U251-SR cells was significantly inhibited, whereas no obvious changes of EphA2 protein expression in U251-P and U251-mSR cells were detected.EphA2 eukaryotic expression vector (pcDNA3.2-DEST-EphA2) and empty vector (pcDNA3.2-DEST) were identified by double-enzymed digestion and sequencing. These two vectors were transfected into BT325 cells by lipofectin medium, respectively; then stably transfected cell lines were established through selection by G418. Based on transfected vectors, the successfully established cell lines were named as BT325-E and BT325-P. RT-PCR and Western blot were performed to investigate the expression of EphA2 mRNA and protein in BT325,BT325-P and BT325-E cells. The electrophoresis showed that pcDNA3.2-DEST-EphA2 vector produced two straps respectively with 7200bp and 1290bp length, and pcDNA3.2-DEST producing two straps respectively with 7100bp and 630bp length, which was consistent with our prediction. Sequencing also proved the correct sequence of these two vectors. RT-PCR demonstrated EphA2 mRNA expression in BT325-E cells was obviously increased, whereas no obvious EphA2 mRNA expression in BT325-P and BT325 cells were detected. Western blot showed EphA2 protein expression in BT325-E cells was markedly increased, whereas no obvious EphA2 protein expression in BT325-P and BT325 cells were detected.Based on the above results, EphA2 RNAi vector and eukaryotic expression vector were successfully transduced into astroglioma cells; subsequently, the expression of EphA2 mRNA and protein was inhibited or induced, which may be excellent cell models that could be used to exploit the effects of EphA2 expression on the biologic features of brain astrogliomas.3. Effects of induced EphA2 overexpression or EphA2 RNAi on the proliferation, apoptosis, invasion and FAK phosphorylation in astroglioma cellsCell counting and colony formation assay were performed to investigate the in vitro proliferation of two groups cells (U251-SR, U251-P, U251 and BT325-E,BT325-P,BT325). The cell growth curve drawn by cell counting showed that U251-SR cells grew significantly slower than U251 and U251-P cells (P<0.05), whereas BT325-E cells grew significantly faster than BT325 and BT325-P cells (P<0.05). The results of colony formation showed that the colony number of U251-SR cells markedly less than those of U251 and U251-P cells (P<0.01), as the colony number of BT325-E cells was significantly more than those of BT325 and BT325-P cells (P<0.01).The cell cycle analysis by flow cytometry (FCM) showed that the G0/G1 phase cells significantly increased, whereas S phase cells markedly decreased in U251-SR cells, as compared with those in U251 and U251-P cells. Compared with BT325 and BT325-P cells, S phase cells markedly increased, whereas G0/G1 phase cells significantly decreased in BT325-E cells.The results of HE staining, Hoechst staining, and transmission electro- microscopy assay showed that the apoptotic cells markedly increased in U251-SR cells compared with those in U251 and U251-P cells, whereas the apoptotic cells markedly decreased in BT325-E cells compared with those in BT325 and BT325-P cells. The quantitative analysis of apoptotic cells by FCM showed that the apoptotic cells in U251-SR increased about 9-fold with the apoptotic rate of 22.3% compared with U251 (2.7%) and U251-P (3.4%) cells; whereas apoptotic cells in BT325-E markedly decreased with the apoptotic rate of 2.0% compared with BT325 (4.9%) and BT325-P (5.3%) cells.The in vitro invasion of two groups of cell was investigated by BD BioCoat? Matrigel? invasion chamber. The results showed that the invasive ratios of U251, U251-P and U251-SR cells were (39.7±2.52)%, (38.3±2.08)% and (22.3±5.03)%, respectively; the ratio of U251-SR was significantly lower than those of other two cell lines (P<0.01). The invasive ratio of BT325, BT325-P and BT325-E cells were (29.7±4.73)%, (29.3±2.52)% and (46.3±10.02)%, respectively; the ratio of BT325-E was significantly higher than other two ratios (P<0.01).The FAK expression and