Thymosin-β4 Attenuates Ethanol-induced Neurotoxicity in Cultured Cerebral Cortical Astrocytes by Inhibiting Apoptosis

Abstract

Thymosin-β4 (Tβ4) is a major actin monomer-binding peptide in mammalian tissues and plays a crucial role in the nervous system in synaptogenesis, neuronal survival and migration, axonal growth, and plastic changes of dendritic spines. However, it is unknown whether Tβ4 is also involved in challenges with external stress such as ethanol-induced neurotoxicity. In the present study, we investigated the effects of Tβ4 on ethanol-induced neurotoxicity in cultured cerebral cortical astrocytes and the underlying mechanisms. Primarily cultured astrocytes were treated with 1 μg/ml Tβ4 2 h prior to administration of 100 mM ethanol for 0.5, 1, 3 and 6 days, respectively. The results showed that ethanol caused neurotoxicity in cultured astrocytes, as shown by declined cell viability, distinct astroglial apoptosis and increased intracellular peroxidation. Tβ4 markedly promoted cell viability, ameliorated the injury of intracellular glial fibrillary acidic protein-immunopositive cytoskeletal structures, reduced the percentage of apoptotic astrocyte and cellular DNA fragmentation, suppressed caspase-3 activity and upregulated Bcl-2 expression, inhibited the accumulation of reactive oxygen species and production of malondialdehyde in ethanol-treated astrocytes in a time-dependent manner. These data indicated that Tβ4 attenuates ethanol-induced neurotoxicity in cultured cortical astrocytes through inhibition of apoptosis signaling, and one of the mechanisms underlying the capacity of Tβ4 to suppress apoptosis may in part be due to its effect of anti-peroxidation.

Introduction

Ethanol is known to be a tetratogen and its abusage can result in the dysfunction of the central nervous system (CNS), growth deficiency, facial malformation as well as the behavioral, learning, sensory and motor disabilities in the fetus (Weinberg 1985; West 1987; Jacobson et al. 1993; Hatten 1999). Chronic ethanolism in the adult is also intimately associated with brain atrophy (Kane et al. 1996). Accumulating evidence indicates that ethanol-induced neurobehavioral dysfunctions may be related to disruptions in the patterns of neuronal and glial developments such as depression of neurogenesis, aberrant migration of neurons and alterations in late gliogenesis and neurogenesis. These changes can further reduce the populations of cortical neurons and glial cells (Gressens et al. 1992; Miller 1992), trigger the biochemical alterations in glial cells and deleterious consequences for neuronal-glial interactions, and eventually lead to damage or apoptosis of these cells.

As the most abundant type of glial cells in the brain, astrocytes provide metabolic and trophic support to neurons, modulate synaptic activities (Takuma et al. 2004) and have a strong capacity to scavenge oxidants and suppress cellular apoptosis. However, when the capacity of cells to eliminate the oxidants is overwhelmed, overproduction of reactive oxygen species (ROS) can cause morphological and functional alterations in the cells, including cellular Ca²⁺ homeostasis and some active molecules tightly associated with neuronal activity (Renau-Piqueras et al. 1989; Goodlett and Horn 2001; Barnham et al. 2004; Hirata et al. 2006). Although astrocytes are more resistant than neurons to the oxidative and neurotoxic stresses and to the chemical and toxic damages in the surrounding environment (Dringen 2000; Takuma et al. 2004), any impairment of astrocytes can dramatically affect neuronal functions. The ethanol-induced detrimental alterations of astrocytes would lead to perturbances in neuron–astroglia interactions and developmental defects of the brain (Valles et al. 1997; Rintala et al. 2001 Signorini-Allibe). Therefore, it is a pivotal solution to seek active molecules that may inhibit or attenuate ethanol-induced neurotoxicity in astrocytes, thus offering an alternative strategy to prevent or treat neurodevelopmental disorders and mental retardation caused by ethanol.

Thymosin-β4 (Tβ4), a peptide of 4.9-kDa, is a major actin-sequestering protein and abundantly exists in the mammalian tissues. Tβ4 has been identified to be involved in many cellular events such as endocytosis, angiogenesis, migration and apoptosis of cells, redistribution of cell adhesion molecules, inflammation, and wound healing (Kamiguchi et al. 1998; Niu and Nachmias 2000; Huff et al. 2001; Sosne et al. 2001, 2002a, b, 2004a; Yang et al. 2008). These events are intimately related to the dynamics of actin de- and repolymerization controlled by actin-binding peptides beta-thymosin. In the nervous system, Tβ4 exerts important roles in the survival, proliferation, differentiation, and motility of neural cells, neurite formation, axonal pathfinding, plastic changes of dendritic spines and synapotogenesis (Fifkova 1985; Border et al. 1993; Otero et al. 1993; Matsuoka et al. 1998; Roth et al. 1999a, b; Huff et al. 2001). The level of Tβ4 is elevated following some brain disorders such as cute brain trauma, focal cerebral ischemia and neurotoxic insults, suggesting the neurotrophic functions of Tβ4 in the pathological responses to brain damages (Vartiainen et al. 1996; Carpintero et al. 1999). Tβ4 is also found to enhance the production of laminin and L1, two major extracellular matrix adhesion molecules (Sosne et al. 2004b; Yang et al. 2008). However, the literature contains limited studies of the effects of Tβ4 on ethanol-induced neurotoxicity in astrocytes and the underlying mechanisms. In the present study, we tackled these issues in primarily cultured rat cerebral cortical astrocytes exposed to ethanol. Our results provided evidence that Tβ4 can attenuate ethanol-induced neurotoxicity in cultured astrocytes by inhibiting astroglial apoptosis, suggesting a potential chemical to prevent or treat ethanol-induced CNS degenerative disorders.

