Experimental Therapy Of Human Glioblastoma Via Targeting 14-3-3zeta

Malignant gliomas are the most common tumors of the central nervous system. As to high mortality, stronger disability and more recurrences, the malignant glioma has been one of principal healthy-threatened diseases. Despite the use of neurosurgery, radiation therapy and chemotherapy, the prognosis still retains dismal. The most malignant type of glioma, glioblastoma, the median survival time for patients is about 12 to 15 months, although current therapy is received. This type of tumor is characterized of high proliferation, escape of apoptosis, tissue invasion and sustained angiogenesis, even with standard treatment, has an early disease recurrence and the median time interval from their first recurrence to death is about 7 months. Thus, identifying one or several markers contributing to the generation of glioblastoma and central to cancer progression is imperative to improve diagnosis and treatment. 14-3-3 protein is a family of highly conserved regulatory molecules, expressed in all eukaryotic cells. There are seven distinct isoforms that are identified in mammals, named beta, eta, zeta, epsilon, gamma, theta and sigma.14-3-3 proteins function by forming homodimers or heterodimers and binding to phosphorylated-serine/threonine motifs on their target proteins. Through recruiting their ligands, 14-3-3 proteins have been implicated to regulate a diverse number of cellular process including mitogenic, apoptosis, cell cycle, stress signaling and invasion. These data suggest that 14-3-3 proteins may affect multiple signal pathways in cancer.Our previous study showed that 14-3-3 protein is over-expessed in human glioma, not or weakly immuno-reactivity in astrocytes. The expression level of 14-3-3 protein is positive associated with the pathological grades of glioma and a poor prognosis in patients with glioma. Targeting 14-3-3 with a general antagonist, difopein, the global inhibition of 14-3-3 functions, or suppressing its expression with siRNA, has induced the apoptosis of glioblastoma cells. This approach suggests an essential role of the 14-3-3 protein in cell survival and provides proof that 14-3-3 inhibition may be able to sensitize cancer cells to apoptosis.To investigate whether an individual or some 14-3-3 isoforms has a dominant role in promoting survival of cancer, we identified the expression of 14-3-3 isoforms in human glioma. 14-3-3beta, eta and zeta were up-regulated and they might provide a survival advantage to tumor cells. Meanwhile, a clinical study about 14-3-3 isofoms and glioblastoma prognosis showed that 14-3-3zeta is a positive prognostic factor. Additionally, the importance of 14-3-3zeta in the formation and progression of cancer has been suggested within some tumors. These reports and our finding showed 14-3-3zeta might play an essential role in glioblastoma. Based on these studies, the effects of 14-3-3zeta are studied from the following six aspects.1. The expression and its significance of 14-3-3zeta in glioblastomaThe expression of 14-3-3zeta protein was detected in 47 cases of formalin-fixed and paraffin embedded archival tumor tissue from patients with glioblastoma, and 12 normal human brain tissues by immunohistochemical ABC method. The results showed that, 14-3-3zeta immunoreactivity was mainly localized in the neuronal somata and processes, and some astrocytes showed only weak immunoreactivity. However, strong 14-3-3zeta immunoreactivity was observed in the majority of glioblastoma (74.5%, 35/47). These results suggested that, 14-3-3zeta was expressed in the majority of human glioblastomas. Connecting with some reports, the up-regulated expression of 14-3-3zeta indicated that 14-3-3zeta may be implicated in glioblastoma genesis and progression, providing a study basis for the next study.2. The relationship between 14-3-3zeta and glioblastoma in prognosisForty-seven patients treated with surgery, radiotherapy, and adjuvant chemotherapy between 2005 and 2007 were divided into 2 groups according to 14-3-3zeta expression in an immunohistochemical study: the 14-3-3zeta positive group (n = 35 patients) and the 14-3-3zeta negative group (n = 12 patients). The clinicopathologic features and survival data for patients in the 14-3-3zeta positive group were compared with data from the patients in the 14-3-3zeta negative group. Kaplan-Meier survival analysis, univariate and multivariate analyses were performed to determine the prognostic factors that influenced patient survival. Patients in the 14-3-3zeta positive group had lower overall survival rates and median survival time than those in the 14-3-3zeta negative group (overall 2-year actuarial survival rates, 8.6% for the 14-3-3zeta positive group vs 16.7% for the 14-3-3zeta negative group; overall 2-year median survival time, 12.9 months for the 14-3-3zeta positive group vs 17.9 months for the 14-3-3zeta negative group, P =0.019). 