HaRas activates the NADPH oxidase complex in human neuroblastoma cells via extracellular signal-regulated kinase 1/2 pathway

Journal of Neurochemistry, 2004, 91, 613–622

doi:10.1111/j.1471-4159.2004.02754.x

HaRas activates the NADPH oxidase complex in human neuroblastoma cells via extracellular signal-regulated kinase 1/2 pathway Rosalba Seru`,* Paolo Mondola,* Simona Damiano,* Silvia Svegliati,  Savina Agnese,à Enrico V. Avvedimentoà and Mariarosaria Santillo* *Dipartimento di Neuroscienze e di Scienze del Comportamento, Sezione di Fisiologia, Universita` ‘Federico II’ di Napoli, Napoli, Italy  Istituto di Clinica Medica Generale, Ematologia ed Immunologia Clinica, Polo Didattico, Ancona, Italy àDipartimento di Biologia e Patologia Cellulare e Molecolare ‘L. Califano’, Universita` ‘Federico II’ di Napoli, Napoli, Italy

Abstract In this study we have investigated the effects of the small GTP-binding-protein Ras on the redox signalling of the human neuroblastoma cell line, SK-N-BE stably transfected with HaRas(Val12). The levels of reactive oxygen species (ROS) and superoxide anions were significantly higher in HaRas(Val12) expressing (SK-HaRas) cells than in control cells. The treatment of cells with 4-(2-aminoethyl) benzenesulfonylfluoride, a specific inhibitor of the membrane superoxide generating system NADPH oxidase, suppressed the rise in ROS and significantly reduced superoxide levels produced by SK-HaRas cells. Moreover, HaRas(Val12) induced the translocation of the cytosolic components of the NADPH oxidase complex p67phox and Rac to the plasma membrane. These effects depended on the mitogen-activated protein kinase kinase/extracellular signal-regulated kinase (MEK/ERK1/2)

pathway, as the specific MEK inhibitor, PD98059, prevented HaRas-mediated increase in ROS and superoxide anions. In contrast, the specific phosphoinositide 3-kinase (PI3K) inhibitors LY294002 and wortmannin were unable to reverse the effects of HaRas(Val12). Moreover, cholinergic stimulation of neuroblastoma cells by carbachol, which activated endogenous Ras/ERK1/2, induced a significant increase in ROS levels and elicited membrane translocation of p67phox and Rac. ROS generation induced by carbachol required the activation of ERK1/2 and PI3K. Hence, these data indicate that HaRas-induced ERK1/2 signalling selectively activates NADPH oxidase system in neuroblastoma cells. Keywords: extracellular signal-regulated kinase 1/ 2, HaRas, muscarinic signalling, NADPH oxidase, neuroblastoma SK-N-BE cells, Rac. J. Neurochem. (2004) 91, 613–622.

Reactive oxygen species (ROS) are small messenger molecules involved in many biological functions, including neuron–neuron and neuron–glia signalling (Atkins and Sweatt 1999). Superoxide is one of the reactive oxygen species involved in the induction of hippocampal long-term potentiation (LTP), a long lasting synaptic potentiation involved in learning and memory functions. ROS also modulate the release of synaptic neurotransmitters (Chen et al. 2001) and control the gating of ion channels (Hidalgo et al. 2002). Ras proteins belong to the Ras superfamily of monomeric small GTP-binding proteins. Such proteins have a pivotal role in the signal transduction that controls cell growth and differentiation, as well as stress responses (Lander et al. 1996; Pennisi 1997). Interestingly, different Ras isoforms generate opposing effects on the redox state of cells. Where

Received April 22, 2004; revised manuscript received July 07, 2004; accepted July 08, 2004. Address correspondence and reprint requests to Mariarosaria Santillo, Dipartimento di Neuroscienze e di Scienze del Comportamento, Sezione di Fisiologia, Universita` di Napoli ‘Federico II’, Via S. Pansini, 5, 80131, Naples, Italy. E-mail: [email protected] Abbreviations used: AEBSF, 4-(2-aminoethyl) benzenesulfonylfluoride; DCF, dichlorofluorescein; DCHF-DA, 5,6-carboxy-2¢,7¢-dichlorofluorescein diacetate; DPI, diphenyleneiodonium; ECL, enhanced chemiluminescence; ERK, extracellular signal-regulated kinase; FBS, foetal bovine serum; GEF, guanine nucleotide exchange factor; GSTRBD, glutathione S-transferase-fused ras binding domain; LTP, longterm potentiation; MEK, mitogen-activated protein kinase kinase; PBS, phosphate buffered saline; PDGF, platelet-derived growth factor; PI3K, phosphoinositide 3-kinase; PMSF, phenylmethanesulfonyl fluoride; PVDF, polyvinylidene difluoride; ROS, reactive oxygen species; SDS– PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; SOD, superoxide dismutase; TBS-T, Tris-buffered saline Tween.


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

613

614 R. Seru` et al.

HaRas increases ROS levels and enhances the cellular sensitivity to oxidative stress-induced apoptosis in transfected cells (Irani et al. 1997; Cuda et al. 2002), the Ki-Ras isoform decreases ROS levels and cellular sensitivity to oxidative stress-induced apoptosis by reducing superoxide levels via an activation of the mitochondrial antioxidant enzyme, Mn-superoxide dismutase (Santillo et al. 2001). Membrane NADPH oxidase complex (EC 1.23.45.3), one of the different superoxide cellular sources, is expressed mainly in phagocytic cells, in which superoxide anions exert their bactericidal functions (for a review, see Babior 1999). Phagocytic NADPH oxidase is a multicomponent complex comprising two integral membrane proteins, gp91 phox and p22 phox, which form the heterodimeric flavoprotein known as cytochrome b558, and the cytosolic components p47phox, p67phox and p40phox. Oxidase activation requires that the cytosolic components translocate to the membrane, where they associate with cytochrome b558. Moreover, another important event, leading to enzyme activation, is the membrane translocation of the small GTPase, Rac. In resting cells, such protein is associated with Rho/GDI (guanine nucleotide dissociation inhibitor) in a cytoplasmatic dimeric complex (Ozaki et al. 2000; Di-Poi et al. 2001). The presence of enzymes having NADPH oxidase activity has been well documented recently in non-phagocytic cell lines (Patterson et al. 1999; Abid et al. 2000). NADPH oxidase like complex has also been found in some primary neurons or glial cells in which the superoxide anions, produced by this enzymatic complex, have been associated with oxidative neuronal injury (Noh and Koh 2000; Tammariello et al. 2000; Hwang et al. 2002; Kim et al. 2002). However, the signalling pathways involved in the activation of neuronal NADPH oxidase complex are still completely unknown. To clarify this issue, using the human neuroblastoma cell line (SK-N-BE) stably tranfected with Ha-Ras (Val12) encoding plasmid, we investigated whether Ras induces ROS levels in neuroblast cells via NADPH oxidase activation. More importantly, we also examined the signalling pathways involved in the process. Interestingly, we found that Ha-Ras(Val12) overexpression in SK-N-BE cells induced ROS and superoxide levels by activating NADPH-oxidase complex through extracellular signal-regulated kinase (ERK)1/2 pathway. Moreover, we demonstrated that endogenous Ras/ERK1/2 activation induced by cholinergic stimulation of SK-N-BE cells, increased NADPH oxidase-mediated ROS levels. Indeed, such findings do highlight a new physiological role played by this enzymatic complex in neuron signalling.

