Study On Therapeutic Potential Of Lenti Virus-mediated RNAi Targeting HMGA2 In Glioblastoma Multiforme

Glioblastoma multiforme (GBM) is one of the most common and lethal tumors in central nervous system (CNS), its annual incidence is up to 3/100 000. Merely in the United States, up to 23 000 patients are diagnosed as GBM every year. According to World Health Organization (WHO) criteria for CNS tumors, GBM was classified as grade IV, the highest malignant grade. Due to the onset site and invasive behavior, it is almost impossible to identify the border of GBM during surgery and achieve gross total resection. Despite of the combination of surgical resection, aggressive radiotherapy and/or chemotherapy, the median survival time for GBM patients are merely 12-15 months. Thus, alternative therapeutic strategy for GBM is highlighted by clinical doctors and oncologists. Recently, with the deeper understanding of the roles of high proliferation of tumor cells in the tumorogenesis of malignancies, gene therapy inhibiting proliferation and invasion is becoming a brand new and promising strategy for malignant tumors. Various preclinical researches have achieved breakthrough in this area. However, proper genetic target for gene therapy is still uncertain. In the huge family of cytokines regulating cellular proliferation, high mobility group (HMG) A2 is one of the most important members.HMG proteins are among the largest and best characterized group of non-histone chromosomal proteins. They are defined as small molecular weight but powerful nuclear proteins with high electrophoretic mobility in polyacrylamide gels. The HMG protein family has been classified into three subfamilies:HMGA, HMGB AND HMGN. Among them, HMGA2 is currently most interesting member under research. With its three "AT-hooks" HMGA2 can be bound to the narrow minor groove of AT-rich regions of DNA, affecting the transcription of target genes by altering chromatin architecture through protein-protein and protein-DNA interactions. HMGA2 is not expressed in human adult normal tissues, but can be detected in many human neoplasms like thyroid carcinoma, colorectal cancers, and ovarian cancers, and was shown to be associated to tumor proliferation, invasiveness and poor prognosis in these human tumors. However, the exact role of HMGA2 played in GBM is unknown.RNA interference (RNAi) is a newly rising and increasing full-grown gene silencing technique, which is considered to be a very effective research tool in gene engineering fields. Employing this technique, we can stably and high-effectively block the expression of target gene to’silencing’ this gene. Double strand RNA (dsRNA) and short hairpin RNA (shRNA) have already been widely and successfully utilized in study of gene function and disease therapy research.The aims of the present study were as follows:Firstly, detect the protein and mRNA expression of HMGA2, MMP-2, Ki-67 in malignant gliomas and investigate the biological relationships between HMGA2 and WHO grades, as well as the correlation among the 3 markers, and further explore the relation between the overexpression HMGA2 and prognoses of GBM patients. Secondly, design and synthesize lentivirus vectors mediated shRNA recombinant targeting HMGA2 that can be stably and high-effectively expressed in human GBM cell (U251). After transfecting U251 cells with those ShRNA mediated by lentivirus, we investigated targeted gene silence effect of this recombinant. In order to investigate whether HMGA2 could be regarded as an effective gene target for GBM tumor gene therapy or not in the future, the variation of cells proliferation and cell invasion caused by HMGA2 knocking down were also detected. Finally, we validated the inhibitory effects of the lentivirus-mediated ShRNA targeting HMGA2 gene in GBM in vivo and exploited the effectiveness of the transfection mediated by liposome microbubbles and ultrasound exposure.Part1. Relationship of HMGA2 expression and biological Behavior in Glioblastoma MultiformeObjectiveHMGA2 is a structurally and functionally unique member of the HMG proteins family. HMGA2 expression in various tumours has been found to be associated with increasing invasiveness and decreasing patient survival. The aim of the present study was to detect the protein expression of HMGA2 in malignant gliomas, investigate the biological relationship between HMGA2 and the proliferation and invasion marker of malignant gliomas, and evaluate the prognostic significance of the overexpression of HMGA2 in GBM patients.Methods1. Sample collection and patient selectionWe retrospectively selected 78 consecutive patients with histologically confirmed glioma who underwent gross total resection in Shandong Provincial Hospital Affiliated to Shandong University between October 2010 and May 2013. Normal brain tissue samples were obtained from 7 spontaneous intracerebral hemorrhage patients. Clinical and pathological information was retrieved from medical records. All the hematoxylin-eosin-stained sections were re-evaluated and classified by two pathologists according to the criteria of WHO histological classification. Twenty