WMC-79, a potent agent against colon cancers, induces apoptosis through a p53-dependent pathway.

WMC-79 is a synthetic agent with potent activity against colon and hematopoietic tumors. In vitro, the agent is most potent against colon cancer cells that carry the wild-type p53 tumor suppressor gene (HCT-116 and RKO cells: GI50<1 nmol/L, LC50 approximately 40 nmol/L). Growth arrest of HCT-116 and RKO cells occurs at the G1 and G2-M check points at sublethal concentrations (10 nmol/L) but the entire cell population was killed at 100 nmol/L. WMC-79 is localized to the nucleus where it binds to DNA. We hypothesized that WMC-79 binding to DNA is recognized as an unrepairable damage in the tumor cells, which results in p53 activation. This triggers transcriptional up-regulation of p53-dependent genes involved in replication, cell cycle progression, growth arrest, and apoptosis as evidenced by DNA microarrays. The change in the transcriptional profile of HCT-116 cells is followed by a change in the levels of cell cycle regulatory proteins and apoptosis. The recruitment of the p53-dependent apoptosis pathway was suggested by the up-regulation of p53, p21, Bax, DR-4, DR-5, and p53 phosphorylated on Ser15; down-regulation of Bcl-2; and activation of caspase-8, -9, -7, and -3 in cells treated with 100 nmol/L WMC-79. Apoptosis was also evident from the flow cytometric studies of drug-treated HCT-116 cells as well as from the appearance of nuclear fragmentation. However, whereas this pathway is important in wild-type p53 colon tumors, other pathways are also in operation because colon cancer cell lines in which the p53 gene is mutated are also affected by higher concentrations of WMC-79.

Caspase-mediated apoptosis and caspase-independent cell death induced by irofulven in prostate cancer cells.

Irofulven (hydroxymethylacylfulvene) is a novel antitumor drug, which acts by alkylating cellular macromolecular targets. The drug is a potent inducer of apoptosis in various types of tumor cells, whereas it is nonapoptotic in normal cells. This study defined molecular responses to irofulven involving mitochondrial dysfunction and leading to death of prostate tumor LNCaP-Pro5 cells. Irofulven caused early (2-5 hours) translocation of the proapoptotic Bax from cytosol to mitochondria followed by the dissipation of mitochondrial membrane potential and cytochrome c release at 4 to 12 hours. These effects preceded caspase activation and during the first 6 hours were not affected by caspase inhibitors. Processing of caspase-9 initiated the caspase cascade at approximately 6 hours and progressed over time. The activation of the caspase cascade provided a positive feedback loop that enhanced Bcl-2-independent translocation and cytochrome c release. General and specific caspase inhibitors abrogated irofulven-induced apoptotic DNA fragmentation with the following order of potency: pan-caspase > or = caspase-9 > caspase-8/6 > caspase-2 > caspase-3/7 > caspase-1/4. Abrogation of caspase-mediated DNA fragmentation failed to salvage irofulven-treated cells from growth inhibition and loss of viability, demonstrating a substantial contribution of a caspase-independent cell death. Monobromobimane, an inhibitor of alternative caspase-independent apoptotic pathway that is mediated by mitochondrial permeability transition, antagonized both apoptosis, measured as phosphatidylserine externalization, and cytotoxicity of irofulven. Collectively, the results indicate that irofulven-induced signaling is integrated at the level of mitochondrial dysfunction. The induction of both caspase-dependent and caspase-independent death pathways is consistent with pleiotropic effects of irofulven, which include targeting of cellular DNA and proteins.

Enhanced topoisomerase II targeting by annamycin and related 4-demethoxy anthracycline analogues.

Targeting topoisomerase II (topo II) is regarded as an important component of the pleiotropic mechanism of action of anthracycline drugs. Here, we show that 4-demethoxy analogues of doxorubicin, including annamycin, exhibit a greater ability to trap topo II cleavage complexes than doxorubicin and some other 4-methoxy analogues. In leukemic CEM cells with wild-type topo II, annamycin induced substantial levels of topo II-mediated DNA-protein cross-links (15-37% of total DNA for 0.5-50 micromol/L drug), whereas doxorubicin-induced DNA-protein cross-links were marginal (0-4%). In CEM/VM-1 cells that harbor mutated, drug-resistant topo II, both 4-methoxy and 4-demethoxy drugs produced marginal DNA-protein cross-links. Annamycin, but not doxorubicin, formed topo II-mediated DNA-protein cross-links also in isolated CEM nuclei. In disparity with the unequal DNA-protein cross-link induction, both drugs induced comparable levels of DNA strand breaks in CEM cells. Compared with CEM, drug cytotoxicity against CEM/VM-1 cells was reduced 10.5- to 13.8-fold for 4-demethoxy analogues but only 3.8- to 5.5-fold for 4-methoxy drugs. Hence, growth inhibition by 4-demethoxy analogues seems more dependent on the presence of wild-type topo II. The enhanced topo II targeting by 4-demethoxy analogues was accompanied by a profound induction of apoptotic DNA fragmentation in leukemic CEM cells. Normal WI-38 fibroblasts, however, were markedly more resistant to annamycin-induced DNA-protein cross-links, apoptosis, and growth inhibition. The enhanced topo II targeting by 4-demethoxy doxorubicin analogues underscores the mechanistic diversity of anthracycline drugs. This diversity needs to be recognized as a factor in responses to drugs such as annamycin and doxorubicin.

