Rapid Disruption of Cellular Integrity of Zinc-treated Astroglia Is Regulated by p38 MAPK and Ca-dependent Mechanisms.

Cultured cortical primary astroglia treated with zinc died while rapidly detached from culture plates, a distinct part of zinc-treated astroglia. In the present study, we investigated the mechanism underlying the rapid change in the morphologic integrity of zinc-treated astroglia. Among the early cellular events occurring in zinc-treated astroglia, strong activation of p38 MAPK and JNK was evident. Although inhibitors of p38 (SB203580 and SB202190) or JNK (SP600125) did not protect zinc-insulted astroglia from cell death, the p38 inhibitors, but not the JNK inhibitor, suppressed actin filament and cell morphology disruption. The Ca(2+) ionophore, A23187, also suppressed actin filament and cell morphology disruption, but not cell death, of zinc-insulted astroglia. However, A23187 did not inhibit p38 MAPK activation in zinc-treated astroglia. Together these results suggest that zinc influx in astroglia results in rapid loss of the morphologic integrity via mechanisms regulated by p38 kinase and/or Ca(2+) signaling.

INTRODUCTION

Acute pathologic conditions, such as cerebral ischemia, induce excess release of free zinc from excitatory synapses, and excessively released zinc causes brain cells to die (Frederickson et al., 2005). Zinc-induced cell death occurs via zinc influx into cells, and zinc influx results in disruption of thiol homeostasis and reactive oxygen species (ROS) stress (Frederickson et al., 1988; Kim et al., 1999a; 1999b; Suh et al., 2000; Kim et al., 2003). In vitro studies have shown that both neurons and astroglia are sensitive to chronic low doses of free zinc (Dineley et al., 2000; Cho et al., 2003; Ralph et al., 2010). Cultured primary astroglia cells challenged with free zinc become elongated in shape, swell and soon thereafter detach from the culture plate. The rapid change and characteristic loss of morphologic integrity of astroglia cells precedes with the events leading to cell death (Cho et al., 2003; Kim et al., 2003). Astroglial cell death induced by zinc is different from that of other insults, such as H2O2 or Fe2+, in that astroglia cells treated with the latter stimulants undergo demise, while most cells are attached on the culture plate. Cell morphology changes, including detachment from the culture plate, might proceed with disruption of cytoskeletal elements. Because zinc-induced astroglia death is initiated by zinc-influx (Ryu et al., 2002; Kim et al., 2003), the rapid and distinct loss of morphologic integrity is likely to occur via zinc-mediated changes in dynamics of cytoskeletal elements. However, it is unclear how this process occurs and whether intracellular cytoskeleton dysregulation is a prerequisite to the cell death process in zinc-insulted cells.

Actin is the major cytoskeletal element in most eukaryotic cells. Failure of actin dynamics leads to disruption of cell morphology (Carlier et al., 1994; Dalle-Donne et al., 2001). Actin dynamics and cell morphology have been shown to be regulated by several factors, including mitogen-activated protein kinases (MAPKs). For instance, the MEK inhibitor, U0126, blocks stellate process extension of fibroblast growth factor 2 (FGF2)-treated astroglia in vitro, while inhibition of the p38 MAPK pathway by SB202190 facilitate stellate process growth and extension of the same cells (Heffron and Mandell, 2005). The p38 MAPK cascade regulates actin polymerization in platelet-derived growth factor-induced cytoskeleton remodeling of aortic smooth muscle cells (Pichon et al., 2004). p38 MAPK activation induces actin re-arrangements in VEGF-treated human endothelial cells (Rousseau et al., 1997). MAPK activated protein kinase 2 regulates actin polymerization in ventilator associated lung injury (Damarla et al., 2009).

Intracellular Ca2+ homeostasis is also important in the maintenance of cell morphology. Calcium stimulates actin filament assembly (Downey et al., 1990; Carlier et al., 1994), and Ca2+-mediated actin filament regulation contributes to growth cone behaviors, synaptic vesicle trafficking in presynaptic terminals, and synaptic plasticity in dendritic spines of postsynaptic neurons (Matus, 2000; Levitan, 2008). Calcium also activates Ca2+-dependent proteases, for examples calpains, that cleave actin-binding proteins, dissociating the anchorage between the plasma membrane and cytoskeleton (Harris and Morrow, 1990; Carragher and Frame, 2002).

