Diethylmaleate and iodoacetate in combination caused profound cell death in astrocytes

Authors

  • Su-Lan Liao,

    1. Department of Education and Research, Taichung Veterans General Hospital, Taichung, Taiwan
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Yen-Chuan Ou,

    1. Division of Urology, Taichung Veterans General Hospital, Taichung, Taiwan
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Cheng-Yi Chang,

    1. Department of Surgery, Fong Yuan Hospital, Taichung, Taiwan
    Search for more papers by this author
  • Wen-Ying Chen,

    1. Department of Veterinary Medicine, National Chung Hsing University, Taichung, Taiwan
    Search for more papers by this author
  • Yu-Hsiang Kuan,

    1. Department of Pharmacology, School of Medicine, Chung-Shan Medical University, Taichung, Taiwan
    Search for more papers by this author
  • Wen-Yi Wang,

    1. Graduate School of Nursing, Hung-Kuang University, Taichung, Taiwan
    Search for more papers by this author
  • Hung-Chuan Pan,

    1. Department of Neurosurgery, Taichung Veterans General Hospital, Taichung, Taiwan
    Search for more papers by this author
  • Chun-Jung Chen

    Corresponding author
    1. Department of Education and Research, Taichung Veterans General Hospital, Taichung, Taiwan
    2. Graduate School of Nursing, Hung-Kuang University, Taichung, Taiwan
    3. Center for General Education, Tunghai University, Taichung, Taiwan
    4. Institute of Biomedical Sciences, National Chung Hsing University, Taichung, Taiwan
    • Address correspondence and reprint requests to Chun-Jung Chen, Department of Education and Research, Taichung Veterans General Hospital; No. 160, Sec. 3, Taichung-Kang Rd., Taichung 407, Taiwan. E-mail: cjchen@vghtc.gov.tw

    Search for more papers by this author

Abstract

Energy failure and oxidative stress have been implicated in the pathogenesis of ischemia. Here, we report a potential link between cytosolic phospholipase A2 (cPLA2) activation and energy failure/oxidative stress-induced astrocyte damage involving reactive oxygen species (ROS), protein kinase C-α (PKC-α), Src, Raf, and extracellular signal-regulated kinase (ERK) signaling and concurrent elevation of endogenous chelatable zinc. Energy failure and oxidative stress were produced by treating astrocytes with glycolytic inhibitor iodoacetate and glutathione chelator diethylmaleate, respectively. Diethylmaleate and iodoacetate in combination caused augmented damage to astrocytes in a time- and concentration-dependent manner. The cell death caused by diethylmaleate/iodoacetate was accompanied by increased ROS generation, PKC-α membrane translocation, Src, Raf, ERK, and cPLA2 phosphorylation. Pharmacological studies revealed that these activations all contributed to diethylmaleate/iodoacetate-induced astrocyte death. Intriguingly, the mobilization of endogenous chelatable zinc was observed in diethylmaleate/iodoacetate-treated astrocytes. Zinc appears to act as a downstream mediator in response to diethylmaleate/iodoacetate treatment because of the attenuating effects of its chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine. These observations indicate that ROS/PKC-α, Src/Raf/ERK signaling and cPLA2 are active participants in diethylmaleate/iodoacetate-induced astrocyte death and contribute to a vicious cycle between the depletion of ATP/glutathione and the mobilization of chelatable zinc as critical upstream effectors in initiating cytotoxic cascades.

image

Energy failure and oxidative stress have been implicated in the pathogenesis of ischemia and cell death. Through the application of the glycolytic inhibitor iodoacetate and the glutathione chelator diethylmaleate, we report a potential link between cytosolic phospholipase A2 (cPLA2) activation and energy failure/oxidative stress-induced astrocyte damage involving reactive oxygen species (ROS), signaling through the kinases PKC-α, Src, Raf, and ERK and concurrent elevation of endogenous chelatable zinc.

