Glycogen synthase kinase-3beta facilitates staurosporine- and heat shock-induced apoptosis. Protection by lithium.

The potential role of glycogen synthase kinase-3beta in modulating apoptosis was examined in human SH-SY5Y neuroblastoma cells. Staurosporine treatment caused time- and concentration-dependent increases in the activities of caspase-3 and caspase-9 but not caspase-1, increased proteolysis of poly(ADP-ribose) polymerase, and induced morphological changes consistent with apoptosis. Overexpression of glycogen synthase kinase-3beta to levels 3.5 times that in control cells did not alter basal indices of apoptosis but potentiated staurosporine-induced activation of caspase-3, caspase-9, proteolysis of poly(ADP-ribose) polymerase, and morphological changes indicative of apoptosis. Inhibition of glycogen synthase kinase-3beta by lithium attenuated the enhanced staurosporine-induced activation of caspase-3 in cells overexpressing glycogen synthase kinase-3beta. In cells subjected to heat shock, caspase-3 activity was more than three times greater in glycogen synthase kinase-3beta-transfected than control cells, and this potentiated response was inhibited by lithium treatment. Thus, glycogen synthase kinase-3beta facilitated apoptosis induced by two experimental paradigms. These findings indicate that glycogen synthase kinase-3beta may contribute to pro-apoptotic-signaling activity, that inhibition of glycogen synthase kinase-3beta can contribute to anti-apoptotic-signaling mechanisms, and that the neuroprotective actions of lithium may be due in part to its inhibitory modulation of glycogen synthase kinase-3beta.

The potential role of glycogen synthase kinase-3␤ in modulating apoptosis was examined in human SH-SY5Y neuroblastoma cells. Staurosporine treatment caused time-and concentration-dependent increases in the activities of caspase-3 and caspase-9 but not caspase-1, increased proteolysis of poly(ADP-ribose) polymerase, and induced morphological changes consistent with apoptosis. Overexpression of glycogen synthase kinase-3␤ to levels 3.5 times that in control cells did not alter basal indices of apoptosis but potentiated staurosporine-induced activation of caspase-3, caspase-9, proteolysis of poly(ADP-ribose) polymerase, and morphological changes indicative of apoptosis. Inhibition of glycogen synthase kinase-3␤ by lithium attenuated the enhanced staurosporine-induced activation of caspase-3 in cells overexpressing glycogen synthase kinase-3␤. In cells subjected to heat shock, caspase-3 activity was more than three times greater in glycogen synthase kinase-3␤-transfected than control cells, and this potentiated response was inhibited by lithium treatment. Thus, glycogen synthase kinase-3␤ facilitated apoptosis induced by two experimental paradigms. These findings indicate that glycogen synthase kinase-3␤ may contribute to proapoptotic-signaling activity, that inhibition of glycogen synthase kinase-3␤ can contribute to anti-apoptotic-signaling mechanisms, and that the neuroprotective actions of lithium may be due in part to its inhibitory modulation of glycogen synthase kinase-3␤.
Glycogen synthase kinase-3 (GSK-3) 1 was initially identified as a kinase that phosphorylates glycogen synthase (1). Subsequent studies have demonstrated that GSK-3 surpasses this function and plays a broad role in cellular metabolism, including contributions to signaling activities, growth, and differentiation (2). GSK-3␤ has been shown to phosphorylate numerous substrates, including several transcription factors such as c-jun, c-myc (3)(4)(5), and heat shock factor-1 (6), cytoskeletal proteins such as the microtubule-associated protein tau (7,8), and the multifunctional protein ␤-catenin (9). Thus it is now evident that the activity of GSK-3␤ influences a wide variety of cellular functions, including multiple signaling systems.