14-3-3zeta positive expression in tumor cells also was correlated with a shorter interval to tumor recurrence (median interval to recurrence, 5.9 months in the 14-3-3zeta positive group vs 8.3 months in the 14-3-3zeta negative group, P = 0.002). Univariate and multivariate analyses showed that 14-3-3zeta positive expression was an independent prognostic factor. 14-3-3zeta positive expression can be used as a potential molecular risk factor in patients with glioblastoma.3. Correlation of 14-3-3zeta level with cell proliferation and apoptosis in human glioblastomaTo investigate the correlation between 14-3-3 IRS (14-3-3 immunoreactivity score) and glioblastoma proliferation and apoptosis, the immunohistochemistry of 14-3-3zeta and Ki67 and TUNNEL staining are programmed in 21 glioblastoma specimens. The percentage of Ki67-positive cells and apoptotic cells were regarded as the proliferation index (PI) and apoptotic index (AI) of the tumor tissues, respectively. The 14-3-3zeta IRS was 6.29±3.57 for 21 cases of tumor specimen. Cell proliferation marker, Ki-67 was found in all tumor specimens, and the PI of all tumor specimens ranged from 27.8 to 82.4% (50.08±13.98%). The PI was positively correlated with 14-3-3zeta IRS (r = 0.870, P<0.001). However, no significant difference was found between the PI in the 14-3-3zeta-positive group (53.15±13.08%) and that in the 14-3-3zeta-negative group (40.26±13.30%; t = 1.916, P = 0.071). Apoptotic cells could be observed in all tumor specimens. The AI of all tumor specimens ranged from 1.0 to 6.4% (3.88±1.41%). The AI was inversely correlated with 14-3-3zeta IRS (r = -0.975, P< 0.001) and AI in the 14-3-3zeta-positive group (3.47±1.19%) was significantly lower than that in the 14-3-3zeta-negative group (5.2±1.31%, t = -2.77, P=0.012). The outcome showed that 14-3-3zeta is involved in glioblastoma proliferation and apoptosis.4. Effects of 14-3-3zeta down-regulation on glioblastoma cell proliferation, apoptosis and invasion in vitroTwo stable cell lines (U251 and U87) were established expressing distinct shRNA through retrovirus approach. The down-regulation of 14-3-3zeta was confirmed in mRNA and protein level by RT-PCR and Western blot. The change of glioblastoma cells with down-regulated 14-3-3zeta was detected by MTT, anchorage-dependent plate colony formation, apoptosis analysis, cell cycle examination and Matrigel transwell. The results from MTT showed that suppressing 14-3-3zeta markedly reduced the cell proliferation, compared with parental glioblastoma cells (PAR) and cells harboring a scrambled shRNA (SCR). The inhibitory rate was 57.5±4.2% for U251-KD cells and 53.6±5.4% for U87-KD cells at day 7, respectively. We next examined the anchorage-independent plate colony formation in the U251 and U87 glioblastoma cells. Compared to U251-PAR and U251-SCR cells, 1.91 and 2.03 fold fewer clones were observed in U251-KD cells, respectively. There were 1.89 and 1.95 fold fewer clones observed in U87-KD cells compared to U87-SCR and U87-PAR glioblastoma cells. The results from FCM showed that the early and late apoptosis ratio of U251-KD cells (10.4±1.8%, 15.1±2.3%) was significantly higher than that in U251-SCR cells (0.8±0.1%, 2.2±0.4%) and U251-PAR cells (0.6±0.1%, 1.8±0.4%; P<0.05). The early and late apoptosis ratio of U87-KD cells (11.4±1.6%, 12.6±2.8%) was significantly higher than that in U87-SCR cells (0.5±0.1%, 1.5±0.4%) and U87-PAR cells (0.3±0.1%, 1.0±0.2%; P<0.05). However, there was no significantly different between U251-PAR cells and U251-SCR cells (or between U87-PAR and U87-SCR; P>0.05). The cell cycle analysis showed that the 14-3-3zeta RNAi group (U251-KD and U87-KD cells) increased in the G0/G1 phase, and reduced in the G2/M and S phase (P<0.05), compared with the parental cells (U251-PAR and U87-PAR) and cells harboring the scrambled shRNA (U251-SCR and U87-SCR). The average count of cells that crossed a matrigel-coated membrane in six high power fields was 46.3±7.1 for the U251-PAR group, 49.2±8.4 for the U251-SCR group and 24.1±3.2 for the U251-KD group. The average count of cell was 57.1±6.9 for the U87-PAR group, 62.2±7.8 for the U87-SCR group and 26.3±3.0 for the U87-KD group. The number of the invaded cells declined notably in U251-KD and U87-KD cells (P<0.05), compared to the control groups, among which no obvious difference was observed in invasion (P>0.05).Our data indicated that 14-3-3zeta depleting reduced cell proliferation, induced apoptosis and cell cycle stasis, decreased the invasive capability and colony-formation. Our data and previous studies support a role for 14-3-3zeta in tumor genesis and progression, including glioblastoma. To further explore the molecular mechanism of growth inhibition caused by 14-3-3zeta suppression, we focus on mRNA and protein expression levels of some important glioma-related genes. 