Abbreviations used
cPLA2

cytosolic phospholipase A2

ERK

extracellular signal-regulated kinase

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

JNK

jun N-terminal kinase

NAC

N-acetyl cysteine

PKC-α

protein kinase C-α

ROS

reactive oxygen species

TPEN

N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine

Accumulating evidence suggests that oxidative stress and energy failure occur in a diversity of neurodegenerative disorders and play crucial roles in their pathogenesis, particularly in cerebral ischemia (Tavazzi et al. 2005; Al-Majed 2011). Cerebral ischemia, mainly resulting from the interruption of cerebral blood circulation, is characterized by the development of neurological deficits because of neuronal dysfunction and/or destruction. Evidence suggests that cell death occurs following ischemia in susceptible brain regions. In the acute phase, cell death in the ischemic core is conventionally considered necrotic, whereas after a short period of cerebral ischemia followed by reperfusion, the cells in the penumbral regions undergo another wave of massive death program (Linnik et al. 1993; Garcia et al. 1995). These ischemia/reperfusion-associated cellular changes are accompanied by a cascade of biochemical events. The deficiency of blood supply primarily causes disturbance in energy metabolism that leads to a decrease in glucose utilization with the consequence of ATP depletion. The resultant energy crisis then triggers disruption of ion pumps, influx of calcium, and excitotoxic changes. Most of these changes are associated with the generation of reactive oxygen species (ROS). Massive production of ROS occurs especially during the subsequent reperfusion stage and this event further potentiates brain damage caused by ischemia (Arai et al. 1986; Sims and Zaidan 1995; Koh et al. 1996; Chan 2001). As energy failure and oxidative stress are increasingly implicated as key mediators of neuronal injury, neuroprotective ATP level boosters and antioxidants are considered to play important roles in a promising approach to limit the extent of neuronal cell loss (Tavazzi et al. 2005; Al-Majed et al. 2006; Al-Majed 2011).

Although biochemical evidence suggests that ATP depletion and oxidative stress might serve as an important primary initiator and augmented factor in the development and expansion of cerebral ischemia/reperfusion injury, respectively, the molecular mechanisms underlying their aggravated actions remain elusive. While alterations in the status of cellular energy level and oxidative burden may cause cell death by drastic energetic failure and oxidative damage, they also may induce cytotoxicity by activating specific signal transduction pathways. Among the potential intracellular signaling molecules, the MAPK family has fundamental roles in both the maintenance of cell survival and the induction of cell death (Xia et al. 1995; Murray et al. 1998; Seo et al. 2001). Studies have shown the concurrence of ATP depletion, oxidative stress, and MAPK activation in the ischemic brain and further implicated their contribution to cerebral ischemia/reperfusion-associated cell death (Tavazzi et al. 2005; Al-Majed et al. 2006; Wu et al. 2008a; Kishimoto et al. 2010; Al-Majed 2011). For this reason it is likely that the increase in MAPK activity participates in ATP depletion- and oxidative stress-mediated cell death.

In the CNS, astrocytes play crucial roles in neuronal energy metabolism and protection from oxidative damage because they serve as a metabolic support for neurons by providing energy intermediates and glutathione. In addition to defect in metabolic support, ischemia-associated energy failure and oxidative stress also render astrocytes unable to uptake and metabolize over-released extracellular neurotransmitters such as glutamate. Even worse, the reversal of glutamate transporters in astrocytes after ischemic insult further elevates glutamate release leading to augmented excitotoxicity (Parpura et al. 2012). Therefore, once the energy metabolism and antioxidant capacity of astrocytes are impaired, they have great potential to impact on neurons as they form a tight functional unit. Compared with the extensive investigations of neurotoxicity, glial cell death mechanisms have received relatively little attention. Therefore, to understand how ATP depletion and oxidative stress affect astrocytes is an important issue for elucidating the impact of ATP depletion and oxidative stress on brain function during pathology. Previously, we reported that the cell death of astrocytes was accompanied by the generation of ROS, depletion of ATP, and increased phosphorylation of MAPKs (Chen et al. 2000; Chen and Liao 2002; Chang et al. 2010; Liao et al. 2011). To extend the scope of our previous findings and relevant studies, we conducted this study to explore whether oxidative stress aggravated ATP depletion-mediated astrocyte death, the upstream regulatory mechanisms of MAPKs activation and their regulation of ATP depletion/oxidative stress-induced astrocyte death via the application of glycolytic inhibitor iodoacetate and glutathione chelator diethylmaleate.

Materials and methods

Materials

N,N,N′,N′-Tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), U0126, SB203580, SP600125, methyl arachidonyl fluorophonate, GW5074, PP2, GF109203X, oxaloacetate, N-acetyl cysteine (NAC), and diphenyleneiodonium chloride were purchased from Tocris (Minneapolis, MN, USA) and iodoacetate and diethylmaleate were obtained from Sigma-Aldrich (St. Louis, MO, USA). The concentrations of pharmacological inhibitors were used according to relevant studies (Chang et al. 2010; Liao et al. 2011).

Cell cultures

The Animal Experimental Committee of Taichung Veterans General Hospital approved the protocol of the animal study. Astrocytes were prepared from cerebral cortices of 1-day-old Sprague–Dawley rats in accordance with previously reported protocols (Chang et al. 2010). After digestion, the dissociated cells were resuspended in Dulbecco's modified Eagle medium/F12 supplemented with 10% fetal bovine serum and were plated on a T-75 flask. Confluent cultures (14–17 days in vitro) were shaken at a speed of 6 g for 24 h to remove less adhesive microglia and other cells. The resultant adherent cells were maintained for another 1 week before replating. Two days after seeding, the obtained astrocytes were used for experiments. For the experiments, the culture medium was changed to a defined buffer containing 116 mM NaCl, 1.8 mM CaCl2, 0.8 mM MgSO4, 5.4 mM KCl, 1 mM NaH2PO4, 14.7 mM NaHCO3, 10 mM HEPES, and 5.5 mM glucose (Liao et al. 2011).

