Primary Glioblastoma Cell Cultures Research Articles

Phosphatidylinositol 3-Kinase Inhibition Broadly Sensitizes Glioblastoma Cells to Death Receptor– and Drug-Induced Apoptosis

The aberrant activity of the phosphatidylinositol 3-kinase (PI3K) pathway has been reported to correlate with adverse clinical outcome in human glioblastoma in vivo. However, the question of how this survival network can be successfully targeted to restore the sensitivity of glioblastoma to apoptosis induction has not yet been answered. Here, we report that inhibition of PI3K by LY294002 broadly sensitizes wild-type and mutant PTEN glioblastoma cells to both death receptor- and chemotherapy-induced apoptosis, whereas mammalian target of rapamycin (mTOR) inhibition is not sufficient to restore apoptosis sensitivity. LY294002 significantly enhances apoptosis triggered by tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), agonistic anti-CD95 antibodies, or several anticancer drugs (i.e., doxorubicin, etoposide, and vincristine) in a highly synergistic manner. In addition, LY294002 cooperates with TRAIL or doxorubicin to suppress colony formation, thus also showing a strong effect on long-term survival. Similarly, genetic knockdown of PI3K subunits p110alpha and/or p110beta by RNA interference (RNAi) primes glioblastoma cells for TRAIL- or doxorubicin-mediated apoptosis. In contrast to PI3K inhibition, pharmacologic or genetic blockade of mTOR by RAD001 (everolimus), rapamycin, or RNAi fails to enhance TRAIL- or doxorubicin-induced apoptosis. Analysis of apoptosis pathways reveals that PI3K inhibition acts in concert with TRAIL or doxorubicin to trigger mitochondrial membrane permeabilization, caspase activation, and caspase-dependent apoptosis, which are abolished by the caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone. Most importantly, PI3K inhibition by LY294002 sensitizes primary cultured glioblastoma cells obtained from surgical specimens to TRAIL- or chemotherapy-induced cell death. By showing that PI3K inhibition broadly primes glioblastoma cells for apoptosis, our findings provide the rationale for using PI3K inhibitors in combination regimens to enhance TRAIL- or chemotherapy-induced apoptosis in glioblastoma.

Migratory neural stem cells for improved thymidine kinase-based gene therapy of malignant gliomas

Gene therapy of glioma based on viral delivery of herpes simplex virus type I thymidine kinase (HSV-TK) has failed in the clinic because of low transduction efficacy. To circumvent this problem, this study evaluated highly migratory HSV-TK-transduced neural stem cells (NSC) for their ability to kill untransduced glioma cells by a gap junction-mediated bystander effect. The admixture of HSV-TK-transduced NSC to U87MG and LN-18 human malignant glioma cell lines at ratios of 1:10 or 1:1 eliminated more than 50% or 90% of glioma cells in the presence of ganciclovir (25 μM). Glioma cell cytotoxicity required cell–cell contact. Similarly, tumor cell cytotoxicity was observed in two of three primary glioblastoma cell cultures, and the presence of this bystander effect correlated with the expression of connexin 43 in the untransduced glioma target cells. In conclusion, we delineate a role for migratory HSV-transfected NSC to eliminate glioma cells purely by means of the bystander effect.

Induction of a CD4+ T regulatory type 1 response by cyclooxygenase-2-overexpressing glioma.

