Apoptosis and nitric oxide release induced by thalidomide, gossypol and dexamethasone in cultured human chronic myelogenous leukemic K-562 cells

The effects of different dosages of thalidomide, gossypol and dexamethasone on the levels of apoptosis and nitric oxide (NO) production were studied in a human chronic myelogenous leukemic cell line, K-562. Increases in the levels of apoptosis were induced by both 15 and 30 microM of thalidomide but only 15 microM significantly increased NO production. All dosages of gossypol used increased the production of NO and all dosages except 200 microM raised the level of apoptosis. After dexamethasone treatment the level of NO either decreased or stayed constant. Thus, some dosages of thalidomide and gossypol concomitantly raise the levels of apoptosis and NO production, but dexamethasone, though it induced apoptosis, had no significant effect on NO production.     

Variable sister-chromatid exchange response in human lymphocytes exposed in vitro to gossypol acetic acid

Gossypol has potential for widespread use as a male oral antifertility agent in humans since it appears to be highly efficacious, with reversible spermatostatic effects and minimal side effects. Furthermore, it is both inexpensive and readily available. Therefore, a thorough understanding of gossypol's genotoxic potential is critical. Although genotoxicity studies have produced conflicting reports, increased sister-chromatid exchange (SCE) and DNA-strand breaks have been reported in human cells exposed to gossypol in vitro. In the present study, SCE was examined in purified human lymphocytes and whole blood cultures exposed to gossypol acetic acid at various concentrations in serum-free medium. A small but statistically significant increase in SCE was observed in pooled analysis of 7 donors in whole blood cultures exposed to 0.70 microM gossypol acetic acid (p less than 0.02). Individual analyses revealed only one donor with a significant SCE response (p less than 0.001). In subsequent experiments, exposure at higher doses had no effect on SCE frequencies. A small but significant increase in SCE was observed in ficoll/hypaque purified lymphocytes exposed to 0.07 and 0.70 microM gossypol acetic acid. Interpretation of SCE data with variable response is discussed.     

Involvement of reactive oxygen species-independent mitochondrial pathway in gossypol-induced apoptosis

Gossypol is a component present in cottonseeds and has been demonstrated to be an effective contraceptive drug in preventing spermatogenesis in mammalian species. In the present, we reported that gossypol could induce apoptosis in human promyelocytic leukemia cells (HL-60), as characterized by DNA fragmentation, poly(ADP) ribose polymerase (PARP) cleavage. The efficacious induction of apoptosis was observed at 50 microM for 6 h. Further molecular analysis showed that gossypol induced the truncation of Bid protein, the loss of mitochondrial membrane potential (DeltaPsi m), cytochrome c release from mitochondria into cytosol, and activation of caspase-3, -8, and -9. However, gossypol did not increase the level of reactive oxygen species (ROS), and antioxidants including N-acetyl cysteine (NAC) and catalase could not block gossypol-induced apoptosis in the HL-60 cells. These data suggest that gossypol induces apoptosis in HL-60 cells through ROS-independent mitochondrial dysfunction pathway.     

Effects of adenosine and deoxyadenosine on PHA-stimulation of lymphocytes of man, horse and pig

1. Adenosine inhibits thymidine and uridine incorporation of PHA-stimulated lymphocytes of man and horse at concentrations higher than 50 and 10 microM, respectively. Deoxyadenosine is inhibitory at concentrations higher than 100 microM. Thymidine and uridine incorporation of porcine lymphocytes are elevated 5-7-fold by 25-100 microM adenosine, deoxyadenosine, inosine and hypoxanthine. Leucine incorporation of PHA-stimulated lymphocytes was affected by adenosine and deoxyadenosine in the same way, but to a lower extent. 2. Effects of adenosine and deoxyadenosine were more pronounced at shorter cultivation times. 3. EHNA potentiated the effects of adenosine and deoxyadenosine on human and equine lymphocytes. With human lymphocytes inhibition by deoxyadenosine and EHNA was higher than by adenosine and EHNA. With porcine lymphocytes only the combination of deoxyadenosine and EHNA was inhibitory. 4. Homocysteine potentiated the inhibition of thymidine incorporation by the combination of adenosine and deoxyadenosine with equine lymphocytes, but not the inhibition of adenosine or deoxyadenosine alone. 5. Adenosine suppressed the PHA-stimulated elevation of PRPP concentrations. With porcine lymphocytes PRPP remained at the level of 0 hr, while with equine lymphocytes PRPP concentration decreased to below that level. 6. The various effects of adenosine and deoxyadenosine on lymphocytes of man, horse and pig can partially be related to differences in adenosine and deoxyadenosine metabolism.     