Materials and methods Materials PD98059, wortmannin and LY294002 were purchased from Calbiochem (San Diego, CA, USA). 4-(2-aminoethyl)

benzenesulfonylfluoride (AEBSF), allopurinol, rotenone, carbachol and betanechol were all purchased from Sigma Chemical Co. (St. Louis, MO, USA). Cell cultures and transfection Human neuroblastoma SK-N-BE cells [American Type Culture Collection (ATCC), LGC- Promochem Teddington, Middlesex, UK] were grown in monolayer in RPMI 1640 medium (Sigma Chemical Co, St. Louis, MO, USA) supplemented with 10% foetal bovine serum (FBS; Sigma Chemical Co, St. Louis, MO, USA), 2 mM L-glutamine, 50 lg/mL streptomicin and 50 IU/mL penicillin; the cells were kept in a 5% CO2/95% air atmosphere at 37C. Stably HaRas expressing SK-N-BE cells (SK-HaRas) were established with Fugene Transfection Reagent (Roche Diagnostics Corporation, Indianapolis, IN, USA). Briefly, semiconfluent cells, grown in 100-mm dishes 24 h after plating, were cotransfected with 4 lg of pCEFL AU5HaRas(Val12) (BamH1, BglII) and with 1 lg of pRSV-neo, used as the selection marker. Control cells were transfected with the selection marker alone. After 72 h, the medium was replaced and 400 lg/mL of G-418 sulfate (Calbiochem, San Diego, CA, USA) was added to the culture medium to start the selection of Ras expressing cells. Western blot analysis of phospho-ERK1/2 and phospho-Akt Semiconfluent cells, kept in 60-mm Petri dishes, were harvested by scraping them into RIPA buffer [50 mM Tris/HCl, pH 7.5, NaCl 150 mM, 1%NP40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate (SDS)] containing 2.5 mM Na-pyrophosphate, 1 mM b-glycerophosphate, 1 mM NaVO4, 1 mM NaF, 0.5 mM phenylmethanesulfonyl fluoride (PMSF), and a cocktail of protease inhibitors (Boheringer–Mannheim Co., Mannheim, Germany). Next, cell lysates were centrifuged at 4C for 10 min at 11 600 g and the pellets were discarded. Fifty micrograms of protein in Laemmli sample buffer (Laemmli 1970) was boiled for 5 min, resolved by 10% SDS–polyacrylamide gel electrophoresis (SDS– PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were then probed either with monoclonal anti-(phospho-ERK1/2) Ig (Santa Cruz Biotechnology, Santa Cruz, CA, USA), or with monoclonal anti-[phospho-Akt(Ser 473)] Ig (Cell Signaling Technology Inc., Beverly, MA, USA) at 1 : 1000 dilution following the manifacturer’s instructions. To normalize for sample loading and protein transfer the membranes were then stripped and reprobed with antibodies at 1 : 1000 dilution against polyclonal total ERK1/2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or polyclonal total Akt (Cell Signaling Technology Inc.), respectively. Protein bands were revealed by enhanced chemiluminescence (ECL) (Amersham Pharmacia Biotech, Milano, Italia.) and quantified by densitometry. Ras activation assay HaRas activity was assayed using a Ras activation assay kit (Upstate Biotechnology, Lake Placid, NY, USA). The kit used glutathione S-transferase (GST)-ras binding domain (RBD) fusion protein, corresponding to the human RBD, residues 1–149 of Raf-1, expressed in Escherichia coli, bound to glutathione agarose. Active Ras proteins were pulled-down following the manufacturer’s instructions. Briefly, cells were lysed in Mg2+ lysis buffer (MLB) containing 25 mM HEPES, pH 7.5, 150 mM NaCl,


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

Ras/ERK pathway and NADPH oxidase in SK-N-BE cells 615

1% Igepal CA-630, 10 mM MgCl2, 1 mM EDTA and 2% glycerol. Ten micrograms of Raf-1 RBD agarose was added to 1 mg of cell lysates and samples were gently rocked at 4C for 30min. The agarose beads were collected by centrifugation at 11 600 g for 5 s at 4C and washed 3· with MLB buffer. Then, the agarose beads were resuspended in an appropriate amount of 2· Laemmli buffer. The amount of active HaRas was then assessed by western blot using anti-HaRas Ig. The pulled-down samples and 50 lg of total lysate proteins in Laemmli sample buffer were resolved by a 12% SDS–PAGE and transferred onto PVDF membrane. The membranes were then blocked for 20 min in Tris-buffered saline Tween (TBS-T) containing 5% dry-fat milk and were incubated overnight at 4C with polyclonal anti-HaRas(c-20) Ig (Santa Cruz Biotechnology) at 1 : 1000 dilution. They were then washed and incubated with a secondary horseradish–peroxidase-linked anti(rabbit IgG) Ig (Amersham Pharmacia Biotech, Milano, Italia) at 1 : 2000 dilution. To control the amount of loaded proteins, the membranes were stripped and reprobed with monoclonal anti-(atubulin) Ig (Amersham Pharmacia Biotech, Milano, Italia) at 1 : 5000 dilution. A horseradish peroxidase-linked, anti-(mouse IgG) Ig (Amersham Pharmacia Biotech) at 1 : 2000 dilution, was used as secondary antibody. Flow cytometric analysis with anti-gp91phox Ig Cells were grown to semiconfluency in 60-mm culture dishes. After trypsin detachment, 5 · 105 cells were suspended in 1 mL of phosphate buffered saline (PBS) and fixed overnight with 1% formaldehyde at room temperature. Next, cells were permeabilized with 0.1% Triton X-100 for 40 min at 4C, washed 4· with 2 mL of PBS containing 2% FBS, 0.01% NaN3, 0.1% Triton X-100 (buffer A), and incubated for 45 min at 4C with 1 : 50 dilution of polyclonal anti-gp91phox Ig (Santa Cruz Biotechnology). The cells were then washed twice with the same buffer and incubated for 45 min at 4C with Cy2-conjugated anti-(rabbit IgG) Ig (Amersham Pharmacia Biotech) at 1 : 50 dilution. Control cells were incubated with Cy2-conjugated anti-(rabbit IgG) Ig alone. After two washes in buffer A, cells were resuspended in PBS and analyzed by flow cytometry using FACSCAN (BD, Heidelberg, Germany) and WINMDI software. Determination of reactive oxygen species ROS levels were determined by the membrane-permeant fluorogenic probe 5,6-carboxy-2¢,7¢-dichlorofluoresceindiacetate, DCHFDA (Molecular Probes, Leiden, the Netherlands). The assay was based on the fluorescence detection of dichlorofluorescein (DCF), formed by ROS-mediated oxidation of the non-fluorescent precursor, dichlorofluorescin. Cells were trypsinized, washed with PBS, resuspended in PBS, and incubated with 10 lM DCHF-DA for 15min at 37C. Intracellular DCF was then analysed by flow cytometry, using FACSCAN (BD, Heidelberg, Germany) and WINMDI software. Determination of superoxide anion levels Superoxide anion release in the culture medium was assayed by the spectrophotometric SOD-inhibitable cytochrome c reduction method (Sambo et al. 1999). Briefly, 15 · 103 cells per well were plated in 24-well plates, were grown for 24 h in a complete medium and then were incubated for 18 h in a medium containing 0.2% FBS.

