Critical Role of TLR4 Response in the Activation of Microglia Induced by Ethanol

Animals

Female Wistar rats (Harlan Ibérica) weighing 200–250 g, C57BL/6 wild-type mice (Harlan Ibérica), and the TLR4^−/− knockout mice (C57BL/6 background, provided by Dr. S. Akira) were used. All animals were kept under controlled light/dark conditions (12/12 h), temperature (23°C), and humidity (60%). After mating, the dams were placed in separate cages during the gestation period. All animal experiments were conducted in accordance with the guidelines approved by the European Communities Council Directive (86/609/ECC) and by Spanish Royal Decree 1201/2005.

Culture of microglial cells

Primary rat and mouse cultures of cortical mixed glial cells were prepared as previously described (25). Cerebral cortices from 1- to 3-day-old neonatal rats or mice were dissected, carefully stripped of their meninges, and digested with 0.25% trypsin for 30 min at 37°C. Trypsinization was stopped by adding an equal volume of culture medium (DMEM/F-12; Invitrogen) with 14% FBS and 0.02% DNase I. The solution was pelleted, resuspended in culture medium, and brought to a single cell suspension by repeated pipetting following by passage through a 90-μm pore mesh. Cells were seeded at density of either 250,000 or 125,000 cells/ml for rat and mouse cultures, respectively. Cells were maintained at 37°C in humidified 5% CO₂/95% air. Medium was replaced every 4–5 days, and cultures were used between 12 and 20 days in vitro.

Rat and mouse microglial cultures were prepared as previously described (26). Briefly, confluent mixed glial cultures were subjected to mild trypsinization (0.05%) in the presence of 0.25 mM EDTA and 0.5 mM Ca²⁺. This results in the detachment of astroglial cells and leaves a population of firmly attached cells identified as >98–99% microglia, as determined by immunofluorescence using anti-CD11b, anti-GFAP, and anti-mitogen-activated protein-2. Twenty-four hours after trypsinization, cells were rinsed and incubated in serum-free medium with N-2 Supplement (Invitrogen) for 18 h, and then microglial cells were treated with either ethanol (10, 50, 100, or 200 mM; Merck Sharp and Dohme), LPS (50 ng/ml; Sigma-Aldrich), or LTA (1 μg/ml purified lipoteichoic acid from Staphylococcus aureus; InvivoGen) for different times. Most of the ethanol concentrations used in the present study were in the range of the blood alcohol levels found among alcoholics (50–125 mM) (27, 28). In some experiments, microglia were treated with 50 mM ethanol for 24 or 48 h. Then, the microglia-conditional medium, with or without 50 mM ethanol, was collected, frozen, and stored at −80°C until used. Alcohol determinations in the medium were assessed with a spectrophotometric assay kit (Sigma-Aldrich).

To evaluate the possible contamination of 99.9% ethanol with LPS, the endotoxin content was determined by the chromogenic Limulus amebocyte lysate test, following the manufacturer’s instructions (Cambrex Bio Science). The endotoxin content in the ethanol solution was ≤2.98 × 10⁻³ pg/ml, which is far below the concentration required to induce microglial activation under our assay conditions. Additionally, to further eliminate LPS-dependent effects in some experiments and as previously described (29, 30), the LPS antagonist, polymyxin B sulfate (PMBS, 10 μg/ml; Sigma-Aldrich), was added to the cell medium 15 min before and during ethanol or LPS treatment, or to supernatants from TLR4^−/− microglia treated with or without ethanol. PMBS is a cationic cyclic polypeptide that binds to the lipid A portion of the LPS in the cell membrane of Gram-negative bacteria, acting as a specific LPS antagonist.

Morphological analyses

Rat microglia were stimulated with 50 mM ethanol for 7 or 24 h. Cells grown on 12-well culture plates were examined in a minimum of three wells for each experiment, and at least three experiments were conducted for each condition. Morphological differentiation was digitally recorded with a camera (model DFC-480; Leica Microsystems) mounted on a Leica microscope (model DM-6000B; Leica Microsystems).