MAPKs are important elements in neuronal and non-neuronal cell death (Wang et al., 1998; Guo and Bhat, 2007). Zinc influx stimulates MAPK pathways in neuronal cells (Seo et al., 2001; An et al., 2005). ERK1/2 signaling leads to mitochondrial dysfunction in extracellular zinc-induced neurotoxicity in rat cultured neurons (He and Aizenman, 2010). In the present study, we explored the mechanisms that underlie rapid changes in the morphologic integrity of zinc-treated astroglia.

MATERIALS AND METHODS

Primary neuronal and astroglial cultures

Primary cortical neuron and astroglia were cultured, as described previously (Cho et al., 2003; Im et al., 2006a). Primary cortical neuronal cultures were prepared from embryonic day 15.5 (E15.5) ICR mouse cortices. Mouse cortices were triturated and dissociated cortical cells were plated in minimal essential medium (MEM) supplemented with 20 mM glucose, 5% fetal bovine serum (FBS), 5% horse serum, and 2 mM glutamine, at a density of 5 hemispheres per plate (4×105 cells per well) onto poly-D-lysine (100 µg/ml) and laminin (4 µg/ml) coated 24-well plates. On day 6 in vitro (DIV6), cytosine arabinoside was added to a final concentration of 10 µM and maintained for 2 days to halt glial growth. FBS and glutamine were not supplemented in the culture medium from day 6, and the medium was changed twice a week after day 8. Cultures in DIV13-15 were used for studies. When treated with zinc and/or drugs, HEPES-buffered salt solution (120 mM NaCl, 5.4 mM KCl, 0.8 mM MgCl2, 1.8 mM CaCl2, 20 mM HEPES, 10 mM NaOH, and 15 mM glucose) was used.

Cortical astroglia cells were cultured, as described previously (Cho et al., 2003; Im et al., 2006b). ICR neonatal mice (day 0-1) were sacrificed and the cortices were isolated. Dissociated cortical cells were plated in MEM supplemented with 20 mM glucose, 5% fetal bovine serum, 5% horse serum, 2 mM glutamine, 50 µg/ml streptomycin, and 50 unit/ml penicillin at a density of 2 hemispheres per plate (5×104 cells per well) onto poly-D-lysine-coated (20 µg/ml) 24-well plates. Cells were maintained at 37℃ in a humidified 5% CO2 incubator. The medium was changed twice a week. The experiments were carried out on cells which were grown to confluence, and obtained after 2 weeks of culturing. Cortical astroglia cultures were characterized by anti-GFAP antibody (Dako, Carpinteria, CA, USA), an astrocyte marker, and isolectin B4 conjugated antibody (Sigma, St. Louis, MO, USA), a microglia marker, as described previously (Cho et al., 2003). Various inhibitors or drugs, including SB203580 and SB202190, were obtained from Calbiochem (La Jolla, CA, USA).

Western blot analysis

Western blot analysis was carried out as described previously (Im et al., 2006a). Briefly, primary astroglia were suspended in lysis buffer containing 150 mM NaCl, 1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM Sodium orthovanadate (Na3VO4), and protease inhibitor cocktails (Roche, Basel, Swiss) in 20 mM Tris-HCl (pH 7.4). Lysates were centrifuged for 10 min at 12,000 rpm at 4℃, and the resulting supernatants were collected. Protein contents were determined using a bicinchoninic acid assay kit (St. Louis, MO, USA). Protein samples were electrophoresed by 10% SDS-PAGE and then transferred to polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA, USA). The blots were blocked with 5% non-fat dried milk for 1 h and incubated with primary antibodies overnight at 4℃. The secondary antibody was incubated for 1 h and specific signals were detected using an enhanced chemiluminescence (ECL) kit (Amersham, Buckinghamshire, UK). Immunoblotting was performed using polyclonal anti-phospho-p38 (1:1,000, Cell signaling, USA), polyclonal anti-phospho-JNK (1:1,000, Cell signaling, USA), polyclonal anti-p38 (1:1,000, Santa Cruz, CA, USA), and monoclonal anti-JNK1 (1:1,000, Pharmingen, USA).