The cells were then incubated in 1 mL of Krebs-Ringer phosphate glucose buffer (145 mM NaCl, 4.8 mM KCl, 0.5 mM CaCl2, 1.2 mM MgSO4, 5.7 mM NaPO4, 5.5 mM glucose, pH 7.4) containing 80 mM ferricytochrome c (type III; Sigma Chemical Co), both in the absence and in the presence of 300 units/mL of superoxide dismutase (SOD) (one unit inhibits by 50% the rate of cytochrome c reduction by the xanthine/xanthine oxidase system). After a 60-min incubation at 37C, the reaction was stopped by placing the culture supernatants on ice. The absorbance was read at 550 nm on a Sclavo Reader SR400 spectrophotometer. Assay of p67phox and Rac membrane translocation Cells, grown to semiconfluence in 100-mm culture dishes in complete RPMI medium, were incubated for 18 h in a medium containing 0.2% FBS and collected by scraping them into a buffer containing 100 mM KCl, 3 mM NaCl, 3,5 mM MgCl2, 1.25 mM EGTA, 10 mM PIPES, 2 mM NaVO4, 10 mM phenylarsine oxide, 5 mM NaF, and the cocktail of protease inhibitors. Cells were then disrupted by sonication (2–10 s pulses at 100 W) and centrifuged at 600 g for 10 min at 4C. Next, the supernatants were centrifuged at 100 000 g for 45min at 4C. The membrane pellet was resuspended in 50 lL RIPA buffer. Fifty micrograms of total cell lysates, cytosol and membrane proteins were resolved by a 12% SDS–PAGE and transferred onto a PVDF membrane. Next, the membrane was blocked in 3% dry-fat milk in TBS-Tween20 (0.05%) and probed with a polyclonal anti-human p67phox or with monoclonal antihuman Rac (clone 23A8) Igs (Upstate Biotechnology, Lake Placid, NY, USA) at 1 : 1000 dilution. Then, the membrane was washed and incubated with a secondary horseradish peroxidase-linked antibody (Amersham Pharmacia Biotech) 1: 2000 and was detected by ECL. To verify whether the subcellular fractionation of cell lysates was successful, 50 lg of total lysate, membrane and cytosolic fractions proteins of control SK-N-BE cells were resolved by 12% SDS–PAGE and transferred onto a PVDF membrane. Then, the membrane was blocked as described above and probed with 1 : 1000 dilution of polyclonal antibodies either against human Cu,Zn superoxide dismutase (Mondola et al. 1996), used as the cytosolic marker, or against platelet-derived growth factor (PDGF) receptor b-subunit (Santa Cruz Biotechnology), used as the membrane marker. Immunoprecipitation followed by immunoblotting experiments SK-N-BE cells, grown to semiconfluence in 100-mm dishes, were incubated for 18 h in 0.2% FBS medium. Then, after a 30-min preincubation with or without MEK-1 inhibitor PD98059 (40 lM), they were stimulated for 2.5 min with 1 mM carbachol. Next, the cells were washed twice with PBS and harvested in cold RIPA buffer containing 2.5 mM Na-pyrophosphate, 1 mM b-glycerophosphate, 1 mM NaVO4, 1 mM NaF, 0.5 mM PMSF, and the cocktail of protease inhibitors. The cells were kept for 15min at 4C and disrupted by repeated aspiration through a 21-gauge needle. Cellular debris was pelleted by centrifugation at 11 600 g for 10 min at 4C. Proteins containing a phosphorylated epitope (S/T)p (ERK1/2 phosphorylation consensus site) were immunoprecipitated from 1 mg of cell lysates by adding an anti-(phospho-Ser/Thr-Pro) Ig at 1 : 100 dilution (Upstate Biotechnology). Samples were rocked gently for 16 h; thereafter 20 lL of protein A/G PLUS-Agarose (Santa Cruz Biotechnology), resuspended in RIPA buffer, was added to immunoprecipitates.


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

Determination of protein content Protein content of total cell lysates and of cell membrane and cytosolic fractions was determined according to the method of Lowry et al. (1951). Statistical analyses Statistical differences were evaluated using a Student’s t-test for unpaired samples.

aR

E SK

-H

-B -N SK

GST-RBD pull down Anti Ha Ras total lysates

Anti α-Tubulin

total lysates

Active Ha Ras (Densitometric Relative Units)

Samples were further rocked for 1 h and then washed thrice in RIPA buffer and once with PBS before the addition of 20 lL Laemmli sample buffer. Immunoprecipitated samples and 50 lg of total lysates in Laemmli sample buffer were then boiled for 5 min and centrifuged for 1 min at 10 600 g at room temperature (22C). The pellets were discarded and supernatants were resoved by 12% SDS–PAGE and transferred onto PVDF membrane. Immunoblotting for p67phox protein was carried out as described above.