Materials and Methods

Primary Cultures and Different Treatments of Astrocytes

Primary cultures of astrocytes were prepared as described (Tomás et al. 2002) with minor modifications. Briefly, the cerebral cortex of 16–18 days old Sprague–Dawley rat embryo (Laboratory Animal Center of the Fourth Military Medical University) was dissected under sterile conditions and rapidly pooled into D-Hank’s balanced sodium salts without Ca²⁺ and Mg²⁺(HBSS; Gibco-Invitrogen, USA). The meningeal tissue was quickly peeled away. The cortical tissue was minced before further enzymatic dissociation with 0.125% trypsin (Gibco-Invitrogen) at 37°C for 25 min, transferred into 10 ml Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco-Invitrogen) supplemented with 20% heat-inactivated fetal calf serum (FCS; Gibco-Invitrogen) to terminate trypsinization, and mechanically dissociated by repeated triturations with a pipette. Dissociated cells were passed through a sterile nylon gauze (80 μm pore size), and seeded in 25-cm² plastic flasks (Nunclon, Denmark) and 35 mm dishes coated with poly-l-lysine (Sigma, USA) at a density of 4 × 10⁵ cells/cm² after counting in a haemocytometer. Cells were suspended in DMEM containing 20% FCS, 2 mM glutamine (Sigma), 25 mM 2-[4-(2-Hydroxyethyl)-1- piperazinyl] ethanesulfonic acid (HEPES, Sigma), 50 U/ml Penicillin (Sigma) and 50 μg/ml streptomycin (Sigma) at 37°C in a humidified 5% CO₂ atmosphere and purified (Yang et al. 2006) after culturing for 48–60 h. These purified cells reached a confluence after culturing for 12–14 days. The primary cells were digested with 0.125% trypsin and 0.02% ethylenediaminetetraacetic acid (EDTA; Gibco-Invitrogen) in HBSS, and reseeded at a density of 4 × 10⁵ cells/cm² onto poly-l-lysine-coated coverslips and 96-well plates (Nunclon), or into 35-mm Petri dishes (Nunclon). The culture medium was refreshed every 3 days.

After incubation in DMEM with 10% FCS for 5 days, cultured cells were divided into three experimental and one control groups: (1) Ethanol Group, in which cells were exposed to 100 mM ethanol (assay G.C., v/v 99.5% diluted in sterile double-distilled water; Bachem, USA); (2) Tβ4 Group, where Tβ4 (1 μg/ml, a gift from Dr. Yan-Ke Chen at the Fourth Military Medical University) was added into the cultures; (3) Ethanol/Tβ4 Group, with the treatment of Tβ4 2 h before ethanol exposure; and (4) Control Group, with neither ethanol nor Tβ4 being added. After different treatments, cells in the four groups were subsequently maintained at 37°C in a humidified 5% CO₂ atmosphere till evaluated at 0.5, 1, 3 and 6 days, respectively. All the following experiments were conducted in a minimum of triplicate preparations.

Assessment of Cell Viability of Cultured Astrocytes

The viability of cultured astrocytes was assessed as the 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyl (MTT) conversion rate in all groups, using a modified MTT assay (Mosmann 1983). MTT (Sigma) was dissolved in saline (5 mg/ml) and diluted directly into the media at a final concentration of 0.2 mg/ml. After incubion at 37°C for 4 h, a portion of MTT was converted into an insoluble purple formazan by the cleavage of tetrazolium ring with dehydrogenase. The medium was gently aspirated and 100 μl dimethyl sulfoxide was added to each well followed by incubation and mechanical shaking for 10 min to dissolve the formazan product. The absorbance expressed as the optical density (OD) was determined on a Dynatech MR4000 microplate reader (Dynatech, USA) with 570 and 630 nm for the measurement and reference wavelengths, respectively.

Morphological Observation and Quantification of Apoptotic Astrocytes

GFAP-Immunofluorescent and Hoechst 33342 Stainings of Astrocytes

To investigate the morphological changes of astroglial cytoskeletal proteins and quantify apoptotic astrocytes, glial fibrillary acidic protein (GFAP)-immunofluorescent and Hoechst 33342 stainings of cultured astrocytes were performed. Cells on coverslips in all groups were fixed with 4% paraformaldehyde (Sigma) for 30 min, treated with 0.02% Triton X-100 (Sigma) and 4% normal goat serum (Sigma) in 0.01 M PBS, incubated with primary mouse monoclonal antibodies against GFAP (1:1500 dilution; Santa Cruz, USA) at 4°C overnight, washed in PBS thrice, incubated with Alexa Fluor^® 488 goat anti-mouse IgG antibody (1:800 dilution; Molecular Probe, USA) for 2 h and Hoechst 33342 (5 μg/ml; Molecular Probes) at room temperature (RT) for 15 min, coverslipped after rinsed in PBS thrice and observed under an Olympus BX-51 fluorescence microscope (Olympus, Japan). All images were captured using a FV10-ASW 1.6 photo system (Olympus, Japan). Hoechst33342-stained apoptotic cells, defined as those with bright blue fragmented nuclei containing one or more lobes of condensed chromatin, were counted and at least 15 fields/coverslip with at least 50 cells/field were examined, yielding n ≥ 400 cells examined per coverslip. Independent cultures were used when each experiment was conducted thrice.

