Low-dose radiation enhances survivin-mediated virotherapy against malignant glioma stem cells

Abstract

To improve the efficacy and selectivity of virotherapy for malignant glioma, we designed a strategy to amplify adenoviral replication in conjunction with radiotherapy using a radioinducible promoter. First, we compared the radiation-inducible activity of FLT-1, vascular endothelial growth factor, DR5, Cox2, and survivin. We then examined the capacity of the optimal promoter to modulate transgene expression followed by E1A activity in vitro and in vivo in a glioma stem cell model. In the presence of radiation, survivin mRNA activity increased 10-fold. Luciferase transgene expression was dose dependent and optimal at 2 Gy. A novel oncolytic adenovirus, CRAd-Survivin-pk7, showed significant toxicity and replication against a panel of passaged and primary CD133(+) glioma stem cells. On delivery of radiation, the toxicity associated with CRAd-Survivin-pk7 increased by 20% to 50% (P < 0.05). At the same time, the level of E1A activity increased 3- to 10-fold. In vivo, treatment of U373MG CD133(+) stem cells with CRAd-Survivin-pk7 and radiation significantly inhibited tumor growth (P < 0.05). At the same time, the level of E1A activity was 100-fold increased versus CRAd-Survivin-pk7 alone. Selected genes linked to radioinducible promoters whose expression can be regulated by ionizing radiation may improve the therapeutic ratio of virotherapy. In this study, we have identified a new radioinducible promoter, survivin, which greatly enhances the activity of an oncolytic adenovirus in the presence of low-dose radiotherapy.

Figures

Figure 1. Real-Time PCR detection of promoter activity after radiation exposure

(A) Five different human promoters: FLT-1, DR5, VEGF, Cox2, and Survivin were assessed for expression of mRNA in U118MG cells 2 hours after exposure to 2Gy radiation using quantitative RT-PCR. The levels of mRNA expression in treated and untreated cells was normalized using GAPDH expression and presented as a bar diagram. Experiments were performed three times in triplicates each. (B) Detection of luciferase expression using replication deficient adenovirus before and after radiation. Adenoviral constructs containing DR5, Survivin or VEGF promoter driving luciferase expression were used for infection of U118MG cells. AdVEGF-Luc and AdDR5-Luc demonstrated an increase of 5–7% in luciferase expression in response to radiation. AdSurvivin-Luc, however, showed an increase from 30% to 70% in response to radiation. (C) Dose–dependent transcriptional activity of survivin promoter measured by luciferase assay. U373MG cells were infected with adenoviral constructs (1000vp/cell) containing the luciferase gene either under control of survivin or CMV promoters. After virus infection (2 h), cells were exposed to 2, 4, 6, 8 or 12 Gy radiation and grown for an additional 24h. Total luciferase activity was measured and values were normalized to amounts of total protein. Experiments were performed twice in triplicates. Mean ± SD is presented. p values were < 0.05 compared to non-radiated control (*).

Figure 2. Cytotoxicity and replication of CRAd-S-pk7 in unsorted brain tumor samples

Brain tumor samples were infected with either AdWT or CRAd-S-pk7 and (A) were assayed for cytotoxicity 72 h after infection using CytoTox Cytotoxicity Assay kit or (B) E1A copy numbers were measured by qPCR. In Figure 2A, the results were normalized to untreated cells and are presented as percentage toxicity. Each experiment was performed in 6 replicates. In Figure 2B, E1A copy number is expressed per ng DNA. Measurements were performed in triplicates. Mean ± SD is presented. * p< 0.05.

Figure 3. Enrichment of CD133-positive cells after radiation in vitro

Patient samples GBM1, GBM2 and the cell line U373MG were exposed to 2Gy radiation and grown for 48h along with unexposed controls. CD133+ subpopulation was determined by flow cytometric analysis. Mean ± SD is presented. * p< 0.05.

Figure 4. Survivin induced killing of CD133- cells in vitro in response to radiation

Cells were infected with CRAd-S-pk7 or AdWT and 24h later exposed to 2Gy of radiation. They were then grown for a 24 hr more in complete media. (A) Cytotoxicity was assayed by LDH release and is presented as percent toxicity and normalized to untreated cells (Mean ± SD is presented, * p<0.05). (B) Viral replication was determined by E1A copy number using qPCR and presented as copies per ng total DNA. Experiments were performed twice in triplicates.

Figure 5. Survivin induced killing of CD133+ cells in vitro in response to radiation

(A) CD133+ enriched tumor cells were infected with either AdWT or CRAd-S-pk7, 24h prior to exposure to radiation (2Gy) followed by 24 hour growth in complete media. Cytotoxicity was measured by LDH assay. (Mean ± SD is presented, * p<0.05) Experiments were performed twice with 6 replicates per condition. (B) Viral replication was determined via E1A copy number measurement using qPCR and presented as copies per ng total DNA. Experiments were performed twice in triplicates.

Figure 6. In vivo tumor growth rate in nude mice

(A) CD133+ enriched U373MG cells were injected s.c. in the right hind leg for tumor formation. After 8 days, when the tumor reached 100mm3, mice were randomly divided in three groups: mock, AdWT or CRAd-S-pk7 treatment and were injected with respective virus at 1000vp/cell or PBS for mock. Twenty four hours later, each group was further divided in two: one group received single dose of radiation of 2Gy, the other group was the non-irradiated control. Tumor volumes were measured for 6 days and the median volumes ± SD were plotted. (B) Viral replication in tumors as measured by E1A qPCR at day six post radiation in each group.