Different pathways of cell killing by gossypol enantiomers

Gossypol, a polyphenolic, aldehyde-containing constituent of cottonseed, produced partial responses (>50% reduction in tumor size) in some patients with advanced cancer and suppressed sperm as an antifertility agent for men. This action in vivo and its novel side effect profile suggest a specific mechanism of the action of gossypol. Using the random homozygous knockout approach of Li and Cohen (1), we developed a cell line resistant to killing by gossypol, but sensitive to methotrexate and doxorubicin. It showed stereospecific resistance to killing by (-) gossypol (ED(50) 4.9 microM) compared with wild type (ED(50) 2.0 microM). The resistant and wild-type cells were equally sensitive to (+) gossypol (ED(50) 8.8 and 8.4 microM, respectively), methotrexate, and doxyrubicin. We conclude that gossypol affects cells by a stereospecific pathway for (-) gossypol, possibly related to its selective effects, and a nonstereospecific pathway for (+) gossypol and higher concentrations of (-) gossypol. Further knowledge about the stereospecific pathway may lead to new therapeutic drugs.     

The effect of gossypol on fast axonal transport and microtubule assembly

Gossypol at micromolar concentrations (2 microM) was found to inhibit axonal transport and a microsomal ATPase activity in the frog sciatic nerve, although axonal microtubules and the neuronal content of AMP, ADP and ATP were not affected. At slightly higher concentrations (30-40 microM), gossypol also inhibited microtubule assembly and neuronal energy metabolism. Gossypol accumulated in the nerve and the results indicate that gossypol may act as a potent neurotoxin.     

Novel effects of gossypol, a chemical contraceptive in man: mobilization of internal Ca(2+) and activation of external Ca(2+) entry in intact cells

The effect of gossypol on Ca(2+) signaling in Madin Darby canine kidney (MDCK) cells was investigated by using fura-2 as a Ca(2+) probe. Gossypol evoked a rise in cytosolic free Ca(2+) levels ([Ca(2+)](i)) concentration-dependently between 2 and 20 microM. The response was decreased by external Ca(2+) removal. In Ca(2+)-free medium pretreatment with gossypol nearly abolished the [Ca(2+)](i) increase induced by carbonylcyanide m-chlorophenylhydrazone (CCCP), a mitochondrial uncoupler, and thapsigargin, an inhibitor of the endoplasmic reticulum Ca(2+) pump; but pretreatment with CCCP and thapsigargin only partly inhibited gossypol-induced Ca(2+) release. Addition of 3 mM Ca(2+) induced a [Ca(2+)](i) increase after pretreatment with 5 microM gossypol in Ca(2+)-free medium. This Ca(2+) entry was decreased by 25 microM econazole, 50 microM SKF96365 and 40 microM aristolochic acid (a phospholipase A(2) inhibitor). Pretreatment with aristolochic acid inhibited 5 microM gossypol-induced internal Ca(2+) release by 55%, but suppression of phospholipase C with 2 microM 1-(6-((17beta-3-methoxyestra-1,3, 5(10)-trien-17-yl)amino)hexyl)-1H-pyrrole-2,5-dione) had no effect. Gossypol (5 microM) also increased [Ca(2+)](i) in human bladder cancer cells and neutrophils. Collectively, we have found that gossypol increased [Ca(2+)](i) in MDCK cells by releasing Ca(2+) from multiple Ca(2+) stores in a manner independent of the production of inositol-1,4,5-trisphosphate, followed by Ca(2+) influx from external space.     

Inhibition kinetics of lactate dehydrogenase isoenzymes by gossypol acetic acid

The effect of gossypol acetic acid, a potent male sterilent was studied on LDH from goat liver (LDH-A4), heart (LDH-B4) and testis (LDH-C4) in vitro. All the preparations of LDH were inhibited by gossypol when the reaction was carried out in pyruvate-lactate (direct) or lactate to pyruvate (reverse) directions. The IC50 of gossypol for the pyruvate oxidation by LDH isozymes varied between 16 and 42 microM in presence of 0.27 mM pyruvate and 0.15 mM NADH at 25 degrees C and pH 7.4 whereas for the lactate oxidation, IC50 was 125 microM in a system containing 3.3 mM lactic acid and 1.8 mM NAD at 25 degrees C and pH 9.0. Reciprocal plots due to Lineweaver-Burk showed that these isozymes are inhibited in a non-competitive manner with respect to pyruvate and lactate, and in a competitive fashion when NAD and NADH were varied as substrates. Ki values of LDH-A4, -B4 and -C4 isozymes in presence of gossypol were 20, 34 and 29 microM against pyruvate; 33, 43 and 45 microM against NADH; 85, 85 and 125 microM against lactate and 94, 108 and 83 microM against NAD respectively.     