BrdU incorporation

Microglial cells, either treated with or without ethanol for 12, 24, and 48 h, were incubated with BrdU labeling reagent for 30 min at 37°C, following the manufacturer’s instructions (Roche Diagnostics). Briefly after postfixation with 70% ethanol, cells were incubated with 4 N HCl for 10 min at room temperature, and the staining with a monoclonal anti-BrdU Ab was coupled with FITC (Roche Diagnostics) for 45 min at 37°C. Finally, samples were analyzed with a Cytomics FC500 flow cytometer (Beckman Coulter).

Phagocytosis

Rat microglial cells were cultured on glass coverslips and incubated with 2-μm-diameter FluoSpheres at 0.01% solid mass (Molecular Probes) at 37°C for 30 min. After rinsing four times with PBS, cells were fixed with methanol for 5 min at −20°C and incubated for 1 h with FITC-labeled tomato lectin (1:500; Sigma-Aldrich). After a 5-min incubation with Hoechst 33342 (2.5 μg/ml in PBS; Molecular Probes), coverslips were mounted in fluorescent mounting medium (Difco). The results were digitally recorded with a camera (Ricoh XR-R 3000) mounted on a Zeiss microscope (Axioskop 2). To verify uptake by phagocytes, Z-scan images through the microglia cells were obtained by a Leica confocal microscope (model TCS-SP2-AOBS) to corroborate that the fluorescein-labeled latex beads had been internalized and were not above the cell or adherent to the cell surface.

Cytokine analysis

TNF-α and IL-1β levels were determined in cell culture media by ELISA kits from Diaclone Research (Bionova Científica) following the manufacturer’s protocols.

Measurement of nitrite production

NO production was assessed by measuring the accumulation of nitrite in the culture medium by the Griess reaction using sodium nitrite as standard. Aliquots of culture medium were mixed with an equal volume of Griess reagent prepared by mixing equal volumes of 1% (w/v) sulfanilamide in 30% acetic acid and 0.5% (w/v) of N-1-naphthylethylenediamine dihydrochloride in 60% acetic acid. Then, absorbance was measured at 540 nm in an automated microplate reader.

Neuronal culture and apoptotic quantification by flow cytometry

Primary cultures of rat or mice cortical neurons were prepared as previously described (31), although some modifications were made for mice cortical neurons. Briefly, cortices from 17-day-old fetal wild-type mice were triturated and digested with 0.05% trypsin for 15 min at 37°C. Trypsinization was stopped with 14% FBS and 0.015% DNase I. Cells were harvested by centrifugation, resuspended in DMEM culture medium (Invitrogen), and passed through a 90-μm pore mesh. Cells were plated at a density of 1 × 10⁶ cells/ml onto poly-d-lysine (12.5 μg/ml)-coated tissue culture wells. After 1 h of incubation in a humidified atmosphere of 5% CO₂/95% air at 37°C, the cell medium was replaced by Neurobasal medium supplemented with B-27 (Invitrogen) and 1% glutamine.

At day 5 in vitro, the rat or mice neuronal medium was removed and replaced by a conditioned medium from microglial cells treated with or without ethanol (50 mM) for 24 or 48 h. Neurons were incubated with microglia-conditioned medium for 1 and 2 days. Then, to analyze apoptosis and to differentiate between apoptosis and necrosis, cells were harvested and incubated with R-PE-conjugated annexin V (Molecular Probes) in combination with the cell-impermeant DNA fluorophore 7-aminoactinomycin D (7-ADD; Molecular Probes) at room temperature for 15 min, following manufacturer’s protocol. Finally, samples were analyzed with a Cytomics FC500 flow cytometer (Beckman Coulter) and data (annexin V⁺ and 7-ADD⁻ cells) were illustrated by CXP software.

