Andrew S. Kraft

The proteasome inhibitor PS-341 overcomes TRAIL resistance in Bax and caspase 9-negative or Bcl-xL overexpressing cells.

We demonstrate that PS-341, a small molecule inhibitor of the proteasome, markedly sensitizes resistant prostate, colon, and bladder cancer cells to TNF-like apoptosis-inducing ligand (TRAIL)-induced apoptosis irrespective of Bcl-xL overexpression. PS-341 treatment by itself does not affect the levels of Bax, Bak, caspases 3 and 8, c-Flip or FADD, but elevates levels of TRAIL receptors DR4 and DR5. This increase in receptor protein levels is associated with the ubiquitination of the DR5 protein. When PS-341 is combined with TRAIL, the levels of activated caspase 8 and cleaved Bid are substantially increased. In Bax-negative TRAIL-resistant HC-4 colon cancer cells, the combination of PS-341 and TRAIL overcomes the block to activation of the mitochondrial pathway and causes SMAC and cytochrome c release followed by apoptosis. Similarly, murine embryonic fibroblasts lacking Bax undergo apoptosis when exposed to the combination of PS-341 and TRAIL; however, fibroblasts lacking Bak are significantly resistant. Taken together, these findings indicate that PS-341 enhances TRAIL-induced apoptosis by increasing the cleavage of caspase 8, causing Bak-dependent release of mitochondrial proapoptotic proteins.

The PIM-1 serine kinase prolongs survival and inhibits apoptosis-related mitochondrial dysfunction in part through a bcl-2-dependent pathway

We have examined potential mechanisms by which the Pim-1 kinase acts as a hematopoietic cell survival factor. Enforced expression of the wild type 33u2009kd (FD/hpim33) and 44u2009kd (FD/mpim44) Pim-1 proteins in murine factor-dependent FDCP1 cells prolonged survival after withdrawal of IL-3, while expression of a dominant negative Pim-1 protein (FD/pimNT81) shortened survival. Following removal of IL-3 FDCP1 cells exhibited loss of mitochondrial transmembrane potential and production of reactive oxygen species, as determined by flow cytometry analysis. The wild type Pim-1 proteins decreased these changes while the dominant negative protein enhanced mitochondrial dysfunction. The antiapoptotic activity of the kinases could not be attributed to modulation of glutathione, catalase, or superoxide dismutase activities. Both the FD/hpim33 and FD/mpim44 cells maintained expression of bcl-2 mRNA following cytokine removal, while a substantial decrease was seen in FD/neo cells. To modulate Bcl-2 protein levels, a bcl-2 antisense RNA construct was coexpressed with the wild type pim-1 cDNAs. FD/hpim33 cells with low cellular Bcl-2 protein levels had shortened cytokine-independent survival compared with FD/hpim33 clones with high Bcl-2 expression. However survival of FD/mpim44 cells after IL-3 withdrawal was substantially independent of cellular Bcl-2 protein levels. The 33u2009kd protein delayed, and the 44u2009kd protein completely prevented enhanced cell death associated with enforced expression of human Bax protein however. Our results suggest that the 33u2009kd Pim-1 kinase may enhance cell survival through cooperation with and regulation of bcl-2. In addition the 44u2009kd kinase may regulate the expression or activity of other pro- and anti-apoptotic members of the bcl-2 family.

Oncogenic Ras Sensitizes Normal Human Cells to Tumor Necrosis Factor-α-Related Apoptosis-Inducing Ligand-Induced Apoptosis

Tumor necrosis factor-α-related apoptosis-inducing ligand (TRAIL) is a cytotoxic cytokine that induces apoptosis in tumor cells but rarely kills normal ones. To determine how normal human cells acquire TRAIL-sensitive phenotype during the process of malignant transformation, we used an experimental system that allows for controlled conversion of human cells from normal to cancerous by introduction of several genes. Human embryonic kidney cells and foreskin fibroblasts were first immortalized by combination of the early region of simian virus 40 and telomerase and then were transformed with oncogenic Ras. Both normal and immortalized cells were resistant to TRAIL-induced apoptosis, whereas Ras-transformed cells were susceptible. Ras transformation enhanced TRAIL-induced activation of caspase 8 by increasing its recruitment to TRAIL receptors. The proapoptotic effects of Ras could be reversed by mutations in its effector loop or by inhibitors of either farnesyl transferase or mitogen-activated protein kinase kinase. The expression of constitutively activated mitogen-activated protein kinase kinase 1 enhanced caspase 8 recruitment and sensitized immortalized human embryonic kidney cells to TRAIL-induced death. These results indicate that in normal human cells the TRAIL-induced apoptotic signal is blocked at the level of caspase 8 recruitment and that this block can be eliminated by Ras transformation, involving activation of the mitogen-activated protein kinase pathway.

Adenovirus-mediated expression of Fas ligand induces apoptosis of human prostate cancer cells

Several laboratories have reported on the apoptotic potentials of human prostate cancer (PC) cell lines in response to crosslinking of Fas (CD95/APO-1) with agonistic anti-Fas antibodies. We have re-evaluated the apoptotic potentials of seven human PC cell lines using the natural Fas ligand (FasL) in place of agonistic antibody. First, PC cell lines were tested in a standard cytotoxicity assay with a transfected cell line that stably expresses human FasL. Next, we developed an adenoviral expression system employing 293 cells that stably express crmA, a poxvirus inhibitor of apoptosis, to analyze the effects of FasL when expressed internally by the PC cell lines. Our data suggest that the apoptotic potentials of these cell lines were greatly underestimated in previous studies utilizing agonistic anti-Fas antibodies. Lastly, adenoviral-mediated expression of FasL prevented growth and induced regression of two human PC cell lines in immunodeficient mice. These preliminary in vivo results suggest a potential use for adenovirus encoding FasL as a gene therapy for PC.