Assessments of cell death

Cell death was assessed by measuring the activity of lactate dehydrogenase (LDH) released in culture medium as described previously (Im et al., 2006a). Culture medium was collected 24 h after drug treatment to use for LDH activity unless indicated otherwise. Twenty-five µl of culture medium was transferred to a microplate and 100 µl of NADH solution (0.3 mg/ml NADH and 0.1 M potassium phosphate, pH 7.4) was added to the medium. Subsequently after 2 min, 25 µl of pyruvate solution (22.7 mM pyruvate and 0.1 M potassium phosphate, pH 7.4) was added. After adding pyruvate solution, the decrease of absorbance at 340 nm, which indicates the conversion of NADH to NAD+, was measured by SpectraMax microplate reader (Molecular Device, Sunnyvale, USA). LDH activity was normalized in that sham-treated culture and culture showing complete cell death were 0% and 100%, respectively, and normalized LDH activity was regarded as an indicator of cell death. Complete cell death was induced by treatment of cells to 2 mM H2O2 for 24 h. Cell death was also confirmed by trypan blue staining, and by morphological changes on phase contrast microscope.

Immunofluorescence-labeled phalloidin staining of F-actin

Visualization of actin filament distribution in cells was performed as described previously (Chae et al., 2006). Cortical cells were fixed by 4% paraformaldehyde solution for 10 min at room temperature and washed in PBS twice. Then the cells were treated with 0.5% triton X-100 in PBS to help penetration of phalloidin. After being washed, the cells were stained with 0.4 µg/ml phalloidin-TRITC (Sigma, St Louis, MO, USA) for 90 min at 4℃. Stained cells were photographed by fluorescence light microscope (Axiovert 200, Carl Zeiss Micro-Imaging, Inc., USA).

Statistical analysis

All data were analyzed using Student t-test for comparisons between two samples or using ANOVA followed by the Newman-Keuls test. A statistical difference was accepted at 5% level.

RESULTS

Zinc-treated astroglia showed strong activation of p38 and JNK MAPKs

Astroglial cells insulted with zinc were elongated in shape after 6 h, then gradually swelled and detached from the bottom of culture plate after 9 h. The rapid loss of morphological integrity in zinc-treated cells proceeded in parallel with the cellular process leading to cell death as demonstrated in previous studies (Cho et al., 2003; Kim et al., 2003). To understand the mechanism underlying the extraordinarily rapid changes in cell's integrity and cell death processes of zinc-treated cells, we tested whether p38 and JNK were involved. As reported previously (Cho et al., 2003; Kim et al., 2003), zinc-dose and zinc-treated time dependent astroglial deaths were observed (Fig. 1A and B). Western blot analysis using anti-phospho-p38 or anti-phospho-JNK showed that both MAPKs were activated in 3 h after zinc challenge in a zinc-dose dependent manner (Fig. 1C). The activation of p38 MAPK by zinc (35 µM) was maximal at ~3 h after zinc challenge, and thereafter were slowly declined (Fig. 1D).

Fig. 1

p38 and JNK MAPKs were activated in zinc-treated astroglia. (A, B) Astroglial cell death by zinc increased in a zinc-dose (A) and zinc-treated time (B) dependent manner. LDH assay was performed after 24 h-exposure of cultures to indicated concentrations of zinc. (C) Western blots showing zinc-dose dependent activation of p38 and JNK MAPKS in astroglia. Assay was performed 3 h after zinc treatment. Anti-phospho-p38 MAPK and anti-phospho-JNK were used to deduce the activation levels of these kinases. (D) Western blots showing time-dependent activation of p38 and JNK MAPKS in zinc (35 µM)-treated astroglia.

p38 MAPK inhibitors suppressed cell morphology disruption of zinc-insulted astroglia

We examined whether p38 and JNK MAPKs were critical players in zinc-induced astroglia death. Addition of p38 MAPK inhibitors, SB202190 or SB203580, in culture media had zinc-treated astroglia on the culture plate with a complete attachment. Moreover, zinc-treated astroglia appeared intact in the presence of SB203580 when examined by a phase-contrast microscopy (Fig. 2A~D). However, it was not the case. The zinc-challenged SB203580-cotreated astroglia cells were strongly stained by trypan-blue (Fig. 2E~H), suggesting that they were not alive. Indeed, LDH assay showed that SB203580 and SB202190 did not protect from cell death of zinc-treated astroglia (Fig. 2I). Neither the JNK inhibitor, SP600125, nor the MEK inhibitor, PD98059, produced protective effects on zinc-induced cell death and cell morphology disruption (Fig. 2I; Table 1).

