TRAIL-related neurotoxicity implies interaction ... - Wiley Online Library

JOURNAL OF NEUROCHEMISTRY

| 2008 | 105 | 1915–1923

doi: 10.1111/j.1471-4159.2008.05291.x

Department of Experimental and Clinical Pharmacology, University of Catania School of Medicine, Catania, Italy

Abstract Tumor necrosis factor related apoptosis inducing ligand (TRAIL) is involved in amyloid beta dependent neurotoxicity via the extrinsic pathway. Recently, several genes modulating TRAIL cytotoxicity have been characterized, providing evidence for a role of wingless-type mouse mammary tumor virus integration site family (Wnt), Jun-N-terminal kinase and other pathways in increased cell susceptibility to the cytokine. We investigated whether neurotoxic effects of TRAIL could be due to modulation of the Wnt signaling pathway. Western blot analysis of Wnt in SH-SY5Y human neuroblastoma cells showed significantly decreased Wnt expression in cultures treated with TRAIL. Correspondingly, both phosphorylation of glycogen synthase kinase 3 beta and degradation of cytoplasmic b-catenin were increased, as well as phosphorylation of the s protein, bringing about the picture of neuronal damage. As a counterproof of the interaction of TRAIL with the

Wnt pathway, the addition of the specific glycogen synthase kinase 3 beta inhibitor SB216763 resulted in rescue of a significant percent of cells from TRAIL-induced apoptosis. The rescue was total when the caspase 8 inhibitor z-IETD-FMK was added in combination with SB216763. Results show that, probably, in addition to triggering caspase signaling, TRAIL also interferes with the Wnt pathway, additionally concurring to neuronal damage. These data suggest that the Wnt pathway substantially contributes to the TRAIL-related neurotoxicity and indicate the TRAIL system as a candidate target for pharmacological treatment of Alzheimer’s disease and related disorders. Keywords: apoptosis, glycogen synthase kinase 3 beta, inflammation, neurodegeneration, pharmacological therapy, signal transduction. J. Neurochem. (2008) 105, 1915–1923.

Although an array of factors have been designated to contribute to the pathogenesis of Alzheimer’s disease (AD), a pivotal role has been attributed to amyloid beta (Ab), an anomalous protein abundantly produced in the AD brain, and organized in insoluble plaques and highly neurotoxic neurofibrillary tangles (Selkoe 2000). Growing evidence demonstrates that neuroinflammation occurs in vulnerable regions of the AD brain, where Ab accumulates fuelling the inflammatory process, which likely starts as a host defense response to damage induced by Ab (Tuppo and Arias 2005). The human neuronal cell-like line originating from the differentiated neuroblastoma cells SH-SY5Y, incubated with Ab, undergoes apoptosis and actively releases substantial amounts of Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL), a proapoptotic cytokine which binds to death receptors belonging to the tumor necrosis factor superfamily, DR4 and DR5 (Pan et al.1997). The proapoptotic effects of TRAIL upon human neurons are associated with activation of the caspase-related signaling pathway

(Cantarella et al. 2003). Degenerating neurons in the AD brain express a wide repertoire of proteins in response to Ab injury (Ho et al. 2005). Molecules such as cycline-dependent kinase 5 (Cdk5) and glycogen synthase kinase 3 beta (GSK-3b), represent the two main enzymes responsible for phosphorylation of the s protein, typically associated with

Received November 23, 2007; revised manuscript received January 15, 2008; accepted January 28, 2008. Address correspondence and reprint requests to Renato Bernardini MD, Department of Experimental and Clinical Pharmacology, University of Catania School of Medicine, Viale Andrea Doria 6, I-95125 Catania, Italy. E-mail: [email protected] Abbreviations used: AD, Alzheimer’s disease; AO, acridine orange; Ab, amyloid beta; Cdk5, cycline-dependent kinase 5; EB, ethidium bromide; ECL, enhanced chemiluminescence; GSK-3b, glycogen synthase kinase 3 beta; JNK, Jun-N-terminal kinase; PCP, planar cell polarity; SDS, sodium dodecyl sulfate; TRAIL, Tumor necrosis factor Related Apoptosis Inducing Ligand; Wnt, wingless-type mouse mammary tumor virus integration site family.