Quantification of DNA Fragmentation

DNA fragmentation, a vital biomarker for apoptosis, was quantified using The Cell Death Detection ELISA kit (Roche Molecular Biochemicals, Germany). The cells were lysed at RT for 30 min and centrifuged at 200 g for 10 min. The supernatants (20 μl) and 80 μl immunoreagent were coincubated in streptavidin-coated mitochondrial transition pore at RT for 2 h. After washing thrice with incubating buffer, the substrate was added and reaction mixtures were incubated at RT for 20 min. The OD was measured on the microplate reader (measurement wavelength 405 nm and reference wavelength 490 nm). DNA fragmentation was expressed as the enrichment factor (EF), calculated with the following formula: EF = mU of the sample (dying or dead cells)/mU of the corresponding control, where mU = absorbance [10⁻³](A _405 nm − A _490 nm).

Determination of Astroglial Apoptosis

Measurement of Caspase-3 Activation

Caspase-3 activation was measured with an ApoAlert Caspase Assay Kit (Clontech Laboratories, USA). Briefly, cultured astrocytes (10⁶) were lysed in the buffer containing 50 mM HEPES (pH 7.4), 100 mM NaCl, 0.1% 3-([3-cholamidopropyl]-dimethylammonio-2-hydroxy-l- propanesulfonate) (Sigma), 1 mM EDTA, 10% glycerol and 10 mM dithiothreitol (Sigma) on an ice bath for 10 min. The lysates were centrifuged at 4°C for another 10 min. The supernatant (50 μl) and 100 μl reaction buffer were mixed in a 96-well plate and incubated in the presence of 5 μl appropriate caspase-3 substrate and 45 μl distilled water at 37°C for 1 h. Negative controls contained all reaction components except for cell lysate. Detection was measured on the microplate reader at 405 nm measurement wavelength. Cell lysate was also used for determination of the total protein with the Bicinchoninic Acid (BCA) Protein Assay, and the microplate reader output corresponding to caspase-3 activation was corrected for the total protein. Caspase activation was reported as a percentage activity in different groups compared with a specific caspase-3-like protein group (arbitrarily set at 100%). Triplicate assays (n = 8 for each assay) were carried out to calculate the average percentage activation.

Measurement of Bcl-2 Level

The level of expressed Bcl-2 proteins was determined using Sandwich ELISA assays according to the protocol of Bcl-2 ELISA kit (Oncogene Research Products, USA). Astrocytes were washed thrice with PBS and lysed in pH 7.4 resuspension buffer containing 50 mM Tris, 5 mM EDTA, 0.2 mM Phenylmethanesulfonyl fluoride, 1 μg/ml pepstatin, and 0.5 μg/ml leupeptin. The proteins were extracted with the antigen extraction reagent provided with the appropriate ELISA kits, and the cleared lysates were used for ELISA. The method included the capture of these proteins by immobilized protein specific mouse monoclonal antibodies bound to the wells. The complex was detected with horseradish peroxidase (HRP)-conjugated streptavidin, and the conversion of chromogenic substrate tetra-methylbenzidine (Oncogene Research Products) was catalyzed with HRP after stopping solution was added. The absorbance, indicative of protein concentration, was measured on the microplate reader at 450 nm. The ELISA was repeated in triplicate (n = 3 for each replicate) to calculate average cytoplasmic Bcl-2 in U/ml ± SEM.

Determination of Peroxidation in Cultured Astrocytes

Intracellular ROS Production

Intracellular production of ROS was measured with 2′, 7′-dichlorofluorescin-diacetate (DCF-DA; Molecular Probes) as the probe (Caro and Cederbaum 2002). Cells were incubated with 30 μM DCF-DA at 37°C for 30 min and treated with experimental reagents at 37°C for additional 30 min. After chilled on ice, cells were washed with ice-cold PBS in darkness, detached from the dishes and resuspended at 1 × 10⁶ cells/ml in 10 mM EDTA-containing PBS. The fluorescence intensity of DCF formed by the reaction of DCF-DA with peroxides of more than 10,000 viable cells from each sample was analyzed by FACS Caliber flow cytometer (Becton Dickinson, USA). Propidium iodide (PI, Molecular Probe) was added to the samples before data collection to gate out dead cells. Similar results were obtained when experiments were repeated for at least three times.

Malondialdehyde (MDA) Concentration

The anti-lipid peroxidation effect of Tβ4 on ethanol-treated astrocytes was evaluated by assaying for the production of thiobarbituric acid reactive substances (TBARS). The amount of MDA was measured by the thiobarbituric acid (TBA) assay to confirm the changes of lipid peroxidation resulting from ethanol-induced generation of ROS. Cells were washed and lysated with cold Tris-HCl (15 mM, pH 7.4) at 4°C for 30 min. The sample was mixed with a TBA reagent consisting of 0.375% TBA and 15% trichloracetic acid in 0.25 M HCl. The reaction mixtures were placed in a boiling water bath for 15 min and centrifuged at 3,000g for 5 min before the absorbance of the supernatant was read on the microplate reader at 535 nm. The MDA concentration of the sample was calculated using an extinction coefficient of 1.56 × 10⁵ M⁻¹ cm⁻¹.