as

616 R. Seru` et al.

25

* 20 15 10 5 0

SK-N-BE

Results

Activated Ha-Ras stimulates NADPH oxidase and induces ROS in neuroblastoma cells To investigate the role of the HaRas pathway in the modulation of the cellular redox state in neuronal-like cells, the human neuroblastoma SK-N-BE cells were stably transfected with an Ha-Ras(Val12) expression vector (Ha-Ras CEFL). HaRas activity was measured in control and SK-HaRas by GST-RBD pull-down and immunoblot with specific anti-HaRas Ig (Fig. 1). The activity of HaRas was 20fold higher in SK-HaRas than in control cells. Immunoblot analysis on total cell lysates showed that exogenous HaRas(Val12) appeared as a lower mobility band above the endogenous protein, due to the presence of an epitope at the NH terminus of the protein. The tag, however, did not significantly modify the GTP-binding activity of the protein compared to the untagged HaRas(Val12). Densitometric analysis showed that HaRas levels were twofold higher in SK-HaRas than in control cells. Flow cytometric analysis of HaRas levels, carried out by indirect immunofluorescence using an antibody that specifically recognised the HaRas isoform, confirmed immunoblot data (data not shown). To test the hypothesis that HaRas activates a superoxide generating NADPH oxidase in SK-N-BE cells, we first determined the expression of the catalytic subunit of the NADPH oxidase complex in SK-N-BE cells, by testing the presence of gp91phox immunoreactivity in these cells. Thus, by indirect immunofluorescence, we labelled SK-N-BE cells with a rabbit anti-(human gp91phox) Ig and Cy2-conjugated anti-rabbit IgG, as secondary antibody. We analysed them by flow cytometry. SK-N-BE cells contain a specific protein interacting with anti-(gp91phox) Ig (Fig. 2a). Flow cytometric analysis of ROS levels, performed with the oxidant-sensitive fluorescent probe DCHF-DA, showed that intracellular ROS levels were twofold higher in HaRas(Val12) expressing

SK-HaRas

Fig. 1 Analysis of HaRas activity and expression in HaRas(Val12) stably transfected SK-N-BE cells (SK-HaRas). HaRas activity was assayed in control and SK-HaRas cells by precipitation of cell lysates with GST-RBD agarose, followed by immunoblotting with anti HaRas antibody. Total HaRas protein levels were measured in all the total lysates of the same samples. To control the amount of protein loaded, the membrane was stripped and reprobed with monoclonal antiatubulin antibody. The values shown in the histogram (mean ± SE of three independent experiments) were obtained by densitometric analysis of GST-RBD pull-down bands. *p < 0.001 vs. SK-N-BE cells.

neuroblastoma cells than in control cells (Fig. 2b). Moreover, the levels of superoxide anions released in the culture medium were 11-fold higher in SK-HaRas cells than in control samples (Fig. 2c). To assess whether a functional NADPH oxidase complex was involved in the generation of superoxide anions, we tested the effect of AEBSF, a specific NADPH oxidase inhibitor (Diatchuk et al. 1997). Treatment of cells with AEBSF, inhibited ROS generation; likewise, AEBSF reduced O2– levels, thus suggesting that in neuroblastoma cells, HaRas(Val12) activates a NADPH oxidase complex similar to that expressed in phagocytic cell lines. Furthermore, to determine whether other superoxide generating systems were also involved, the SK-HaRas cells were treated either with rotenone (a mitochondrial oxidase inhibitor) or allopurinol (a xanthine oxidase pathway inhibitor) (Fig. 2c). Both drugs did not change superoxide levels in these cells. These data suggest that HaRas(Val12) selectively stimulated a membrane NADPH oxidase complex. To determine more precisely the action site of HaRas(Val12), we measured p67phox membrane translocation. Accordingly, we performed a subcellular fractionation of the cells and then we tested the purity of the cytosolic and membrane fractions by immunoblotting with antibodies against specific protein markers. Cu,Zn superoxide dismutase (SOD), an enzyme almost exclusively localized in the cytosol, was used as


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

e l

br an em

m

ys at e

so

PDGFR β

4 9

(b) aR

E

Anti gp91phox

-B

Control

membrane

2.5

p67phox

cytosol total lysate

2

6 5 4 3

**

2 1

SK

**

SK

as aR -H

SK

as aR -H + F BS E A

+A -H EB aR SF as

-N

-H SK

SK

SK-HaRas

SK

E

as

(c) 0.5

membrane

SK-HaRas +AEBSF Rac

3

cytosol total lysate

Rac membrane levels (Densitometric relative units)

1

SK-N-BE

O2-. levels (nmoles/60min/1x105 cells)

7

BE N-

1.5

0

(c)

*

8

0

-B

DCF fluorescence (Arbitrary Relative Units)

*

aR

(b)

p67phox membrane levels (Densitometric relative units)

2

-H

104

to

6

-N

102 103 FL1-Height

CuZn SOD

SK

101

8

0

0 100

10

as SK +A -H EB aR SF as

Mean cell fluorescence (Arbitrary units)

Events

Anti gp91phox

to ta ll

(a)

Control

cy

(a)

SK

64

Ras/ERK pathway and NADPH oxidase in SK-N-BE cells 617

4

*

3.5 3 2.5 2 1.5

**

1 0.5 0

2.5

BE N-

*

SK

2 1.5

**

1 0.5

S + K-H AE a BS Ra F s SK + H al lo aR pu a rin s SK ol + -H ro a te R no as ne

SK -N -B E SK -H aR as

0

Fig. 2 NADPH oxidase-dependent increase of ROS and superoxide levels in SK-HaRas cells. (a) Immunoreactivity for gp91phox in SK-N-BE cells was evidenced by indirect immunofluorescence and flow cytometric analysis using primary antibodies against human gp91phox and Cy2-conjugated anti-rabbit IgG as secondary antibodies. Control was treated with secondary antibodies alone. Ten thousand cells were counted in triplicate for each sample. The histogram shows the mean ± SE values of three independent experiments. (b) ROS levels measured as DCF fluorescence by flow cytometry and (c) spectrophotometric analysis of superoxide levels using the SODinhibitable cytochrome c reduction method. Cells were incubated for 18 h in medium containing 0.2% FBS. For ROS detection, the NADPH oxidase inhibitor AEBSF (40 lM) was added to the cells 15 min before the DCHF-DA incubation. In the assay of superoxide release in the medium, AEBSF (40 lM), allopurinol (100 lM) and rotenone (50 lM) were added directly to the test buffer. Values are means ± SE of three independent experiments performed in duplicate. *p < 0.001 vs. SK-N-BE; **p < 0.001 vs. SK-HaRas.

cytosolic marker; whereas, the b-subunit of the membrane PDGFR was used as membrane marker (Fig. 3a). The same cell fractions were tested with anti-(human p67phox) Ig. In HaRas(Val12) expressing cells, a significant fraction of p67phox (7.2-fold higher than in control samples) was bound

SK

as aR -H

SK

as aR -H + F BS AE

Fig. 3 Translocation of p67phox and Rac proteins to plasma membrane in SK-HaRas cells. Cells were incubated for 18 h in medium containing 0.2% FBS before harvesting them for subcellular fractionation experiments. (a) Membrane and cytosolic fractions and total lysates of SK-N-BE cells were first analyzed for western blot for Cu,Zn SOD and PDGFR b-subunit as cytosolic and membrane fraction markers, respectively. The same fractions were analyzed with a western blot using anti-p67phox (b) or anti-Rac (c) Igs. Cells were incubated with 40 lM AEBSF for 15min. The histograms show the mean ± SE values of membrane p67phox and Rac protein levels, obtained by the densitometric analysis of three independent experiments. *p < 0.001 vs. SK-N-BE; **p < 0.001 vs. SK-HaRas.