seven of the 78 glioma specimens were classified as DA. Twenty five and 26 tumor specimens belonged to AA and GBM, respectively. None of the patients had received chemotherapy or radiotherapy prior to surgery. All AA and GBM patients received postoperative radiotherapy or chemotherapy, while none DA patients received postoperative radiotherapy or chemotherapy. The mean age of patients (female 37, male 41) was 43 years old. Written consent to use stored specimens was obtained from all living patients. The study was approved by the Research Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong University.2. Immunohistochemical staining procedureThe protein expression of HMGA2 is detected according to the standard protocol of streptavidin-perosidase (S-P) method with UltraSensitive S-P kit. Examine the protein expression of HMGA2, MMP-2 and Ki-67 in malignant gliomas of different WHO grades, and analyze the correlation among those markers.3. Analysis of immunohistochemical resultsEach slide was evaluated using light microscopy by 4 independent observers blinded to clinicopathological background of patient samples. Sections upon which there were differing evaluations were re-evaluated simultaneously by observers using a double-headed microscope. HMGA2 staining was categorized on the basis of percentage of positive cells on the slide (0, no expression;+, expression rate 1-30% or weak staining;++, expression rate 31-65% or moderate staining;+++, expression rate 65% or strong staining). Expression of Ki-67 and MMP-2 was also recorded by using the percentage of immunopositive cells.MIB-1 LI was calculated as the percentage of immunopositive glioma cells. Two hundred neoplastic cells were counted in 5 separate areas for each specimen in which positive nuclear staining was evenly distributed.4. Examination of mRNA level and protein expression of HMGA2, MMP-2 and Ki-67Total mRNA and qRT-PCR was performed to examine the mRNA expression level in 7 normal brain tissues and 29 malignant gliomas. Western blot analysis was performed to explore the protein expression in these samples.5. Survival analysisExpression of HMGA2 was analyzed as a dichotomous covariate in survival analysis, with HMGA2 low expression (--+) versus HMGA2 over-expression (+ +-+++). Two groups of GBM patients received postoperative follow-up.6. Statistical analysisStatistical analysis was performed using SPSS 19.0. Fisher’s exact test and the nonparametric test were used to analyze HMGA2 expression in malignant gliomas with different WHO classification. Spearman rank correlation analysis was used to analyze the correlation of HMGA2 expression with expression of Ki-67, MMP-2, and with age and sex. The relation between HMGA2 expression and PFST was analyzed by using the Kaplan-Meier method and confirmed using the Log-rank test. Results were reported as being statistically significant if P-values< 0.05. All quantitative data are presented as means ± SD.ResultsImmunohistochemical results showed that immunopositive cases accounted for 92.3%(72/78) of all gliomas and no immunopositive cases was observed in non-tumorous specimens. The expression level of HMGA2 protein significantly correlated with malignancy of gliomas. It was significantly higher in GBM (P= 0.007) and in AA (P= 0.037) than in DA, whereas there was no significant difference between GBM and AA (P= 0.862); most interestingly, HMGA2 expression was also significantly correlated with expression of Ki-67 (r= 0.415, P< 0.01) and MMP-2 (r = 0.363, P< 0.01), but not with sex (r=-0.040, P= 0.747) and age (r=-0.073, P= 0.525). Spearman correlation analysis showed that there was significant correlation between HMGA2 nuclear expression and tumor grades. Moreover, qRT-PCR results were similar to those from immunohistochemical analysis:HMGA2 gene expression was significantly higher in GBM (2.56 folds, P< 0.01) and WHO grade Ⅲ gliomas (1.98 folds, P= 0.029) than in WHO grade Ⅱ gliomas, whereas there was no significant difference between GBM and WHO grade Ⅲ gliomas (1.29 folds, P= 0.118). HMGA2 gene expression was correlated with gene expression of Ki-67 (r= 0.386, P= 0.039) and MMP-2 (r= 0.432, P= 0.019). The Western blot showed that the protein expression of HMGA2 in GBM and AA (P< 0.01) was higher than that in DA, however, compared to qRT-PCR results, the protein expression of HMGA2 in GBM was higher than AA (P= 0.028).Conclusions1. There was no expression in normal brain tissues while high expression of HMGA2 in malignant gliomas.2. There was a statistically difference in HMGA2 expression between gliomas of different WHO grades. HMGA2 was highly associated with MMP-2 and Ki-67, which suggested relations between HMGA2 expression and invasiveness and proliferation.3. GBM patients with overexpression of HMGA2 presented poorer prognosis than those with low-expression of HMGA2. HMGA2 could serve as a useful tool for the diagnosis of malignant gliomas and prognosis of GBM patients. HMGA2 gene also might be an effective target for GBM gene therapy.2. There was a statistically difference in HMGA2 expression between gliomas of different WHO grades. HMGA2 was highly