Statistical Analysis

All data were presented as mean ± SEM at different time points for each group. The statistical significance was determined using One-way analysis of variance (ANOVA) with SPSS10.0 software. A value of P < 0.05 was considered statistically significant.

Results

Tβ4 Ameliorates Cell Viability of Ethanol-Treated Astrocytes

Similar OD values (P > 0.05) as cell viability were found between Tβ4 Group (0.337 ± 0.031, 0.382 ± 0.033, 0.486 ± 0.036, and 0.550 ± 0.040) and Control Group (0.335 ± 0.026, 0.379 ± 0.020, 0.488 ± 0.050, and 0.580 ± 0.030) at all four time points. Although no difference (P > 0.05) in the OD values could be detected between Ethanol (0.289 ± 0.022) and Control (0.335 ± 0.026) Groups at 0.5 day, the OD values were significantly suppressed in Ethanol Group at 1 (0.274 ± 0.040, P < 0.05), 3 (0.279 ± 0.033, P < 0.01), and 6 (0.267 ± 0.036, P < 0.001) days in comparison with Control Group at the corresponding time points (0.379 ± 0.020, 0.487 ± 0.050, and 0.580 ± 0.030). Tβ4 did not effectively promote the viability of ethanol-exposed astrocytes at 0.5 day (0.314 ± 0.036 and 0.289 ± 0.022 for Ethanol/Tβ4 and Ethanol Groups, respectively, P > 0.05), but markedly elevated OD values were observed in Ethanol/Tβ4 Group at 1 (0.346 ± 0.052, P < 0.05), 3 (0.393 ± 0.021, P < 0.05), and 6 (0.408 ± 0.067, P < 0.01) days compared to those in Ethanol Group (0.274 ± 0.040, 0.279 ± 0.033, and 0.267 ± 0.036, Fig. 1) at the corresponding time points.

Fig. 1.

The OD values showing the viability of astrocytes in different groups. Astrocyte viability was assessed by MTT assay after cultured astrocytes were treated with 100 mM ethanol and/or 1 μg/ml Tβ4 for 0.5, 1, 3 and 6 days, respectively. Data are expressed as means ± SEM (n = 3, each an average of 10 wells with cells per conditions).* P < 0.05, ** P < 0.01 and *** P < 0.001 in comparison with Control Group

Tβ4 Attenuates Ethanol-Induced Apoptosis in Astrocytes

Tβ4 Reduces Ethanol-Induced Injury of Cytoskeletal Proteins

In Control Group, no marked difference could be detected in the morphology and distribution of GFAP-immuopositive filamentous network within astrocytes among the four time points. The fluorescence intensity of GFAP intermediate filaments was homogeneously spread throughout the cytoplasm and processes with branching of different degrees, and intercellular GFAP-immunoreactive products exhibited relatively intense filamentous distribution. Almost all the GFAP-positive cells were polygonal with thick processes or irregular with flat cytoplasm (Fig. 2a, b). Intracellular appearance and distribution of expressed GFAP within astrocytes in Tβ4 Group were similar with those in Control Group (Fig. 2c, d), and few alterations in the intensity of GFAP were found among all time points. In Ethanol Group, however, the intensity of intracellular GFAP decreased markedly (Fig. 2e) and homogeneous GFAP network broke into dotted substances in the cytoplasm at 0.5 and 1 day (Fig. 2f). When the culture time extended to 3 and 6 days, these glial microfilaments in the cytoplasm were progressively disrupted into punctate structures and most cells were polygonal or flat with short processes (Fig. 2g, h). In Ethanol/Tβ4 Group, the distribution of GFAP remained bright and homogeneous and the glial filament network in cytoplasm was easily seen at 0.5 and 1 day points (Fig. 2i, j). Although GFAP filaments became discrete and relatively decreased at 3 and 6 day points, they still had a predominant filamentous network in the cytoplasm (Fig. 2k, l).

Fig. 2.

Photomicrographs of GFAP-immunoreactive astrocytes with different treatments for 0.5, 1, 3 and 6 days, respectively, depicting effects of Tβ4 on distribution and content of GFAP in astrocytes in normal (a and b), Tβ4-treated (c and d) (cultures for 1 and 3 days are not shown for no appreciable differences in GFAP distribution pattern compared with that of 0.5 and 6 days), ethanol-exposed at 0.5 (e), 1 (f), 3 (g) and 6 (h) days and Tβ4/Ethanol-treated groups at each corresponding time points as shown in panels i, j, k, and l. The distributions of GFAP strongly stained structures in astrocytes are indicated by the triangle arrowheads, while lightly stained or missing substances in astrocytes by the long arrows. Asterisks represent disrupted glial filaments with tiny spot structures. Scale bar 150 μm