to the membrane complex (Fig. 3b). In contrast, the cytosolic content of p67phox in SK-HaRas cells was significantly lower than that in control cultures, whereas the p67phox levels in the total lysates did not change. These data demonstrate that in these cells, HaRas(Val12) stimulates p67phox membrane translocation. AEBSF was able to displace the p67phox from the membrane of SK-HaRas cells, thus confirming that reduction of ROS and superoxide levels by AEBSF in SK-HaRas cells resulted from the specific inhibition of the NADPH oxidase complex. We have also measured Rac membrane translocation in SK-HaRas cells. Although HaRas(Val12) did not change Rac protein levels in total lysates, it did induce a significant membrane partition (3.2-fold) of the protein (Fig. 3c). In this case, we were not able to detect differences in the cytosolic levels of Rac protein because of the high concentration of the protein in the cytosol compared to membrane fraction. Also Rac membrane translocation was inhibited by the specific NADPH oxidase inhibitor AEBSF,


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

Endogenous Ras activation by cholinergic stimulation of SK-N-BE cells activates NADPH-oxidase dependent ROS production To analyse the effects of endogenous Ras pathway activation on NADPH oxidase complex, we stimulated SK-N-BE cells with the cholinergic agonist carbachol. Under these conditions, the Ras/ERK1/2 pathway is activated through the stimulation of M3 muscarinic receptors that act on Ras via Gq/11/phospholipase C/protein kinase Ce pathway (Kim et al. 1999). The treatment of SK-N-BE cells with 1 mM carbachol for 2.5min induced a twofold increase in active HaRas levels, measured as GST-RBD pull-down (Fig. 5a). Cholinergic stimulation also induced phospho-ERK1/2 and phospho-Akt levels (Figs 5b and c), and the PI3K inhibitor LY294002 partially inhibited carbachol-induced ERK1/2 phosphorylation. To test whether endogenous Ras could activate the NADPH complex, we stimulated neuroblastoma cells either

**

1.0 0.5

0.8 0.6

**

0.4 0.2 0 SK-N-BE

SK-HaRas +LY294002

SK-HaRas +PD98059

SK-HaRas

0

1.0

SK-HaRas +LY294002

1.5

1.2

SK-HaRas +PD98059

(b) *

SK-HaRas

P-Akt (Densitometric relative units)

2.5 2.0

P-ERK1/2

P-Akt

ERK1/2

Akt

1.4

(c)

1.2

DCF fluorescence (Arbitrary Relative Units)

1.0 0.8

**

0.6

**

0.4 0.2 0

-

PD98059 (µM)

1

LY294002 (µM)

-

-

wortmannin (nM)

-

-

(d)

O2-. levels (nmoles/60min/1x105 cells)

A NADPH oxidase complex in SK-N-BE cells is activated by HaRas(Val12) via ERK1/2 pathway Ras can activate multiple signalling pathways. The major signals activated by Ras are Raf/MEK/ERK and PI3K pathways (Rodriguez-Viciana et al. 1997). As a PI3Kdependent activation of the NADPH oxidase complex has been found in phagocytes (Coffer et al. 1998), we evaluated the involvement of both ERK1/2 and PI3K pathways in HaRas-mediated NADPH oxidase activation in SK-N-BE cells. Accordingly, we measured ERK1/2 and Akt phosphorylation levels in control and SK-HaRas cells by immunoblot analysis. ERK1/2 was constitutively activated by HaRas(Val12) overexpression. MEK-1 inhibitor PD98059 but not PI3K inhibitor LY204002 significantly reduced the p-ERK1/2 levels (Fig. 4a). Conversely, HaRas(Val12) did not activate PI3K/Akt pathway, as p-Akt levels in SK-HaRas cells were not significantly different from those of control cells. However, as expected, LY294002 decreased p-Akt levels in SK-HaRas cells. Next, we evaluated the effects of the specific ERK1/2 or PI3K pathway inhibitors on ROS and superoxide production in SK-HaRas cells. The dose– response effect of PD98059 and of two different PI3K pathway inhibitors (LY294002 and wortmannin) on ROS levels was evaluated to exclude effects of these specific inhibitors on other kinase pathways. Treatment of SK-HaRas cells with PD98059 significantly reduced, in a dose-dependent manner, cellular ROS levels. On the other hand, neither LY294002 or wortmannin displayed any effect in the range of concentrations used. Furthermore, superoxide release in the medium was significantly reduced by PD98059; by contrast, treatment of SK-HaRas cells with LY294002, did not change superoxide levels.

(a)

SK-N-BE

thereby demonstrating that Rac translocation depended on the activation of the enzymatic complex.

P-ERK1/2 (Densitometric relative units)

618 R. Seru` et al.

40

80

-

-

-

-

-

-

-

1

40

80

-

-

-

-

-

-

-

-

1

5

10

-

3 2.5 2 1.5

**

1 0.5 0

s

Ra

Ha

SK

s s Ra 9 Ra 2 Ha 05 -Ha 400 8 K 9 D9 S LY2 +P +

SK

Fig. 4 ROS and superoxide levels are induced by HaRas in SK-N-BE cells via ERK1/2 pathway. (a) Western blot analysis of phosphoERK1/2 and (b) phospho-Akt levels in SK-HaRas cells in the absence or in the presence of the MEK-1 inhibitor PD98059 (30min, 40 lM) and of the PI3K inhibitor LY294002 (15 min, 40 lM). The histograms show the values (mean ± SE) obtained by densitometric analysis of p-ERK1/2 and p-Akt bands normalized for total ERK1/2 and Akt levels, respectively, of three independent experiments. (c) Flow cytometric analysis of ROS levels (DCF fluorescence) in SK-HaRas cells preincubated for 30min with increasing concentrations of the MEK inhibitor PD98059, or for 15min with increasing concentration of PI3K inhibitors, LY294002 and wortmannin, before the DCHF-DA incubation. (d) Spectrophotometric measure of superoxide levels in SK-HaRas cells treated with PD98059 and LY204002. The inhibitors were added directly to the test buffer at the concentration of 40 lM. All the experiments were performed on SK-HaRas cells after 18 h of incubation in medium containing 0.2% FBS. All values are means ± SE of three independent experiments performed in duplicate. *p < 0.001 vs. SK-N-BE; **p < 0.001 vs. SK-HaRas.

with carbachol, or with betanechol, a specific muscarinic agonist. We then determined ROS levels; both carbachol and betanechol significantly increased ROS levels in SK-N-BE cells. Carbachol-induced ROS levels were