associated with MMP-2 and Ki-67, which suggested relations between HMGA2 expression and invasiveness and proliferation.3. GBM patients with overexpression of HMGA2 presented poorer prognosis than those with low-expression of HMGA2. HMGA2 could serve as a useful tool for the diagnosis of malignant gliomas and prognosis of GBM patients. HMGA2 gene also might be an effective target for GBM gene therapy.Part2. Design, synthesis, and screen effective lentivirus-mediated ShRNA targeting HMGA2 geneObjectiveThree shRNA vectors aiming at HMGA2 were designed and synthesized using public website service. Two effective ones were screened out, packaged by lentivirus and extracted for the next step.Methods1. Design and synthesize shRNA vector and package by lentivirusOperating RNAi sequence designing software was downloaded from public website. Based on RNAi structure principle, we designed 3 different inhibitory shRNA vectors targeting directly at HMGA2 labeled as 1#,2#,3# and one negative control vector, labeled as 0#. These different vectors were collected for qRT-PCR and sequence analysis. Results showed that these four vectors are successfully constructed. The vectors were packaged with lentivirus using lentivirus kit and human embryonic kidney cell (HEK293T) and extracted for the transfection of cells.2. Transfect U251 cells with shRNA vectorl#、2# respectively.Objective cells (U251) were cultured in good condition and planted in 6-well plate one day before transfecting. According to group designing, Vectorl#,2# and 0# were separately added into the plate for transfection.3. Screen stable transfectants with pyromycinCells were trypsinized and plated into 1000 cells/ml, and concentration ranges (0.2μg/ml -5μg/ml) pyromycin is added to the culture medium. We picked out the lowest concentration pyromycin when all cells die within 10-14 days. Pyromycin was added 24 hours after transfection. With Cells metabolic activity and pyromycin concentration decreasing, the culture medium containing pyromycin was changed in 3-5 days. At this time the drug concentration may be reduced to 0.2μg/ml. Fluorescence microscope was employed to observe GFP expression.14 days after selection, cell colonies were isolated and expanded for testing.4. Detect interference effects and screen effective vectorsThe vectors with transfection efficiency over 90% were picked out. We detected the HMGA2 RNA level of these cells with qRT-PCR method to inspect the depressing results of different shRNA vectors and screen 1 effective ones.Results1. The transfection efficiency of shRNA vector 1#,2#,3# were 89.4±6.5%, 93.6±5.7%,78.6±12.3% respectively. The transfection efficiency of negative control vector 0# was 96.1±5.4%.2. All these 3 shRNA vectors could decrese the gene presentation. Compared to negative control vector 0#, the silencing efficiency of 1#,2#,3# were 78.3%,82.4%, 50.6% respectively. Considering transfection efficiency and silencing efficiency,1# and 2# were better than 3#.Conclusions1. In this part, we successfully designed and synthesized 3 specific targeting HMGA2 shRNA vectors.2. We managed to pick out 2 effective shRNA vectors of these three. These results provided us reliable evidence to carry on RNAi experiment targeting HMGA2 in U251 cells.Part3. Cell proliferation and invasion level variation of GBM cells by RNA interfering HMGA2 geneObjectiveUsing constructed the conditionally replicating vector 1#、2# containing shRNA targeting human HMGA2 gene which could transfect U251 effectually and selectively and, we investigate the effect of the vector silencing on expression of mRNA and protein of HMGA2 gene in U251 cells on propagation and apoptosis of GBM cell lines U251.Methods1. Investigation of pituitary tumor cell proliferation-ability variation by CCK-8Stable transfection GBM cells (U251) were cultured in good condition. According to group designing (shRNA recombinants and controls), cells were planted in 96-well plate.1 day later, CCK-8 method was employed to contrast the changes of cell proliferation cycle every 24 hours.2. Examination of GBM cell invasion level variation by matrigel assayStable transfection GBM cells (U251) were cultured in good condition. According to group designing (shRNA recombinants and controls), cells were collected. Matrigel assay was employed to contrast the changes of cell apoptosis levels.3. Investigation of GBM cell migration level variation by transwell assayStable transfection GBM cells (U251) were cultured in good condition. According to group designing (shRNA recombinants and controls), cells were collected. Transwell assay was employed to contrast the changes of cell apoptosis levels.4.Examination of phospholytion level of Erkl/2, Jnk and p-38 MAPK by western blotThe proteins were extracted from 3 U251 cell line groups. Western blot was employed to examine phospholytion level of Erkl/2, Jnk and p-38 MAPK proteins.Results1. The proliferation suppressing effects of RNAi targeted to HMGA2 were observed by CCK-8 assay. The proliferation inhibitory rates of 1# and 2# were 45.2% and 40.3%, respectively, which suggested that GBM cells’ proliferation ability