Profound negative regulatory effects by resveratrol on vascular smooth muscle cells: a role of p53-p21(WAF1/CIP1) pathway

We investigated the role of resveratrol, a polyphenol rich in red wine, in cell cycle progression and apoptosis of vascular smooth muscle cells (VSMCs). Resveratrol inhibited the growth of human aortic VSMCs at concentrations as low as 1 microM. This was due to the profound dose-dependent inhibition of DNA synthesis by resveratrol. DNA synthesis was more effectively inhibited when cells were pretreated with resveratrol. Resveratrol caused a dose-dependent increase in intracellular p53 and p21(WAF1/CIP1) levels. At lower concentrations (6.25-12.5 microM), resveratrol effectively blocked cell cycle progression of serum-stimulated VSMCs without inducing apoptosis, while the higher concentration of resveratrol (25 microM) selectively induced apoptosis in the same VSMCs. Intriguingly, however, the same high concentration of resveratrol could not induce apoptosis in quiescent VSMCs. These differential biological effects of resveratrol on quiescent and proliferating VSMCs suggest that resveratrol may be capable of selectively eliminating abnormally proliferating VSMCs of the arterial walls in vivo.     

Effect of gossypol on intracellular Ca2+ regulation in human hepatoma cells

Gossypol is a natural toxicant present in cottonseeds, and is hepatotoxic to animals and human. The effect of gossypol on cytosolic free Ca2+ levels ([Ca2+]i) in HA22/VGH human hepatocytes was explored using fura-2 as a fluorescent Ca2+ indicator. Gossypol increased [Ca2+]i in a concentration-dependent manner with an EC50 value of 2 microM. The Ca2+ signal was reduced by removing extracellular Ca2+ or by 10 microM La3+, but was not affected by nifedipine, verapamil or diltiazem. Pretreatment with 1 microM thapsigargin (an endoplasmic reticulum Ca2+ pump inhibitor) to deplete the endoplasmic reticulum Ca2+ partly reduced 10 microM gossypol-induced Ca2+ release; and conversely pretreatment with gossypol abolished thapsigargin-induced Ca2+ release. The Ca2+ release induced by 10 microM gossypol was not changed by inhibiting phospholipase C with 2 microM U73122 or by depleting ryanodine-sensitive Ca2+ stores with 50 microM ryanodine. Together, the results suggest that in human hepatocytes, gossypol induced a [Ca2+]i increase by causing store Ca2+ release from the endoplasmic reticulum in a phospholipase C-independent manner, and by inducing Ca2+ influx.     

Cholesterol side-chain cleavage and 11 beta-hydroxylation are inhibited by gossypol in adrenal cortex mitochondria

Cholesterol side-chain cleavage and 11 beta-hydroxylation were assessed in isolated adrenal cortex mitochondria by formation of pregnenolone and corticosterone, respectively, in the presence and absence of gossypol. Pregnenolone biosynthesis was inhibited when increasing concentrations of gossypol were added. The control value of 4 nmol min-1 mg-1 dropped to 2 nmol min-1 mg-1 with 30 microM of the drug in the incubation medium. A more pronounced inhibitory effect was observed upon 11 beta-hydroxylation of steroids; I50 was 11 microM. Seventy-five percent of corticosterone production was impaired when 30 microM of gossypol were present. Bovine serum albumin prevented and reversed the inhibitory action of the drug. Kinetic studies showed a linear mixed type inhibition, suggesting a direct action of the drug upon the enzymatic complex. This study demonstrates a direct inhibitory effect of gossypol upon the steroidogenic enzymes located in the inner mitochondrial membrane of the adrenal cortex.     

