Staurosporine-induced neuronal apoptosis

Increased surface phosphatidylserine is an early marker of neuronal apoptosis

Neuronal apoptosis is the subject of intense investigation and is beginning to be understood in some molecular detail. In the present study, we show that PC12 cells, like certain other cell types, redistribute phosphatidylserine (PS) from the inner leaflet to the outer leaflet of the plasma membrane early in the process of apoptosis. The externalised PS can be readily visualised by incubating intact cells with a fluorescent derivative of the protein annexin V. When apoptosis is blocked with an inhibitor of interleukin-1beta-converting-enzyme-like proteases, the increased annexin binding is also blocked. Fluorescent annexin V binding provides a rapid and convenient way to identify apoptotic neurones.

Susceptibility to apoptosis is enhanced in immature cortical neurons

The susceptibility of cortical neurons to two forms of apoptotic death was compared with susceptibility to excitotoxic death during development in vitro (DIV 4-21). Murine cortical cultures were exposed for 48 h to the phosphatase inhibitor cyclosporine, the protein kinase inhibitor staurosporine or the excitotoxin N-methyl-D-aspartate (NMDA). Susceptibility to apoptosis induced by staurosporine or cyclosporine was maximal between DIV 4-10 and declined from DIV 10 through 18. The opposite pattern was observed with susceptibility to NMDA receptor-mediated excitotoxic necrosis, which was minimal at DIV 6 and progressively increased through DIV 21.

Characterization of the cell death process induced by staurosporine in human neuroblastoma cell lines

Staurosporine is a potent and non-specific inhibitor of protein kinases. There is also evidence of staurosporine being a potent inducer of apoptosis. In several human neuroblastoma cell lines (SH-SY5Y, NB69, IMR-5 and IMR-32) we have found 100 nM staurosporine to induce cell death in half the population (EC50). Electron microscopy of these cells, fluorescence microscopy after Hoechst-33258 staining of chromatin and agarose-electrophoresis of DNA, show two different types of cell death. SH-SY5Y and NB69 die by apoptosis and display all the characteristic features of it. IMR-5 and IMR-32 lack some of these features and a ladder pattern of DNA degradation is not found. Different morphological types of apoptosis have been described during the development of vertebrates; the possibility of finding a similar diversity in cell culture is suggested. On the other hand, staurosporine is a potent promoter of neurite outgrowth. In all the neuroblastoma cell lines we have tested, neurite-promoting and cell death-inducing staurosporine concentrations mostly overlap. This fact has not been reported before, probably because of an early versus late timing of these two different phenomena. The neuritogenic effect has prompted the suggestion that staurosporine could be a prototype of drugs for neurodegenerative diseases; the present study raises several concerns about such a proposal.

Apoptotic Vascular Endothelial Cells Become Procoagulant

Whereas unperturbed endothelial cells provide potent anticoagulant properties, exposure to inflammatory and atherogenic stimuli can rapidly lead to a procoagulant behavior. Because recent studies provide evidence that apoptosis of vascular cells may occur under conditions such as atherosclerosis and inflammation, we investigated whether apoptotic endothelial cells may contribute to the development of a prothrombotic state. In this report, it is shown that both adherent and detached apoptotic human umbilical vein endothelial cells (HUVECs) become procoagulant. Apoptosis was induced by staurosporine, a nonspecific protein kinase inhibitor, or by culture in suspension with serum deprivation. Both methods resulted in similar findings. As assessed by flow cytometric determination of annexin V binding, HUVECs undergoing cell death exhibited typically a more rapid exposure of membrane phosphatidylserine (PS) than DNA fragmentation. Depending on the stage of apoptosis, this redistribution of phospholipids was found to induce an increase of the activity of the intrinsic tenase complex by 25% to 60%. Although apoptotic cells did not show antigenic or functional tissue factor (TF ) activity, when preactivated with lipopolysaccharide, TF procoagulant activity increased by 50% to 70%. At 8 hours after apoptosis induction, antigenic thrombomodulin, heparan sulfates, and TF pathway inhibitor decreased by about 83%, 80%, and 59%, respectively. The functional activity of these components was reduced by about 36%, 52%, and 39%, respectively. Moreover, the presence of apoptotic HUVECs led to a significant increase of thrombin formation in recalcified citrated plasma. In conclusion, apoptotic HUVECs, either adherent or in suspension, become procoagulant by increased expression of PS and the loss of anticoagulant membrane components.