Much still remains to be learned about the regulation of GSK-3␤ activity and its role as a modulator of signaling cascades that determine cell fate. Although often considered to be a constitutively active enzyme, GSK-3␤ can be both activated and inhibited. Activation has been shown to occur subsequent to phosphorylation of Tyr 216 (10) and recently by transient increases in intracellular calcium (11). Inhibition of GSK-3␤ can be induced by activation of the Wnt pathway (12) or by agents that activate a signaling cascade that commences when growth factors or insulin bind to their respective receptors (see Ref. 13 for review), resulting in the recruitment and activation of phosphatidylinositol 3-kinase. Activated phosphatidylinositol 3-kinase catalyzes the production of phosphatidylinositol 3,4,5-trisphosphate, which binds the pleckstrin homology domain of Akt (also known as protein kinase B) to bring it into close proximity with phosphoinositide-dependent kinase-1. The juxtaposition of phosphoinositide-dependent kinase-1 to Akt on the membrane facilitates the phosphorylation and activation of Akt by phosphoinositide-dependent kinase-1 (14). Subsequently, Akt dissociates from the membrane and can phosphorylate Ser 9 of GSK-3␤, which inhibits its activity (15). Activation of the phosphatidylinositol 3-kinase/Akt-signaling pathway protects cells from pro-apoptotic stimuli as well as reducing the activity of GSK-3␤. For example, activators of phosphatidylinositol 3-kinase and Akt, such as insulin-like growth factor-1, platelet-derived growth factor (16,17), and interleukin-2 (18) and -3 (19,20), protect cells from a variety of apoptotic insults. Thus, the signaling mechanism that is associated with inhibition of GSK-3␤ is also associated with antiapoptotic outcomes. Akt-mediated cell protection has been attributed to processes other than inhibition of GSK-3␤, such as by phosphorylation of the proapoptotic Bcl family member Bad (21) or by preventing the release of cytochrome c from mitochondria (22), but it is not known whether or not inhibition of GSK-3␤ contributes to the anti-apoptotic effects of Akt activity. There is some evidence of the converse, that activation of GSK-3␤ contributes to pro-apoptotic signaling, as it was recently found that overexpression of GSK-3␤ in Rat-1 and PC12 cells stimulated apoptosis (23). Considering the potentially important role of GSK-3␤ in regulating apoptosis, it was of great interest to note that lithium was recently discovered to inhibit of GSK-3␤ (24,25). Lithium is used therapeutically for the treatment of bipolar disorder, and although it has been used in the psychiatric domain for many years, its influences at the biochemical level are only beginning to be elucidated (26). One of the most intriguing findings is that lithium confers protection to neurons against pro-apoptotic stimuli such as glutamate-induced excitotoxicity (27), C2-ceramide (28), radiation (29), and ischemia (30). Taken together, these findings raise the possibilities that GSK-3␤ contributes to apoptotic-signaling cascades and that inhibition of GSK-3␤ contributes to the neuroprotective properties of lithium.
The enzymes that ultimately carry out the command for apoptosis are the cysteine proteases known as caspases. Caspases, which are zymogens, are typically cleaved autocatalytically or by other caspases from inactive procaspase proteins to produce activated enzymes (31,32). Caspase-3, also called CPP32, is activated by many pro-apoptotic stimuli and is an early step in the execution phase of apoptosis (33). The activation of caspase-3 commences after apoptotic signals induce the release of cytochrome c from the mitochondrial intermembrane space (34), which subsequently associates with apoptotic protease-activating factor-1 and procaspase-9 to form the "apoptosome" (35). This complex formation stimulates the oligomerization of procaspase-9 and its autocatalytic activation. The effect of caspase-9 activity is the proteolytic activation of downstream caspases such as caspase-3 (36), which in turn proteolyzes the DNA-binding protein poly(ADP-ribose) polymerase (PARP) (37) and other proteins. Hence the measurement of caspase-3 activity can serve as a biochemical marker for the execution phase of apoptosis.
The goal of this investigation was to test if GSK-3␤ activity modulates apoptosis using the neuronal model system of human neuroblastoma SH-SY5Y cells. Apoptosis was generated using staurosporine, which previously has been demonstrated to induce apoptosis in these and other cells (38 -40), and heat shock, a method widely used to cause cell stress (41,42). GSK-3␤ activity was increased by overexpression, and GSK-3␤ activity was decreased by using lithium. The results show that overexpression of GSK-3␤ alone did not induce apoptosis, but it sensitized cells to apoptosis caused by exposure to staurosporine or to heat shock, and that inhibition of GSK-3␤ by lithium attenuated activation of caspase-3.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfection of Cells-Human neuroblastoma SH-SY5Y cells were grown in continuous culture RPMI media (Cellgro, Herndon, VA) containing 10% horse serum (Life Technologies), 5% fetal clone II (Hyclone, Logan, UT), 2 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. Stably transfected cells were maintained with transfection-maintenance medium that contained 100 g/ml G418 (Geneticin); other components were the same as continuous culture media. Cells were maintained in humidified, 37°C incubators with 5% CO 2 . Cells were plated at a density of 10 5 cells/60-mm dish 48 h before apoptosis-inducing treatments.