14-3-3zeta inhibition induced down-regulation of oncogene, Cox-2, c-Myc, Survivin andβ-catenin, as well as invasion-related gene, E-caderin. Cell cycle-associated gene, Cyclin D1, was no difference after 14-3-3zeta inhibition, although the cell cycle arrest was indeed observed in the study. These results indicated that the biological function change is a complicated and multiple-gene involved process after 14-3-3zeta RNAi, and the further study on molecular mechanism should be detected in the future study. The treatment targeting 14-3-3zeta may be a potential therapeutic intervention for glioblastoma.5. Stable knockdown of 14-3-3zeta inhibited tumor growth in vivoCells (5×10~6 cells in 100 l of PBS) were implanted by subcutaneous injection into the flank of each 4-week old (20-22 g) athymic BALB/c nude mouse (Center of Laboratory Animal, Fourth Military Medical University) to establish the glioma xenografts model. All animal experiments complied with the international guidelines for the care and treatment of laboratory animals. The mice were divided randomly into 3 groups: U87-PAR, U87-SCR and U87-KD cell groups. The tumor growth was measured with calipers every week from the first week to the sixth week. As reported previously, tumor volumes in mice were measured with a slide caliper and recorded using the formula: volume = a×b2/2, where"a"stands for the larger while"b"stands for the smaller of the two dimensions. By the sixth week, mice injected with U87-KD cells showed a statistically significant decrease in average tumor size compared with all other groups (P<0.05). The tumor of each mouse was removed on the sixth week and weighed. 14-3-3zeta RNAi significantly decreased the solid tumor mass compared to the control group (U87-PAR, 1.63±0.37g; U87-SCR, 1.58±0.41g; U87-KD, 0.73±0.26g; P<0.05). To determine whether tumor growth inhibition is caused by proliferation slowing/or apoptosis cells in tumor from nude mice. As a result, cells in U87-KD group have lower proliferation and higher apoptosis ratio than these in U87-SCR and U87-PAR group. The results were consistent with that from human brain glioblastoma tissues. Thus, we concluded that 14-3-3zete RNAi-induced growth inhibition may be due to slow proliferation and to induce apoptosis in vitro and in vivo. 6. The expression and potential biological significance of 14-3-3zeta in GBM stem cellsIn order to determine whether 14-3-3zeta was present in GBM stem cells, we first established GBM stem cell lines from human GBM tumor specimens. Although some GBM stem cell lines were established, this work focused on two lines. These cell lines possessed the expected characteristics of GBM stem cell. Both cell lines grew as non-adherent neurospheres in culture and were continuously passaged. Both GBM stem cell lines expressed the NSC markers CD133 and Nestin. Both GBM stem cell lines grew as adherent monolayers and began to exhibit morphology consistent with that of astrocytes and neurons on the conditions established for differentiation. Furthermore, with cell diferetiation, the astrocytic marker GFAP and the neural markerβ- tubulinⅢwere induced. This demonstrated that the tumor-derived cell lines retentioned the ability to differentiate, a defining feature of cancer stem cells. We also confirmed that the GBM stem cells generated tumors when as few as 1000 cells were implanted in the nude mice. We found that 14-3-3zeta was positively expressed in these cells as shown by immunostaining. 14-3-3zeta protein may play an important role in the GBM stem cells, but the functions of 14-3-3zeta protein further are needed to studying in glioblastoma stem cells.In summary, our study firstly showed that 14-3-3zeta protein played an important role in the glioblastoma genesis, progression and invasion. The mechanism may involve the anti-apoptotic effect of 14-3-3zeta proteins in the tumorigenesis and development of glioblastoma. 14-3-3zeta may be considered as a novel, potential and promising target for glioblastoma treatment.