Staurosporine- and H-7-induced cell death in SH-SY5Y neuroblastoma cells is associated with caspase-2 and caspase-3 activation, but not with activation of the FAS/FAS-L-caspase-8 signaling pathway

Apoptotic cell death is induced in SH-SY5Y neuroblastoma cells following exposure to the protein kinase inhibitors staurosporine (100 nM) and 1-(5-Isoquinolinesulfonyl)-2-methylpiperazine: H-7 (100 microM). This is associated with reduced levels of PARP 117 kDa and with the concomitant formation of PARP-cleaved products of 89 kDa that result from caspase-3 activation. The process is inhibited with DEVD-fmk, a potent caspase-3 (and caspase-8) inhibitor, thus indicating that staurosporine- and H-7-induced cell death in SH-SY5Y is mediated by caspase activation. Increased caspase-2- and caspase-3-like activities, but not caspase-9-like activity, were demonstrated by monitoring proteolysis of the corresponding colorimetric substrates. Caspase-2 activity peaked at 6 h, whereas caspase-3 peaked at 12 h in parallel with the maximal loss of cell viability. No modifications in the expression levels of Fas and Fas-L were observed by Western blotting. Furthermore, no activation of caspase-8 was elicited by colorimetric assays through the process of apoptosis of neuroblastoma cells. These findings indicate that the Fas/Fas-L-caspase-8 pathway of cell death signaling is not involved in staurosporine- and H-7-induced apoptosis in SH-SY5Y neuroblastoma cells.     

Caspases-2, -3, and -7 are involved in thapsigargin-induced apoptosis of SH-SY5Y neuroblastoma cells

Caspase-2 has been reported to play a role in the cell death observed under a number of different conditions; however, it is unclear whether caspase-2 plays a role in cell death triggered by endoplasmic reticulum (ER) stress. The purpose of this study was to determine whether caspase-2 is involved in SH-SY5Y neuroblastoma cell death caused by thapsigargin-induced ER stress. Thapsigargin treatment (1 microM, 16 hr) stimulated the proteolytic processing of caspases-2, -3, and -7, suggesting that these caspases are activated by ER stress. The role of these caspases in thapsigargin-induced cell death was examined by using cell-permeable caspase inhibitors. In the absence of pretreatment with caspase inhibitors, thapsigargin (0.1 microM, 20 hr) reduced the number of viable cells to 53.9% +/- 3.3% of starting-time control. Pretreatment for 90 min with either the pan-caspase inhibitor Z-VAD-FMK or the caspase-2-selective inhibitor Z-VDVAD-FMK inhibited thapsigargin-stimulated cell death, resulting in the number of viable cells being 115.6% +/- 5.3% (P < 0.001) and 69.3% +/- 2.9% (P < 0.01), respectively, of starting-time control. Neither the caspase-3- and -7-selective inhibitor Z-DEVD-FMK nor the caspase-9-selective inhibitor Z-LEHD-FMK significantly affected thapsigargin-stimulated cell death. An anticaspase-12-reactive protein was also identified in SH-SY5Y cells, but thapsigargin had no effect on proteolysis of this protein. These data demonstrate that caspases-2, -3, and -7 are involved in ER stress-mediated death of SH-SY5Y cells.     

Transient oxidative stress in SH-SY5Y human neuroblastoma cells results in caspase dependent and independent cell death and tau proteolysis

The effects of an oxidative insult on cell survival and tau metabolism were investigated in human neuroblastoma SH-SY5Y cells. In this treatment paradigm cells were exposed to the membrane permeant oxidant tert-butylhydroperoxide (tBHP) for 40 min, returned to fresh media and cell survival/death was monitored during the post-treatment period. Cell viability decreased significantly by 6 hr after tBHP exposure, and by 24 hr lactate dehydrogenase (LDH) release was 40.1 +/- 8.8% in tBHP treated cells compared to 8.1 +/- 4.7% in control cells. This oxidative stress paradigm also resulted in significant activation of caspase-3 by 2 hr post-treatment and nuclear apoptotic morphology. Furthermore, tBHP treatment also resulted in delayed tau proteolysis that was first evident 2 hr post-treatment. Treatment of the cells with the general caspase inhibitor Boc-Asp(OMe)-Fluoromethylketone (BAF) completely inhibited caspase-3 activation in response to tBHP, and delayed, but did not prevent cell death. BAF treatment also decreased tau proteolysis. In vitro, recombinant tau was readily proteolyzed by active recombinant caspase-3 into a stable breakdown product. Further tau in the cell lysates was cleaved by active recombinant caspase-3 at a rate, and to an extent similar to that observed for the well-established caspase-3 substrate poly(ADP-ribose)polymerase (PARP). These results suggest that oxidative stress-induced cell death occurs through both caspase-dependent and-independent pathways, and that tau is likely an in situ substrate of caspase-3.     