Staurosporine differentiated human SH-SY5Y neuroblastoma cultures exhibit transient apoptosis and trophic factor independence

The use of chemically differentiated neuroblastoma cells in the study of neuronal function has become a common alternative to primary neuronal cell cultures in recent years, particularly in the area of cell death. Staurosporine, a nonselective protein kinase inhibitor, has been demonstrated to be a particularly strong inducer of differentiation in the SH-SY5Y human neuroblastoma cell line. However, at present, no data exist on the long-term effects of this compound. We have compared the effects of staurosporine with 12-O-tetradecanoyl phorbol-13 acetate and retinoic acid in terms of long-term cell viability and neuronal function in the SH-SY5Y cell line. In the presence of serum, staurosporine-treated cells underwent apoptosis, which ultimately resulted in total cell loss. In contrast, when cultured in defined serum-free medium, a cessation of apoptosis occurred after approximately 1 week, at which point viability could be maintained in excess of 1 month. The addition of aurintricarboxylic acid, which has been demonstrated to prevent apoptosis in a variety of cell models, completely prevented both apoptosis and differentiation in staurosporine-treated cells both under serum-supplemented and serum-free conditions. Apoptosis was not prevented by the protein synthesis inhibitor, cycloheximide. The removal of staurosporine from the culture medium after 3 weeks had no effect on cellular morphology, function, or proliferation, indicating that the attained neuronal phenotype was terminal. Voltage-gated calcium channel sensitivity, used as a measurement of neuronal function, was highest in staurosporine-treated cells. On the basis that apoptosis and neurotrophin independence are hallmarks of the maturation of dorsal root ganglion neurons, results suggest that staurosporine-differentiated SH-SY5Y cells may bear a similar phenotype to that found in vivo. Furthermore, this model may provide for an excellent means of obtaining a stable and homogenous population of postmitotic monoaminergic neurons for investigating neuronal function and differentiation.

Caspase inhibition selectively reduces the apoptotic component of oxygen-glucose deprivation-induced cortical neuronal cell death

Cultured mouse cortical neurons undergo apoptosis when exposed to staurosporine. The cell-permeable caspase inhibitor Z-Val-Ala-Asp fluoromethylketone (Z-VAD.FMK) attenuated this death, without altering overall protein synthesis. Z-VAD.FMK also attenuated cortical neuronal apoptosis induced by removal of serum. However, Z-VAD.FMK did not attenuate the excitotoxic necrosis induced by 5-min exposure to 100 microM NMDA, 24-h exposure to 100 microM kainate, or 90-min exposure to oxygen-glucose deprivation. We have previously shown that blockade of the excitotoxic component of oxygen-glucose deprivation-induced neuronal death with glutamate antagonists unmasks an apoptotic death. Treatment with Z-VAD.FMK, but not the cathepsin-B protease inhibitor Z-Phe-Ala fluoromethylketone (Z-FA.FMK), also attenuated this oxygen-glucose deprivation-induced neuronal apoptosis. These data support the idea that brain caspases mediate the apoptotic component of oxygen-glucose deprivation-induced neuronal death and raise the possibility that combining caspase inhibitors with glutamate antagonists might attenuate brain damage induced by hypoxic-ischemic insults in vivo.

Alzheimer-associated presenilin-2 confers increased sensitivity to apoptosis in PC12 cells

Presenilin-2 is a gene of unknown function recently identified based upon linkage with some forms of familial Alzheimer's disease. To investigate potential effects of PS-2 on cell viability, rat pheochromocytoma (PC12) cells were stably transfected with cDNA constructs encoding either full-length human PS-2 or, for comparison, mouse Bcl-X(L). Overexpression of PS-2 conferred increased sensitivity to the apoptotic stimuli staurosporine and hydrogen peroxide. In contrast, Bcl-X(L) overexpression significantly reduced cell death induced by these stimuli. These results suggest that one function of PS-2 may involve modulation of cell viability.

