Tumor necrosis factor-related apoptosis-inducing ligand-mediated apoptosis in established and primary glioma cell lines.

OBJECT: Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a member of the TNF cytokine family, which mediates programmed cell death (apoptosis) selectively in tumor cells. The selective tumoricidal activity of TRAIL is believed to be modulated by agonistic (DR4 and DR5) and antagonistic receptors (DcR1 and DcR2), which appear to compete for ligand binding. Because TRAIL is expressed in a wide range of tissues, including brain, kidney, and spleen, and seems consistently to induce cell death in tumor cells, the cytokine has been identified as a promising approach for selectively inducing tumor cell death. In this study, the authors examine the importance of TRAIL's receptors in both its selectivity for tumor cells and its ability to induce apoptosis.
METHODS: The authors first examined sensitivity to TRAIL and expression of TRAIL receptors in four established and four primary cultured glioma cell lines by using viability and fluorescent apoptosis assays. They then evaluated DR5 expression and JNK, caspase 3, and caspase 7 activation by conducting immunoblot analyses. Reverse transcriptase-polymerase chain reaction (RTPCR) was performed to study expression of DR4, DR5, DcR1, and DcR2. The DR5 transcripts from one TRAIL-sensitive, one partially TRAIL-resistant, and one TRAIL-resistant cell line were subsequently sequenced. The expression of TRAIL receptors in normal and glial brain tumor pathological specimens were then compared using immunohistochemistry. Finally, to study the direct effects of DR5 on glioma cells, the authors conducted transient and stable transfections of the full-length DR5 transcript into glioma cells with and without preestablished overexpression of the antiapoptotic gene bcl-2. The established glioma cell lines T98G and U87MG, and all primary cell lines, were apoptotic at greater than or equal to 100 ng/ml TRAIL. The A172 cells, by contrast, were susceptible only with cycloheximide, whereas U373MG cells were not susceptible to TRAIL. The JNK, caspase 3, and caspase 7 activity evaluated after treatment with TRAIL showed that TRAIL-sensitive cell lines exhibited downstream caspase activation, whereas TRAIL-resistant cells did not. The DR5 sequences in T98G, A172, and U373MG cell lines were identical to published sequences despite these differences in sensitivity to TRAIL. The RT-PCR performed on extracts from the eight glioma cell lines showed that all expressed DR5. Immunohistochemistry revealed ubiquitous expression of DR5 in glioma specimens, with an associated lack of decoy receptor expression. Normal brain specimens, by contrast, stained positive for both DR5 and DcR1. Overexpression of DR5 under both transfection conditions resulted in cell death in all three cell lines. The previously seen resistance of U373 cells to TRAIL was not observed. Apoptotic cell death was confirmed using DNA fragmentation in T98G cell lines and fluorescent miscroscopy in all cell lines. The T98G cells stably transfected with bcl-2 before DR5 overexpression were protected from cell death.
CONCLUSIONS: The authors conclude that DR5 represents a promising new approach to directly activating the intrinsic caspase pathway in glioma cells. The fact that TRAIL-resistant gliomas do not express decoy receptors suggests a mechanism of resistance unique from that proposed for normal tissues. The overexpression of DR5 induced apoptotic cell death in glioma cells without TRAIL and was able to overcome the resistance to TRAIL demonstrated in U373 cells. The Bcl-2 protects cells from DR5 by acting downstream of the receptor, most likely at the level of caspase activation.