Neuroprotective effect of N-acetylcysteine on neuronal apoptosis induced by a synthetic gingerdione compound: Involvement of ERK and p38 phosphorylation

Chemicals and Reagents

N-acetylcysteine (NAC) and U0126 were purchased from Sigma (St. Louis, MO). SB303580, SB202190, and FTI 277 were obtained from Calbiochem (La Jolla, CA). Phospho-p38 MAPK, p38 MAPK, phospho-ERK, and ERK antibodies were purchased from Cell Signaling Technology (Beverly, MA). At DIV 4-6, cortical neurons were treated with various concentrations of I₆, SB 303580, SB 202190, FTI 277, or U0126. When culture was co-treated with I₆ and SB303580, SB202190, FTI277, or U0126, the final concentration of DMSO was <0.6%. We have shown that DMSO at this concentration had no effect on the cell viability.

Primary Cortical Neuron Culture

Primary cortical cells were prepared as described previously (Xia et al., 1996). In brief, cerebral cortices were dissected from postnatal (P0–P2) Sprague-Dawley rats and dissociated in 0.025% trypsin at 37°C for 60 min. Trypsin digestion was stopped by trypsin inhibitor and DNase I. After passage through a Pasteur pipette several times, cortical cells were plated at a density of 5 × 10⁴ per well in 96-well plates for cell viability assay, 1 × 10⁶ per plate in 35 mm plates with coverslips for TUNEL reaction and Hoechst staining, or 3 × 10⁶ per plate in 60 mm plates for Western blotting. All plates and coverslips were coated with poly-l-lysine and laminin. The neurons were cultured in neuro basal medium supplement with 5% FBS, 35 mM glucose, 1 mM l-glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. Cytosine-d-arabinofuranoside (4 μM) was added at Day 2 in vitro (DIV 2) after plating to prevent the proliferation of non-neuronal cell.

TUNEL Reaction

Neurons were washed with PBS, fixed in 4% paraformaldehyde, and permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate. After washing with PBS, neurons were labeled (60 min, 37°C) with a fluorescein TUNEL reagent mixture according to the manufacturer's suggested protocol (Roche Diagnostics, Indianapolis, IN). Neuronal cultures were mounted on slides and examined by fluorescence microscopy.

Hoechst33342, PI, and Double Staining of phosho-p38 MAPK and Hoechst 33342

Apoptotic or necrotic cell death was characterized by use of Hoechst 33342 and propidium iodide (PI; Sigma-Aldrich) double staining. Twenty-four hours after I₆ treatment, cells were stained with 10 μg/ml Hoechst 33342 and 10 μg/ml PI for 30 min at 37°C. After being washed with PBS twice, cells were fixed with 4% PFA in PBS for 30 min at 25°C. Cells were mounted with Aqua Poly/Mount (Polysciences Inc., Warrington, PA), and imaged on a digidata camera attached to a fluorescence microscope.

Cortical cells on 8-well Lab-Tek chamber slides were fixed with 4% PFA in PBS for 30 min at 25°C, followed by permeabilization using 100% methanol for 5 min at 25°C. Cells were then rinsed twice with PBS and preincubated for 1 hr at 25°C in blocking buffer (2% bovine serum albumin and 0.1% Triton X-100 in PBS). Cells were incubated overnight at 4°C in blocking buffer containing anti-phospho-p38 MAPK antibody (1:50; Cell Signaling Technology), rinsed in PBS, and incubated for 2 hr at 25°C with fluorescein-con jugated anti-rabbit immunoglobulin (1:200; ICN/Cappel, Aurora, OH). The nucleus was stained with 10 μg/ml Hoechst 33342 and imaged on a digidata camera attached to a fluorescence microscopy.

MTT (3-(4,5-Dimethylthianol-2-yl)-2,5 Diphenyl Tetrazolium Bromide) Assay

The colorimetric MTT reduction assay, that measures cell viability, was carried out as described previously (Mosmann, 1983). This method assesses mitochondrial activity by measuring the ability of cultured cells to convert yellow MTT to the purple formazan dye. Cortical neurons in 96-well plate were incubated with MTT (125 μg/ml) in growth medium (without phenol red) for 4 hr at 37°C. The precipitated formazan was solubilized with SDS (25 mg/ml) and quantified spectrophotometrically at a wavelength of 570 nm. Data were presented as the percentage of survival relative to vehicle-treated control culture.