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

Ras/ERK pathway and NADPH oxidase in SK-N-BE cells 619

-

+

GST-RBD pull down Anti Ha Ras total lysates

total lysates

1.5 1.0 0.5 0

Anti-α tubulin

Control

2.5

(a)

* 2.0

DCF fluorescence (Arbitrary Relative Units)

Carbachol

Active Ha Ras (Densitometric Relative Units)

2.5

(a)

Carbachol

*

2

* 1.5 1

**

0.5

-

+ -

+ + -

+ +

+ -

+ +

+ + -

P-Akt (fold induction)

-

-

ho ta be

ca

ne

ac

+ membrane

p67phox -

+ -

+ + -

cytosol

+ +

total lysates

2.0

Carbachol

-

+

1.5

membrane

**

1.0

Rac

cytosol total lysates

0.5

0 Carbachol PD98059 LY294002

-

+ -

5

+ +

+ + -

Fig. 5 Activation of Ras, ERK1/2 and PI3K/Akt pathways by cholinergic stimulation of SK-N-BE cells. (a) HaRas activity in control and carbachol treated SK-N-BE cells. Cells were treated with 1 mM carbachol for 2.5 min before being harvested for HaRas activity assay (see legend of Fig. 1). The values shown in the histogram (mean ± SE of three independent experiments) were obtained by densitometric analysis of GST-RBD pull-down bands. To control the amount of loaded proteins, the membrane was stripped and reprobed with monoclonal anti-a-tubulin Ig. (b) Western blot analysis of phosphoERK1/2 and (c) phospho-Akt levels in SK-N-BE cells after cholinergic stimulation with 1 mM carbachol for 2.5 min in the absence or presence of the MEK inhibitor PD98059 and of the PI3K inhibitor LY294002. The inhibitors were used as described in the legend of Fig. 4a and b. All the experiments were performed on SK-N-BE cells after 18 h of incubation in medium containing 0.2% FBS. *p < 0.001 vs. controls; **p < 0.001 vs. carbachol treated cells.

sensitive to the NADPH oxidase inhibitor AEBSF (Fig. 6a), as demonstrated in HaRas(Val12) expressing cells. The rise in ROS levels after carbachol stimulation was accompanied by p67phox and Rac membrane translocation: a fourfold increase in p67phox and a 2.3-fold increase in Rac membrane levels were observed after carbachol stimulation (Fig. 5b). Pre-treatment of the cells either with PD98059 (MEK-1 inhibitor) or with LY294002 (PI3K inhibitor) prevented the increase in carbacholinduced ROS levels in SK-N-BE cells (Fig. 6c). The effects of PD98059 and LY204002 were additive (Fig. 6c), thus indicating that ERK1/2 and PI3K signal independently to the NADPH oxidase complex.

(c)

*

4 3 2 1 0

control

*

*

Akt Carbachol PD98059 LY294002

(b) Carbachol

**

2.5

P-Akt

rb ca

2

0 Carbachol PD98059 LY294002

(c)

l

l ro nt co **

4

P67phox membrane levels (densitometric relative units)

Carbachol PD98059 LY294002

6

Rac membrane levels (densitometric relative units)

ERK1/2

8

DCF fluorescence (Arbitrary Relative Units)

P-ERK1/2

P-ERK 1/2 (fold induction)

(b)

*

ch ol r b +A a c EB h SF ol

0 10

carbachol

3

* 2

1

0

control

carbachol

2.5

*

2 1.5

**

** **

1 0.5 0

Carbachol PD98059 LY294002

-

+

-

+ +

-

+

-

+

+ + +

Fig. 6 Cholinergic stimulation of SK-N-BE cells activates NADPH oxidase complex in SK-N-BE cells through ERK1/2 and PI3Kinase pathways. (a) NADPH oxidase-dependent ROS induction upon cholinergic stimulation of SK-N-BE cells. ROS levels were measured as DCF fluorescence by flow cytometry after a 2.5 min incubation with 1 mM carbachol or 5 mM betanechol, which were added to the DCHF-DA loaded cells. AEBSF was used as described in the legend of Fig. 2. (b) p67phox and Rac membrane translocation after carbachol treatment. Cells were stimulated for 5 min with 1 mM carbachol before harvesting them for subcellular fractionation. The histograms show the mean ± SE values of membrane p67phox and Rac protein levels obtained by the densitometric analysis of three independent experiments. (c) ROS levels were measured as DCF fluorescence by flow cytometry after a 2.5 min incubation with 1 mM carbachol, added to the DCHF-DA loaded cells. Cells were preincubated with 40 lM of PD98059 and LY294002 for 30 and 15min, respectively, before DCHF-DA incubation. All the experiments were performed on SK-N-BE cells after 18 h of incubation in medium containing 0.2% FBS. *p < 0.001 vs. control; **p < 0.001 vs. carbachol treated cells.


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

620 R. Seru` et al.

ERK1/2-induced phosphorylation of p67phox upon cholinergic stimulation of SK-N-BE cells To identify the ERK1/2 molecular target responsible for the Ras/ERK1/2-dependent activation of the NADPH oxidase complex, SK-N-BE cells were stimulated with carbachol in the presence or in the absence of the specific MEK-1 inhibitor PD98059, and then cell protein extracts were immunoprecipitated with an antibody against the phosphorylated epitope (S/T)p (ERK1/2 phosphorylation consensus site) and were immunoblotted with antip67phox antibody. Figure 7 shows that p67phox is recognised by the antibody. PD98059 significantly reduced the signal, and the same signal was not present in immunodepleted samples (Fig. 7, lane 1). These data indicate that p67 phox is a bona fide ERK1/2 substrate and suggest that its phosphorylation stimulates NADPH oxidase activity. Discussion

We have found that ERK1/2, activated by HaRas, induces NADPH oxidase complex and ROS levels in the human neuroblastoma cell line SK-N-BE. The cytosolic subunit p67phox is a direct ERK1/2 target, as, upon cholinergic-

I.B. Anti p67phox Carbachol PD98059

+ -

-

+ -

+ + I.P. anti phospho Ser/Thr-Pro

I.D.

total lysates

Phospho-p67phox (densitometric arbitrary units)

600

*

500 400 300 200

**

100 0

Carbachol PD98059

-

+ -

+ +

Fig. 7 ERK1/2-dependent phosphorylation of p67phox upon cholinergic stimulation of SK-N-BE cells. Cells were incubated for 18 h in medium containing 0.2% FBS and were then stimulated for 2.5 min with 1 mM carbachol before harvesting them for immunoprecipitation with antibody directed against phospho Ser/Thr-Pro, MPM2 and immunoblot with anti-p67phox Ig. The MEK inhibitor PD98059 (40 lM) was added to the cells 30min prior to the carbachol treatment. The first lane (I.D.) shows the immunodepleted carbachol-treated sample. The histograms show the mean ± SE values obtained by the densitometric analysis of three independent experiments. The lower bands show the immunoblot of p67phox of total lysates. *p < 0.001 vs. control; **p < 0.001 vs. carbachol treated cells.