Evaluation of the cytotoxicity, cytostaticity and genotoxicity of argentatins A and B from Parthenium argentatum (Gray)

Argentatins A and B are abundant triterpenes present in Parthenium argentatum. Both compounds have shown cytotoxic properties on K562, MCF-7, PC-3, HCT-15 and U251 human cancer cell lines. Furthermore the cytotoxic, cytostatic and genotoxic effects of the argentatins on proliferating lymphocytes were evaluated using cytokinesis-block micronucleus test. Argentatin A had no cytostatic properties, but it was cytotoxic for proliferating lymphocytes at a concentration of 25 microM (P < 0.005). On the other hand, argentatin B showed significant cytostatic effects (P < 0.001) at concentrations of 5 to 25 microM and it did not show cytotoxic effects at the same concentrations. Neither argentatin showed genotoxic effects in terms of micronucleus frequency in human lymphocytes. According to these results the argentatins are not able to cause injury on DNA by clastogenic or aneugenic mechanisms.     

Gossypol modulation of nucleotide metabolizing enzymes in the reproductive tract of male rats

The activities of several pivotal nucleotide metabolizing enzymes from the testis and vasal sperm of rats treated for 7 wk with 0, 20 or 30 mg X kg X day gossypol acetic acid were examined. Total testicular lactate dehydrogenase (LDH) activity increased 40% above control in the highest treatment group examined. However, the specific activity of the testis-specific isozyme of LDH, LDH-C4, decreased to 50 and 20% of control in the 20 and 30 mg X kg X day treatment groups, respectively. Basal soluble adenylate cyclase from a 100,000 X g supernatant of testis homogenate exhibited a 25% decrease in activity only in the 30-mg treatment group. Basal adenylate cyclase activity in the testicular membrane fraction increased 20 to 30% above control in response to gossypol administration. Testis membranes from the 20- and 30-mg treatment group exhibited a 2- and 4-fold greater activation of adenylate cyclase by guanine nucleotides. In vitro dose-response curves showed a half-maximal inhibitory concentration (IC50) for inhibition of soluble testicular adenylate cyclase by gossypol of 400 microM in each treatment group. Caudal epididymal sperm adenylate cyclase activity decreased to 25% of control levels in gossypol-treated animals, and the in vitro sensitivity of the enzyme to the inhibitory effects of gossypol increased 4-fold. IC50 values for gossypol inhibition of sperm adenylate cyclase decreased from 200 microM in control animals to 75 and 50 microM in the 20 and 30 mg X kg X day treatment groups, respectively. Cyclic adenosine 3':5' monophosphate phosphodiesterase activity in caudal sperm increased 6-fold in the 20- and 30-mg treatment groups. These results demonstrate that nucleotide metabolizing enzymes in sperm are major targets for the actions of gossypol and provide a possible mechanism for the inhibition of normal sperm function by this compound.     

1,3-Butadiene and its epoxides induce sister-chromatid exchanges in human lymphocytes in vitro

Sister-chromatid exchanges (SCEs) were induced in human lymphocytes by 1,3-butadiene and its epoxides 3,4-epoxy-1-butene and 1,2:3,4-diepoxybutane. After a pulse treatment of 2 h, 1,3-butadiene produced a weak but reproducible increase in SCEs both with and without S9 mix. The response was similar in cultures of whole blood and of isolated lymphocytes. The 2 epoxide metabolites of butadiene, studied in whole-blood lymphocyte cultures without exogenous metabolic activation, were highly active SCE inducers. The lowest effective concentrations of butadiene, monoepoxybutene, and diepoxybutane were 2000 microM, 25 microM and 0.5 microM, respectively. A slight but dose-dependent increase in SCEs was also observed without an exogenous metabolic system after a 48-h treatment with 1,3-butadiene. Already the lowest concentration tested (500 microM) was effective. Again, the response was similar in cultures of whole blood and isolated lymphocytes, suggesting that the lymphocytes are capable of metabolically activating 1,3-butadiene.     

Cadmium- and chromium-induced oxidative stress, DNA damage, and apoptotic cell death in cultured human chronic myelogenous leukemic K562 cells, promyelocytic leukemic HL-60 cells, and normal human peripheral blood mononuclear cells