Western Blot Assay

Cells were lysed in a lysis buffer containing 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholic acid, 0.1% SDS, 1 mM phenyl methyl sulfonyl fluoride, and 100 μg/ml leupeptin. Lysates were centrifuged at 19,720 × g for 10 min. Supernatants were collected, subjected to electrophoresis on 8.5% SDS-polyacrylamide gel, and transferred to a nitrocellulose membrane. The blot was incubated in 5% nonfat dry milk for 60 min, reacted with primary antibodies overnight at 4°C, and then incubated with HRP-conjugated secondary antibodies for 1 hr at room temperature. Immunoreactivity was detected by using the Western blot chemiluminescence reagent system (Perkin-Elmer, Boston, MA). Films were exposed at different time points to ensure the optimum density, but not saturated.

Statistical Analysis

Data were expressed as mean ± SEM. Analysis of variance followed by Newman-Keuls test was used for statistical comparisons. Levels of P < 0.05 were considered to be of statistical significance.

I₆ Induces Cell Apoptosis in Cortical Neuronal Cultures

After a 5-day maturation period in vitro, cortical neurons were treated with I₆ for 24 hr and cell viability was assessed using the MTT metabolism assay. As shown in Figure 1A, cell survival was inversely correlated with I₆ concentrations in the range of 20–180 μM. One-way ANOVA showed a significant effect for the drug treatment (F_(6,47) = 258.1, n = 6–8 experiments in each concentration, P < 0.0001). Post-hoc comparison showed the differences between control and 60, 80, 120, and 180 μM (P < 0.001). No significant difference was seen between control and 20 and 40 μM, (P > 0.05). The concentration that caused 50% cell death was approximately 60–80 μM. Therefore, 60–80 μM of I₆ was used in the following experiments. The vehicle (0.6% DMSO) alone had no effect on the cell viability. I₆-treated neurons displayed cell body shrinkage, neurite retraction, and neural process fragmentation. To determine whether the reduced cell viability was due to apoptosis, cortical neurons were stained with Hoechst 33342 and propidium iodide (PI). Intracellular penetration of hydrophilic PI occurs if cell death is due to membrane disruption or necrosis (Bal-Price and Brown, 2000; Hamabe et al., 2003). In I₆-treated neurons, very few neurons are PI positive, but many neurons are intensively stained with Hoechst 33342, indicating that after I₆ treatment neurons underwent apoptosis (Fig. 1B). To further confirm I₆-induced apoptosis, TUNEL assay was conducted on the sister cultures. As seen in Figures 1C and D, treatment with I₆ dose-dependently increased the number of darkly stained TUNEL-positive cells (F_(3,16) = 27.69, n = 5 in each concentration, P < 0.0001).

To determine whether glial cells would influence the survival of neurons, we tested the effect of I₆ on the glials cell viability. We stained our cultured cells with neuronal marker Neu-N and glial cell marker GFAP and the results showed that approximate 10% cells were astrocytes. We made a purified culture of astrocytes and examined the effect of I₆ astrocytes. Similar to the action on neurons, I₆ induced cell death in astrocytes in a concentration-dependent manner. The concentration that caused 50% cell death was approximately 70 μM (data not shown).

Prevention of I₆-Mediated Neuronal Apoptosis by N-Acetylcysteine

We examined the effect of NAC on I₆-induced cell death. Cortical neurons were pre-treated with this drug for 30 min before 80 μM I₆ was added. The results depicted that NAC was able to prevent cell death caused by I₆ (Fig. 2A). A total of 1, 5, and 10 mM of NAC increased the survival rate from 54.8 ± 1.5% to 59.5 ± 2.1%, 71.6 ± 2.0%, and 85.0 ± 2.1% respectively (F_(3,60) = 53.72, n = 16 in each concentration, P < 0.0001). We further tested whether NAC rescued cortical neurons from death caused by I₆ using TUNEL reaction. Cortical neurons were treated with I₆ (80 μM) for 24 hr in the presence of NAC. Figures 2B and C show that NAC reduced TUNEL positive cells in a dose-dependent manner (F_(2,12) = 26.5, n = 5 in each concentration, P < 0.0001).