Protective effect of 17-beta-estradiol in human neurocellular models of lead exposure

The developing nervous system has long been recognized as a primary target site for lead (Pb)-induced toxicity. Pb-exposure causes cognitive dysfunction, growth retardation, hyperactivity and neurochemical deficits in animals and humans. In the present study the effects of 17-beta-estradiol on human SH-SY5Y neuroblastoma cells in culture exposed to low-levels of Pb were assessed. The cells were exposed to Pb (0.01-10 microM) for 48 h and cell proliferation was determined by the MTT reduction assay. Pb significantly inhibited the proliferation and growth of neuroblastoma cells in a concentration-dependent manner. A 50% inhibition (IC50) in the proliferation of cells was observed with 5 microM Pb. Exposure of cells to Pb (5 microM) for 48 h resulted in a significant increase (+732% of control) in caspase-3 activity, an indicator of apoptosis and total cellular prostaglandin E2 level (+1180% of control), marker of programmed cell death/neuronal cell loss. Pretreatment with 17-beta-estradiol (10 nM) effectively blocked the effects of Pb on caspase-3 activity but not prostaglandin E2 level. Further, Pb but not 17-beta-estradiol in a concentration (0.1-10 microM)-dependent manner effectively decreased (38-84%) the cellular concentration of glutathione (GSH), an important intracellular antioxidant. However, the effect of Pb on GSH level was effectively blocked when pretreated with 17-beta-estradiol. The data indicate that even low concentrations of Pb can be detrimental and potentially toxic to the developing brain. In conclusion, these results suggest that at least some of the neurotoxic effects of Pb may be mediated by apoptosis, which by pretreatment with 17-beta-estradiol can be prevented. This study further confirms previous reports of 17-beta-estradiol acting as a neuroprotective and antiapoptotic agent during induced toxic stress conditions.     

Citicoline increases glutathione redox ratio and reduces caspase-3 activation and cell death in staurosporine-treated SH-SY5Y human neuroblastoma cells

Citicoline, or CDP-choline, is an essential endogenous intermediate in the biosynthesis of phosphatidylcholine that may act as a neuroprotector in several models of neurodegeneration. The present study analyses the effects of citicoline in the paradigm of staurosporine-induced cell death in human SH-SY5Y neuroblastoma cells. Citicoline reduces apoptosis induced by 100 nM staurosporine for 12 h in SH-SY5Y cells. This effect is higher with pre-treatment of 60 mM citicoline for 24 h after staurosporine challenge. Moreover, citicoline treatment restores glutathione redox ratio diminished after staurosporine challenge. Finally, citicoline also reduces the expression levels of active caspase-3 and specific PARP-cleaved products of 89 kDa resulting from staurosporine exposure when citicoline is added to the culture medium 24 h before staurosporine. These findings demonstrate that citicoline affects the staurosporine-induced apoptosis cell-signalling pathway by interacting with the glutathione system and by inhibiting caspase-3 in SH-SY5Y human neuroblastoma cells.     

Fluazinam-induced apoptosis of SH-SY5Y cells is mediated by p53 and Bcl-2 family proteins

A number of epidemiological studies have demonstrated a strong association between the incidence of neurodegenerative disease and pesticide exposure. Fluazinam (FZN) is a preventative fungicide from the pyridinamine group that was introduced in the 1990 s and that quickly established itself as a new standard for the control of blight caused by Phytophthora infestans in potatoes. We used human neuroblastoma SH-SY5Y cells to investigate mechanisms of neuronal cell death in response to FZN and showed that FZN was cytotoxic to SH-SY5Y cells in a concentration- and time-dependent manner. Additionally, we showed that FZN treatment significantly decreased the neuron numbers including dopaminergic neurons and mitochondrial complex I activity. The cytotoxic effects of FZN were associated with an increase in reactive oxygen species (ROS) generation because pretreatment with N-acetyl cysteine, an anti-oxidant, reduced cell death. We showed that neuronal cell death in response to FZN was due to apoptosis because FZN increased cytochrome C release into the cytosol and activated caspase-3 through the accumulation of p53. FZN also reduced the levels of Bcl-2 protein but increased the levels of Bax. Our results provide insight into the molecular mechanisms of FZN-induced apoptosis in neuronal cells.     