Cytotoxicity assessment

Cell damage was assessed by measuring the activity of lactate dehydrogenase by colorimetric detection of formazan, using a lactate dehydrogenase diagnostic kit (Promega, Madison, WI, USA).

Measurement of ATP content

Intracellular ATP levels were measured using a luciferin/luciferase-based method in accordance with previously reported protocols (Chen and Liao 2002). The cells (96-well) were washed with cold phosphate-buffered saline and extracted with 100 μL of cell lysate buffer [100 mM NaCl, 20 mM Tris-HCl (pH 8.0), 1 mM EDTA, 1% NP-40, 0.5% sodium deoxycholate, and 0.05% sodium dodecyl sulfate]. Then, 1 μL of the lysate was diluted to 100 μL with water and mixed with 100 μL of luciferase/luciferin reagent. After 10 s, the light emitted was recorded using a luminometer at 562 nm and integrated over 5 s. Relative ATP levels were normalized by the protein contents.

Measurement of glutathione content

Intracellular levels of glutathione were measured as previously reported (Chen and Liao 2003). The assayed cells were washed with cold phosphate-buffered saline, incubated at 37°C for 20 min with 0.05 mM monochlorobimane (Molecular Probes, Eugene, OR, USA), then washed and dispersed with a defined buffer. The fluorescence of monochlorobimane was measured using a fluorometer (Ex 355 nm and Em 460 nm).

Measurement of ROS

Intracellular oxidative stress was assayed by measuring intracellular oxidation of dichlorofluorescein, as described previously (Liao et al. 2011). Cells were loaded with 5 μM 2′,7′-dichlorofluorescein diacetate (Molecular Probes) at 37°C for 30 min, washed, and subjected to the treatment. The fluorescence signal of oxidized 2′,7′-dichlorofluorescein was measured using a fluorometer (Ex 485 nm and Em 510 nm).

Measurement of intracellular zinc content

Cells were scraped off and homogenized in 1 mL of homogenization buffer (0.32 M sucrose, 5 mM HEPES, pH 7.4). Precipitation of cytosolic proteins was performed by adding trichloroacetic acid (10% w/v) to the supernatants. The amount of zinc was estimated in 1 mL of homogenization buffer containing 0.001% (w/v) N-(6-methoxy-8-quinolyl)-p-toluenesulfonamide (Molecular Probes) (Chen and Liao 2003). N-(6-methoxy-8-quinolyl)-p-toluenesulfonamide fluorescence was measured using a fluorometer (Fluoroskan Ascent; Labsystems, Helsinki, Finland) with excitation at 355 nm and emission at 510 nm.

Western blot

After separation, electronic transfer, and blocking (Chang et al. 2010), the membranes were incubated with the indicated antibodies against glyceraldehyde 3-phosphate dehydrogenase (GAPDH, Sigma-Aldrich), Src, phospho-Src (Tyr-416), ERK, phospho-ERK (Tyr-204), c-Jun N-terminal kinase (JNK), phospho-JNK (Thr-183/Tyr-185), p38, phospho-p38 (Tyr-182), cytosolic Ca2+-dependent phospholipase A2 (cPLA2), phospho-cPLA2 (Ser-505), Raf, phospho-Raf (Ser-338), phospho-Raf (Tyr-340), phospho-Raf (Ser-259), and protein kinase C-α (PKC-α) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The blots were developed using enhanced chemiluminescence western blotting reagents. The intensity of each signal was determined by a computer image analysis system (IS1000; Alpha Innotech Corporation, San Leandro, CA, USA).

Membrane fractionation

Membrane fractionation was carried out as previously reported (Liao et al. 2011). The assayed cells were resuspended in lysis buffer [20 mM Tris-HCl (pH 8.0), 0.4 mM phenylmethylsulfonyl fluoride, 20 μM leupeptin, 0.005 U/mL aprotinin, and 2 μg/mL soybean trypsin inhibitor] and homogenized, and a crude membrane and cytosolic fraction were separated by centrifugation at 50 000 g for 30 min at 4°C. Proteins obtained from cytosolic and membrane fractions were kept at −70°C until use.

Statistical analysis

The data are expressed as mean ± SD. Statistical analysis was carried out using one-way anova, followed by Dunnett's test to assess the statistical significance between treated and untreated groups through all experiments. A level of p < 0.05 was considered statistically significant.