2008 The Authors Journal Compilation 2008 International Society for Neurochemistry, J. Neurochem. (2008) 105, 1915–1923

1915

1916 | G. Cantarella et al.

the increased apoptotic rate in the AD brain (Liu et al. 2002). With regard to GSK-3b, the activity of this enzyme is generally down-regulated by different signal transduction pathways (Wang et al. 2006), including the wingless-type mouse mammary tumor virus integration site family (Wnt) pathway (Brown and Moon 1998). Physiologically, inhibition of GSK-3b is associated with reduced phosphorylation of s protein and accumulation of b-catenin (Lucas et al. 2001). Recently, Aza-Blanc et al. (2003), using an RNAibased forward genomics approach aimed to understand the biological mechanism of TRAIL-induced apoptosis, have identified a variety of known and previously uncharacterized genes which modulate TRAIL activity, and concluded that Wnt, Jun-N-terminal kinase (JNK) and other pathways play a relevant role in the susceptibility of HeLa cells to TRAIL. In the light of the powerful effects of TRAIL in mediating Ab neurotoxicity (Cantarella et al. 2003), as well as of such relationships between the TRAIL system and the Wnt signaling pathway (Aza-Blanc et al. 2003), it appeared of interest to investigate whether TRAIL, in addition to setting into motion its extrinsic pathway, could additionally modulate the Wnt signaling pathway to elicit its neurotoxic effect. To do so, we analyzed possible involvement of single molecular events of the Wnt pathway in SH-SY5Y differentiated human neuroblastoma cells treated with TRAIL.

Materials and methods Reagents Tissue culture media were from Invitrogen Srl (San Giuliano Milanese, Italy). TRAIL was purchased from Alexis Biochemicals (San Diego, CA, USA). The GSK-3b inhibitor SB216763 [3-(2,4dichlorophenyl)-4-(1-methyl-1H-indol-3-yl)-1H-pyrrole-2,5-dione] Tocris Cookson Inc. (Ellisville, MO, USA). The caspase 8 inhibitor z-IETD-FMK was purchased from Alexis Biochemicals. Primary antibodies were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA) unless otherwise specified. Cell cultures The human neuroblastoma cell line SH-SY5Y was grown in 1 : 1 modified Eagle’s medium: Ham’s F12 supplemented with 10% (v/v) fetal bovine serum, 2 mmol/L glutamine, 50 lg/mL penicillin, and 100 lg/mL streptomycin and kept at 37C in humidified 5% CO2/ 95% atmosphere. For differentiation, cultured cells were treated for 1 week with 10 lmol/L retinoic acid. Western blot analysis Cells were harvested in 100 lL of lysis buffer containing 50 mmol/ L Tris, pH 7.6, 150 mmol/L NaCl, 5 mmol/L EDTA, 1 mmol/L phenyl methyl sulfonyl fluoride, 0.5 lg/lL leupeptin, 5 lg/lL aprotinin, and 1 lg/mL pepstatin. The samples were centrifuged at 15 000 g for 30 min at 4C. The resulting supernatants were isolated and protein concentration was determined by the Bradford (1976) method. Cellular protein (30 lg) was mixed with an a equal volume of sodium dodecyl sulfate (SDS) loading buffer [20% glycerol,

100 mmol/L Tris–HCl (pH 6.8), 200 mmol/L dithiothreitol, 4% SDS, and 0.1% bromophenol blue], boiled for 5 min, and separated by electrophoresis in 8% and 12% polyacrylamide gel. Proteins were transferred from acrylamide gels onto Hybond enhanced chemiluminescence (ECL) nitrocellulose membranes (Amersham Italia, Milan, Italy). The membranes were blocked with 5% milk in phosphate buffered saline Tween-20 and then incubated at 4C overnight with rabbit anti-Wnt-1 polyclonal antibody, or mouse antiGSK-3b monoclonal antibody, or mouse anti-p-GSK-3b monoclonal antibody (BD Transductions Laboratories, Franklin Lakes, NJ, USA), or rabbit anti-Cdk5 polyclonal antibody, or goat anti-p-Cdk5 polyclonal antibody, or rabbit anti-Tau polyclonal antibody, or rabbit anti-p-Tau polyclonal antibody (Calbiochem, San Diego, CA, USA), or rabbit anti-JNK-1 polyclonal antibody, or mouse p-JNK monoclonal antibody, or rabbit anti-b-tubulin polyclonal antibody, and horse radish peroxidase-conjugated anti-rabbit or anti-mouse or antigoat IgG secondary antibody (Amersham Italia), then detected by ECL (Amersham Italia), chemiluminescence assay (ECL; Amersham Italia). For determination of cytoplasmic and nuclear b-catenin it was used NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology, Inc., Rockford, IL, USA). Nuclear and cytoplasmic protein extract (30 lg) was mixed with an a equal volume of SDS loading buffer [20% glycerol, 100 mmol/L Tris–HCl (pH 6.8), 200 mmol/L dithiothreitol, 4% SDS, and 0.1% bromophenol blue], boiled for 5 min, and separated by electrophoresis in 10% polyacrylamide gel. Proteins were transferred from acrylamide gels onto Hybond ECL nitrocellulose membranes (Amersham Italia). The membranes were blocked with 5% milk in phosphate buffered saline Tween-20 and then incubated at 4C overnight with a mouse anti-b-catenin monoclonal antibody (BD Transductions Laboratories) or with rabbit anti-b-tubulin polyclonal antibody (for normalization of cytoplasmic proteins), or with mouse antiproliferating cell nuclear antigen (for normalization of nuclear proteins), and with the secondary peroxidase-conjugated anti-mouse or anti-rabbit antibodies and finally detected by ECL (Amersham Italia S.r.l.). For validation of blot data, densitometric analysis was performed on immunoblots by using IMAGEJ analysis software (NIH, Bethesda, MD, USA). Cell viability analysis Cell viability was determined by 3-[4,5 dimethylthiazol-2-yl]-2,5diphenyltetrazolium bromide assay. Briefly, SH-SY5Y differentiated cells were plated on 96-well cultured plates at a density of 5 · 103 cells per well. After the treatment, cell viability was measured by the reduction of 3-[4,5 dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide solution (0.5 mg/mL). After 3 h incubation at 37C, this solution was removed, and produced blue formazan crystals were solubilized in dimethylsulfoxide. The optical density of the formed blue formazan was measured at 545 nm (ref. 630 nm). All tests were performed in triplicate, at least twice. Assay of caspase 3 and 8 activity Hundred micrograms of protein lysate were incubated with 200 lmol/L caspase 3 substrate (Ac-DEVD-pNA) or caspase 8 substrate (Ac-IETD-pNA) (Caspase Colorimetric Substrate Set; Alexis) at 37C. Substrate hydrolysis was determined as absorbance change at 405 nm in a microplate reader and fold increase in caspase activity was determined by comparing these results with the level of