Ablation of the chemokine monocyte chemoattractant protein-1 delays retrograde neuronal degeneration,attenuates microglial activation,and alters expression of cell death molecules

The mechanisms regulating retrograde neuronal degeneration and subsequent death of thalamic neurons following cortical injury are not well understood. However, the delay in the onset of retrograde cell death and observed morphological changes are consistent with apoptosis. Our previous studies demonstrated that monocyte chemoattractant protein-1 (MCP-1), a beta-chemokine that attracts cells of monocytic origin to sites of injury, is rapidly and specifically expressed in the lateral geniculate nucleus following visual cortical lesions. To determine the potential role of MCP-1 in retrograde degeneration, the present study examined the effect of genetic deletion of MCP-1 (MCP-1 KO or -/-) or its high affinity receptor CCR2 (CCR2 KO or -/-) on thalamic microglial activation and neuronal cell death following aspiration lesions of the visual cortex in adult mice. Deletion of the MCP-1 gene delayed microglial activation and transiently improved the survival of thalamic neurons. Deletion of the CCR2 receptor resulted in a significant increase in apoptosis as measured by nucleosomal fragmentation after injury compared to wild-type mice, but did not alter neuron survival, suggesting that glial apoptosis is increased in the receptor knockout mice. Investigation of Bcl-2, Bax, Fas, Fas ligand (FasL) and activated caspase-3, key regulators of apoptosis that can be modulated by cytokines, revealed complex alterations of mRNA and protein levels in MCP-1(-/-) and CCR2(-/-) mice. As examples, Bcl-2 protein was detected in wild-type, but not in MCP-1(-/-) mice. Caspase-3 activity was higher in MCP-1(-/-) mice compared to wild-type and CCR2(-/-) mice at 5 days after injury. High levels of activated caspase-3 correlate with the beginning of a period of delayed, but rapid cell death in the thalami of MCP-1(-/-) mice. In summary, our data strongly suggest that MCP-1 is involved in early microglial response to axotomy and that modulation of this chemokine could provide a novel strategy for improved neuronal survival following injury to the central nervous system.     

DNA damage induces cdk2 protein levels and histone H2B phosphorylation in SH-SY5Y neuroblastoma cells

DNA damage and activation of the cell cycle have been implicated in numerous neurodegenerative diseases, including Alzheimer disease, Parkinson's disease, and amyotrophic lateral sclerosis. To better understand the role of cell cycle proteins in DNA-damage induced neuronal cell death, we examined various cell cycle proteins during camptothecin-induced death of human neuroblastoma cells. We report a rapid induction of p53 and increased expression of p21, concurrent with reduced levels of many cell cycle proteins that regulate G1 to S phase cell cycle progression. However, we found increased levels of cdk2 and cyclin E, and formation of a cyclin E-cdk2-p21 protein complex. DNA damage failed to induce activation and progression of the cell cycle. Finally, camptothecin-induced neuronal cell death occurred concurrent with phosphorylation of histone H2B. Pretreatment of cells with cdk inhibitor olomoucine impeded cdk2-cyclin E accumulation, but not the induction of p53. Olomucine concurrently delayed histone H2B phosphorylation, caspase-3 activation and cell death. These findings suggest that DNA-damage of differentiated neuroblastoma cells induces a rapid p53-mediated inhibition of cell cycle progression and induction of cdk2-cyclin E, followed by caspase-3 activation, phosphorylation of histone and cell death.     