Sodium dichromate [Cr(VI)] and cadmium chloride [Cd(II)] are both cytotoxic and mutagenic. This study examined the toxic and apoptotic potentials of these two cations on three cell types in vitro, namely, human chronic myelogenous leukemic (CML) K562 cells, promyelocytic leukemic HL-60 cells, and normal human peripheral blood mononuclear cells. The cells were incubated with 0-100 microM concentrations of the two cations for 0, 24, or 48 hours at 37 degrees C. Both Cr(VI) and Cd(II) induced changes in intracellular oxidized states of cells, which were detected using laser scanning confocal microscopy. Cell cycle modulation and apoptosis of the K562 cells by Cr(VI) and Cd(II) were determined by flow cytometry. Significant decreases in the G2/M phase were observed in the Cr(VI) and Cd(II) treated CML cells compared with untreated cells. At 12.5 microM, Cr(VI) induced greater apoptosis in K562 cells as compared with Cd(II). In the K562 cells, 2.2- and 3.0-fold increases in DNA fragmentation were observed following incubation with 12.5 and 25 microM Cr(VI), respectively, and 1.2- and 1.7-fold increases in DNA fragmentation were observed with Cd(II). Furthermore, approximately 2.7- and 4.9-fold increases in cytochrome c reduction were observed following incubation with 12.5 and 25 microM Cr(VI), respectively, and 1.6- and 3.3-fold increases in cytochrome c reduction were observed with Cd(II), demonstrating enhanced production of superoxide anion. Approximately 3.1 to 6-fold increases in hydroxyl radical production were observed following incubation of the K562 cells with these cations at 12.5 and 25 microM concentrations. These results in K562 cells were compared with promyelocytic leukemic HL-60 cells and normal human peripheral blood mononuclear cells. More pronounced effects were observed on K562 and HL-60 cells, and much lesser effects were observed on normal human peripheral blood mononuclear cells. The results demonstrate that both cations are toxic, producing oxidative tissue damage and apoptosis. Furthermore, more drastic effects were observed on K562 and HL-60 cells as compared with normal human peripheral blood mononuclear cells.     

Arachidonic acid cytotoxicity in leukocytes: implications of oxidative stress and eicosanoid synthesis

Arachidonic acid (AA)-induced cytotoxicity was evaluated in leukocytes: the human leukemia cell lines HL-60, Jurkat and Raji and in rat lymphocytes. Such cytotoxicity was dose- and time-dependent. At concentrations below 5 microM, AA was not toxic; at 10-400 microM, AA induced apoptosis and at concentrations beyond 400 microM, necrosis. The minimum exposure time to trigger cell death was of around 1 h, but the effect was increased by longer exposure times until 6-24 h. Apoptosis was morphologically characterized by a decrease in cell and nuclear volume, chromatin condensation and DNA fragmentation and the presence of lipid bodies, without changes in organelle integrity. Biochemically, AA-induced apoptosis was associated with internucleosomal fragmentation and caspase activation, evaluated by PARP cleavage and the use of a caspase inhibitor. Necrosis was characterized by increased cell volume, presence of loose chromatin, appearance of vacuoles, loss of membrane integrity and of the definition of organelles. The apoptotic effect of AA was studied as to oxidative-reductive imbalance and the participation of eicosanoids. Apoptotic AA treatment was accompanied by an increase in the quantity of thiobarbituric acid reactive substances (TBARS), low-level chemiluminescence and in the glutathione disulfide/reduced glutathione ratio, indicating oxidative stress. The addition of tocopherol, ascorbate, prostaglandin E2 and lipoxygenase inhibitors delayed cell death, whereas the inhibition of cyclooxygenase promoted AA-induced cell death. Cell treatment with AA was accompanied by increased cellular production of LTB4. AA, therefore, is cytotoxic at physiological and supraphysiological concentrations, causing apoptosis and necrosis. Cell treatment with apoptotic concentrations of AA involves oxidative stress and changes in eicosanoid biosynthesis.     

Effects of resveratrol on the G(0)-G(1) transition and cell cycle progression of mitogenically stimulated human lymphocytes

Resveratrol (RSV) has been suggested to have cancer preventive properties, on the basis that it suppresses proliferation and induces apoptosis in various tumor cells. Here we test its cytostatic effects on peripheral blood human lymphocytes. RSV (up to 50 microM) had no detectable effects on resting lymphocytes. With the mitogen phytohemagglutin (PHA), however, RSV elicited concentration- and time-dependent responses in lymphocytes. RSV (>/=50 microM) prevented cell entry into the cell cycle, resulting in 99% suppression at 100 microM. The arrested lymphocytes following 24h treatment with 50 microM RSV had minimal RNA content, the feature characteristic of G(0) cells, and were blocked at the stage past the induction of cyclins D2 and D3 and prior to induction of cyclin E. Prolonged treatment (72h) of PHA-stimulated lymphocytes with 100 microM RSV showed a pronounced decrease in the expression of pRb, cyclins E and B, and reduction in p34cdc2 and PCNA. The activation-induced apoptosis was also reduced in the presence of >/=50 microM RSV. These data suggest that studies designed to test RSV efficacy as a chemopreventive agent should include evaluation of its immunomodulatory effect revealed by suppression of lymphocyte stimulation as well as its effect on apoptosis of stimulated lymphocytes.     