The human brm protein is cleaved during apoptosis: The role of cathepsin G

The human brm (hbrm) protein (homologue of the Drosophila melanogaster brahma and Saccharomyces cervisiae SNF-2 proteins) is part of a polypeptide complex believed to regulate chromatin conformation. We have shown that the hbrm protein is cleaved in NB4 leukemic cells after induction of apoptosis by UV-irradiation, DNA damaging agents, or staurosporine. Because hbrm is found only in the nucleus, we have investigated the nature of the proteases that may regulate the degradation of this protein during apoptosis. In an in vitro assay, the hbrm protein could not be cleaved by caspase-3, -7, or -6, the “effector” caspases generally believed to carry out the cleavage of nuclear protein substrates. In contrast, we find that cathepsin G, a granule enzyme found in NB4 cells, cleaves hbrm in a pattern similar to that observed in vivo during apoptosis. In addition, a peptide inhibitor of cathepsin G blocks hbrm cleavage during apoptosis but does not block activation of caspases or cleavage of the nuclear protein polyADP ribose polymerase (PARP). Although localized in granules and in the Golgi complex in untreated cells, cathepsin G becomes diffusely distributed during apoptosis. Cleavage by cathepsin G removes a 20-kDa fragment containing a bromodomain from the carboxyl terminus of hbrm. This cleavage disrupts the association between hbrm and the nuclear matrix; the 160-kDa hbrm cleavage fragment is less tightly associated with the nuclear matrix than full-length hbrm.

Treatment of human prostate cancer cells with dolastatin 10, a peptide isolated from a marine shell-less mollusc

Dolastatin 10 is an anticancer peptide isolated from the sea hare, Dolabela auricularia, which is currently in phase I trials.

Regulation of myeloid cell growth by distinct effectors of Ras

To examine the biochemical pathways by which activated Ha-Ras(G12V) (Ha-RasV12) induces factor-independent growth of myeloid cells, Ha-Ras effector loop mutations, including Y40C, T35S, and E37G, were analysed in a mouse factor-dependent myeloid cell line, WT19. Expression of a single effector loop mutant, Ha-Ras(G12V, Y40C) (Ha-RasV12C40), inhibited factor-withdrawal apoptosis, suggesting that activation of the phosphatidylinositol 3′-kinase (Pl3K) pathway is essential to prevent cell death. Neither Ha-Ras (G12V, T35S) (Ha-RasV12S35), which activates the Raf1 signaling pathway, nor Ha-Ras(G12V, E37G) (Ha-RasV12G37), which stimulates the RalGDS pathway, did not have significant effects on factor-withdrawal apoptosis of myeloid cells. Although Ha-RasV12C40 inhibited apoptosis, it did not stimulate entry into the cell cycle. Cell lines containing the combination of Ha-RasV12G37 and Ha-RasV12C40 were capable of factor-independent cell growth, while expression of the other combinations of the Ha-Ras effector mutants were not. The combined expression of Bcl-2 and Ha-RasV12G37 was not sufficient to stimulate factor independent growth, suggesting that Ha-RasV12C40 activates additional signals, besides blocking apoptosis, which are critical for factor-independent growth of myeloid cells. In factor-starved myeloid cells, inducible expression of Ha-RasV12G37 results in decreased level of p27Kip1 protein, a cyclin-dependent kinase inhibitor (CKI). These data suggest that the factor-independent growth of myeloid cells requires the activation of at least two pathways, one inhibiting factor-withdrawal apoptosis, and another causing cell cycle progression.

Ligation of membrane IgM stimulates a novel c-Jun amino-terminal domain kinase activity in Daudi human B cells

Stress-activated protein kinases (SAPK; also known as JNK for c-Jun N-terminal kinase) phosphorylate Ser63 and Ser73 in the amino-terminus of the c-Jun protein and potentiate its transcriptional activity. We have analysed phosphorylation of GST fusion proteins containing the c-Jun N-terminal domain by lysates of Daudi human B lymphoblastoid cells stimulated with medium or anti-IgM. Crosslinking membrane IgM (mIgM) results in an increase in phosphorylation of GST-c-Jun (5-89) in an antibody dose-dependent manner. The kinase activity specifically phosphorylates the c-Jun N-terminal domain since it does not phosphorylate GST or GST-JunB. The activity preferentially phosphorylates the substrate that contains the sites for in vivo phosphorylation by SAPK/JNK and requires the delta domain of c-Jun, which is also required for SAPK/JNK activity. However, the c-Jun N-terminal kinase activity induced by mIgM ligation is not precipitatable with anti-SAPK/JNK antibodies. In addition, unlike SAPK/JNKs, the mIgM-dependent c-Jun N-terminal kinase activity is not detectable in assays for renaturable kinase activity (in-gel assay) or in assays that test activities that bind to c-Jun (solid-phase assay). The increased phosphorylation of c-Jun N-terminal domain in response to mIgM ligation is unlikely to be due to mIgM-activated ERKs as it was not suppressed by a selective MEK inhibitor. Thus, the mIgM-induced activity is distinct from the known SAPK/JNKs and may represent a novel mechanism for c-Jun phosphorylation in response to mIgM engagement in human B cells.