Cell cycle effects and induction of premitotic apoptosis by irofulven in synchronized cancer cells.

Unlike postmitotic cell death, direct premitotic apoptosis diminishes the risk of clonal selection and allows for the elimination of slowly growing cancer cells. This study characterized the ability to induce premitotic apoptosis by irofulven (hydroxymethylacylfulvene), a novel alkylating drug which targets cellular DNA and proteins. Irofulven effects were examined in HeLa-derived BH2 cancer cells with conditional overexpression of antiapoptotic Bcl-2. Cells were synchronized in either early S or in G(1). Following 12 h exposure to irofulven, cells that were originally in early S accumulated in late S or remained in early S phase (at 0.5 and 2.5 muM drug, respectively). Drug treatment of cells in the G(1) cohort prevented their entry into the S phase. Significant apoptosis was detected based on the appearance of sub-G(1) particles and cells with DNA strand breaks in both G(1) and S cohorts. Apoptotic cells were mostly recruited from the G(1)/S border ("G(1)" cohort) and from the S phase ("early S" cohort). All the cell cycle and apoptotic effects were only marginally affected by Bcl-2 overexpression. Similar results were obtained with irofulven-treated synchronized cultures of leukemic CEM cells. Collectively, these observations indicate that irofulven-treated cells become committed to death early. Neither active DNA replication nor traverse through mitosis are necessary for irofulven-induced cell death. The ability to promote direct premitotic apoptosis is likely to play a role in the consistently potent apoptotic effects of irofulven and its ability to cause tumor regression in vivo.

Apoptosis induction by the dual-action DNA- and protein-reactive antitumor drug irofulven is largely Bcl-2-independent.

The overexpression of Bcl-2 is implicated in the resistance of cancer cells to apoptosis. This study explored the potential of irofulven (hydroxymethylacylfulvene, HMAF, MGI 114, NSC 683863), a novel DNA- and protein-reactive anticancer drug, to overcome the anti-apoptotic properties of Bcl-2 in HeLa cells with controlled Bcl-2 overexpression. Irofulven treatment resulted in rapid (12hr) dissipation of the mitochondrial membrane potential, phosphatidylserine externalization, and apoptotic DNA fragmentation, with progressive changes after 24hr. Bcl-2 overexpression caused marginal or partial inhibition of these effects after treatment times ranging from 12 to 48hr. Both Bcl-2-dependent and -independent responses to irofulven were abrogated by a broad-spectrum caspase inhibitor. Despite the somewhat decreased apoptotic indices, cell growth inhibition by irofulven was unaffected by Bcl-2 status. In comparison, Bcl-2 overexpression drastically reduced apoptotic DNA fragmentation by etoposide, acting via topoisomerase II-mediated DNA damage, but had no effect on apoptotic DNA fragmentation by helenalin A, which reacts with proteins but not DNA. Irofulven retains its pro-apoptotic and growth inhibitory potential in cell lines that have naturally high Bcl-2 expression. Collectively, the results implicate multiple mechanisms of apoptosis induction by irofulven, which may differ in time course and Bcl-2 dependence. It is possible that the sustained ability of irofulven to induce profound apoptosis and to block cell growth despite Bcl-2 overexpression may be related to its dual reactivity with both DNA and proteins.

Preferential targeting of apoptosis in tumor versus normal cells.

Elimination of cancer cells by early apoptosis is preferred over other forms of cell growth inhibition. Apoptosis directly leads to tumor regression and reduces risks of selecting more aggressive and/or drug-resistant phenotypes that are often responsible for tumor regrowth and treatment failure. Although DNA damage by anticancer drugs is commonly recognized as an apoptotic stimulus, there is enormous variability in the magnitude and timing of such effects. Especially potent and rapid apoptosis seems to be a hallmark of various alkylating anticancer drugs that are regarded as DNA-reactive agents but are observed to react mainly with cellular proteins. Our studies with such dual-action drugs (irofulven, oxaliplatin) suggest that not only DNA damage, but also protein damage, contributes to apoptosis induction. DNA damage is well known to initiate death-signaling pathways leading to mitochondrial dysfunction. Protein damage, in turn, can distort cell redox homeostasis, which facilitates apoptosis execution. Such dual effects can be particularly lethal to tumor cells, which tend to function under pro-oxidative conditions. In contrast to tumor cells that are highly susceptible, normal cells show marginal apoptotic responses to the dual action drugs. This protection of normal cells might reflect their greater ability to buffer pro-oxidative changes and quickly restore redox homeostasis, despite substantial drug uptake and macromolecular binding. Importantly, by targeting the death process at multiple points, DNA- and protein-damaging drugs can be less vulnerable to various bypass mechanisms possible with single targets. The reviewed studies provide a proof of concept that differential apoptosis targeting in cancer versus normal cells can be a basis for tumor selectivity of anticancer drugs.