Role of the Wnt pathway in TRAIL neurotoxicity | 1917

the uninduced control. All tests were performed in triplicate, at least twice. Acridine orange/ethidium bromide staining Morphological assessment of apoptotic cells was performed using the acridine orange (AO)/ethidium bromide (EB) double-staining method. SH-SY5Y differentiated cells were seeded in 24-well microtiter plates (10 000 cells per well). After 24-h incubation, cells were pre-treated with SB216763 (5 lmol/L) and after 24-h treated with TRAIL (100 ng/mL), and incubated at 37C in a 5% CO2 atmosphere for 48 h. Freshly isolated SH-SY5Y cells were harvested in an Eppendorf centrifuge tube, spun for 5 min at 106 g and suspended in phosphate-buffered saline containing fluorescence dye AO/EB (both AO and EB were at the concentration of 100 mg/L in phosphate-buffered saline). Then, cells were prepared and dropped onto slides. The morphology of the cells was observed under fluorescence light microscope (DMIRB, Leica, Germany) and photographed.

(a)

(b)

Statistical analysis of results Results were analyzed by the one-way ANOVA, followed by the Fisher’s Least Significant Difference test. Significance was set at a p < 0.05.

Results TRAIL modulates Wnt, p-GSK-3b kinase, b-catenin, and p-s protein expression in SH-SY5Y cells In view of the toxicity of TRAIL upon SH-SY5Y cells (Cantarella et al. 2003), as well as of the core role of the Wnt pathway in neurodegenerative processes (Inestrosa et al. 2002), we first investigated whether TRAIL could modulate the expression of Wnt. Western blot analysis of the Wnt protein in SH-SY5Y cells treated with TRAIL showed that the expression of the Wnt protein decreased significantly after 16, 24, and 48 h of incubation (Fig. 1a and c). As inhibition of the Wnt-related signal transduction encompasses activation of the downstream GSK-3b kinase, we verified whether phosphorylation of the latter could vary relatedly to the TRAIL-modulated Wnt expression. Western blot analysis showed a time-dependent increase of the phosphorylated form of the GSK-3b protein in SH-SY5Y cells incubated with TRAIL. Expression of p-GSK-3b augmented after 6 h incubation, increased thereafter, and reached its peak at 48 h (Fig. 1b and c). The breakdown of bcatenin follows phosphorylation of GSK-3b. In fact, incubation of SH-SY5Y cells with TRAIL resulted in degradation of b-catenin marked by significant reduction of its expression, either in the cytoplasm or in the nuclei, as shown by the western blot analysis performed on both cytosolic and nuclear extracts. In both fractions, the decrease of intact bcatenin was inversely related with GSK-3b phosphorylation at all time points studied (Fig. 2). A downstream consequence of b-catenin degradation is hyperphosphorylation of the s protein, a process associated with microtubule damage

(c)

Fig. 1 Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL) modulates the protein expression of wingless-type mouse mammary tumor virus integration site family (Wnt)-1 and p- glycogen synthase kinase 3 beta (GSK-3b) in SH-SY5Y cells. (a) Densitometric analysis of western blot for time-dependent Wnt-1 protein expression in the differentiated human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by the Fisher’s LSD test). (b) Densitometric analysis from western blot for timedependent phosphorylated GSK-3b (p-Tyr216) protein expression in human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by the Fisher’s test). (c) Western blot analysis for time-dependent Wnt-1 (top blot), b-tubulin (second blot), GSK-3b (third blot), and the phosphorylated form of GSK-3b (p-Tyr216) (bottom blot) protein expression in the human neuroblastoma cell line SH-SY5Y treated with TRAIL.