NAC Induces ERK Phosphorylation in Cortical Neurons

ERK activation by NAC plays an important role in preventing neuronal death after withdrawal of trophic support (Yan and Greene, 1998). To determine whether ERK pathway was involved in the action of NAC, cortical neurons were treated with NAC (10 mM) for various periods and cellular extract were blotted with antibodies directly against the active form of ERKs. Figure 3A shows that the increase in ERK phosphorylation was transient that peaked at 10 min after application and subsided within 60 min (F_(7,16) = 19.57, n = 3 in each time point, P < 0.0001). Newman-Keuls t-tests showed that differences existed between control and 2-, 10-, and 30-min time points. No significant difference was detected between control and 0.5 min or control and 60-, 120-, and 180-min time points (P > 0.05). No change was observed when blotted membrane was reprobed with an antibody that recognizes ERK independently of its phosphorylation state, suggesting that the observed pERK increment was not due to an increase in the total amount of ERK. In addition, as shown in Figure 3B, I₆ did not affect pERK by itself (95.7 ± 2.4%, n = 5). As control, an MEK inhibitor U0126 was added to test its effect on ERKs phosphorylation. ERK phosphorylation was decreased to below the baseline by U0126 (59.3 ± 4.0%, n = 5, P < 0.01) (Fig. 3C), suggesting the constitutive activation of ERKs.

To determine whether ERK activation was involved in neuroprotective effect of NAC, we pre-treated the neurons with U0126 for 30 min before adding NAC and I₆. Figure 4 shows that U0126 attenuated the effect of NAC in a concentration-dependent manner (F_(3,28) = 288.7, n = 8 in each concentration, P < 0.0001). Higher concentrations of U0126 (10 and 15 μM) completely abolished the effect of NAC such that there was no difference in survival between neurons treated with I₆ and I₆ + NAC + U0126 (P = 0.98).

In many instances, ERK activation is mediated via the GTP-binding Ras protein (Waskiewicz and Cooper, 1995). To investigate the involvement of Ras-ERK cascade in neuroprotective effect of NAC, a farnesyl transferase inhibitor was applied. Figure 5A shows that NAC increased ERK phosphorylation to 142.5 ± 10.1% of control (n = 5, P < 0.01) and the effect of NAC was blocked by FTI277 (100.8 ± 2.5%, n = 5). Furthermore, FTI277 attenuated the neuroprotective effect of NAC in a concentration-dependent manner (F_(3,28) = 43.52, n = 8 in each concentration, P < 0.0001) (Fig. 5B).

NAC Induces p38 MAPK Phosphorylation in Cortical Neurons

P38 MAPK is involved in cell death observed in Alzheimer's disease (Zhu et al., 2000), transient forebrain ischemia (Takagi et al., 2000), and apoptosis induced by various stimuli (Kawasaki et al., 1997; De Zutter and Davis, 2001). We examined the role of p38 MAPK in I₆-induced cell death. Activation of p38 MAPK was determined by immunoblotting with antibody specific for phosphorylated p38 (Thr¹⁸⁰ and Tyr¹⁸²). Figure 6A shows that pre-treatment of cortical neurons with I₆ for 1 hr did not affect the phosphorylated level of p38 (95.6 ± 2.4%, n = 3). Adding NAC (10 mM) to the medium with or without I₆ increased significantly phosphorylated level of p38 to 224.4 ± 6.5% (n = 3, P < 0.001) and 234.8 ± 11.0 (n = 3, P < 0.001) of control respectively. Figure 6B shows that the increase in p38 phosphorylation was transient that peaked at 10 min after application and subsided within 180 min (F_(7,16) = 52.0, n = 3 in each time points, P < 0.001). Newman-Keuls t-tests showed that differences existed between control and 0.5, 2, 10, 30, 60, and 120 min time points. No significant difference was detected between control and 180 min time points (P > 0.05). No change was observed when blotted membrane was reprobed with an antibody that recognizes p38 independently of its phosphorylation state, suggesting that the observed p38 phosphorylation was not due to an increase in the total amount of ERK. SB203580 is a specific p38 MAPK inhibitor (Lee et al., 1999). Figure 6C shows that NAC-induced p38 phosphorylation was blocked by SB203580.