Cyclosporine induces neuronal apoptosis and selective oligodendrocyte death in cortical cultures

Cyclosporine is used clinically as an immunosuppressant, but carries a risk of central nervous system toxicity due to undefined mechanisms. We examined the ability of cyclosporine exposure to kill cultured mouse cortical neurons and glia. Mixed neuron/glial cultures exposed to 1 to 20 microM cyclosporine for 24 to 48 hours developed concentration-dependent neuronal death, with most neurons destroyed by 20 microM cyclosporine. This neuronal death was characterized by cell body shrinkage and blebbing, chromatin condensation, and internucleosomal DNA fragmentation, consistent with apoptosis. Neuronal death was reduced by addition of cycloheximide, brain-derived neurotrophic factor, or insulin-like growth factor I but not N-methyl-D-aspartate- or AMPA-type glutamate receptor antagonists. Oligodendrocytes were more sensitive to cyclosporine-induced damage than were neurons, but astrocytes were relatively resistant. Oligodendrocyte death was accompanied by positive TUNEL (terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end-labeling) staining and was attenuated by application of ciliary neurotrophic factor or insulin-like growth factor I but not glutamate receptor antagonists. Present observations raise the possibility that the central nervous system toxicity syndrome associated with cyclosporine may be caused by the drug-induced death of oligodendrocytes and neurons.

TGF-beta 1 protects hippocampal neurons against degeneration caused by transient global ischemia. Dose-response relationship and potential neuroprotective mechanisms

BACKGROUND AND PURPOSE

Transforming growth factor-beta 1 (TGF-beta 1) has been shown to rescue cultured neurons from excitotoxic and hypoxic cell death and to reduce infarct size after focal cerebral ischemia in mice and rabbits. The present study investigated the effects of TGF-beta 1 in a different pathophysiological setting and the delayed neuronal death of hippocampal pyramidal cells after transient global ischemia in rats, and evaluated the potential mechanisms of the neuroprotective activity of TGF-beta 1.

METHODS

Transient forebrain ischemia was induced in male adult Wistar rats with bilateral occlusion of both common carotid arteries combined with systemic hypotension for 10 minutes. Seven days after ischemia, brains were perfusion-fixed and stained for histological evaluation. TGF-beta 1 or vehicle was injected intracerebroventricularly (ICV; 0.5, 4, and 50 ng) or intrahippocampally (4 ng) 1 hour before ischemia. For in vitro studies, hippocampal neurons were derived from E17 rat embryos and cultured for 10 to 14 days. Cells were exposed to (1) S-nitrosocysteine (SNOC; 30 mumol/L) to induce nitric oxide-induced oxidative injury and (2) staurosporine (0.03 mumol/L) to induce apoptotic cell death.

RESULTS

Transient forebrain ischemia caused extensive degeneration of CA1 hippocampal pyramidal cells in vehicle-treated control animals. Ischemic injury was not significantly reduced after ICV administration of 0.5 ng TGF-beta 1 (71 +/- 7% damaged neurons versus 84 +/- 3% in vehicle-treated controls: n = 9 and 11, respectively; P = .07, Mann-Whitney U test). Administration of 4 ng TGF-beta 1 reduced the percentage of damaged CA1 pyramidal cells from 71 +/- 10% in controls to 52 +/- 7% in TGF-beta 1-treated animals (n = 11 and 12, respectively; P = .04). TGF-beta 1 (4 ng) also produced significant protection when injected directly into the hippocampal tissue. In contrast, ICV administration of 50 ng TGF-beta 1 failed to show a protective effect in two separate sets of experiments. In vitro, a 24-hour pretreatment of the cultured hippocampal neurons with TGF-beta 1 (0.1 to 10 ng/mL) significantly inhibited both nitric oxide and staurosporine neurotoxicity. Posttreatment with TGF-beta 1 (10 ng/mL) also inhibited staurosporine neurotoxicity but actually potentiated nitric oxide-induced neuronal injury.