For stable transfection, SH-SY5Y cells were replated one day before electroporation. The cells were removed from the dish with continuous culture medium containing 0.05% of trypsin, diluted with continuous culture medium, and centrifuged for 5 min at 250 ϫ g. Cells were washed, resuspended in 1 ml HEPES buffer (0.14 M NaCl, 25 mM HEPES, 0.075 mM Na 2 HPO 4 , pH was adjusted to 7.05 with 10 M NaOH), and incubated with 10 g of HA-GSK-3␤ in pcDNA3.1 (generously provided by Dr. J. R. Woodgett, University of Toronto) on ice for 10 min. Electroporation was carried out with a Bio-Rad Pulse II electroporator set at 0.25 kV and 960 microfarads as described previously (43). After electroporation, cells were incubated on ice for 10 min and then mixed with 10 ml of continuous culture medium and plated onto a 100-mm Corning dish. After 48 h the medium was replaced with transfectionmaintenance medium, and the cells were maintained in transfectionmaintenance medium for approximately 1 month until only cells resistant to Geneticin survived. Cells were cloned and screened for expression of HA-GSK-3␤.
Collection of Lysates-For immunoblotting, cells in 60-mm plates were washed twice with phosphate-buffered saline and were lysed with 100 l of lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 1 mM sodium orthovanadate, 100 M phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml aprotinin, 5 g/ml pepstatin, and 0.2% Nonidet P-40). For caspase activity, cells were lysed with 100 l of lysis buffer without sodium orthovanadate. The lysates were collected in microcentrifuge tubes, sonicated, and centrifuged. Protein concentrations in the supernatants were determined using the bicinchoninic acid (BCA) method (Pierce). The lysates were stored at Ϫ80°C until used either for immunoblotting or measuring caspase activity.
GSK-3␤ Activity-The activity of GSK-3␤ was measured essentially as described previously (44,45). Cells were lysed in immunoprecipitation lysis buffer (20 mM Tris, pH 7.5, 0.2% Nonidet P-40, 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 1 mM sodium orthovanadate, 100 M phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml aprotinin, 5 g/ml pepstatin, 1 nM okadaic acid, 100 mM sodium fluoride, and 1 mg/ml glycogen). The lysates were sonicated in microcentrifuge tubes for 10 s on ice and centrifuged at 20,800 ϫ g for 15 min. After the protein concentration was determined, 100 g of protein (1 g/l) was precleared with 40 l of protein G-Sepharose beads for 3 h at 4°C, then incubated with 2 g of monoclonal GSK-3␤ antibody (Pharmingen/ Transduction Laboratories) overnight at 4°C with gentle agitation. The immobilized immune complexes were washed three times with immunoprecipitation lysis buffer and once with 20 mM Tris, pH 7.5, 5 mM MgCl 2 , 1 mM dithiothreitol. Kinase activity was assayed in a total volume of 15 l of kinase buffer containing 20 mM Tris, pH 7.5, 5 mM MgCl 2 , 1 mM dithiothreitol, 250 M ATP, 1.4 Ci of [␥-32 P]ATP, and 100 M phosphoglycogen synthase peptide-2 (YRRAAVPPSPSLSRH-SSPHQSEDEEE) (Upstate Biotechnology, Inc., Lake Placid, NY). Glycogen synthase (Ala21) peptide-2 was used as negative control. The samples were incubated at 30°C for 30 min, the reaction tubes were centrifuged for 1 min, and triplicate 5-l aliquots of reaction supernatants were spotted onto 1 cm ϫ 2 cm P81 filter paper. The filter papers were washed 4 times in 0.5% phosphoric acid for a total time of 1 h, rinsed in 95% ethanol, air-dried, and counted in a liquid scintillation counter. The efficiency of GSK-3␤ immunoprecipitation was determined by immunoblotting for GSK-3␤.