decreased, and the growth ability of tumor cells was depressed remarkably.2. Invasion of U251 cell line transfected with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.3. Invasion of U251 cell line transfected with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.4. Phospholytion level of Jnk and p-38 MAPK proteins but not ERK 1/2 proteins were inhibited after transfection with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.Conclusions1. HMGA2 playd an important role in the process of GBM cells’ survival and proliferation indeed. After HMGA2 silenced, invasion and migration ability of tumor cell as depressed remarkably.2. The RNA interference technology would be important in the application value of pituitary tumor gene therapy in the future. HMGA2 could probably be considered as an effective gene target for pituitary tumor gene therapy.3. Jnk and p38 MAPK but not ERK signaling pathways might participate in the proliferation, migration and invasion rgulated by HMGA2 genes.Results1. The proliferation suppressing effects of RNAi targeted to HMGA2 were observed by CCK-8 assay. The proliferation inhibitory rates of 1# and 2# were 45.2% and 40.3%, respectively, which suggested that GBM cells’ proliferation ability decreased, and the growth ability of tumor cells was depressed remarkably.2. Invasion of U251 cell line transfected with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.3. Invasion of U251 cell line transfected with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.4. Phospholytion level of Jnk and p-38 MAPK proteins but not ERK 1/2 proteins were inhibited after transfection with lentivirus-mediated ShRNA targeting HMGA2 was significantly inhibited.Conclusions1. HMGA2 playd an important role in the process of GBM cells’ survival and proliferation indeed. After HMGA2 silenced, invasion and migration ability of tumor cell as depressed remarkably.2. The RNA interference technology would be important in the application value of pituitary tumor gene therapy in the future. HMGA2 could probably be considered as an effective gene target for pituitary tumor gene therapy.3. Jnk and p38 MAPK but not ERK signaling pathways might participate in the proliferation, migration and invasion rgulated by HMGA2 genes.Part 4 Exploiting Inhibitory Effects of the lentivirus-mediated RNAi targeting HMGA2 gene in GBM model of nude mice by a novel combination:microbubbles and ultrasound exposureObjectiveTo validate the inhibitory effects of the lentivirus-mediated RNAi targeting HMGA2 gene in Ovarian Cancer in vivo and to exploit the effectiveness of the transfection mediated by liposome microbubbles and ultrasound exposureMethodsmethods:24 Balb/c nude mice were subcutaneously inoculated human U251 GBM cells. The mice were randomly divided into 4 groups when transplantation tumor was established:UM+ShRNA (microbubbles+ShRNA+ultrasound) group, LV group (ShRNA), MB group (microbubble+MB, no ShRNA)and NC group (blank control, natural saline), with 6 mice each group. For MB+ShRNA group,400μL mixture of microbubble/ShRNA/(with 2.5ug ShRNA) was inject into the center of the tumor of every mice plus ultrasound exposure, the parameter was set with a frequency 2.5 MHz and the intensity 2.0w/cm2.The exposure was given 30 seconds twice with 30 seconds interval. For LV group,400μL of lentivirus solution was injected into the center of the tumor of every mice. For MB group,400μL of microbubble solution was injected into the center of the tumor of every mice with the same ultrasound exposure as the UM+ShRNA group. For BC group,400μL natural saline was inject into the center of the tumor of every mice, no ultrasound. The tumor volume was calculated according to V= n ab2/6 (a=length, b=width) every three days from the beginning of treatment.15 days later, the nude mice were killed and the weight of the tumor was weighed.Results1. The volume of nude mice tumor in NC group, LV group, MB group and MBShRNA group were 298.7±26.93,303.3±31.44,234.7±17.90,181.7±24.61, respectively. The volume of the ShRNA +UM group was obviously statistically smaller than UM group and BC group (LSD method, p<0.01).2. The weight of nude mice tumor were 0.47±0.06g,0.49±0.08g,0.36±0.03g, 0.25±0.03g, respectively. The weight of the ShRNA+MB group was obviously statistically smaller than MB group and BC group (Tukey method, p<0.01).3. The dissection of the nude mice showed that:There is some thing similar to membrane surrounding the pale, smaller-sized tumor of ShRNA+MB group. The boundary between tumor and normal tissue was obvious; But in MB and NC group, the tumor was red and larger, abundant in blood supply and mingled with the normal tissue, cannot be easily dissected.Conclusion(1) We successfully set up the nude mice model of subcutaneously incubated GBM U251 cells.(2) We successfully disposed of the nude mice model of subcutaneously incubated U251 GBM with Optison(?) ultrasound contrast agent microbubbles and ultrasound.(3) We found that the lentivirus-mediated RNAi targeting HMGA2 gene greatly decrease the growth of the tumor of the U251 GBM cells in vivo with ultrasound exposure and microbubbles injection.