Results

Effects of diethylmaleate and iodoacetate on astrocyte viability

Astrocytes, prepared from cerebral cortices of 1-day-old Sprague–Dawley rats, expressed increasing level of glial fibrillary acidic protein, glutamine synthetase, and astrocytic excitatory amino acid transporter-1, but expressed decreasing level of neuronal excitatory amino acid transporter-3 as they grew in vitro. The obtained astrocytes were characterized by their immunoreactivity of glial fibrillary acidic protein (more than 95%) and flattened, polygonal appearance (Figure S1). To explore the potential effects of energy failure and oxidative stress on the cell viability of astrocytes, experimental conditions were produced by treating cells with iodoacetate (an agent can modify thiol group and inhibit GAPDH) and diethylmaleate (an agent can modify thiol group and chelate glutathione), respectively (Schmidt and Dringen 2009; Zheng and Bizzozero 2010). Their selective effects on targeted GAPDH activity are shown in Figure S2. Sustained exposure to diethylmaleate (Fig. 1a) and iodoacetate (Fig. 1b) caused a progressive decline of intracellular ATP levels, reduced intracellular glutathione levels, and elevated ROS production in astrocytes and these alterations were accompanied by cell death, respectively. However, no elevation of ROS generation was detected after 6 h of treatment with a high concentration of diethylmaleate or iodoacetate, possibly because of apparent cytotoxicity (Fig. 1a and b). We then wanted to examine whether diethylmaleate or iodoacetate had an aggravating effect on astrocyte death. As shown in Fig. 1c and d, a non-toxic concentration of iodoacetate augmented diethylmaleate-induced astrocyte death and vice versa. At first, iodoacetate and diethylmaleate primarily depleted their specific effector targets ATP (Fig. 2a) and glutathione (Fig. 2b), respectively. However, depletion of both ATP and glutathione was detected in astrocytes when treated with each inhibitor at higher toxic concentrations or longer exposure. Intriguingly, non-toxic concentrations of iodoacetate (5 μM) and diethylmaleate (50 μM) co-treatment caused remarkable depletion of ATP (Fig. 2a) and glutathione (Fig. 2b) and cell death (Fig. 2c). These findings show that oxidative stress caused by diethylmaleate had an augmented effect on astrocyte death in the presence of energy failure caused by iodoacetate and vice versa. Follow-up studies were done by treating astrocytes with iodoacetate (5 μM) and diethylmaleate (50 μM) to elucidate their action characteristics.

Figure 1.

Iodoacetate and diethylmaleate caused astrocyte damage. Astrocytes were treated with various concentrations of diethylmaleate (a) and iodoacetate (b) over time. Cell damage was evaluated by measuring lactate dehydrogenase (LDH) efflux. Cellular content of reduced glutathione (GSH) and ATP and the generation of reactive oxygen species were measured. The level of GSH, ATP, and fluorescence in medium control was defined as 100%. *p < 0.05 and **< 0.01 versus each medium control, = 4. Astrocytes were treated with various concentrations of diethylmaleate in the absence or presence of iodoacetate (5 μM) (c). Astrocytes were treated with various concentrations of iodoacetate in the absence or presence of diethylmaleate (50 μM) (d). Cell damage (3 h) was evaluated by measuring LDH efflux. **< 0.01 versus each control, = 4.

Figure 2.

Combinatory effects of iodoacetate and diethylmaleate on astrocytic alterations. Astrocytes were treated with various concentrations of iodoacetate or diethylmaleate or iodoacetate plus diethylmaleate over time. Intracellular ATP levels (a) and cellular content of GSH (b) were measured. The level of ATP and GSH in medium control was defined as 100%. Cell damage was evaluated by measuring lactate dehydrogenase (LDH) efflux (c). *< 0.05 and **< 0.01 versus each medium control, = 4.

Roles of MAPKs in diethylmaleate/iodoacetate-induced astrocyte death

To elucidate whether MAPKs are involved in the development of diethylmaleate/iodoacetate-induced astrocyte death, potential MAPK alterations, and contributions were examined. As shown in Fig. 3a, there were significant expressions of phosphorylated ERK, p38, and JNK in diethylmaleate/iodoacetate-treated astrocytes when compared to control and single treatment alone. The pharmacological inhibitor U0126 (against ERK) significantly attenuated diethylmaleate/iodoacetate-induced astrocyte damage, whereas, within the tested concentrations, SB203580 (against p38) and SP600125 (against JNK) had negligible effects on diethylmaleate/iodoacetate-evoked cell injury (Fig. 3b). Concentration curves of pharmacological inhibitors against diethylmaleate/iodoacetate-induced astrocyte death are shown in Figure S3. These results show that ERK but not p38 or JNK is an active participant in the diethylmaleate/iodoacetate-induced astrocyte death program.

Figure 3.