(a)

(a)

(b)

(b)

(c)

(c)

Fig. 2 Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL) modulates cytoplasmic b-catenin degradation in SH-SY5Y cells. (a) Densitometric analysis from western blot for time-dependent cytosolic b-catenin protein expression in the differentiated human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by a Fisher’s LSD test). (b) Densitometric analysis from western blot for time-dependent nuclear b-catenin protein expression in the human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by the Fisher’s LSD test). (c) Western blot analysis for time-dependent cytosolic b-catenin (upper blot), b-tubulin (second blot), nuclear bcatenin (third blot), or PCNA (lower blot) protein expression in the human neuroblastoma cell line SH-SY5Y treated with TRAIL.

(Lucas et al. 2001). In fact, phosphorylation of the s protein occurred already after 6 h incubation of SH-SY5Y cells with TRAIL, increased significantly at 16 h, reached a peak at 24 h, and plateaued up to 48 h (Fig. 3a and c).

Fig. 3 Expression of p-tau and p- cycline-dependent kinase 5 (Cdk5) in SH-SY5Y cells treated with Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL). (a) Densitometric analysis from western blot for time-dependent phosphorylated Tau (Ser199, Ser202) protein expression in the differentiated human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by a Fisher’s LSD test). (b) Densitometric analysis from western blot for time-dependent phospho-Cdk5 (Tyr15) protein expression in the human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (one-way ANOVA followed by the Fisher’s LSD test). (c) Western blot analysis for time-dependent tau-1 (top blot), phospho-tau (Ser199, Ser202) (second blot), Cdk5 (third blot), or its phosphorylated form (bottom blot) in the human neuroblastoma cell line SH-SY5Y treated with TRAIL.



Both Cdk5 and JNK contribute to TRAIL-induced hyperphosphorylation of the s protein Cycline-dependent kinase 5 is among protein factors contributing to hyperphosphorylation of s protein (Cruz et al. 2003). Western blot analysis of extracts from SHSY5Y cells treated with TRAIL, indeed revealed a timedependent phosphorylation of Cdk5, which appeared after 16 h of incubation of cells with TRAIL and reached its peak after 48 h (Fig. 3b and c). JNK is a proline-directed kinase with the ability to phosphorylate the s protein and to regulate the stability of transcription factors, such as early immediate response genes, including c-jun (Reynolds et al. 1997). In this context, it was noteworthy that a similar time pattern of phosphorylation appeared in the western blot analysis performed for the stress kinase JNK, which was activated after 6 h incubation of SH-SY5Y cells with TRAIL and reached its maximum after 48 h of incubation (Fig. 4a and b).

(a)

The specific GSK-3b inhibitor SB216763 partially rescues SH-SY5Y cells from TRAIL-induced death and has an effect additive to that of the caspase inhibitors The above data suggested a role for GSK-3b in mediating TRAIL neurotoxicity. Thus, it appeared of interest to provide further evidence in support of such hypothesis. To do so, we added the specific GSK-3b inhibitor SB216763 to SH-SY5Y cell cultures incubated with TRAIL and measured cell death rate. Maximal killing effect of TRAIL upon SH-SY5Y cells occurred after 48 h whereas cell death was partially prevented in cultures pre-treated with SB216763 (Fig. 5a). Such effect was paralleled by increased phosphorylation of GSK-3b by TRAIL, significantly attenuated in cultures coincubated with SB216763 (Fig. 5b). As SB216763 was able to only partially prevent apoptosis in cells treated with TRAIL, we attempted to clarify whether activation of the caspase cascade could account for the remaining part of such potent proapoptotic effect of TRAIL. Treatment of SH-SY5Y cells with TRAIL resulted in a significant increase of caspase 8 activity. Maximal activity was observed after 6 h of incubation and decreased significantly thereafter to reach a minimum at 48 h (Fig. 6a). Similarly, TRAIL induced a significant increase of caspase 3 activity, which reached a maximum at 24 h and decreased at 48 h (Fig. 6b). However, the activity of both caspase 8 and 3 was decreased in the

(a)

(b)

(b) GSK-3

p-GSK-3β β (p-Tyr216)

Fig. 4 Jun-N-terminal kinase (JNK-1) activation in SH-SY5Y cells treated with Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL). (a) Densitometric analysis of western blot for timedependent phosphorylated JNK-1 protein expression in the differentiated human neuroblastoma cell line SH-SY5Y treated with TRAIL. Results are expressed as percent variation versus control values. Vertical bars are mean + SE; *p < 0.05 versus control (oneway ANOVA followed by Fisher’s test). (b) Western blot analysis for time-dependent JNK-1 (upper blot), or p-JNK-1 (lower blot) protein expression in the human neuroblastoma cell line SH-SY5Y treated with TRAIL.