Morphologic Assessment of Nucleus and Quantification of Apoptotic Astrocytes

Hoechst 33342-stained nuclei in Control (Fig. 3a, b) and Tβ4 (Fig. 3c, d) Groups were similarly oval in shape and exhibited homogeneously bright blue fluorescence. No significant difference (P > 0.05) was revealed in the low percentages of apoptotic cells (5.20 ± 0.84, 4.97 ± 0.76, 5.80 ± 0.49, and 5.76 ± 0.81% for Control Group; 3.85 ± 0.50, 4.20 ± 0.70, 4.97 ± 0.65, and 5.8 ± 0.55% for Tβ4 Group) between the two groups at all time points. More apoptotic cells appeared in Ethanol Group (Fig. 3e, f, g, h), and the percentage of apoptotic cells increased dramatically from 16.00 ± 5.20, 26.40 ± 9.70, 39.50 ± 10.20 to 44.60 ± 15.90% when the culture time was prolonged from 0.5, 1, 3 to 6 days (Fig. 3m), suggesting a time-dependent increase in the percentage of ethanol-induced apoptotic astrocytes. However, such an increased percentage was suppressed markedly (P < 0.001) in Ethanol/Tβ4 Group (16.50 ± 5.90, 13.50 ± 3.10, 10.90 ± 2.90, and 7.20 ± 2.20%; Fig. 3i, j, k, l) when compared to those (16.00 ± 5.20, 26.40 ± 9.70, 39.50 ± 10.50, and 44.60 ± 15.90%) in Ethanol Group at the corresponding time points. However, these suppressed percentages of apoptotic cells were still significantly higher (P < 0.05) than those in Control and Tβ4 Groups (Fig. 3m).

Fig. 3.

Photomicrographs of Hoechst 33342-stained astrocytes in the four groups. Astrocytes at 0.5 (a) and 6 (b) day time points in Control Group, astrocytes at 0.5 (c) and 6 (d) days in Tβ4 Group (cultures for 1 and 3 days are not shown for similar reasons described in Fig. 2), astrocytes at 0.5, 1, 3 and 6 days in Ethanol Group (e, i, f and j) and Ethanol/Tβ4 Group (g, k, h and l) are shown with the triangle arrowheads pointing to the normal astrocytes and the long arrows to clusters of fragmented nuclei in apoptotic cells. Scale bar 200 μm. The histogram illustrating the percentages of apoptotic astrocytes with different treatments was presented in M. Ethanol significantly reduces the percentages of apoptotic astrocytes at all four different time points, while such declined percentages are remarkably promoted by the treatment of Tβ4. *, ** and *** represent significant differences of P < 0.05, P < 0.01 and P < 0.001, respectively, in comparison with Control Group

Tβ4 Attenuates Ethanol-Induced DNA Fragmentation

DNA fragmentation is vital biomarker of apoptosis. To further investigate in detail whether Tβ4 could attenuate ethanol-induced astrocyte apoptosis, DNA fragmentation was thus quantified. As shown in Fig. 4, similar EF values (P > 0.05) of DNA fragmentation were found between Control (1.000 ± 0.065, 1.030 ± 0.076, 0.993 ± 0.020, and 0.971 ± 0.075) and Tβ4 (0.927 ± 0.040, 0.937 ± 0.090, 0.910 ± 0.051, and 0.941 ± 0.063) Groups at all time points. However, the EF values were about 1.3- to 2.3-fold higher in Ethanol Group (1.243 ± 0.055, 1.367 ± 0.156, 1.627 ± 0.070, and 1.981 ± 0.067) and 1.1- and 1.4- fold lower (1.107 ± 0.088, 1.190 ± 0.0350, 1.323 ± 0.121, and 1.390 ± 0.059) in Ethanol/Tβ4 Group than those in Control Group at the corresponding time points.

Fig. 4.

Quantification of DNA fragmentation by ELISA. Astrocytes were treated with ethanol (100 mM) and/or Tβ4 (1 μg/ml) for 0.5, 1, 3, and 6 days. Quantitative data of EF value are presented as means ± SEM (from three independent trials and about 30 wells for each test). * P < 0.05, ** P < 0.01 and *** P < 0.001 represent significant difference, respectively, in comparison with Control Group

Tβ4 Attenuates Astroglial Apoptosis by Suppressing Caspase-3 Activity and Upregulating Bcl-2

Tβ4 Suppresses Ethanol-Induced Activity of Caspase-3

Similar caspase-3 activities were detected in Control (100.0 ± 0.9, 101.3 ± 1.8, 102.0 ± 4.7, and 104.0 ± 4.2) and Tβ4 (98.7 ± 1.2, 97.7 ± 12.5, 98.3 ± 8.2, and 98.5 ± 4.5) Groups at all time points. Astroglial capase-3 activities in Ethanol Group (139.0 ± 5.4, 145.0 ± 5.6, 176.3 ± 14.0, and 219.7 ± 10.4) were 1.4- to 2.2-fold higher than the control level at all time points. This ethanol-induced stronger activity of capase-3 decreased significantly (about 1.1- to 1.4-fold of the control level and 40% reduction of the increased caspase-3 activity induced by ethanol) in Tβ4/Ethanol Group (112.0 ± 5.3, 121.0 ± 3.1, 133.3 ± 4.6, and 144.3 ± 6.5; Fig. 5).

Fig. 5.