Combinatory effects of iodoacetate and diethylmaleate on cellular signaling. Astrocytes were treated with medium, iodoacetate (5 μM), diethylmaleate (50 μM), or in combination. The collected proteins (3 h) were subjected to western blot analysis using antibodies against phosphorylated (P-) and non-phosphorylated ERK, p38, Jun N-terminal kinase (JNK), cytosolic phospholipase A2 (cPLA2), Raf, and Src, and membrane-associated protein kinase C-α (PKC-α) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). One representative blot from three independent experiments is shown (a). Astrocytes were pre-treated with medium, U0126 (ERK inhibitor), SB203580 (p38 inhibitor), SP600125 (JNK inhibitor), methyl arachidonyl fluorophonate (MAFP) (cPLA2 inhibitor), GW5074 (Raf inhibitor), PP2 (Src inhibitor), GF109203X (PKC inhibitor), or diphenyleneiodonium chloride (antioxidant) for 30 min before iodoacetate plus diethylmaleate exposure. Cell damage (6 h) was evaluated by measuring lactate dehydrogenase (LDH) efflux (b). Astrocytes were treated with medium, iodoacetate, diethylmaleate, or in combination. The generation of reactive oxygen species (3 h) was measured. The intensity of fluorescence in medium control was defined as 100% (c). **< 0.01 versus medium control and ##< 0.01 versus iodoacetate/diethylmaleate control, = 4.

Role of cPLA2 in diethylmaleate/iodoacetate-induced astrocyte death

cPLA2 has been shown to be involved in astrocyte death. Evidence shows that the enzymatic activity of cPLA2 is regulated post-translationally via direct phosphorylation by several serine/threonine kinases, including ERK (Gabryel et al. 2007; Chang et al. 2010; Liao et al. 2011). To further elucidate the contribution of ERK-associated cytotoxic mechanisms, we wanted to determine the activation and role of cPLA2 in diethylmaleate/iodoacetate-treated astrocytes. We found that diethylmaleate/iodoacetate treatment up-regulated serine phosphorylation in cPLA2 (Fig. 3a). A pharmacological inhibitor of cPLA2, methyl arachidonyl fluorophonate, attenuated diethylmaleate/iodoacetate-induced astrocyte damage (Fig. 3b). These results suggest that enhanced cPLA2 activity is implicated in astrocyte death after exposure to diethylmaleate/iodoacetate.

Roles of Raf in diethylmaleate/iodoacetate-induced astrocyte death

As ERK activation plays an important role in the transmission of the cytotoxic signal of diethylmaleate/iodoacetate, we next sought to determine which upstream signaling molecules involved in the activation of ERK are up-regulated in diethylmaleate/iodoacetate-treated astrocytes. As the most extensively characterized regulator of the ERK signaling pathway is Ras/Raf (Pawson 1995), we wanted to determine its activity and contribution to cell survival in diethylmaleate/iodoacetate-treated astrocytes. Generally, Raf activity is regulated by its membrane translocation and phosphorylation in critical motif residues (Wu et al. 2008a). Diethylmaleate/iodoacetate treatment increased Raf phosphorylation in serine 338 and tyrosine 340, two phosphorylated residues that positively regulate Raf activity. Phosphorylation in serine 259 residue results in auto-inhibition of Raf activity, and this phosphorylation event was decreased in diethylmaleate/iodoacetate-treated astrocytes (Fig. 3a). In parallel, inhibition of Raf signaling by the inhibitor GW5074 attenuated diethylmaleate/iodoacetate-induced cell damage (Fig. 3b). These findings show that diethylmaleate/iodoacetate-induced astrocyte death is accompanied by elevated activity of Raf.

Roles of Src and PKC in diethylmaleate/iodoacetate-induced astrocyte death

Intracellular signaling cascades like the PKC- and Src-dependent pathways have also been reported to activate ERK (Chang et al. 2005; Wu et al. 2008b; Wang et al. 2009). We therefore examined the involvement of these kinases in diethylmaleate/iodoacetate-induced cell damage. As shown in Fig. 3a, diethylmaleate/iodoacetate elevated Src and PKC-α activity, as evidenced by increased phosphorylation and membrane translocation, respectively. Treatment with Src inhibitor PP2 or PKC inhibitor GF109203X attenuated diethylmaleate/iodoacetate-induced astrocyte death (Fig. 3b). These results show that diethylmaleate/iodoacetate-induced cell damage is associated with increased Src and PKC activity.

Roles of ROS in diethylmaleate/iodoacetate-induced astrocyte death

Evidence indicates that ROS are crucial regulators of astrocyte viability (Chang et al. 2010; Liao et al. 2011). To determine whether ROS contribute to diethylmaleate/iodoacetate-induced cell death in astrocytes, we assessed the generation of ROS and the effect of antioxidants on diethylmaleate/iodoacetate-induced cell death. We found that diethylmaleate/iodoacetate treatment increased intracellular redox potential change in astrocytes (Fig. 3c). Treatment with the antioxidant diphenyleneiodonium chloride significantly attenuated diethylmaleate/iodoacetate-induced cell death (Fig. 3b). These results suggest that ROS may mediate cell death caused by diethylmaleate/iodoacetate treatment.