Fig. 5 SB216763 (5 lmol/L) partially rescues cells from Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL)-dependent and inhibits glycogen synthase kinase 3 beta (GSK-3b) phosphorylation. (a) Differentiated human neuroblastoma SH-SY5Y cell viability following 48 h treatment with TRAIL (100 ng/mL), the GSK-3b (pTyr216) inhibitor SB216763 (5 lmol/L), or combination of the two compounds. Vertical bars are mean + SE; *p < 0.05 versus control; p < 0.05 versus TRAIL (one-way ANOVA followed by the Fisher’s LSD test). (b) Western blot analysis of GSK-3b (upper blot) and p-GSK-3b (p-Tyr216) protein (lower blot) in SH-SY5Y cells after 48 h treatment with TRAIL (100 ng/mL), the GSK-3b inhibitor SB216763 (5 lmol/L), or combination of the two compounds.



was inducing apoptotic death in SH-SY5Y cells, they were treated with TRAIL and processed with both the apoptosisrevealing morphologic methods AO and EB. TRAIL-induced death was of the apoptotic type. In fact, fluorescence data (with both AO and EB) confirmed that whereas most of SHSY5Y cells treated with TRAIL (48 h incubation) showed typical features of apoptosis, the latter were significantly reduced in cells treated at the same time with SB216763 (Fig. 7, photographs). Similar results were obtained after incubation of cells with 5 mmol/L LiCl (data not shown).

Discussion

Fig. 6 The glycogen synthase kinase 3 beta (GSK-3b) inhibitor SB216763 (5 lmol/L) inhibits caspase 8 and 3 activity in SH-SY5Y cells treated with Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL) in a time-dependent way. (a) Time-course of caspase 8 activity in the differentiated human neuroblastoma SH-SY5Y cell line treated with TRAIL (100 ng/mL), or the combination of TRAIL (100 ng/ mL) and the GSK-3b inhibitor SB216763 (5 lmol/L). (b) Time-course of caspase 3 activity in the human neuroblastoma SH-SY5Y cell line treated with TRAIL (100 ng/mL), or the combination of TRAIL (100 ng/ mL) and the GSK-3b inhibitor SB216763 (5 lmol/L). Empty bars: values from cells treated with TRAIL; filled bars: values from cells treated with TRAIL + SB216763. Vertical bars are mean + SE; *p < 0.05 versus control; p < 0.05 versus TRAIL (one-way ANOVA followed by the Fisher’s LSD test).

presence of SB216763 (Fig. 6a and b). To test whether the effects of TRAIL on SH-SY5Y cells were due to a concerted action of both signaling pathways, Wnt and caspase 8, cultures treated with TRAIL were pre-incubated with the specific caspase 8 inhibitor z-IETD-FMK, alone or in combination with SB216763. In fact, significant rescue of SH-SY5Y cells from TRAIL-dependent death was observed in cultures treated with the specific caspase 8 inhibitor zIETD-FMK; a similar result was obtained in cells incubated with SB216763 alone. Interestingly, co-incubation of SHSY5Y cells with both inhibitors, resulted in an additive effect, which promoted total rescue of cells from TRAIL induced death (Fig. 7, histogram). To confirm that TRAIL

Abnormalities of the Wnt signaling pathway are implicated in human brain diseases, including autism (Wassink et al. 2001), schizophrenia (Cotter et al. 1998), and AD (Inestrosa et al. 2002; De Ferrari et al. 2003). Here, we demonstrate the involvement of the Wnt signaling pathway in TRAILmediated neurotoxicity. Treatment of SH-SY5Y cells with TRAIL resulted in cell death associated with a decrease in the Wnt protein expression. These data support both the hypotheses that: (i) Wnt is involved in neuron survival (Malaterre et al. 2007); and (ii) TRAIL is a promoter of apoptosis in neurons (Huang et al. 2005). Consistently, the Wnt signaling pathway is related to TRAIL-induced death in tumor cells. In addition, the former has been demonstrated to play a role in increasing susceptibility to the cytotoxic effects of TRAIL (Aza-Blanc et al. 2003). Activation of Wnt is associated with a non-functional GSK-3b, which prevents degradation of b-catenin and repression of genes related to Wnt activation (Willert and Nusse 1998). When GSK-3b is blocked, two major pathophysiologic and pharmacological implications occur relatedly to AD: not only b-catenin accumulates in the cytoplasm and translocate to the nucleus, where it converts transcription factors from repressors into activators (Behrens et al. 1996), but also the phosphorylation of s protein is reduced (Takashima et al. 1998). In fact, treatment of cells with TRAIL implied GSK-3b phosphorylation, as well as ubiquitin-mediated degradation of cytosolic b-catenin. b-catenin expression was decreased also in the nuclear fractions, suggesting that, possibly because of the cytoplasmic degradation process, lesser amounts of the protein had been transferred to the nuclei. Intact b-catenin is transferred into the nucleus of differentiated cells (Behrens et al. 1996), unlike it occurs in tumors, where some phosphorylation sites of the protein are mutated (van Es et al. 2003). Hyperphosphorylation of s protein was associated with TRAIL-induced GSK-3b phosphorylation. It is known that the neuronal scenario typical of the Alzheimer’s brain, initially characterized by activation of GSK-3b and b-catenin breakdown, continues with s protein hyperphosphorylation (Mandelkow and Mandelkow 1998), a sequence of events which brings about repression of the Wnt-related genes involved in cell survival (Aberle et al. 1997). Thus,