Fig. 2

The p38 MAPK inhibitor, SB203580, suppressed zinc-induced morphology disruption and detachment from the culture plate. (A~H) Phase contrast (A~D) or trypan-blue stained (E~H) astroglia with sham treatment (A, E) or astroglia treated with zinc (B, F), with zinc and SP600125 (C, G), or with zinc and SB203580 (D, H). Photomicrographs were taken 24 h after zinc (35 µM)-treatment in the presence of SP600125 (20 µM) or SB203580 (20 µM). Scale bar represents 100 µm. (I) LDH assay showing cell death levels of zinc (35 µM)-treated astroglia in the presence of PD98059 (PD: 20 µM), SP600125 (SP: 20 µM), or SB203580 (SB: 20 µM). LDH assay was performed 24 h after zinc treatment.

Table 1

Effects of various cellular factor inhibitors on morphological disruption and cell death of zinc-treated astroglia

Assays were performed 24 h after Zn2+ (35 µM)-treatment. Inhibitors were applied to cultures 1 h prior to the start of Zn2+ treatment

The treatment of primary astroglia with Cd2+ (10 µM) or H2O2 (300 µM) also activated p38 MAPK and caused to cell death (data not shown). However, SB203580 did not prevent both morphological disruption and cell death induced by Cd2+ or H2O2 (Table 2). In addition, SB203580 did not suppress morphological disruption and cell death of zinc (100 µM)-treated C6 glioma cells and of zinc (100 µM)-treated NIH3T3 fibroblasts (Table 3). These results raise the possibility that SB203580-dependent morphological protection is a process that is a distinct property of zinc-treated primary astroglia.

Table 2

Effects of the p38 MAPK inhibitor, SB203580, on morphological disruption and cell death of zinc-treated C6 glioma and NIH3T3 cells

Assays were performed 24 hrs after Zn2+ treatment. SB203580 was applied to cultures 1 h before Zn2+ treatment. *SB203580 concentration was 20 µM.

Table 3

Effects of the p38 MAPK inhibitor, SB203580, on morphological disruption and cell death of H2O2- or cadmium-treated primary astroglia culture

Assays were performed 24 hrs after Zn2+ treatment. SB203580 was applied to cultures 1 h before Zn2+ treatment. *SB203580 concentration was 20 µM

Supplement of intracellular Ca2+ helped the maintenance of cell morphology of zinc-insulted astroglia

We searched for other cellular signaling pathways that produce morphological disruption of zinc-treated astroglia. Inhibitors of ROS generating enzymes, namely the COX-2 inhibitor NS398 (30 µM), the NADPH oxidase inhibitor DPI (1 µM), the xanthine oxidase inhibitor allopurinol (5 µM), and the NOS inhibitor N-nitro-l-arginine methyl ester (L-NAME) (1 mM), and the free radical trapping agent phenyl-alpha-tert-butyl nitrone (PBN) (200 µM) did not affect zinc-induced morphological changes. In addition, the caspase inhibitor zVAD (10 µM), the PARP inhibitor benzamide (1 mM), the protein phosphatase 2A inhibitor okadaic acid (OKA; 1 µM), the NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC; 100 µM), protein kinase-A inhibitor H89 (1 µM), the Ca2+-calmodulin-dependent protein kinase (CAMK) inhibitor KN62 (1 µM), and the protein kinase C (PKC) inhibitor GF109203X (GFX; 1 µM) were also ineffective (Table 1).