Involvement of endogenous N-methyl(R)salsolinol in Parkinson's disease: induction of apoptosis and protection by (-)deprenyl

An endogenous dopamine-derived N-methyl(R)salsolinol has been suggested to be involved in the pathogenesis of Parkinson's disease. In Parkinson's disease, the level of N-methyl(R)salsolinol increased in cerebrospinal fluid and the high activity of a synthesizing enzyme, (R)salsolinol N-methyltransferase, was detected in lymphocytes. This isoquinoline induced apoptotic DNA damage in human dopaminergic neuroblastoma SH-SY5Y cells. Among catechol isoquinolines, only N-methylsalsolinol induced apoptosis in the cells, and the scavengers of hydroxyl radicals and antioxidants suppressed DNA damage, suggesting that reactive oxygen species initiate apoptosis. The isoquinoline activated caspase-3 like proteases and a caspase-3 inhibitor protected the cells from DNA damage. (-)Deprenyl, but neither clorgyline nor pargyline, prevented apoptotic cell death. The mechanism of the protection was due to stabilization of mitochondrial membrane potential reduced by the toxin. In Parkinson's disease apoptosis may be induced in dopamine neurons by this endogenous neurotoxin, and (-)deprenyl may protect them from apoptotic death process.     

Effect of high-dose methylprednisolone treatment on CCR5 expression on blood cells in MS exacerbation

OBJECTIVES: Therapy of acute exacerbations of multiple sclerosis (MS) with high-dose intravenous methylprednisolone (IVMP) has shortened the recovery period after relapses, but the mechanisms responsible for the beneficial effects of IVMP in attacks have not been clearly established. Our purpose was to analyze the effect of IVMP on the expression of chemokine receptor 5 (CCR5) protein in blood in acute MS exacerbation. MATERIALS AND METHODS: Blood samples were collected from 10 patients with an acute MS exacerbation and the levels of CCR5 on CD4(+) and CD8(+) T cells and CD14(+) monocytes were analyzed by using flow cytometry before IVMP, 24 h, 1 and 3 weeks after commencement of treatment. RESULTS: During the 3-week period the percentages of CCR5-expressing CD4(+) T cells and CD8(+) T cells tended to decrease (P = 0.09 and 0.05, respectively), but the effect did not reach statistical significance. No marked changes were found in the percentage of CCR5-expressing CD14(+) cells. CONCLUSIONS: A tendency to a reduction of CCR5-expressing CD4(+) and CD8(+) blood cells induced by IVMP suggests inhibition of their potential to transmigrate into the central nervous system, which is consistent with the short-term beneficial effect of IVMP in acute exacerbation of MS.     

RNA interference targeting Akt promotes apoptosis in hypoxia-exposed human neuroblastoma cells

Overactivation of the PI3 kinase/Akt pathway plays an essential role in the development and progression of various tumors. Akt is a key component of this pathway and hyperactivated in different tumors including neuroblastoma and glioma. In the present study, we tested the therapeutic efficacy of siRNA targeting Akt in inducing apoptotic cell death in NBFL cells (a human neuroblastoma cell line) subjected to anoxia/reoxygenation (A/R), a process that has been shown to modulate growth and progression of malignant tumors. We observed that siRNA targeting Akt effectively induced apoptotic cell death in NBFL cells (as determined by TUNEL assay and activated caspase-3 immunoreactivity) under normoxic conditions, an effect that was greatly enhanced under conditions of A/R. These findings underscore the importance of Akt signaling in promoting survival of neuroblastoma cells and may have potential therapeutic applications.     

Survival motor Neuron protein regulates apoptosis in anin vitro model of Spinal muscular atrophy

Progressive spinal muscular atrophy (SMA), the most prevalent hereditary lower motor neuron disease, is caused by mutations in the telomeric copy of the survival of motor neuron (SMN1) gene. Unlike other cells, lower motor neurons cannot tolerate low levels of smn protein. However, it is unclear as to the nature of the cell death involved. There is evidence that lower motor neurons undergo apoptosis in SMA, leading to muscle weakness and wasting. This study investigated whether SMN1 regulation in a motor neuron model affected indices of apoptotic cell death. Decreased smn expression in neuroblastoma hybrid (NSC-34) cell lines by small interfering RNA (siRNA) was demonstrated at the mRNA and protein level. Smn-depleted cells showed elevated caspase-3 activity, decreased cell viability and increased percentage of TUNEL positive cells. Conversely, NSC-34 cell smn overexpression by adenoviral gene transfer decreased staurosporine-induced caspase-3 elevation and mitigated induced cell toxicity as assessed by 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. However, increased smn expression by itself did not increase cell viability. These data suggest not only that decreased smn levels increase apoptosis in an in vitro model of SMA, but also that increased smn can protect against neural injury.     