Fig. 7 Combination of SB216763 and the caspase 8 inhibitor z-IETD-FMK completely prevent apoptosis induced by Tumor necrosis factor Related Apoptosis Inducing Ligand (TRAIL). Histogram: differentiated human neuroblastoma SH-SY5Y cell viability following 48 h treatment with TRAIL (100 ng/mL), the glycogen synthase kinase 3 beta (GSK-3b) inhibitor SB216763 (5 lmol/L), the caspase 8 inhibitor z-IETDFMK (2 lmol/L), or various combinations of the compounds (respectively: TRAIL + SB216763; TRAIL + z-IETD-FMK; TRAIL + z-IETD-FMK + SB216763). Vertical bars are mean + SE; *p < 0.05 versus control; p < 0.05 versus TRAIL; Dp < 0.05 versus TRAIL + SB216763 (one-way ANOVA followed by the Fisher’s LSD test). Photographs: Acridine orange (upper panels) and ethidium bromide (lower panels) tests for apoptosis in differentiated human neuroblastoma SH-SY5Y cells treated with (from the left to the right): Control cultures; cultures treated with TRAIL 100 ng/mL; TRAIL (100 ng/mL) + SB216763 (5 lmol/L).

these data suggest that TRAIL somehow mimics the ‘amyloid cascade’ set into motion by metabolites of the amyloid precursor protein (Hardy and Higgins 1992), which inhibits Wnt signaling thus activating downstream events (Inestrosa et al. 2002, 2007) that account, for example, for the occurrence of neurofibrillary tangles in neurons (Hardy 2003). Interestingly, TRAIL induced time-dependent phosphorylation of Cdk5. Besides the s protein, other factors may contribute to the development of neurofibrillary tangles, such as Cdk5 (Sobue et al. 2000), a protein physically interacting with several epitopes of s protein, thought to be transformed in neurofibrillary tangles when undergoing hyperphosphorylation (Flaherty et al. 2000). Thus, in accordance with previous data (Cantarella et al. 2003), it appears plausible to hypothesize that TRAIL additively contributes to the neurotoxic cascade in a manner redundant with Ab. In addition, a role for Cdk5 in the progression of neurofibrillary pathology has been further confirmed by recent reports of transgenic mice displaying aberrant p25-Cdk5 activity (Cruz et al. 2003). The Wnt signaling pathway may be related to other signal transduction mechanisms (Dale 1998), including JNK, a stress kinase phosphorylated during TRAIL-induced caspase activation (Cantarella et al. 2004). Treatment of SHSY5Y cells with TRAIL resulted in phosphorylation of JNK which occurred in timely accordance with the molecular events reported above. It has been reported that Wnt is able to activate three different pathways: the Wnt/Ca2+ pathway; the so-called planar cell polarity (PCP) pathway, and the

canonical Wnt pathway. The PCP pathway, acting through activation of small GTPases, in turn activates JNK (van Es et al. 2003). In light of these data, it appears plausible that TRAIL, besides the caspase-dependent activation of JNK (Cantarella et al. 2004), may induce a caspase independent activation of the stress kinase, as specifically reported, for example, for the PCP pathway (Yamanaka et al. 2002). Such dual mechanism set into motion by TRAIL, could account for its powerful proapoptotic effect on neuronal cells, which might be fueled by multiple merging pathways, eventually implying potentiation of JNK and/or other kinases (Jurewicz et al., 2006). In this line, it is noteworthy that the toxic effect of TRAIL upon oligodendrocytes is abrogated by an antiTRAIL antibody, in a way dependent upon prevention of JNK phosphorylation (Cantarella et al. 2004). These data strongly suggest that interactions between TRAIL and GSK3b are critical to neurotoxicity. Additional support to this hypothesis arises from TRAIL-induced SH-SY5Y cell death, significantly reduced by the GSK-3b inhibitor SB216763 and lithium chloride. The fact that SB216763 partially prevented TRAIL-induced SH-SY5Y cell apoptosis confirmed the relevance of the role of GSK-3b in this process (Kim et al. 2003; Lai et al. 2006). In fact, drugs that either enhance the Wnt pathway or increase responsiveness of tyrosine kinase receptors could limit the loss of Wnt function occurring in late AD (Caricasole et al. 2003). TRAIL not only promoted activation of GSK-3b system, but also activated caspases. Thus, it is plausible to hypothesize that the proapoptotic