CONCLUSIONS

We demonstrated that TGF-beta 1 in a surprisingly low dose range has the capacity to reduce injury to CA1 hippocampal neurons caused by transient global ischemia in rats. This protective action could well be associated with the antioxidative and antiapoptotic effects of TGF-beta 1 demonstrated in vitro.

Induction of neuronal apoptosis by camptothecin, an inhibitor of DNA topoisomerase-I: evidence for cell cycle-independent toxicity

Camptothecin is an S-phase-specific anticancer agent that inhibits the activity of the enzyme DNA topoisomerase-I (topo-I). Irreversible DNA double-strand breaks are produced during DNA synthesis in the presence of camptothecin, suggesting that this agent should not be toxic to nondividing cells, such as neurons. Unexpectedly, camptothecin induced significant, dose-dependent cell death of postmitotic rat cortical neurons in vitro; astrocytes were more resistant. Aphidicolin, an inhibitor of DNA polymerase alpha, did not prevent camptothecin-induced neuronal death, while death was prevented by actinomycin D and 5,6-dichloro-1-beta-D-ribofuranosyl benzimidazole as well as cycloheximide and anisomycin, inhibitors of RNA and protein synthesis, respectively. Camptothecin-induced neuronal death was apoptotic, as characterized by chromatin condensation, cytoplasmic shrinking, plasma membrane blebbing, and fragmentation of neurites. DNA fragmentation was also confirmed by the use of the in situ DNA end labeling assay. In addition, aurintricarboxylic acid, an inhibitor of the apoptotic endonuclease, partially protected against camptothecin-induced neuronal death. The toxicity of stereoisomers of a camptothecin analogue was stereospecific, demonstrating that toxicity was a result of inhibition of topo-I. The difference in sensitivity to camptothecin between neurons and astrocytes correlated with their transcriptional activity and level of topo-I protein expression. These data indicate important roles for topo-I in postmitotic neurons and suggest that topo-I inhibitors can induce apoptosis independent of DNA synthesis. We suggest a model based on transcriptionally mediated DNA damage, a novel mechanism of action of topo-I poisons.

The role of zinc in selective neuronal death after transient global cerebral ischemia

Zinc is present in presynaptic nerve terminals throughout the mammalian central nervous system and likely serves as an endogenous signaling substance. However, excessive exposure to extracellular zinc can damage central neurons. After transient forebrain ischemia in rats, chelatable zinc accumulated specifically in degenerating neurons in the hippocampal hilus and CA1, as well as in the cerebral cortex, thalamus, striatum, and amygdala. This accumulation preceded neurodegeneration, which could be prevented by the intraventricular injection of a zinc chelating agent. The toxic influx of zinc may be a key mechanism underlying selective neuronal death after transient global ischemic insults.

Contrasting effects of protein synthesis inhibition and of cyclic AMP on apoptosis in the developing retina

The role of protein synthesis in apoptosis was investigated in the retina of developing rats. In the neonatal retina, a ganglion cell layer, containing neurons with long, centrally projecting axons, is separated from an immature neuroblastic layer by a plexiform layer. This trilaminar pattern subsequently evolves to five alternating cell and plexiform layers that constitute the mature retina and a wave of programmed neuron death sweeps through the layers. Apoptosis due to axon damage was found in ganglion cells of retinal explants within 2 days in vitro and was prevented by inhibition of protein synthesis. Simultaneously, protein synthesis blockade induced apoptosis among the undamaged cells of the neuroblastic layer, which could be selectively prevented by an increase in intracellular cyclic AMP. Both the prevention and the induction of apoptosis among ganglion cells or neuroblastic cells, respectively, occurred after inhibition of protein synthesis in vivo. The results show the coexistence of two mechanisms of apoptosis within the organized retinal tissue. One mechanism is triggered in ganglion cells by direct damage and depends on the synthesis of proteins acting as positive modulators of apoptosis. A distinct, latent mechanism is found among immature neuroblasts and may be repressed by continuously synthesized negative modulators, or by an increase in intracellular cyclic AMP.