Caspase Activity-Fluorometric assays were conducted in 96-well clear-bottom plates, and all measurements were carried out in triplicate wells. To each well 200 l of assay buffer (20 mM HEPES, pH 7.5, 10% glycerol, 2 mM dithiothreitol) was added. Peptide substrates for caspase-3 (Ac-DEVD-AMC), caspase-9 (Ac-LEHD-AMC), or caspase-1 (Ac-YVAD-AMC) (Alexis Biochemicals, San Diego, CA) were added to each well to a final concentration of 25 ng/l. When the caspase-3 inhibitor (Ac-DEVD-CHO) was used, it was added at a concentration of 2.5 ng/l immediately before the addition of the caspase-3 substrate. Cell lysates (20 g of protein) were added to start the reaction. Background fluorescence was measured in wells containing assay buffer, substrate, and lysis buffer without the cell lysate. Assay plates were incubated at 37°C for 1 h for measurement of caspase-3 and 3 h for measurement of caspase-1 and caspase-9, based on preliminary measurements of the time course (0.5 to 6 h) of caspase activities. Fluorescence was measured on a fluorescence plate reader (Bio-Tek, Winooski, VT) set at 360 nm excitation and 460 nm emission. Caspase activity was calculated as ((mean AMC fluorescence from triplicate wells) Ϫ (background fluorescence))/g of protein.
Nuclear Staining-Cells were cultured on poly-D-lysine-coated glass coverslips placed in 35-mm culture plates. After treatments, medium was removed, and the cells were fixed in 2% paraformaldehyde for 1 h at room temperature. After two washes with phosphate-buffered saline, cells were stained with 10 ng/l Hoechst 33342 (Molecular Probes, Eugene, OR) for 1 h at room temperature. The coverslips were rinsed twice with phosphate-buffered saline and then mounted onto glass slides using Immu-Mount (Shandon, Pittsburgh, PA). The slides were examined by fluorescence microscopy (Nikon) set at 400ϫ magnification. To score the number of cells exhibiting apoptotic morphology, a minimum of 300 cells were counted/coverslip.

Induction of Caspase-3 Activity and Apoptosis-SH-SY5Y
cells were treated with staurosporine and then examined for characteristics associated with apoptosis, including activation of caspases, proteolysis of PARP, and changes in cellular mor-

GSK-3␤ Facilitates Apoptosis
phology. Caspase-3 activity increased concentration dependently after treatment with 0.1 to 1 M staurosporine (Fig. 1A). Examination of the time-dependent activation of caspases induced by 0.5 M staurosporine revealed that caspase-3 activity began to increase approximately 1 h after staurosporine treatment and reached maximum activation after 4 -5 h (Fig. 1B). The specificity of the caspase-3 assay was confirmed by the use of a caspase-3 inhibitor, which resulted in the complete inhibition of caspase-3 activity. Caspase-9 and caspase-1 were also evaluated to test if other caspases were activated by staurosporine treatment. Caspase-9 activity began to increase after 1 h of staurosporine treatment and reached a maximum within 3 h (Fig. 1C), but caspase-1 was not activated. The time-dependent proteolytic cleavage of PARP to the 85-kDa fragment after treatment with 0.5 M staurosporine corresponded to the timedependent activation of caspase-3 (Fig. 1D). Examination of cells treated with 0.5 M staurosporine for 8 h and stained with Hoechst 33342 revealed the characteristic morphology associated with apoptosis, such as nuclear condensation and cell shrinkage (Fig. 1E).
The staurosporine concentration-dependent activation of caspase-3 was measured 3 h after treatment in vector-transfected cells and in the four subclones of HA-GSK-3␤-transfected cells. Basal caspase-3 activities in vector-and HA-GSK-3␤-transfected cells were not different from control SH-SY5Y cells. In vector-transfected cells, activation of caspase-3 by staurosporine (Fig. 3A) was equivalent to that obtained in untransfected SH-SY5Y cells (Fig. 1). In marked contrast, overexpression of HA-GSK-3␤ greatly potentiated staurosporineinduced activation of caspase-3 in all four cells lines overexpressing GSK-3␤. The caspase-3 activity in HA-GSK-3␤transfected cells averaged 280% and 250% of that in vectortransfected cells after treatment with 0.1 and 0.5 M staurosporine, respectively (Fig. 3A). Furthermore, caspase-9 activity in HA-GSK-3␤-transfected cells averaged 180% and 245% of that in vector-transfected cells after treatment with 0.1 and 0.5 M staurosporine, respectively (Fig. 3B). The potentiation of staurosporine-induced caspase-3 activity by GSK-3␤ overexpression was confirmed by a greater degree of PARP proteolysis (Fig. 3C) and greater numbers of cells exhibiting apoptotic morphology (Fig. 3, D and E). These results indicate that overexpression of HA-GSK-3␤ increases the sensitivity of cells to staurosporine-induced apoptosis.