Tβ4 inhibition of ethanol-induced increase in caspase-3 activity. Astrocytes were treated with ethanol (100 mM) for 0.5, 1, 3, and 6 days in the presence or absence of 1 μg/ml Tβ4. Data are expressed as mean ± SEM of three independent trials. P < 0.05 was used as criterion for the significance of differences between the means. * P < 0.05, ** P < 0.01 and *** P < 0.001, respectively, compared with Control

Tβ4 Promotes Bcl-2 Expression

No significant difference (P > 0.05) in the concentration of cytosolic Bcl-2 was shown between Control (16.3 ± 1.1, 16.4 ± 1.4, 16.7 ± 2.2, and 15.8 ± 1.1) and Tβ4 (17.3 ± 1.4, 17.5 ± 1.4, 17.2 ± 1.1, and 16.6 ± 2.3) Groups at all time points. The Bcl-2 concentration decreased significantly (10.4 ± 2.4, 9.1 ± 1.3, 8.1 ± 2.2, and 7.1 ± 1.5, P < 0.05) in Ethanol Group and increased significantly (15.4 ± 1.3, 14.7 ± 1.2, 13.8 ± 2.1, and 12.8 ± 1.1; P < 0.001) in comparison with those in Control Group (Fig. 6).

Fig. 6.

Quantifications of Bcl-2 protein concentration in astrocytes in different groups. Note that ethanol exposure results in significant decrease in Bcl-2 level compared to normal culture, while Tβ4 significantly suppress a decrease in the level of cytosolic Bcl-2 proteins resulted from ethanol. Data are means ± SEM from three to four separate experiments. * P < 0.05, ** P < 0.01 and *** P < 0.001 represent significant differences, respectively, in comparison with its Control

Tβ4 Suppresses Peroxidation in Ethanol-Treated Astrocytes

Tβ4 Decreases Ethanol-Induced Accumulation of ROS

To test whether Tβ4 has the capacity to suppress ROS production in astrocytes after exposure to ethanol, we measured changes in intracellular ROS production by DCF formation rate. As shown in Fig. 7. Similar mean DCF fluorescence intensity (P > 0.05) were found in Control and Tβ4 Groups at all time points (100.0 ± 5.5, 101.0 ± 1.2, 102.3 ± 8.2, and 97.3 ± 7.0 for Control Group; 105.0 ± 1.4, 101.0 ± 3.8, 98.3 ± 9.8, and 101.9 ± 16.7 for Tβ4 Group). Increased DCF intensity was 1.7- to 2.3-fold increased in Ethanol Group (162.8 ± 9.6, 169.2 ± 18.9, 196.9 ± 16.2, and 228.0 ± 17.2) when compared to Control Group. Such increased DCF intensities were reduced by about 45% in Ethanol/Tβ4 Group (109.6 ± 6.1, 124.6 ± 5.8, 144.2 ± 7.9, and 153.3 ± 21.0).

Fig. 7.

Effect of Tβ4 on ROS generation in astrocytes exposed to ethanol. The fluorescence intensity of DCF was measured after astrocytes were exposed to ethanol and Tβ4 + ethanol (100 mM) for 0.5, 1, 3, and 6 days, respectively. All data are presented as means ± SEM for three independent experiments with triplicate determinations. * P < 0.05, ** P < 0.01 and *** P < 0.001 represent significant differences, respectively, compared with its Control

Tβ4 Reduces Ethanol-Induced Lipid Peroxidation

No significant difference in the MDA level in astrocytes was observed at all time points between Tβ4 and Control Groups. Whereas, the MDA levels in Ethanol Group were promoted about 1.4- to 2-fold when compared to the control cultures, and Tβ4 significantly suppressed ethanol-induced MDA generation by about 35% at all the time points (Table 1).

Table 1.

Protective effects of Tβ4 against ethanol-induced lipid peroxidation

Groups	0.5 day (nmol MDA/mg protein)	1 day (nmol MDA/mg protein)	3 day (nmol MDA/mg protein)	6 day (nmol MDA/mg protein)
Control	2.28 ± 0.26	1.97 ± 0.21	2.09 ± 0.29	2.08 ± 0.26
Tβ4	2.19 ± 0.22	1.94 ± 0.23	2.07 ± 0.17	1.98 ± 0.29
Ethanol	3.04 ± 0.31**	3.48 ± 0.27***	3.94 ± 0.12***	3.89 ± 0.32***
Ethanol + Tβ4	2.34 ± 0.23Δ	2.74 ± 0.25ΔΔ	2.85 ± 0.25ΔΔ	2.98 ± 0.28ΔΔ

MDA intracellular content was measured by a colorimetric assay after astrocytes were incubated for 0.5, 1, 3, and 6 days with ethanol (100 mM) in the presence of Tβ4. Changes in MDA levels were evaluated by the production of TBARS. Data are means ± SEM from three separate experiments

** P < 0.01 and *** P < 0.001 represent significant differences, respectively, in comparison with Control group (untreated cells)

^ΔP < 0.05, and ^ΔΔ P < 0.01 represent significant differences in comparison with Tβ4 Group

Discussion

Ethanol is one of the common chemicals that cause toxic damages to developing neural cells. Heavy ethanol consumption during the pregnancy or adult leads to mental retardation and neurobehavioral disorders. In the present study, treatment of ethanol at 100 mM resulted in neurotoxicity in cultured astrocytes, as shown by declined cell viability, distinct astroglial apoptosis and increased intracellular peroxidation. Only one concentration (100 mM) for the ethanol stress treatment was applied in the present study. The reasons why this particular concentration was used are: (1) ethanol has a dose-dependent effect of the oxidative stress. Ethanol at a concentration up to 100 mM can cause cell apoptosis, which is very close to an in vivo model of chronic ethanolism. When ethanol concentration exceeds 100 mM, however, cell death happens more severely and resembles that in a model of acute ethanol injury; and (2) 100 mM has been frequently used for many studies of ethanol stress to efficiently activate apoptotic pathways (Holownia et al. 1997; Hirata et al. 2006). Although a number of mechanisms underlying ethanol-induced neurotoxicity and subsequent neurodegeneration have been identified, how ethanol causes apoptosis of ethanol-exposed cortical astrocytes remains elusive. Reliable clinical therapies for ethanol neurotoxicity are not available.