mediated Ras/ERK1/2 pathway activation, p67phox is phosphorylated at ERK1/2 phosphorylation consensus sites. Neuroblastoma cells (SK-N-BE) express a NADPH– oxidase complex that is similar functionally to that found in phagocytes. In fact, these cells showed a clear gp91phox immunoreactivity. Moreover, the superoxide anions, released in the culture medium of SK-N-BE cells stably transfected with HaRas(Val12) encoding plasmid, were reduced markedly by a specific NADPH oxidase inhibitor, AEBSF (Fig. 2). This inhibitor interferes with the assembly of the cytosolic and membrane components of the complex, probably by chemical modification of the cytochrome b558 molecule (Diatchuk et al. 1997). The specificity of the AEBSF inhibitor on NADPH oxidase complex was also demonstrated by its ability to displace the cytosolic subunit p67phox from the membrane in SK-HaRas cells (Fig. 3b). AEBSF and DPI, the latter a less specific but widely used NADPH oxidase inhibitor, were equally effective in inhibiting HaRas(Val12)-induced ROS (data not shown), thus, demonstrating that a functional NADPH oxidase system constitutes the main source of ROS in HaRas(Val12) expressing cells. Translocation of the GTP-binding protein Rac to the membrane is another essential step in NADPH complex activation. This event was markedly inhibited by AEBSF (Fig. 3c), thus revealing that in SK-N-BE cells Rac translocation is strictly correlated with NADPH oxidase activation. Several reports have indicated that PI3K pathway modulates phagocyte NADPH oxidase. The cytosolic subunits p47phox and p40phox, discovered recently as the binding partners of p67phox, bind the lipid products of PI3K, phosphatidylinositol-3, 4-bisphosphate [PtdIns (3,4)P2], and phosphatidylinositol-3-phosphate [PtdIns(3)p], respectively, to their PX domains: a module found in a variety of proteins associated with the cell membrane. (Ellson et al. 2001; Kanai et al. 2001). Moreover, there is evidence that PI3K stimulates the Rac guanine nucleotide exchange factors (GEF) VAV (Han et al. 1998). Apparently, in neuroblastoma cells, NADPH oxidase activity was not regulated by HaRas(Val12) via PI3K signalling, because the PI3K/Akt pathway was not constitutively activated in SK-HaRas cells. Moreover, two different inhibitors of the PI3K pathway, LY294002 and wortmannin did not affect HaRas(Val12)dependent ROS production (Fig. 4). Therefore, in human neuroblastoma cells, ERK1/2 pathway seems to be the main signal activated by Ras responsible for NADPH oxidase activation. In addition, our data indicate that HaRas significantly increases the fraction of Rac in the membranes. Recently, Tiam1, a Rac-GEF (Fleming et al. 2000), has been found to be a direct Ras effector. A Tiam1 region shares a significant sequence homology with the Ras-binding domain (RBD) of Raf1. As a result, the endogenous Tiam1 and Ras form a


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

Ras/ERK pathway and NADPH oxidase in SK-N-BE cells 621

complex in the cells that contribute to the formation of RacGTP in a PI3K-independent manner (Lambert et al. 2002). We have also studied the effects of endogenous Ras pathway activation by cholinergic stimulation of SK-N-BE cells on NADPH oxidase complex. We demonstrated that the cholinergic agonist carbachol activates Akt and ERK1/2 (Figs 5b and c). Moreover, cholinergic stimulation increased ROS levels by NADPH oxidase complex activation (Fig. 6a and b). The PI3K inhibitor LY204002 or PD98059 reduced carbachol-induced ROS (Fig. 6c). These data demonstrate that cholinergic stimulation of ROS levels requires both PI3K and ERK1/2 activation (Fig. 5b). As Ha-Ras(Val12) expression in SK-N-BE cells stimulates ROS levels exclusively via ERK1/2 (Figs 4c and d), we suggest that, in these cells, PI3K stimulates HaRas, which induces ERK1/2 and ROS. In SK-N-BE cells ERK1/2-mediated p67phox phosphorylation may be involved in NADPH oxidase activation by Ras/ERK1/2 pathway. In fact, p67phox reacts specifically with antibodies that recognise phosphorylated ERK1/2 consensus sites (Ser-Thr/Pro) and the signal was abolished by the Mek1 inhibitor (Fig. 7). In effect, although muscarinic neurotransmission is involved in cognitive function (Bartus et al. 1982; Lebrun et al. 1990), the physiological role, as well as the mechanisms whereby muscarinic receptor signalling acts in cognitive function, still remains unclarified. We do know, however, that carbachol produces LTP in rat hippocampal slices (Auerbach and Segal 1994), thus suggesting the existence of a substantial correlation between the cholinergic system and memory functions. In addition to tetanic LTP, muscarinic LTP is also modulated by ROS. Consequently, having demonstrated that NADPH oxidase induces an increase in ROS levels in SK-N-BE cells after muscarinic stimulation, we suggest that NADPH oxidase represents a muscarinic signalling target and mediates some of the biological effects in vivo. Acknowledgements Special thanks to Mr Lucio Cammarota for his skilful technical assistance and to Dr Paola Merolla for her English editorial assistance.

References Abid M. R., Kachra Z., Spokes K. C. and Aird W. C. (2000) NADPH oxidase activity is required for endothelial cell proliferation and migration. FEBS Lett. 486, 252–256. Atkins C. M. and Sweatt J. D. (1999) Reactive oxygen species mediate activity-dependent neuron-glia signalling in output fibers of the hippocampus. J. Neurosci. 19, 7241–7248. Auerbach J. M. and Segal M. (1994) A novel cholinergic induction of long-term potentiation in rat hippocampus. J. Neurophysiol. 72, 2034–2040. Babior B. M. (1999) NADPH oxidase: an update. Blood 93, 1464–1476.