Lithium Attenuates Staurosporine-induced Caspase-3 Activation-Cells were pretreated with lithium to inhibit GSK-3␤ to test if inhibition of GSK-3␤ attenuates staurosporine-induced activation of caspase-3 in control or HA-GSK-3␤-transfected cells. The large differences in the concentration of staurosporine that activated caspase-3 in control and HA-GSK-3␤transfected cells made it necessary to use different protocols for each cell line, so caspase-3 was measured in untransfected cells 4 h after treatment with 0.5 M staurosporine and in HA-GSK-3␤-transfected cells 3 h after treatment with 0.1 M staurosporine. Untransfected SH-SY5Y cells were pretreated with 1.25, 2.5, and 5 mM lithium for 24 h and then treated with 0.5 M staurosporine for 4 h. Pretreatment of control SH-SY5Y cells with 5 mM lithium significantly decreased staurosporine-induced caspase-3 activity by 28% (Fig. 4A). HA-GSK-3␤-transfected cells were pretreated for 24 h (Fig. 4B) or were treated chronically for 7 days (Fig. 4C) with 1.25, 2.5, or 5 mM lithium. Pretreatment for 24 h with 5 mM lithium significantly reduced the staurosporine-induced caspase-3 activity by 40%. Chronic treatment with 1.25, 2.5, and 5 mM lithium resulted in reductions of 23%, 31%, and 65%, respectively, of caspase-3 activity induced by staurosporine. These results demonstrate that the facilitatory effect of GSK-3␤ on staurosporine-induced caspase-3 activity is attenuated by lithium.
GSK-3␤ Potentiates Heat Shock-induced Activation of Caspase 3-To test if the potentiation by HA-GSK-3␤ and attenuation by lithium of caspase-3 activation occur with another apoptotic paradigm, the effects of these treatments were measured in cells subjected to heat shock (30 min at 45°C) followed by incubation at 37°C, a model widely used to study the responses of cells to stress (41,42). In vector-transfected cells caspase-3 activity increased to 425% and 975% that of the basal activity following 30 min and 90 min of incubation at 37°C, respectively (Fig. 5A). In HA-GSK-3␤-transfected cells, caspase-3 activity increased to 1500% and 3100% that of the basal activity after 30 and 90 min of incubation at 37°C, respectively. Pretreatment of HA-GSK-3␤-transfected cells with 1.25, 2.5, and 5 mM Li for 24 h reduced the heat shockinduced caspase-3 activation by 13%, 25%, and 56%, respectively. These results demonstrate that overexpression of GSK-3␤ potentiates the activation of caspase-3 after treatment with heat shock and that pretreatment with lithium, a GSK-3␤ inhibitor, attenuates this effect. DISCUSSION In marked contrast to its modest beginning as a regulator of glycogen synthesis, GSK-3␤ has been found to participate in a remarkable number of signaling pathways, which, based on the findings reported here, include the ultimate decision between cell death and survival. Thus, cells overexpressing GSK-3␤ succumbed much more rapidly to the apoptosis-inducing actions of staurosporine or exposure to elevated temperature than did cells with a normal complement of GSK-3␤. Furthermore, lithium, an inhibitor of GSK-3␤, counteracted the facilitation of apoptosis caused by overexpression of GSK-3␤. These findings demonstrate that initial steps in the apoptotic signaling involving activation of caspases can be influenced by GSK-3␤.
Staurosporine is one of the most commonly used agents to experimentally induce apoptosis, and apoptosis occurs in essentially all cell types exposed to appropriate concentrations of staurosporine, suggesting that it activates a cell death program common to all cells (38). Staurosporine previously has been reported to induce apoptosis in human neuroblastoma SH-SY5Y cells (40,46). In these cells, 0.5 M staurosporine was reported to increase caspase-3 activity, result in PARP proteolysis, and cause morphological changes indicative of apoptosis

GSK-3␤ Facilitates Apoptosis 7587
within a few hours of treatment (40), effects also observed in this investigation. Furthermore, we observed approximately a linear concentration-dependent activation of caspase-3 within the range of 0.1 to 1 M staurosporine. In cells overexpressing GSK-3␤, the staurosporine concentration-dependent activation of caspase-3 and of caspase-9 was shifted to the left, so greater activation of these caspases were achieved after treatment with low concentrations of staurosporine. These findings indicated that GSK-3␤ facilitated staurosporine-induced apoptosis. This conclusion was further confirmed by the findings of greater PARP proteolysis and morphological changes denoting apoptotic cells in GSK-3␤-transfected cells than control cells after exposure to staurosporine. Furthermore, the GSK-3␤ inhibitor, lithium, attenuated the facilitation of staurosporine-induced caspase-3 activity in cells overexpressing GSK-3␤. Taken together, these results demonstrate that GSK-3␤ is a facilitator of the apoptosis-signaling cascade induced by staurosporine.