Role of zinc in diethylmaleate/iodoacetate-induced astrocyte death

Evidence suggests that energy failure and/or oxidative stress can alter cellular functions via the mobilization of intracellular zinc (Matsui et al. 2010). We found that treatment with iodoacetate and diethylmaleate in combination elevated the intracellular chelatable zinc level. A selective zinc chelator, TPEN, decreased basal chelatable zinc level in astrocytes and inhibited diethylmaleate/iodoacetate-mobilized intracellular chelatable zinc levels (Fig. 4a). TPEN also attenuated diethylmaleate/iodoacetate-induced astrocyte damage (Fig. 4b). In parallel with the protective action, TPEN attenuated diethylmaleate/iodoacetate-induced astrocyte alterations, including ROS generation (Fig. 4c), protein phosphorylations in ERK, p38, JNK, cPLA2, Raf serine-338, Raf tyrosine-340, Raf serine-259, and Src and PKC-α membrane translocation (Fig. 5). These results demonstrate that iodoacetate and diethylmaleate co-treatment might trigger ERK-associated astrocyte death involving the mobilization of intracellular chelatable zinc.

Figure 4.

Iodoacetate and diethylmaleate mobilized intracellular zinc. Astrocytes were pre-treated with medium or N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) (Zn chelator) for 30 min before iodoacetate, diethylmaleate, or in-combination exposure. The level of intracellular chelatable zinc (1 h) was determined by N-(6-methoxy-8-quinolyl)-p-toluenesulfonamide (TSQ). The TSQ fluorescence in medium control was defined as 100% (a). Cell damage (6 h) was evaluated by measuring lactate dehydrogenase (LDH) efflux (b). The generation of reactive oxygen species (3 h) was measured. The intensity of fluorescence in medium control was defined as 100% (c). **< 0.01 versus medium control and ##< 0.01 versus the iodoacetate/diethylmaleate group, = 4.

Figure 5.

Effects of N,N,N′,N′-tetrakis(2-pyridyl-methyl)ethylenediamine (TPEN) on iodoace-tate/diethylmaleate-induced cellular signaling. Astrocytes were pre-treated with medium or TPEN (Zn chelator) for 30 min before iodoacetate/diethylmaleate exposure. The collected proteins (3 h) were subjected to western blot analysis using antibodies against phosphorylated (P-) and non-phosphorylated ERK, p38, Jun N-terminal kinase (JNK), cytosolic phospholipase A2 (cPLA2), Raf, and Src, and membrane-associated protein kinase C-α (PKC-α) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). One representative blot from three independent experiments is shown.

Role of ATP depletion and oxidative stress in zinc-induced astrocyte death

Energy failure, intracellular glutathione depletion, and the consequences of oxidative stress are well-characterized alterations in zinc-treated cells (Sheline et al. 2000; Chen and Liao 2003). We therefore decided to explore whether these events also played roles in ZnCl2-induced astrocyte death. Sustained exposure to ZnCl2 caused damage to astrocytes in a time- and concentration-dependent manner (Fig. 6a). Zinc chloride-induced astrocyte death was accompanied by a progressive decline of ATP (Fig. 6b) and glutathione (Fig. 6c). As shown in Fig. 6d, pre-treatments with the energy substrate pyruvate and the glutathione precursor NAC attenuated ZnCl2-induced cell death. In contrast, iodoacetate and diethylmaleate augmented cell death in the presence of non-toxic concentrations of ZnCl2. In addition to pyruvate, pre-treatments with other energy substrates such as citrate, lactate, and oxaloacetate also alleviated ZnCl2-induced cell death (Figure S4). These findings suggest that ATP and glutathione depletion might play an active role in astrocyte death caused by ZnCl2.

Figure 6.

Zinc caused astrocyte damage. Astrocytes were treated with various concentrations of ZnCl2 over time. Cell damage was evaluated by measuring lactate dehydrogenase (LDH) efflux (a). Intracellular ATP levels (b) and cellular content of GSH (c) were measured. The level of ATP and GSH in medium control was defined as 100%. Astrocytes were pre-treated with medium, pyruvate (energy substrate), iodoacetate, N-acetyl cysteine (NAC) (glutathione precursor), or diethylmaleate for 3 h. After pre-treatments, the cells were then changed to fresh media and treated with ZnCl2 (0 or 50 μM). Cell damage (6 h) was evaluated by measuring LDH efflux (d). *< 0.05 and **< 0.01 versus medium control and ##< 0.01 versus the ZnCl2-treated group, = 4.