Characterization of catechol-thioether-induced apoptosis in human SH-SY5Y neuroblastoma cells

Recent work has highlighted the involvement of a dopamine derivative, 5-S-cysteinyl-dopamine (CysDA), in neurodegeneration and apoptotic cell death. In this paper we study in further detail the apoptotic process activated by this catechol-thioether derivative of dopamine in SH-SY5Y neuroblastoma cells. CysDA activates a cascade of events by an initial perturbation of Calcium homeostasis in the cell. Cell treatment with the catechol-thioether induces an immediate rise in intracellular Ca(2+) concentration, as demonstrated by a shift in the indo-1 dye emission spectrum, and a sustained high calcium concentration at long times of incubation. Fluorescence microscopy data show that the treatment of cells induces mitochondrial transmembrane potential depolarization, a clear evidence of the onset of apoptotic process. Programmed cell death activation is also demonstrated by cytochrome c release from the mitochondria, by an increased activity of both caspase-8 and -9 and by the poly(ADP-ribose)polymerase (PARP-1) cleavage, yielding the typical 86 kDa fragment due to caspase-3 activity. Overall, our data support the hypothesis that CysDA may induce apoptotic death in neuronal cells, via an initial perturbation of calcium homeostasis in the cytosol.     

Arachidonic acid peroxides induce apoptotic Neuro-2A cell death in association with intracellular Ca(2+) rise and mitochondrial damage independently of caspase-3 activation

The present study aimed at understanding the effects of arachidonic acid peroxides on neuronal cell death using the mouse neuroblastoma cell line, Neuro-2A cells. Arachidonic acid peroxides were produced by ultraviolet (UV) radiation. UV-radiated arachidonic acid significantly reduced Neuro-2A cell viability at concentrations of more than 0.1 muM, with being more potential than non-radiated arachidonic acid. Nuclei of Neuro-2A cells killed with UV-radiated arachidonic acid were reactive to Hoechst 33342, a marker of apoptosis, and the effect was much greater than that achieved with non-radiated arachidonic acid. UV-radiated arachidonic acid persistently increased intracellular Ca(2+) concentrations and dissipated mitochondrial membrane potential in Neuro-2A cells. UV-radiated arachidonic acid-induced Neuro-2A cell death, whereas it was not affected by a pancaspase inhibitor or a caspase-3 inhibitor, was significantly inhibited by an inhibitor of caspase-1, -8, or -9. The results of the present study suggest that arachidonic acid peroxides induce apoptotic neuronal cell death in association with intracellular Ca(2+) rise and mitochondrial damage, in part via a caspase-dependent pathway regardless of caspase-3.     

Proapoptotic effects of tau cleavage product generated by caspase-3

Using an in vitro translation assay to screen a human brain cDNA library, we isolated the microtubule-associated protein Tau and determined it to be a caspase-3 substrate whose C-terminal cleavage occurred during neuronal apoptosis. DeltaTau, the 50-kDa cleavage product, was detected by Western blot in apoptotic cortical cells probed with anti-PHF-1 and anti-Tau-5 antibodies, but not anti-T-46 antibody which recognizes the C-terminus. Overexpression of DeltaTau in SK-N-BE2(C) cells significantly increased the incidence of cell death. Staurosporine-induced Tau cleavage was blocked by 20 microM z-Asp-Glu-Val-Asp-chloromethylketone, a caspase-3 inhibitor, and in vitro, Tau was selectively cleaved by caspase-3 or calpain, a calcium-activated protease, but not by caspases-1, -8, or -9. (D421E)-Tau, a mutant in which Asp421 was replaced with a Glu, was resistant to cleavage by caspase-3 and tended to suppress staurosporine-induced cell death more efficiently than did wild-type Tau in both transient and stable expression systems. Finally, the incidence of DeltaTau-induced cell death was augmented by expression of Abeta precursor protein (APP) or Swedish APP mutant. Taken together, these results suggest that the caspase-3 cleavage product of Tau may contribute to the progression of neuronal cell death in Alzheimer's disease.     