Next, we examined whether cell morphology of zinc-treated astroglia can be regulated by Ca2+. Co-treatment of astroglia with A23187 (0.1 µM), a Ca2+ ionophore, for 24 h did not inhibit zinc-induced cell death (Fig. 3A), but did suppress cell morphology disruption, thus leaving zinc-treated cells attached on the culture plate (Fig. 3B~E). Treatment with the Ca2+ chelator, BAPTA (10 µM) suppressed cell morphology disruption and partially cell death of zinc-insulted astroglia (Fig. 3F~L). Together, these results suggest that Ca2+ is important for the maintenance of cell morphology of zinc-insulted astroglia.

Fig. 3

The Ca2+ ionophore, A23187, suppressed cell morphology disruption of zinc-treated astroglia. (A) LDH assay showing cell death levels of astroglia treated with A23187 (0.1 µ M), zinc (35 µM), zinc (35 µM)+A23187 (0.1 µM), or zinc (35 µM)+SB203580 (20 µM). LDH assay was performed after 24 h of drug treatment. (A) A23187; SB, SB203580. (B~E) Trypan-blue stained astroglia treated with zinc (35 µM), zinc (35 µM)+SB203580 (20 µM), A23187 (0.1 µM) or zinc (35 µM)+A23187. Photomicrographs were taken after 24 h of zinc treatment in the presence of SB203580 or A23187. (F~K) Trypan-blue stained astroglia with sham treatment (F, G), with zinc (H, I), or with zinc and SB203580 (J, K) in the presence or absence of the Ca2+ chelator, BAPTA (10 µM). Photomicrographs were taken 24 h after treatment of zinc (35 µM) and SB203580 (20 µM). (L) LDH assay showing cell death levels of astroglia treated with zinc (35 µM) or zinc (35 µM)+SB203580 (20 µM) in the presence or absence of BAPTA (10 µM). LDH assay was performed after 24 h of drug treatment. Scale bar represents 100 µm. SB, SB203580. The data are presented as means±S.E.M. (n=6~12). **Denotes difference at p<0.05.

Zinc influx disrupted actin filament organization in astroglia, which was reversed by p38 inhibitors or A23187

Because cell morphology of zinc-treated astoglia was retained in the presence of SB203580 or A23187, we extended our efforts to visualize the actin organization in zinc-treated astoglia. Zinc-treated astoglia were stained with TRITC-conjugated phalloidin which labels actin fibers. Microscopic examination revealed that actin fibers in untreated normal astroglia were distributed throughout the cytoplasm, along with a mildly preferential localization at the plasma membrane. After 9~10 h of zinc treatment, actin filament distribution in the cytoplasm was disorganized by leaving numerous patched clumps within cells. However, in the presence of SB203580, zinc-induced actin filament disruption disappeared, and instead preferential distribution of actin filament at the cell periphery was strongly reinforced, giving rise to a distinct actin filament-ring along the plasma membrane (Fig. 4A~C).

Fig. 4

SB203580 and A23187 suppressed actin-filament disruption in zinc-treated astroglia. (A~E) Photomicrographs showing actin-filament distributions in zinc-treated astroglia stained with TRITC-labeled phalloidin. Sham control (A) and astroglia treated with Zn (35 µM) alone (B), Zn (35 µM)+SB203580 (20 µM) (C), A23187 (0.1 µM) alone (D), or zinc (35 µM)+A23187 (0.1 µM) (E). Actin-filament disruptions were visualized after 10 h of zinc treatment. Scale bar represents 100 µm.

Because the Ca2+ ionophore, A23187, also maintained cell morphology of zinc-insulted astroglia (Fig. 3B), we examined whether A23187 stabilized actin filament organization. Phalloidin-staining of zinc-treated astroglia showed that A23187 (0.1 µM) protected actin filament disruption of zinc-treated astroglia so that the resulting actin filament distribution was indistinguishable from that displayed by untreated astroglia. A23187 did not produce actin-ring formation along the plasma membrane (Fig. 4).

Next, we tested whether Ca2+-mediated process acts over the p38 MAPK pathway in zinc-insulted astroglia. Western blot data showed that A23187 did not suppress p38 MAPK activation in zinc-treated astroglia (Fig. 5). These results suggest that Ca2+-mediated process is not in the up-stream of p38 MAPK-mediated process.

Fig. 5

A23187 did not suppress p38 MAPK activation in zinc-treated astroglia. Western blots showing p38 MAPK activation levels in zinc-treated astroglia. Western blot analysis was performed after 3 h of treatment with zinc (40 µM) alone or zinc (40 µM)+A23187 (A: 0.1 µM).