Our results extend those of Pap and Cooper (23), who recently presented evidence that GSK-3␤ is proapoptotic. In their studies, transient transfection of GSK-3␤ in PC12 cells and Rat-1 fibroblasts caused 60 to 70% of cells to spontaneously undergo apoptosis with no additional treatment within 48 h in a caspase-3-dependent manner. Our results differ from those of Pap and Cooper (23) in that a 3.5-fold overexpression of GSK-3␤ alone failed to alter caspase-3 activity or cell survival, a difference that may be due to differential cell susceptibilities to the action of GSK-3␤ or to differences in the expression levels of GSK-3␤, which were not reported by Pap and Cooper (23). Regardless of this difference, both studies clearly support the conclusion that GSK-3␤ is an important modulator of cell survival.
In addition to staurosporine-induced apoptosis, overexpression of GSK-3␤ potentiated, and lithium treatment attenuated, caspase-3 activation following subjection of cells to heat shock. A variety of strategic responses are initiated in cells to limit the deleterious consequences of stressors, as modeled by heat shock, such as alterations in the activities of signaling protein kinases and induction of the expression of heat shock proteins (see Ref. 47 for review). In SH-SY5Y cells as well as many other types of cells, the responses to a 30-min period of heat shock (45°C) are adequate to support survival; thus, little caspase-3 activation was observed following heat shock in control cells in this study. Both activation of phosphatidylinositol 3-kinase (48) and inactivation of GSK-3␤ (49) have been reported to follow heat shock, responses that may contribute, along with others, to protection from heat shock. The results of the present study suggest that adequate inactivation of GSK-3␤ is critical for cell survival, as excessive GSK-3␤ resulted in a massive activation of caspase-3 following heat shock, whereas inhibition of GSK-3␤ by lithium in cells overexpressing HA-GSK-3␤ protected cells from heat shock-induced caspase-3 activation.
The results presented here suggest two complementary conclusions, that activation of GSK-3␤ facilitates apoptosis, and that the anti-apoptotic actions of agents that stimulate the phosphatidylinositol 3-kinase/Akt pathway (17) may be due in part to the inhibitory effect of phosphatidylinositol 3-kinase/ Akt signaling on GSK-3␤ activity. The mechanism by which decreased GSK-3␤ activity contributes to neuroprotection and increased GSK-3␤ activity contributes to apoptosis remains unknown. One possible mechanism is the regulation by GSK-3␤ of ␤-catenin. Increased GSK-3␤ activity facilitates degradation of ␤-catenin (50), and reduced ␤-catenin and the associated reduction in the activity of Tcf/Lef transcription factors has been linked to decreased cell survival (51)(52)(53). Additionally, we have found that activation of the heat shock factor-1 transcription factor and the associated expression of heat shock protein-70, which is known to protect against cell death (54,55), were impaired by expression of GSK-3␤ and restored by lithium treatment (56). However, since GSK-3␤ affects a large number of signaling systems, further investigation is necessary to identify those that are critical for its facilitatory action on apoptosis.
Regardless of the mechanism by which GSK-3␤ facilitates apoptosis, it is evident that inhibition of GSK-3␤ by lithium reduced caspase-3 activation after both staurosporine and heat shock treatments of GSK-3␤-transfected cells. Although lithium has been reported to affect a variety of other targets (26) that cannot completely be discounted as contributory, the accentuated protective effects of lithium in GSK-3␤-transfected cells suggests that inhibition of GSK-3␤ accounts for the protection from apoptosis conferred by lithium. During the last few years several studies (reviewed in Ref. 26), especially those from Chuang and co-workers, have shown that lithium protects neurons from the deleterious effects of a wide variety of insults, such as ischemia (30) and activation of excitatory amino acid receptors (27,57). The findings in this study raise the possibility that these neuroprotective actions of lithium may occur at least in part because of its capacity to inhibit GSK-3␤.
In summary, the results reported here demonstrate for the first time that modest increases in GSK-3␤ facilitate apoptosis in two model systems, including apoptosis induced by staurosporine and by heat shock. Furthermore, the inhibitory effect of lithium on GSK-3␤ and its attenuation of GSK-3␤-facilitated apoptosis suggest that some of the widely reported neuroprotective effects of lithium may result from this action.