In Response: In their Letter, Kyritsis et al. discussed the interesting question of the contribution of up-regulation of the apoptosis-inducing tumor necrosis factor–related apoptosis inducing ligand (TRAIL) receptors, TRAIL-R1 (DR4) and TRAIL-R2 (DR5), to the sensitization of tumor cells to TRAIL-induced apoptosis by chemotherapeutic drugs. In glioma cell lines, the authors found that the level of TRAIL-R2 expression was not predictive for TRAIL sensitivity. Furthermore, they showed that inhibition of PKB/Akt sensitized cells to TRAIL-induced death (1).
This is in line with our data from primary glioma cells which displayed considerable TRAIL resistance despite TRAIL-R2 expression and proper TRAIL death-inducing signaling complex formation upon receptor triggering (2). This result pointed to mechanisms of TRAIL resistance in these tumor cells which were independent of TRAIL death receptor expression. As emphasized by Puduvalli et al., PKB/Akt, which is an important survival signal, has been shown to modulate cFLIP expression in different tumor cells (3) and, indeed, we found that down-regulation of cFLIP alone was sufficient to sensitize tumor cells to TRAIL (2, 4). Interestingly, proteasome inhibition by bortezomib led to the down-regulation of cFLIPL in primary glioma cells (2) and in other tumor entities (4) and, consequently, to TRAIL sensitization (5). Because proteasome inhibitors deregulate the degradation and/or expression of quite a number of intracellular apoptosis regulators, numerous proteins have been found to play a role as effectors of bortezomib-mediated TRAIL sensitization, e.g., enhanced expression of proapoptotic BH-3–only proteins (6) or reduced levels of prosurvival Bcl-2-family members (7). Accordingly, apart from decreased cFLIP expression, we found elevated Bax and/or Bak expression in sensitized primary glioma cells, and blocking the mitochondrial pathway of apoptosis induction by Bcl-2 rendered bortezomib-pretreated glioma cells more TRAIL-resistant (2).
As mentioned by Kyritsis et al., the role of TRAIL-R1/R2 up-regulation in TRAIL sensitization by several chemotherapeutic drugs is highly debated, and in fact, many studies do not go beyond the mere correlation between receptor up-regulation and TRAIL sensitization. To study the contribution of up-regulated TRAIL receptors, the activation of these additional death receptors must be blocked specifically. By carefully titrating TRAIL-R2–directed short interfering RNA, Inoue et al. could keep TRAIL-R2 expression of depsipeptide-sensitized chronic lymphocytic leukemia cells constant, which could, however, still be sensitized for TRAIL-induced apoptosis, showing that up-regulation of TRAIL-R2 is dispensable for TRAIL sensitization by histone deacetylase inhibitors (8). In our “wash kill” experiment, we triggered TRAIL receptors which were already present on the cell surface and sensitized them with bortezomib afterwards. Although up-regulated TRAIL-R1/R2 could not have been activated under these conditions, bortezomib still sensitized for TRAIL-induced apoptosis. Nevertheless, re-addition of TRAIL together with bortezomib could further potentiate TRAIL sensitization showing the partial contribution of up-regulated TRAIL death receptors to bortezomib-mediated TRAIL sensitization in primary glioma cells (2). Furthermore, we inhibited the trans-Golgi transport of TRAIL-R1 or TRAIL-R2 by Brefeldin A in hepatoma cells, which blocked their bortezomib-induced up-regulation but not the activation of preexisting TRAIL death receptors at the cell surface. Again, tumor cells could still be sensitized for TRAIL-induced apoptosis. In summary, our studies show that bortezomib releases a decisive blockade of the TRAIL signaling pathway which is independent of TRAIL-R1/R2 up-regulation (2, 4, 9). This blockade could either occur at the death-inducing signaling complex due to high expression of cFLIP (10) or further downstream in the apoptosis signaling pathway (e.g., by Bax deficiency or high expression of Bcl-2 or XIAP; refs. 11, 12).
Taken together, whereas overcoming the intracellular TRAIL blockade is indispensable for TRAIL sensitization, the chemotherapy-induced increase in TRAIL death receptors is not necessary for but could contribute to TRAIL sensitization (2, 4, 8, 9, 13). In this context, it would be quite interesting to test whether a cell line with a moderate increase in TRAIL-R2 surface expression, comparable to the level achieved by chemotherapeutic drug treatment, e.g., by carefully fine-tuned stable overexpression, would be sufficient to confer TRAIL sensitivity to primary glioma cells.