Ascorbic acid protects SH-SY5Y neuroblastoma cells from apoptosis and death induced by beta-amyloid

beta-Amyloid causes apoptosis and death in cultured neurons that may be mediated by generation of reactive oxygen species. Since ascorbic acid concentrations are relatively high in brain, we tested whether and how this antioxidant might protect cultured SH-SY5Y neuroblastoma cells from apoptotic cell death. SH-SY5Y cells did not contain ascorbate in culture but readily took it up to achieve intracellular concentrations several-fold those of GSH. Treatment of cells with 2-10 microM beta-amyloid(25-35) decreased both intracellular ascorbate and GSH without affecting rates of ascorbate transport, which suggests that the peptide induces an oxidant stress in the cells. Overnight culture of cells with 10-20 microM beta-amyloid(25-35) induced apoptosis in SH-SY5Y cells when measured as externalization of phosphatidylserine by annexin V binding, as DNA fragmentation in the TUNEL assay, and as caspase-3 activity in cell lysates. Pre-loading cells with ascorbate substantially prevented apoptosis measured by these assays as well as cell death. In addition to preventing apoptosis, ascorbate loading of SH-SY5Y cells also decreased basal rates of generation of endogenous beta-amyloid. Together, these results support the notion that beta-amyloid induces apoptosis and death in neurons due to oxidant stress and suggest that intracellular ascorbate effectively prevents this toxicity.     

NMDA-sensitive neurons profoundly influence delayed staurosporine-induced apoptosis in rat mixed cortical neuronal cultures

We have investigated cell killing in cultured rat embryonic cortical neurons exposed to the protein kinase inhibitor staurosporine, the excitatory amino acid N-methyl-D-aspartate (NMDA), or a combination thereof. Our data indicate that there are several populations of neurons that differ in their response to these agents. Cultures exposed to NMDA undergo cell death typified by lactate dehydrogenase (LDH) leakage which is likely primarily necrotic in that little caspase-3 activation or oligonucleosome formation is observed even when followed for 48 h. Cells exposed to staurosporine (STS) exhibit rapid, extensive activation of caspase-3 with coincident LDH leakage, oligonucleosome formation and TUNEL staining. Both LDH leakage and oligonucleosome content were significantly more elevated at 48 h than at 20 h with STS treatment while caspase-3 activity peaked early (8-20 h) and declined markedly by 48 h. Deletion of NMDA-responsive neurons by pre-treatment of the cultures with NMDA for 4 days prevented the late phase (20-48 h) increases in LDH leakage and oligonucleosomes in the remaining neuronal population. Caspase-3 activity was also completely abolished by NMDA pre-treatment. These results indicate that cells susceptible to acute NMDA-induced toxicity can be killed by non-apoptotic means when exposed to NMDA; however, they undergo a delayed, apoptotic death when exposed to STS. Interestingly, removal of NMDA-responsive cells prevents the processing of procaspase-3; thus, STS-induced apoptosis in cells resistant to NMDA-mediated killing proceeds independent of caspase-3 activation. The data indicate that nearly all neurons in these mixed cultures can undergo apoptosis in response to appropriate stimuli such as STS but that the temporal nature, and the pathways activated in response to STS, vary amongst the subpopulations of neurons. These findings may help to explain the simultaneous appearance of features of both apoptosis and necrosis observed in vivo following cerebral ischemia.     

Caspase activation pathways induced by staurosporine and low potassium: role of caspase-2

Apoptotic death is a physiological process with regulatory mechanisms that are under the control of different molecules such as caspases. These are classified as initiators, such as caspases-8 and -9, and effectors, such as caspases-3 and -7. The participation of caspase-2 in the effector phase of apoptosis has been commonly observed in many cell types; however, it is able to act as an initiator caspase, depending on the apoptotic stimulus. Cerebellar granule cells (CGCs) undergo apoptosis when they are transferred from high potassium (K25) to low potassium (K5); this process seems to be mediated by caspase-3 activation. Staurosporine (STS), a full strength inhibitor of kinase proteins, also induces apoptosis in these cells. To characterize the caspase cascade induced by two stimuli in the same cell type we studied the activation of different caspases in CGCs treated with STS or K5. We found that both K5 and STS induce the activation of caspase-3. This result was confirmed by the proteolytic cleavage of poly (ADP-ribose) polymerase (PARP), an endogenous caspase-3 substrate. Caspase-2 was activated preferentially by STS, which showed a temporal course suggesting that this caspase was induced before caspase-3. The initiator caspase-9 was also activated by both K5 and STS, as well as cytochrome-c release. The results obtained in this study suggest that STS and K5 induced different activation caspase pathways for apoptotic cell death of CGCs.