phosphorylation of two groups of cells was investigated by immunoprecipitation assay. The results showed that the level of FAK phosphorylation in U251-SR cells was lower than that in U251 and U251-P cells (P<0.01); whereas the level of FAK phosphorylation in BT325-E cells was higher than that in BT325 and BT325-P cells (P<0.01). The above results suggested that the overexpression of EphA2 was closely related with proliferation, invasion and anti-apoptosis in human astrogliomas, presumably through FAK phosphorylation.4. Impact of induced EphA2 overexpression and EphA2 RNAi on tumorigenesis, proliferation and apoptosis of transplanted human astroglioma in nude miceU251 cells were inoculated in flank subcutaneous tissue of nude mice to establish xenograft models of human brain astroglioma. On 15th day postinoculation, a 20μl of PBS (abbreviated as U251-PBS group), Lipofectamine2000+RPMI 1640 without serum (abbreviated as U251-Lipo-RPMI 1640 group), Lipofectamine2000+pSilencer (abbreviated as U251-Lipo-P group) or Lipofectamine2000 +pSilencer-EphA2-SR (abbreviated as U251-Lipo-SR group) was injected into transplanted tumor twicely in one week in situ, respectively. Growth status of tumors was continuously observed and tumor volume was measured termly, and the tumor growth curve was drawn. On 30th day, the animals were killed and the tumor weight was determined. Subsequently, the transplanted tumors were obtained to perform HE staining, EphA2 or Ki67 immunohistochemical staining and TUNEL staining. The PI and AI value were also calculated. The results showed that, compared with U251-PBS, U251-Lipo-RPMI 1640 and U251-Lipo-P group, the tumor growth in U251-Lipo-SR group was markedly slower. HE staining for each group of tumor specimen indicated that apoptotic cells increased in U251-Lipo-SR group in comparison with other three groups. EphA2 immunohistochemical staining detected significant down-regulation of EphA2 protein expression in U251-Lipo-SR group. The PI of transplanted tumor specimens in four groups were (71.26±4.54)%, (70.58±5.68)%, (66.64±4.94)% and (40.62±4.39)%; PI in U251-Lipo-SR group was significantly decreased. In the above four groups, the AI were (2.40±0.43)%, (2.22±0.76)%, (2.16±0.47)% and (11.78±1.56)%; as the AI in U251-Lipo-SR group was significantly increased.BT325,BT325-P and BT325-E cells were inoculated in flank subcutaneous tissue of nude mice to establish xenograft models of human brain astroglioma, respectively. The tumor growth was determined and growth curve of transplanted tumor was drawn. 30 days later, nude mice were killed and resected tumors were investigated by HE staining, EphA2/Ki67 immunohistochemical staining, as well as TUNEL staining. The AI and PI value of tumor specimens were also calculated. The results showed that, compared with BT325 and BT325-P group, the tumorigenesis in BT325-E group was markedly rapid. EphA2 immunohistochemical staining showed that, the EphA2 protein expression in BT325-E group was obviously up-regulated. The PI value in BT325, BT325-P and BT325-E group were (51.74±13.98)%, (48.10±13.21)%, and (78.64±14.87)%, respectively. PI in BT325-E group was significantly increased. AI value were (3.58±0.61)%, (4.32±0.59)% and (1.78±0.33)% in three groups, and AI in BT325-E group was obviously lower. The above results suggested in situ injection of EphA2 RNAi plasmids could inhibit the growth of transplanted astroglioma in nude mice and induced overexpression of EphA2 could promote the growth of transplanted tumors, which not only underlied the fact EphA2 may play a important role in the malignant pregression of brain astrogliomas, but also indicated a new strategy for gene tharapy aiming at malignant brain astrogliomas.In summary, EphA2 overexpression may be closely related with aggressive proliferation, invasion and anti-apoptosis of human brain astrogliomas. EphA2 is not only a valuable biomarker of brain astrogliomas, but also a promising candidate target for gene therapy aiming at malignant astrogliomas. |
PDF Full Text Request