PGE(2), synthesized by cyclooxygenase-2 (COX-2)-overexpressing tumor, is known to contribute to cellular immune suppression in cancer patients, but the mechanism remains unclear. We report the mechanism of a CD4(+) T regulatory type 1 (Tr1) induction by CD11c(+) mature dendritic cells (DCs) that phagocytose allogeneic and autologous COX-2-overexpressing glioma. A human glioma cell line, U-87MG, and primary cultured glioblastoma cells (MG-377) overexpressed COX-2. We did not detect IL-10Ralpha expression in these gliomas, and rIL-10 did not suppress their COX-2 expression. Exposure to COX-2-overexpressing glioma induced mature DCs to overexpress IL-10 and decreased IL-12p70 production. These DCs induced a Tr1 response, which is characterized by robust secretion of IL-10 and TGF-beta with negligible IL-4 secretion by CD4(+) T cells, and an inhibitory effect on admixed lymphocytes. Peripheral CD4(+) T cell populations isolated from an MG-377 patient also predominantly demonstrated a Tr1 response against MG-377 cells. Selective COX-2 inhibition in COX-2-overexpressing gliomas at the time of phagocytic uptake by DCs abrogated this regulatory response and instead elicited Th1 activity. COX-2 stable transfectants in LN-18 (LN-18-COX2) also induced a Tr1 response. The effect of a COX-2 inhibition in LN-18-COX2 is reversible after administration of PGE(2). Taken together, robust levels of PGE(2) from COX-2-overexpressing glioma, which is unresponsive to IL-10 within the local microenvironment, may cause DCs to secrete high levels of IL-10. These results indicate that COX-2-overexpressing tumors induce a Tr1 response, which is mediated by tumor-exposed, IL-10-enhanced DCs.

Soluble TNF-alpha receptors are constitutively shed and downregulate adhesion molecule expression in malignant gliomas.

The regulation of adhesion molecule expression in malignant gliomas by tumor necrosis factor-alpha (TNF) and soluble TNF receptors (TNFR) was examined in the malignant glioma cell line A-172 and in 2 primary glioblastoma cell cultures (LA-492 and LA-567). A-172 cells expressed only the p55 TNF receptor transcripts and protein. The 2 primary cell cultures expressed both the p55 and p75 TNF receptors. In A-172 cells and in 1 of 2 primary glioma cell cultures, TNF upregulated the expression of ICAM-1 and VCAM-1, A-172 and both primary glioma cultures also shed their TNF receptors in the absence of activation by stimulating agents. Soluble p55 (sp55) receptors, but not soluble p75 (sp75) receptors, were found to reduce the TNF induced VCAM-1 and ICAM-1 expression in both the glioma cell line and the primary cell culture. Immunostaining of malignant glioma sections confirmed the presence of soluble TNFR and adhesion molecule expression in glioma cells in situ. These data suggest that soluble TNF receptors may play a role in the mechanism by which malignant gliomas downregulate the effects of infiltrating immune-competent cells.

Protein and messenger RNA expression of connexin43 in astrocytomas: implications in brain tumor gene therapy.

The expression of connexin43, the primary gap-junction constituent of glial cells, was evaluated at the messenger RNA and protein levels in different grades of astrocytoma to investigate the relevance of gap junctions in herpes simplex virus-thymidine kinase (HSV-tk)-mediated gene therapy of brain tumors. Transduction of the retroviral-mediated HSV-tk gene into tumor cells with subsequent administration of ganciclovir has recently been used as an experimental therapeutic strategy for treatment of brain tumors. One aspect of this approach is the bystander effect, which augments the efficacy of this therapeutic approach. Glioblastoma cells with minimum levels of connexin43 protein were transfected with a connexin43 complementary DNA. These cells manifested a marked increase in the in vitro bystander effect, supporting the contention that the in vitro bystander effect is a consequence of metabolic cooperation between cells mediated by gap junctions. To assess relative levels of gap-junction protein expression in the relevant tumor type, we examined primary astrocytomas, primary astrocytoma cell cultures, and glioblastoma cell lines. Although most astrocytoma tumor samples expressed connexin43, they differed in the level of expression, with the greatest variation exhibited in high-grade astrocytomas. Primary glioblastoma cell cultures and established glioblastoma cell lines also displayed some variability in connexin43 levels. In aggregate, our results anticipate that glioblastomas will have a varied bystander effect during HSV-tk gene therapy depending on the level of connexin43 expression.