Discussion

In this study, we found that iodoacetate and diethylmaleate aggravated astrocyte death partly through ERK- but not JNK- and p38-mediated mechanisms. The activation of redox-sensitive ERK increases its kinase activity, which in turn phosphorylates several cytoplasmic and nuclear proteins critical in regulating cell survival and death. cPLA2 is an important enzyme for arachidonic acid release in cells and for modulating a number of intracellular processes. cPLA2 and the release of arachidonic acid from membrane phospholipids and subsequent initiation of the arachidonic acid cascade have been implicated in the pathology of cerebral ischemia and in astrocyte death (Gabryel et al. 2007; Kishimoto et al. 2010). Together with our previous findings (Chang et al. 2010; Liao et al. 2011), the results of this study showed that ERK contributed to diethylmaleate/iodoacetate-induced astrocyte death involving cPLA2, suggesting a role for ERK/cPLA2 in energy failure/oxidative stress-induced astrocyte death.

The regulation of ERK phosphorylation requires a balance between the activity of protein kinases and protein phosphatases. Our data showed that ERK was activated by diethylmaleate/iodoacetate, manifested by increased phosphorylation levels. Generally, the Ras/Raf cascade is a common upstream kinase for the phosphorylation of ERK (Pawson 1995; Wu et al. 2008a). The activation of the Ras/Raf/ERK cascade occurs in neurodegenerative diseases, including ischemia (Guo et al. 2006; Wu et al. 2008a). Results of biochemical and pharmacological studies demonstrated the initiation of the Ras/Raf/ERK cascade in diethylmaleate/iodoacetate-treated astrocytes and suggested that the activation of the Ras/Raf/ERK pathway and accompanying ERK substrate cPLA2 were associated with diethylmaleate/iodoacetate-induced cell death in astrocytes.

The activation of PKC and Src family tyrosine kinases has been implicated in brain injury and in vitro cell death (Manzerra et al. 2001; Guo et al. 2006; Henriksson et al. 2007; Chang et al. 2010; Liao et al. 2011). Although evidence demonstrates that the activation of PKC or Src also shows cytoprotective potential (Cordey and Pike 2006; Aras et al. 2009), our results indicated that the activation of PKC-α and Src played a detrimental role in diethylmaleate/iodoacetate-induced astrocyte death. Evidence further indicates that PKC and Src family tyrosine kinases are upstream activators that transmit signals to the Ras/Raf/ERK cascade (Chang et al. 2005; Wang et al. 2009). Generally, the small GTP-binding protein Ras is recruited to the membrane in response to stimulation (Pawson 1995). The transmission of signals from Ras is propagated by sequential phosphorylation and activation of kinases. The classical Ras-mediated pathway involves the binding of Raf and subsequent phosphorylation of Raf at serine 338, which in turn activates ERK (Wu et al. 2008a). Negative regulation of Raf has also been reported, in that phosphorylation of serine 259 can inactivate Raf probably by promoting the formation of a Raf auto-inhibitory complex via association with 14-3-3 (Wu et al. 2008a; Han and Meier 2009). Besides, the Ras-independent pathways also lead to ERK activation. For example, PKC and Src have been identified as activators of ERK via the stimulation of Raf kinase (Kolch et al. 1993; Wu et al. 2008b). Ras-associated Raf serine-338 phosphorylation was detected in diethylmaleate/iodoacetate-treated astrocytes. We also had observed an increased phosphorylation of Raf at tyrosine 340, indicating a potential involvement of Src. Consistent with these findings, our results showed that the activation of PKC and Src could lead to the activation of the Ras/Raf/ERK signaling cascade in diethylmaleate/iodoacetate-treated astrocytes. Having established the involvement of the PKC- and Src-related Ras/Raf/ERK signaling cascades in diethylmaleate/iodoacetate-induced astrocyte death, we next attempted to identify the determinant events by which these signaling cascades were up-regulated. It has been suggested that ROS generation and consequent oxidative stress can activate PKC, Src, and Ras/Raf/ERK signaling (Aikawa et al. 1997; Noh and Koh 2000). Indeed, in our system, ROS and oxidative stress appeared to be key triggers to initiate and propagate diethylmaleate/iodoacetate-induced astrocyte death.