Bartus R. T., Dean R. L., Beer B. and Lippa A. S. (1982) The cholinergic hypothesis of geriatric memory dysfunction. Science 217, 408– 414. Chen B. T., Avshalumov M. V. and Rice M. E. (2001) H2O2 is a novel, endogenous modulator of synaptic dopamine release. J. Neurophysiol. 85, 2468–2476. Coffer P. J., Geijsen N., M’rabet L., Schweizer R. C., Maikoe T., Raaijmakers J. A., Lammers J. W. and Koenderman L. (1998) Comparison of the roles of mitogen-activated protein kinase kinase and phosphatidylinositol 3-kinase signal transduction in neutrophil effector function. Biochem. J. 329, 121–130. Cuda G., Paterno R., Ceravolo R., Candigliota M., Perrotti N., Perticone F., Faniello M. C., Schepis F., Ruocco A., Mele E. et al. (2002) Protection of human endothelial cells from oxidative stress: role of Ras-ERK1/2 signalling. Circulation 105, 968–974. Diatchuk V., Lotan O., Koshkin V., Wikstroem P. and Pick E. (1997) Inhibition of NADPH oxidase activation by 4-(2-aminoethyl)benzenesulfonyl fluoride and related compounds. J. Biol. Chem. 272, 13 292–13 301. Di-Poi N., Faure` J., Grizot S., Molnar G., Pick E. and Dagher M. C. (2001) Mechanism of NADPH oxidase activation by the Rac/RhoGDI complex. Biochemistry 40, 10 014–10 022. Ellson C. D., Gobert-Gosse S., Anderson K. E., Davidson K., Erdjument-Bromage H., Tempst P., Thuring J. W., Cooper M. A., Lim Z-Y., Holmes A. B. et al. (2001) PtdIns(3)P regulates the neutrophil oxidase complex by binding to the PX domain of p40phox. Nat. Cell. Biol. 3, 679–682. Fleming I. N., Gray A. and Downes C. P. (2000) Regulation of the Rac1specific exchange factor Tiam1 involves both phosphoinositide 3-kinase-dependent and – independent components. Biochem. J. 351, 173–182. Han J., Luby-Phelps K., Das B., Shu X., Xia Y., Mosteller R. D., Krishna U. M., Falck J. R., White M. A. and Broek D. (1998) Role of substrates and products of PI 3-kinase in regulating activation of Racrelated guanosine triphosphatases by Vav. Science 279, 558–560. Hidalgo C., Aracena P., Sanchez G. and Donoso P. (2002) Redox regulation of calcium release in skeletal and cardiac muscle. Biol. Res. 35, 183–193. Hwang J. J., Choi S. Y. and Koh J. Y. (2002) The role of NADPH oxidase, neuronal nitric oxide synthase and poly (ADP ribose) polymerase in oxidative neuronal death induced in cortical cultures by brain-derived neurotrophic factor and neurotrophin-4/5. J. Neurochem. 82, 894–902. Irani K., Xia Y., Zweier J. L., Der Sollott S. J. C. J., Fearon E. R., Sundaresan M., Finkel T. and Goldschmidt-Clermont P. J. (1997) Mitogenic signalling mediated by oxidants in Ras-transformed fibroblasts. Science 275, 1649–1652. Kanai F., Liu H., Field S. J., Akbary H., Matsuo T., Brown G. E., Cantley L. C. and Yaffe M. B. (2001) The PX domains of p47phox and p40phox bind to lipid products of PI (3) K. Nat. Cell Biol. 3, 675–678. Kim J. Y., Yang M. S., Oh C. D., Kim K. T., Ha M. J., Kang S. S. and Chun J. S. (1999) Signalling pathway leading to an activation of mitogen-activated protein kinase by stimulating M3 muscarinic receptor. Biochem. J. 337, 275–280. Kim S. H., Won S. J., Sohn S., Kwon H. J., Lee J. Y., Park J. H. and Gwag B. J. (2002) Brain-derived neurotrophic factor can act as a pronecrotic factor through transcriptional and translational activation of NADPH oxidase. J. Cell Biol. 159, 821–831. Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Lambert J. M., Lambert Q. T., Reuther G. W., Malliri A., Siderovski D. P., Sondek J., Collard J. G. and Der C. J. (2002) Tiam1 mediates


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622

622 R. Seru` et al.

Ras activation of Rac by a PI (3) K-independent mechanism. Nat. Cell Biol. 4, 621–625. Lander H. M., Milbank A. J., Tauras J. M., Hajjar D. P., Hempstead B. L., Schwartz G. D., Kraemer R. T., Mirza U. A., Brian T. C., Burk S. C. and Quilliam L. A. (1996) Redox regulation of cell signalling. Nature 38, 380–381. Lebrun C., Durkin T. P., Marighetto A. and Jaffard R. A. (1990) Comparison of the working memory performances of young and aged mice combined with parallel measures of testing and drug-induced activations of septo-hippocampal and nbm-cortical cholinergic neurones. Neurobiol. Aging 11, 515. Lowry O. H., Rosbrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin-Phenol reagents. J. Biol. Chem. 193, 265–275. Mondola P., Annella T., Santillo M. and Santangelo F. (1996) Evidence for secretion of cytosolic CuZn superoxide dismutase by HepG2 cells and human fibroblasts Int. J. Biochem. Cell Biol. 28, 677– 681. Noh K. M. and Koh J. Y. (2000) Induction and activation by zinc of NADPH oxidase in cultured cortical neurons and astrocytes. J. Neurosci. 20, RC111. Ozaki M., Deshpande S. S., Angkeow P., Bellan J., Lowenstein C. J., Dinauer M. C., Goldsmith-Clermont P. J. and Irani K. (2000) Inhibition of the Rac1 GTPase protects against nonlethal ischemia/

reperfusion-induced necrosis and apoptosis in vivo. FASEB J. 14, 418–429. Patterson C., Ruef J., Madamanchi N. R., Barry-Lane P., Hu Z., Horaist C., Ballinger C. A., Brasier A. R., Bode C. and Runge M. S. (1999) Stimulation of a vascular smooth muscle cell NAD(P)H oxidase by thrombin. Evidence that p47 (phox) may partecipate in forming this oxidase in vitro and in vivo. J. Biol. Chem. 274, 19 814–19 882. Pennisi E. (1997) Superoxides relay Ras protein’s oncogenic message. Science 275, 1567–1568. Rodriguez-Viciana P., Warne P. H., Khwaja A., Marte B. M., Pappin D., Das P., Waterfield M. D., Ridley A. and Downward J. (1997) Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell 89, 457–467. Sambo P., Jannino L., Candela M., Salvi A., Donini M., Dusi S., Luchetti M. M. and Gabrielli A. (1999) Monocytes of patients with systemic sclerosis (scleroderma) spontaneously release in vitro increased amounts of superoxide anion. J. Invest. Dermatol. 112, 78–84. Santillo M., Mondola P., Seru R., Annella T., Cassano S., Ciullo I., Tecce M. F., Iacomino G., Damiano S., Cuda G. et al. (2001) Opposing functions of Ki- and Ha-Ras genes in the regulation of redox signals. Curr. Biol. 11, 614–619. Tammariello S. P., Quinn M. T. and Estus S. (2000) NADPH oxidase contributes directly to oxidative stress apoptosis in nerve growth factor deprived sympathetic neurons. J. Neurosci. 20, RC53.


2004 International Society for Neurochemistry, J. Neurochem. (2004) 91, 613–622