effect of TRAIL on SH-SY5Y cells is resulting from possible pleiotropic interaction with intracellular signaling related to apoptosis. The caspase pathway represents a major executor of TRAIL apoptotic effects. TRAIL activation of caspases was paralleled by decreased cell viability, whereas incubation with inhibitors of caspases resulted in rescue of cells from death. It is noteworthy that the number of cells rescued from death was in the latter case higher than that resulting from the GSK-3b blocking effects of SB216763. As a counterproof, the contemporary incubation of cells with both SB216763 and caspase inhibitors resulted in an additive effect, confirming that TRAIL induces at least two pathways at the same time. Moreover, part of the caspase-activating effects of TRAIL appear mediated by the GSK-3b system itself, as one can assume by data indicating that activated caspase levels decrease in cells treated with SB216763. Finally, when cells were incubated with both inhibitors of caspases and SB216763, the rescuing effect upon cell death was virtually complete, confirming that TRAIL-dependent apoptosis of SH-SY5Y cells is mostly dependent upon interaction of the cytokine with both the Wnt and the caspase signaling pathways. Morphologic apoptosis tests confirmed the biochemical data, showing that in all cases, TRAILinduced death was of the apoptotic type. In summary, we supplied the proofs that the TRAILinduced apoptosis occurs as a consequence of setting into motion of two different mechanisms. In fact, apoptosis of SH-SY5Y cells was associated to both inhibition of Wntrelated signaling and activation of the caspase cascade. Moreover, the effects of TRAIL upon the two signaling pathways appear redundant and eventually concur to apoptosis of human neural cells in an additive fashion. These results could, at least in part, account for the potent neurotoxic effect of TRAIL. Finally, pharmacological modulation of such complex machinery of molecules involved in TRAIL-related neurotoxicity might bring about potential tools for AD treatment.

Acknowledgements This work has been supported by a PRIN grant from MIUR, Italian Ministry for Research; FAR grant from the University of Catania.

References Aberle H., Bauer A., Stappert J., Kispert A. and Kemler R. (1997) Betacatenin is a target for the ubiquitin-proteasome pathway. EMBO J. 16, 3797–3804. Aza-Blanc P., Cooper C. L., Wagner K., Batalov S., Deveraux Q. L. and Cooke M. P. (2003) Identification of modulators of TRAIL-induced apoptosis via RNAi-based phenotypic screening. Mol. Cell 12, 627–637. Behrens J., von Kries J. P., Kuhl M., Bruhn L., Wedlich D., Grosschedl R. and Birchmeier W. (1996) Functional interaction of beta-catenin with the transcription factor LEF-1. Nature 382, 638–642.

Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72, 248–254. Brown J. D. and Moon R. T. (1998) Wnt signaling: why is everything so negative? Curr. Opin. Cell Biol. 10, 182–187. Cantarella G., Uberti D., Carsana T., Lombardo G., Bernardini R. and Memo M. (2003) Neutralization of TRAIL death pathway protects human neuronal cell line from b-amyloid toxicity. Cell Death Diff. 10, 134–141. Cantarella G., Risuglia N., Lombardo G., Lempereur L., Nicoletti F., Memo M. and Bernardini R. (2004) Protective effects of estradiol on TRAIL-induced apoptosis in a human oligodendrocytic cell line: evidence for multiple sites of interactions. Cell Death Differ. 11, 503–511. Caricasole A., Copani A., Caruso A., Caraci F., Iacovelli L., Sortino M. A., Terstappen G. C. and Nicoletti F. (2003) The Wnt pathway, cell-cycle activation and beta-amyloid: novel therapeutic strategies in Alzheimer’s disease? Trends Pharmacol. Sci. 24, 233–238. Cotter D., Kerwin R., al-Sarraji S., Brion J. P., Chadwich A., Lovestone S., Anderton B. and Everall I. (1998) Abnormalities of Wnt signalling in schizophrenia – evidence for neurodevelopmental abnormality. Neuroreport 9, 1379–1383. Cruz J. C., Tseng H. C., Goldman J. A., Shih H. and Tsai L. H. (2003) Aberrant Cdk5 activation by p25 triggers pathological events leading to neurodegeneration and neurofibrillary tangles. Neuron 40, 471–483. Dale T. C. (1998) Signal transduction by the Wnt family of ligands. Biochem. J. 329, 209–223. De Ferrari G. V., Chacon M. A., Barria M. I., Garrido J. L., Godoy J. A., Olivares G., Reyes A. E., Alvarez A., Bronfman M. and Inestrosa N. C. (2003) Activation of Wnt signaling rescues neurodegeneration and behavioral impairments induced by beta-amyloid fibrils. Mol. Psychiatry 8, 195–208. van Es J. H., Barker N. and Clevers H. (2003) You Wnt some, you lose some: oncogenes in the Wnt signaling pathway. Curr. Opin. Genet. Dev. 13, 28–33. Flaherty D. B., Soria J. P., Tomasiewicz H. G. and Wood J. G. (2000) Phosphorylation of human tau protein by microtubule-associated kinases: GSK3beta and cdk5 are key participants. J. Neurosci. Res. 62, 463–472. Hardy J. (2003) The relationship between amyloid and tau. J. Mol. Neurosci. 20, 203–206. Hardy J. A. and Higgins G. A. (1992) Alzheimer’s disease: the amyloid cascade hypothesis. Science 256, 184–185. Ho G. J., Drego R., Hakimian E. and Masliah E. (2005) Mechanisms of cell signaling and inflammation in Alzheimer’s disease. Curr. Drug Targets Inflamm. Allergy 4, 247–256. Huang Y., Erdmann N., Peng H., Zhao Y. and Zheng J. (2005) The role of TNF related apoptosis-inducing ligand in neurodegenerative diseases. Cell Mol. Immunol. 2, 113–122. Inestrosa N., De Ferrari G. V., Garrido J. L., Alvarez A., Olivares G. H., Barria M. I., Bronfman M. and Chacon M. A. (2002) Wnt signaling involvement in beta-amyloid-dependent neurodegeneration. Neurochem. Int. 41, 341–344. Inestrosa N. C., Varela-Nallar L., Grabowski C. P. and Colombres M. (2007) Synaptotoxicity in Alzheimer’s disease: the Wnt signaling pathway as a molecular target. IUBMB Life 59, 316–321. Jurewicz A., Matysiak M., Andrzjak S. and Selmaj K. (2006) Trailinduced death of human adult oligodendrocytes is mediated by JNK pathway. GLIA 53, 158–166. Kim H. S., Kim E. M., Lee J. P. et al. (2003) C-terminal fragments of amyloid precursor protein exert neurotoxicity by inducing glycogen synthase kinase-3beta expression. FASEB J. 17, 1951–1953.