Another interesting finding in this study was the mobilization of endogenous chelatable zinc in diethylmaleate/iodoacetate-treated astrocytes. The attenuating effects of TPEN revealed that endogenous chelatable zinc was one effector that could mediate diethylmaleate/iodoacetate's actions. Zinc dyshomeostasis has been recognized as an important mechanism for cell death in brain injury. Accumulating evidence shows that hypoglycemia-, hypoxia-, and oxidative stress-induced cell death are attributable to free chelatable zinc (Suh et al. 2004; Lee et al. 2009; Lee and Koh 2010; Matsui et al. 2010; Udayabanu et al. 2012). Previously, we had demonstrated that ZnCl2 initiated several cellular alterations and caused apparent astrocyte death without the significant involvement of apoptosis and autophagy. Zinc chloride triggered PKC-α- and Src-related Ras/Raf/ERK signaling pathways through ROS-mediated mechanisms to activate cPLA2 and consequent astrocyte death (Chang et al. 2010; Liao et al. 2011). Although we demonstrated the association of diethylmaleate/iodoacetate-induced astrocyte death with generation of ROS, phosphorylations in Src, Raf, ERK, and cPLA2, and membrane translocation of PKC-α, their potential mutual interaction and cross-talk were not precisely demonstrated in this study. In accordance with relevant studies, our findings showed that diethylmaleate/iodoacetate was able to trigger biochemical cascades similar to those initiated by zinc in astrocytes, thus suggesting that ROS-associated and PKC-α- and Src-related Ras/Raf/ERK/cPLA2 signaling pathways might also play crucial roles in diethylmaleate/iodoacetate-induced astrocyte death.

Growing evidence suggests that elevated intracellular free zinc levels may play a critical role in inducing cell death. Zinc is an endogenous molecule that may cause diversity of cellular alterations such as glycolytic inhibition and glutathione depletion and contribute to cell death following cerebral ischemia (Koh et al. 1996; Stork and Li 2009). The supplementation of energy metabolic substrates and glutathione precursors provoking a beneficial effect against ZnCl2-induced astrocyte death showed that the depletion of ATP and glutathione was not only a concurrent event of cell death but also a possible downstream mediator playing an active role in inducing astrocyte death. The importance of ATP and glutathione depletion, two well-characterized alterations in zinc-treated cells, is recognized in zinc-induced cell death (Sheline et al. 2000; Chen and Liao 2003). Studies have suggested a mutual interaction between glutathione depletion and energy failure. Glutathione depletion can result in lower intracellular ATP levels and the reduction in glutathione is one possible consequence of energy failure (Vesce et al. 2005; Schmidt and Dringen 2009). We had observed a similar phenomenon in diethylmaleate/iodoacetate-treated astrocytes, a treatment condition mimicking ATP and glutathione depletion. Unlike iodoacetate, diethylmaleate itself had a negligible effect on GAPDH activity. Intriguingly, diethylmaleate further augmented iodoacetate-induced reduction in GAPDH activity. Because the depletion of ATP could be directly related to the inhibition of GAPDH activity, the augmented inhibition of GAPDH activity might partly explain the potentiating effect of each on ATP levels and cytotoxicity. Therefore, a vicious cycle of cascades between energy failure, oxidative stress, and zinc toxicity might occur.

Energy failure, oxidative stress, and elevation of zinc level occur in pathological conditions and are increasingly implicated as key mediators of ischemic brain injury (Choi and Koh 1998; Côté et al. 2005; Tavazzi et al. 2005; Al-Majed et al. 2006; Al-Majed 2011). In addition to being a concurrent event of cell death, ATP and glutathione depletion might also play a role in driving further cellular alterations and cell death, at least in part, involving a zinc-related mechanism. As with ZnCl2 (Chang et al. 2010; Liao et al. 2011), our findings suggest a potential link between cPLA2 activation and the depletion of ATP and glutathione (Fig. 7). Diethylmaleate/iodoacetate may cause cPLA2 activation through several mechanisms and ROS-associated PKC-α- and Src-related Ras/Raf/ERK signaling may be one. These observations indicate that ROS/PKC-α, Src/Ras/Raf/ERK signaling and cPLA2 are active participants in astrocyte death and that the depletion of ATP/glutathione and zinc are critical upstream effectors in initiating cytotoxic cascades. In this study, the cellular sources of ROS generation and zinc mobilization were not answered. Evidence suggests that ROS might trigger zinc release from PKC (Korichneva et al. 2002). Therefore, the mitochondria, PKC, zinc finger proteins, and metallothionine might be potential sources for their production, respectively. However, these issues remain open for further investigations.

Figure 7.

A possible mechanism of cytotoxicity elicited by iodoacetate/diethylmaleate treatment of astrocytes is proposed. This schematic diagram indicates the signaling molecules employed in mediating astrocyte death after iodoacetate/diethylmaleate treatment. Some additional signaling molecules and cascades have been omitted for the sake of clarity. [Zn2+]i represents intracellular free zinc. The label ↑ in each molecule indicates its activity was stimulated and the label ↓ in each molecule indicates its content was decreased.

Acknowledgements

This study was supported by grants from Taichung Veterans General Hospital and Hung-Kuang University (TCVGH-HK958005), Taichung, Taiwan. The authors declare that there are no conflicts of interest.

Ancillary