DISCUSSION

The present study demonstrates that the Ca2+ ionophore A23187 and the p38 MAPK inhibitors (SB203580 and SB202190) suppressed rapid cell morphology disruption that distinctly occurs in zinc-treated astroglia. These results suggest that Ca2+ signaling and/or the p38 MAPK pathway are mechanism(s) that regulate actin filament dynamics in zinc-treated astroglia. A23187 produced a completely protection of cell morphology and actin-filament stabilization in zinc-treated astroglia, whereas SB203580 produced similar stabilization effects with leaving abnormally intensified actin filament-ring formation along the cell membrane, suggesting that in zinc-insulted astroglia, zinc influx-induced disturbance of cell morphology might be diverse.

The present study demonstrates that supplement of Ca2+ into zinc-treated astroglia using the Ca2+ ionophore, A23187, protected cell morphology disruption of zinc-treated astroglia. Therefore, we speculate that zinc influx disrupts Ca2+ homeostasis or Ca2+ availability in zinc-treated astroglia, which in turns affects actin filament destabilization, resulting in rapid loss of the morphological integrity. Disruption of Ca2+ availability or substitution of Ca2+ by Zn2+ might affect the function of Ca2+-dependent proteases or Ca2+ regulating factors such as calreticulin and calbibdin. For examples, calpains are Ca2+-dependent proteases that cleave actin-binding proteins (Harris and Morrow, 1990; Carragher and Frame, 2002). Calreticulin potentially has a role in cell adhesion and maintenance of intracellular Ca2+ homeostasis (Michalak et al., 1998). Zinc influx induces functional disruption of thiolcontaining cellular factors (Frederickson et al., 1988; Kim et al., 2003) and increase of ROS stress (Kim et al., 1999a; 1999b; Suh et al., 2000). Our results add the evidence that zinc influx in astroglia causes a disturbance of Ca2+ homeostasis.

Concerning the effects of SB203580 on cell morphology and actin filament dynamics, it may be possible that the p38 MAPK pathway normally plays an inhibitory role in actin filament stabilization in zinc-treated astroglia. The notion that inhibition of p38 MAPK pathway by SB203580 enhances actin filament stabilization is consistent with the previous report that inhibition of the p38 MAPK pathway by SB202190 facilitated stellate process growth and extension of FGF2-treated primary astroglia (Heffron and Mandell, 2005). Stellate process extension of astroglia is an important aspect of reactive astrogliosis that occurs in a variety of pathological stimuli including trauma, ischemia, and neurodegenerative diseases (Chen and Swanson, 2003; Pekny and Nilsson, 2005). During the changes of stellate process extension of astroglia, actin filament breakdown, reassembly and stabilization should take place rapidly. Consistent with this view, reactive astrogliosis express high levels of p38 MAPK (Che et al., 2001; Piao et al., 2003). Many brain tumors arise from astoglia and metastasis of brain tumors may undergo a phase in which cell adhesion and cell morphology changes are critical steps. The results of the present study may raise the possibility that the p38 MAPK pathway is an important regulator of morphological behaviors of brain cells with an astroglia-origin.

A23187 did not suppress p38 MAPK activation in zinc-treated astroglia and both factors distinctly regulate actin-filament organization. Therefore, we speculate that SB203580/SB202190 target(s) and Ca2+ signaling players independently work to stabilize cell morphology. Cell morphology preservation of zinc-treated astroglia by SB203580 and SB202190 is likely produced by stabilization of actin-filament organization by inhibiting p38 MAPK. However, we do not rule out the possibility that SB203580/SB202190 targets unidentified cellular factors regulating actin-filament organization. Our preliminary data showed that cell morphology stabilization, but no cell death protection, of zinc-treated astroglia can be also achieved by NDGA (nordihydroguaiaretic acid), which possess a lipoxygenase (LOX) inhibition activity and an anti-oxidant property (Konno et al., 1990; Arteaga et al., 2005), and causes GSH depletion (Im and Han, 2007). Regarding that A23187, SB202190 and NDGA might distinctly work within cells, the possibility that they may act on a common pathway still remains to be explored further in the future.