Lai J. S., Zhao C., Warsh J. J. and Li P. P. (2006) Cytoprotection by lithium and valproate varies between cell types and cellular stresses. Eur. J. Pharmacol. 539, 18–26. Liu F., Iqbal K., Grundke-Iqbal I. and Gong C. X. (2002) Involvement of aberrant glycosylation in phosphorylation of tau by cdk5 and GSK3beta. FEBS Lett. 530, 209–214. Lucas J. J., Hernandez F., Gomez-Ramos P., Moran M. A., Hen R. and Avila J. (2001) Decreased nuclear beta-catenin, tau hyperphosphorylation and neurodegeneration in GSK-3beta conditional transgenic mice. EMBO J. 20, 27–39. Malaterre J., Ramsay R. G. and Mantamadiotis T. (2007) Wnt-Frizzled signalling and the many paths to neural development and adult brain homeostasis. Front. Biosci. 12, 492–506. Mandelkow E. M. and Mandelkow E. (1998) Tau in Alzheimer’s disease. Trends Cell Biol. 8, 425–427. Pan G., O’Rourke K., Chinnaiyan A. M., Gentz R., Ebner R., Ni J. and Dixit V. M. (1997) The receptor for the cytotoxic ligand TRAIL. Science 276, 111. Reynolds C. H., Utton M. A., Gibb G. M., Yates A. and Anderton B. H. (1997) Stress-activated protein kinase/c-jun N-terminal kinase phosphorylates tau protein. J. Neurochem. 68, 1736–1744. Selkoe D. J. (2000) The origins of Alzheimer’s disease: A is for amyloid. JAMA 283, 1615–1617.

Sobue K., Agarwal-Mawal A., Li W., Sun W., Miura Y. and Paudel H. K. (2000) Interaction of neuronal Cdc2-like protein kinase with microtubule-associated protein tau. J. Biol. Chem. 275, 16673– 16680. Takashima A., Honda T., Yasutake K., Michel G., Murayama O., Murayama M., Ishiguro K. and Yamaguchi H. (1998) Activation of tau protein kinase I/glycogen synthase kinase-3beta by amyloid beta peptide (25–35) enhances phosphorylation of tau in hippocampal neurons. Neurosci. Res. 31, 317–323. Tuppo E. E. and Arias H. R. (2005) The role of inflammation in Alzheimer’s disease. Int. J. Biochem. Cell Biol. 37, 289–305. Wang Q., Zhou Y., Wang X. and Evers B. M. (2006) Glycogen synthase kinase-3 is a negative regulator of extracellular signal-regulated kinase. Oncogene 25, 43–50. Wassink T. H., Piven J., Vieland V. J. et al. (2001) Evidence supporting WNT2 as an autism susceptibility gene. Am. J. Med. Genet. 105, 406–413. Willert K. and Nusse R. (1998) Beta-catenin: a key mediator of Wnt signaling. Curr. Opin. Genet. Dev. 8, 95–102. Yamanaka H., Moriguchi T., Masuyama N., Kusakabe M., Hanafusa H., Takada R., Takada S. and Nishida E. (2002) JNK functions in the non-canonical Wnt pathway to regulate convergent extension movements in vertebrates. EMBO Rep. 3, 69–75.