To determine whether p38 was involved in neuroprotective effect of NAC, we applied the p38 specific inhibitors, SB203580, and SB202190. As shown in Figure 7A, SB203580 alone had no effect on survival at concentrations up to 10 μM but significantly and dose-dependently attenuated neuroprotective effect of NAC. Similar result was observed with SB202190. SB202190 alone had no effect on the survival at concentrations up to 25 μM but significantly and dose-dependently attenuated the effect of NAC (Fig. 7B). To determine whether constitutive activation of p38 provided neuroprotection against I₆ toxicity, SB203580 was added in the presence of I₆ (40 μM). Cell survival rate was 100.6 ± 0.5% (n = 16) in I₆ (40 μM) and was decreased to 94.1 ± 2.9% (n = 16), 77.0 ± 2.6% (n = 16), and 75.7 ± 3.2% (n = 16) respectively after adding 1, 5, or 15 μM of SB203580 to the medium (F_(3,60) = 24.26, P < 0.001) (Fig. 7C). Thus, SB203589 dose-dependently enhanced I₆-induced cell death, suggesting a neuroprotection against I₆ toxicity provided by the constitutive activation of p38.

Finally, we examined whether cells with phospho-p38 positive have minor risk to undergo nuclear fragmentation. Cells were treated with NAC (10 mM) and I₆ (80 μM) for 24 hr. NAC and I₆ were then washed out and 24 hr later NAC was added to activate p38 MAPK and the cells were stained with anti-phospho-p38 and Hoechst dye. As shown in Figure 7D, phospho-p38 positive cells displayed very little nuclear condensation.

Cross-Talk Between p38 MAPK and ERK

Apoptosis triggered by KCl-withdrawal in cerebellar granule cells decreased ERK activity but did not activate p38 MAPK (Watson et al., 1998). Ceramide-induced apoptosis in cortical neurons was mediated by an increase in p38 phosphorylation and not by the decrease in ERK activation (Willaime et al., 2001). To our knowledge, this is the first study showing activation of both p38 MAPK and ERK by the neuroprotective concentrations of NAC. Does NAC activate ERK and p38 MAPK sequentially or in parallel? We addressed this issue by treating cortical neurons with NAC (10 mM) to induce p38 phosphorylation and examined whether p38 phosphorylation was affected by Ras or MEK inhibitors. Figure 8A shows that NAC increased p38 phosphorylation to 228.9 ± 27.7% of control and pre-treatment of neurons with FTI277 (10 μM) completely abolished NAC-induced p38 phosphorylation (99.2 ± 11.9%, n = 3). Similarly, NAC increased p38 phosphorylation to 251.9 ± 9.3% of control and pre-treatment of neurons with U0126 (10 μM) completely abolished NAC-induced p38 phosphorylation (117.6 ± 25.2%, n = 3). P38 inhibitor SB 203580 did not influence NAC-induced ERK phosphorylation such that there was no difference in ERK phosphorylation between NAC (140.5 ± 9.3%, n = 5) and NAC + SB 203580 (142.5 ± 8.2%, n = 5). These results suggest that ERK is an upstream regulator of p38 MAPK signaling.

I₆ Increases Proteolysis of Pro-Caspase-3

Caspases are key mediators of cell death and caspase-3 is an executioner for the death program in cortical neurons in response to various noxious insults (Enari et al., 1998; Gorman et al., 1998; Thornberry and Lazebnik, 1998;). We examined whether I₆-induced cell death was dependent on caspase-3 activation. Constitutive expression of the 32-kDa pro-caspase-3 protein was detected in controls. After treatment with I₆, the expression of 17-kDa cleaved fragment of caspase-3 gradually increased (Fig. 9). There was no detectable change in relative levels of pro-caspase-3 protein. These results suggest that after treatment with I₆ proteolysis of pro-caspase-3 occurs with maintenance of relative pro-caspase-3 protein level.