Gossypol Induces Death Receptor-5 through Activation of the ROS-ERK-CHOP Pathway and Sensitizes Colon Cancer Cells to TRAIL

Development of resistance to TRAIL, an apoptosis-inducing cytokine, is one of the major problems in its development for cancer treatment. Thus, pharmacological agents that are safe and can sensitize the tumor cells to TRAIL are urgently needed. We investigated whether gossypol, a BH3 mimetic that is currently in the clinic, can potentiate TRAIL-induced apoptosis. Intracellular esterase activity, sub-G₁ cell cycle arrest, and caspase-8, -9, and -3 activity assays revealed that gossypol potentiated TRAIL-induced apoptosis in human colon cancer cells. Gossypol also down-regulated cell survival proteins (Bcl-x_L, Bcl-2, survivin, XIAP, and cFLIP) and dramatically up-regulated TRAIL death receptor (DR)-5 expression but had no effect on DR4 and decoy receptors. Gossypol-induced receptor induction was not cell type-specific, as DR5 induction was observed in other cell types. Deletion of DR5 by siRNA significantly reduced the apoptosis induced by TRAIL and gossypol. Gossypol induction of the death receptor required the induction of CHOP, and thus, gene silencing of CHOP abolished gossypol-induced DR5 expression and associated potentiation of apoptosis. ERK1/2 (but not p38 MAPK or JNK) activation was also required for gossypol-induced TRAIL receptor induction; gene silencing of ERK abolished both DR5 induction and potentiation of apoptosis by TRAIL. We also found that reactive oxygen species produced by gossypol treatment was critical for TRAIL receptor induction and apoptosis potentiation. Overall, our results show that gossypol enhances TRAIL-induced apoptosis through the down-regulation of cell survival proteins and the up-regulation of TRAIL death receptors through the ROS-ERK-CHOP-DR5 pathway.     

Effect of gossypol on trypomastigotes and amastigotes of Trypanosoma cruzi

Bloodstream Trypanosoma cruzi trypomastigotes isolated from infected mice undergo reduction of motility and structural damages after 5 to 45 min exposure to gossypol at concentrations ranging from 5 to 50 microM. When 1% serum albumin is added to the incubation medium, no alterations of parasites are observed, even with 100 microM gossypol. Intracellular T. cruzi amastigotes in infected Vero cell cultures exposed to 5 microM gossypol for 2 h do not show changes. Incubation with 5 microM gossypol for 48 h produces complete disruption of host cells; however, the amastigotes they contain show only minor alterations. The observations indicate that, in protein-rich media, gossypol is complexed into associations which have no activity on the different forms of the T. cruzi biological cycle.     

Cepharanthine activates caspases and induces apoptosis in Jurkat and K562 human leukemia cell lines.

Cepharanthine (CEP) is a known membrane stabilizer that has been widely used in Japan for the treatment of several disorders such as anticancer therapy-provoked leukopenia. We here report that apoptosis was induced by low concentrations (1-5 microM) of CEP in a human leukemia T cell line, Jurkat, and by slightly higher concentrations (5-10 microM) in a human chronic myelogenous leukemia (CML) cell line K562, which expresses a p210 antiapoptotic Bcr-Abl fusion protein. Induction of apoptosis was confirmed in both Jurkat and K562 cells by DNA fragmentation and typical apoptotic nuclear change, which were preceded by disruption of mitochondrial membrane potential and were induced through a Fas-independent pathway. CEP treatment induced activation of caspase-9 and -3 accompanied by cleavage of PARP, Bid, lamin B1, and DFF45/ICAD in both Jurkat and K562 cells, whereas caspase-8 activation and Akt cleavage were observed only in Jurkat cells. The CEP-induced apoptosis was completely blocked by zVAD-fmk, a broad caspase inhibitor. Interestingly, CEP treatment induced remarkable degradation of the Bcr-Abl protein in K562 cells, and this degradation was prevented partially by zVAD-fmk. When used in combination with a nontoxic concentration of herbimycin A, lower concentrations (2-5 microM) of CEP induced obvious apoptosis in K562 cells with rapid degradation or decrease in the amount of Bcr-Abl and Akt proteins. Our results suggest that CEP, which does not have bone marrow toxicity, may possess therapeutic potential against human leukemias, including CML, which is resistant to anticancer drugs and radiotherapy.