Tβ4, a major actin-sequestering protein capable of crossing the blood-brain barrier (Tijerina et al. 1997), plays a widely biochemical roles in fishes (Roth et al. 1999a) and mammalians (Condon and Hall 1992). In the nervous system, Tβ4 not only enhances neuronal survival and neurite outgrowth by up-regulating L1 expression, but induces the expression of nerve growth factor after injury (Turrini et al. 1998; Popoli et al. 2007; Yang et al. 2008). Tβ4 also exerts a critical role in the development and differentiation of the CNS (Condon and Hall 1992), and is needed for glial activation, restoration, and plastic changes occurring in the neurites of surviving neurons following excitotoxic damage in the forebrain. In the present study, we found that 1 μg/ml Tβ4 attenuated severe ethanol-induced neurotoxicity through its effects of anti-apoptosis to suppress caspase-3 activity and upregulate the expression of Bcl-2, leading to a promoted viability of astrocytes, improved intracellular changes of GFAP-immunopositive cytoskeletal proteins, and reduced percentage of Hoechst33342-stained apoptotic astrocytes and DNA fragmentation. However, Tβ4 treatment alone showed no protective effect on cultured astrocytes because no significant difference could be observed between Tβ4 and Control Groups in any of our measurements and assays. Our findings were consistent with the previous studies (Niu and Nachmias 2000; Sosne et al. 2004a; Ho et al. 2007) and provided a novel insight into a potential therapeutic value of Tβ4 for the prevention and treatment of ethanol-induced degenerative disorders in the CNS. There was also only one concentration (1 μg/ml) for Tβ4 which was used as protective treatment for ethanol stress. The choice of this concentration was based on the following reported studies: (1) Tβ4 possesses a promotive effect on neural cell survival and neurite outgrowth at the concentration of 1 μg/ml, as demonstrated in our previous study (Yang et al. 2008); and (2) Tβ4 inhibits apoptosis of corneal epithelial cells at the concentration of 1 μg/ml (Sosne et al. 2004a).

Severe apoptosis of cultured astrocytes was induced by ethanol in this study. The percentage of Hoechst 33342-stained apoptotic astrocytes dramatically increased in Ethanol Group, and homogeneously organized GFAP-positive filamentous structure in the astrocytes collapsed and changed to dotted substances within the first culture day and punctate materials with even longer culture time. The occurrence of significantly augmented internucleosomal DNA fragmentation also indicated that ethanol could severely devastate the cytoskeletal elements in astrocytes. Apoptosis is characterized by the activation of endogenous endonucleases with subsequent cleavage of chromatin DNA into internucleosomal fragments. Thus, the evaluation of internucleosomal fragmentation of DNA has been widely accepted as one of the best-characterized biochemical markers for apoptosis. Furthermore, ethanol-induced mitochondrial damage and dysfunction can cause single or double DNA strand breaks (Vallés et al. 1996; Guerri et al. 2001). Therefore, one of the reasons for ethanol-induced changes in GFAP-immunoreactive structure might be the deleterious perturbance of ethanol on the rate of GFAP transcription, mRNA stability, and DNA methylation of GFAP in astrocytes. Tβ4 significantly reduced the percentage of apoptotic astrocytes, the morphological changes of GFAP-immunoreactive cytoskeletal proteins, and DNA fragmentation, indicating a capacity of Tβ4 to inhibit ethanol-induced apoptosis of astrocytes. Possible mechanisms underlying the improving effect of Tβ4 on damaged cytoskeletal proteins were proposed as follows: (1) exogenous Tβ4 might enter the cytoplasm through rapid internalization, stabilize the process of GFAP expression and suppress ethanol-induced alterations by remodeling actin cytoskeletons following actin-sequestering; (2) reduced deleterious alterations in GFAP expression may in part be due to L1 up-regulation by Tβ4, because Tβ4 can upregulate L1 expression which in turn can effectively ameliorate cell apoptosis (Ramanathan et al. 1996); and (3) Tβ4 facilitates bindings of G-actin and ADP/ATP-Ca²⁺ through the interactions of Tβ4 with G-actin, thus suppress the increased Ca²⁺ level and decreased ADP/ATP-actin in ethanol-exposed cells and further cause the activation of casein kinase II and extracellular signal-regulated kinase2 (ERK2) cascades. Transcriptional and translational activation of viable molecule expression develop consequently. Overexpression of Tβ4 can decrease the amount of E-cadherin and increase the accumulation of hypophosphorylated β-catenin in the cytoplasm, thus resulting in translocation of phosphor-β-catenin to the nucleus and activating transcription factors in the Wnt signaling pathway which triggers gene expression of some molecules (Brieger et al. 2007; Bednarek et al. 2008).

