Eric Maynard Davis

BAY 43-9006 is a kinase inhibitor that induces apoptosis in a variety of tumor cells. Here we report that treatment with BAY 43-9006 results in marked cytochrome c and AIF release into the cytosol, caspase-9, -8, -7, and -3 activation, and apoptosis in human leukemia cells (U937, Jurkat, and K562). Pronounced apoptosis was also observed in blasts from patients with acute myeloid leukemia. These events were accompanied by ERK1/2 inactivation and caspase-independent down-regulation of Mcl-1. Inducible expression of a constitutively active MEK1 construct did not prevent Mcl-1 down-regulation, suggesting that this event is not related to MEK/ERK pathway inactivation. Furthermore, BAY 43-9006 did not induce major changes in Mcl-1 mRNA levels monitored by real-time PCR or Mcl-1 promoter activity demonstrated by luciferase reporter assays, but it did enhance Mcl-1 down-regulation in actinomycin D-treated cells. Inhibition of protein synthesis by cycloheximide or proteasome function with MG132 and pulse-chase studies with [35S]methionine demonstrated that BAY 43-9006 did not diminish Mcl-1 protein stability, nor did it enhance Mcl-1 ubiquitination, but instead markedly attenuated Mcl-1 translation in association with the rapid and potent dephosphorylation of the eIF4E translation initiation factor. Finally, ectopic expression of Mcl-1 in leukemic cells markedly inhibited BAY 43-9006-mediated cytochrome c cytosolic release, caspase-9, -7, and -3 activation, as well as cell death, indicating that Mcl-1 operates upstream of cytochrome c release and caspase activation. Together, these findings demonstrate that BAY 43-9006 mediates cell death in human leukemia cells, at least in part, through down-regulation of Mcl-1 via inhibition of translation.

Sorafenib is a multikinase inhibitor that induces apoptosis in human leukemia and other malignant cells. Recently, we demonstrated that sorafenib diminishes Mcl-1 protein expression by inhibiting translation through a MEK1/2-ERK1/2 signaling-independent mechanism and that this phenomenon plays a key functional role in sorafenib-mediated lethality. Here, we report that inducible expression of constitutively active MEK1 fails to protect cells from sorafenib-mediated lethality, indicating that sorafenib-induced cell death is unrelated to MEK1/2-ERK1/2 pathway inactivation. Notably, treatment with sorafenib induced endoplasmic reticulum (ER) stress in human leukemia cells (U937) manifested by immediate cytosolic-calcium mobilization, GADD153 and GADD34 protein induction, PKR-like ER kinase (PERK) and eukaryotic initiation factor 2alpha (eIF2alpha) phosphorylation, XBP1 splicing, and a general reduction in protein synthesis as assessed by [35S]methionine incorporation. These events were accompanied by pronounced generation of reactive oxygen species through a mechanism dependent upon cytosolic-calcium mobilization and a significant decline in GRP78/Bip protein levels. Interestingly, enforced expression of IRE1alpha markedly reduced sorafenib-mediated apoptosis, whereas knockdown of IRE1alpha or XBP1, disruption of PERK activity, or inhibition of eIF2alpha phosphorylation enhanced sorafenib-mediated lethality. Finally, downregulation of caspase-2 or caspase-4 by small interfering RNA significantly diminished apoptosis induced by sorafenib. Together, these findings demonstrate that ER stress represents a central component of a MEK1/2-ERK1/2-independent cell death program triggered by sorafenib.

B246 Sorafenib, also known as BAY 43-9006, is a multi-kinase inhibitor that induces apoptosis in human leukemia and other malignant cells. Recently, we demonstrated that sorafenib diminishes Mcl-1 protein expression by inhibiting translation through a MEK1/2/ERK1/2 signaling-independent mechanism, and that this phenomenon plays a key functional role in sorafenib lethality (Rahmani et al., Journal of Biologic Chemistry 280:35217-27, 2005). Here, we report that inducible expression of constitutively active MEK1 fails to protect U937 human leukemia cells from sorafenib-mediated lethality, indicating that sorafenib-induced cell death is unrelated to MEK1/2/ERK1/2 pathway inactivation. Notably, treatment with sorafenib induced ER stress manifested by immediate cytosolic calcium mobilization, GADD153 and GADD34 protein induction, PERK and eIF2α phosphorylation, XBP1 splicing, and a general reduction in protein synthesis as assessed by [ 35 S] methionine incorporation. Notably, genetic disruption of PERK (e.g., transfection of cells with DN-PERK), but not PKR, HRI, or CGN2, attenuated eIF2α phosphorylation, identifying PERK as the primary eIF2α kinase in this setting. These events were accompanied by the pronounced generation of reactive oxygen species (ROS) through a mechanism dependent upon cytosolic calcium mobilization and a significant decline in GRP78/Bip protein levels. Interestingly, enforced expression of IRE1α or XBP1 splice variant (XBP1s) significantly reduced sorafenib-mediated apoptosis, whereas knockdown of IRE1α or XBP1, disruption of PERK activity, or inhibition of eIF2α phosphorylation enhanced sorafenib-mediated lethality. Such observations suggest that these events comprise a cytoprotective component of the sorafenib-induced ER stress response. Finally, downregulation of caspase-2 or caspase-4 by siRNA or shRNA respectively significantly diminished apoptosis induced by sorafenib in these cells, providing further support for the notion that the lethal actions of sorafenib involve induction ER stress. Together, these findings demonstrate that ER stress represents a central component of a MEK1/2/ERK1/2-independent cell death program triggered by sorafenib. They also raise the possibility that induction of ER stress by sorafenib in human leukemia cells may play a role in determining interactions between this kinase inhibitor and other targeted agents.