Our present study demonstrated that ethanol-induced apoptosis of the cortical astrocytes was markedly inhibited by Tβ4 through reduction of caspase-3 activity and upregulation of Bcl-2 expression. Caspases play an important role in the apoptotic process in both extrinsic and intrinsic pathways (Markus 2000), and caspase-3 acts as an apoptotic executor in both pathways. Caspase-3 is activated by many noxious chemicals like ethanol (Bachis et al. 2003) and certain pathological situations such as the formation of ROS and the disruption of intracellular Ca²⁺ homeostasis. Caspase-3 activity is also increased in the brain with neurodegenerative diseases, and elimination of such stimuli inhibits the activation of caspase-3 (Lee et al. 2002). Some molecules are capable of preventing caspase-3 activation (Bachis et al. 2003). Tβ4 constitutes a highly conserved family of actin-monomer-binding proteins involved in many physiologic processes in the cells, repairs damaged cytoskeletal structures and redistributes cell adhesion molecules. The effect of Tβ4 to suppress caspase-3 activation also reflected its regulatory function in the stabilization of Ca²⁺ homeostasis and activation of signal-regulated kinase pathways in actin polymerization.

Bcl-2 family members are involved in neural apoptosis, axonal regeneration, cellular differentiation and survival. Bcl-2 also regulates mitochondrial membrane potential and mitochondrial pore transition, releases Cytochrome C and other mitochondrial factors (Kane et al. 1993; Reyes et al. 1993; Ellerby et al. 1996), and possesses the properties of antioxidant, free radical scavenger, and regulator of cellular redox state against oxidative stress (Shimizu et al. 1999; Amstad et al. 2001). The present study showed that ethanol significantly decreased the intracellular level of Bcl-2 and such decreased level was significantly promoted after Tβ4 treatment, indicating that one of the explanations for the anti-apoptosis effect of Tβ4 might be the activation of Bcl-2 signaling pathway. Our results were in good agreement with an earlier observation that Tβ4 inhibited ethanol-induced activation of caspase-3 and up-regulated Bcl-2 expression in human corneal epithelial cells (Sosne et al. 2004a), suggesting that Tβ4 be involved in the decreased activity of phosphatidyl inositol 3-kinase prosurvival pathways as well as in the promoted astroglial survival by the expression of genes including Bcl-2. Tβ4 is also reported to protect against focal ischemic brain damage via the up-regulation of Bcl-2 (Vartiainen et al. 1996).

Ethanol at 100 mM can elevate the level of ROS generated directly via mitochondrial respiration and result in severe impairment of cellular structures and functions, such as disruption of plasma membrane integrity (Butterfield et al. 1997), abnormal cross-linking of cytoskeletal proteins (Stadtman 1990) and dysfunctions of cytoplasmic RNA and nuclear and mitochondrial DNA (Mecocci et al. 1994; Nunomura et al. 1999). Ethanol at the same concentration also significantly promoted the levels of ROS accumulation and MDA production in astrocytes in the present study. Therefore, it is not surprising that antioxidants and free radical scavengers have been used to attenuate ethanol-induced neurotoxicity in vitro and in vivo (Reyes et al. 1993; Guerri et al. 1994; Montoliu et al. 1995; Heaton et al. 2000; González et al. 2007; Salazar et al. 2008) and to prevent the damage of lipid peroxidation and consequent neurodegenerative diseases (Reyes et al. 1993). One previous observation showed that Tβ4 attenuated membrane lipid peroxidation and thus protected injured human cornea epithelial cells (Ho et al. 2008). In the present study, we also detected that Tβ4 can significantly eliminate ethanol-induced ROS accumulation and MDA production in astrocytes. These findings suggested that the anti-apoptosis effect of Tβ4 on ethanol-induced astrocytes be partially related to the capacity of Tβ4 to eliminate cellular peroxidation.

Mitochondria are the major source for generating ROS and also a main target to be attacked by oxidative stress. Mitochondrial injury derived from oxidative damage can cause necrosis by depleting ATP and apoptosis by inducing the release of mitochondrial factors such as Cytochrome C, which activates the caspase cascade. The damage of mitochondrial membrane has been proposed to be a key mechanism by which lipid peroxidation causes the decrease in cell viability. Therefore, another possible mechanism involved in anti-apoptosis effects of Tβ4 is the inhibition of mitochondrial injury. The time points of no later than 6 days (0.5, 1, 3 and 6 day incubation) were selected to carry out the experimental analysis because: (1) most astrocytes arranged in the confluent sheets at an appropriate plating density show contact inhibition and consequent decease in cell viability or even degeneration of the cells when cultured for more than 6 days; and (2) no changes in cell viability and DNA damage in cultured rat astrocytes were found after the treatment of ethanol at different concentrations of 20, 50, or 100 mM when incubation time was limited within 6 h (Signorini-Allibe et al. 2005).

Taken together, this report shows that a complicated interaction between ethanol, lipid peroxidation, and apoptosis might exist and contribute to AST toxic damage, and even death. Meanwhile, Tβ4 prevents ethanol-induced cortical astrocyte toxic damage through its antioxidative and nervous protective pathway. The protection is also associated with increase in levels of Bcl-2, a critical anti-apoptotic molecule that protects against alcohol-induced astrocyte damages. Therefore, it is highly unlikely that a single mechanism can account for the Tβ4 anti-apoptosis. The present preliminary study on the mechanism of protective effect of Tβ4 on ethanol-induced apoptosis in AST may provide new insights into the future study on Tβ4 properties of anti-apoptosis.