Immunoprecipitation

Microglial cells were treated with lysis buffer (1% Nonidet P-40, 20 mM Tris-HCl (pH 8), 4 mM sodium chloride, 40 mM sodium fluoride, and protease inhibitors) for 30 min on ice. Cellular extracts were precleaned with protein G-agarose for 1 h and then incubated overnight with anti-IRAK polyclonal Ab or with anti-STAT-1 polyclonal Ab (Santa Cruz Biotechnology) or preimmune serum. Then, protein G-agarose beads were added and samples were incubated for 1 h. Precipitated beads were solubilized by boiling in SDS buffer, and samples were analyzed by Western blot.

Western blot analysis

After ethanol, LPS, or LTA treatment, microglial cells were dissolved in lysis buffer for 30 min on ice. To analyze p65, subunit of NF-κB, the nuclear fraction was isolated as previously described (32). An equal amount of cell lysate of each sample was loaded onto SDS-PAGE and then blotted onto polyvinylidene difluoride membrane. Membranes were blocked with 5% dried milk in Tris-buffered saline containing 0.1% Tween 20 (TBST), and then incubated overnight with the following Abs: anti-lamin A/C, anti-phospho (P)-IRAK1 (Ser³⁷⁶), and anti-P-IRF-3 (Ser³⁹⁶) from Cell Signaling Technology; and anti-IRAK1, anti-P-ERK-1/2, anti-ERK, anti-P-p38, anti-p38, anti-P-JNK, anti-JNK, anti-inducible NO synthase (iNOS), anti-cyclooxygenase-2 (COX-2), anti-p65, anti-P-serine, anti-IRF-1, anti-IRF-3, and anti-GAPDH from Santa Cruz Biotechnology. Membranes were then incubated with the appropriate secondary Ab, and proteins were visualized with either ECL system (ECL Plus, Amersham Pharmacia Biotech) or alkaline phosphatase conjugate (Sigma-Aldrich). Band densities were quantified using the AlphaImager 2200 software (Alpha Innotech).

Native PAGE for IRF-3 detection

To detect dimeric IRF-3, cells were lysed in whole microglia cell extract. After addition of native sample buffer (without SDS and 2-ME), cell extracts were run on a 10% separating gel and 4% stacking gel without SDS. Gels were pre-run for 20 min at 15 mÅ at 4°C. Electrophoresis was conducted at 15 mÅ and 4°C using a Tris-glycine running buffer (27 mM Tris (pH 8.3)/186 mM glycine), resulting in a separation time of ∼150 min. After electrophoretic separation, gels were incubated in 0.1% SDS Tris-glycine transfer buffer (25 mM Tris (pH 8.3)/25 mM glycine). Monomeric and dimeric forms and phosphorylated IRF-3 were detected by Western blot.

RNA isolation and quantitative RT-PCR

mRNA of the gene of interest was quantified by quantitative real-time RT-PCR. Total RNA from mouse microglial cells treated with or without ethanol (50 mM) was isolated using the TRI reagent method (Sigma-Aldrich), as described in the manufacturer’s protocol. Reverse transcription of 0.5–1 μg of total RNA was performed using the Transcriptor First Strand cDNA synthesis kit (Roche Diagnostics). The results were standardized via the amplification product of cyclophilin A. The sequences of both forward and reverse primers were 5′-CCACCACGCTCTTCTGTCTAC-3′ and 5′-CTCCTCCACTTGGTGGTTTG-3′ corresponding to TNF-α, 5′-TCCAAGAAAGGACGAACATTC-3′ and 5′-TGAGGACATCTCCCACGTCAA-3′ corresponding to INF-β, and 5′-CGAGCTGTTTGCAGACAAAG-3′ and 5′-AGATGCCAGGACCTGTATGC-3′ corresponding to cyclophilin A. Quantitative two-step RT-PCR was performed with the LightCycler 480 detection system (Roche Diagnostics). Amplifications of target and housekeeping (cyclophilin) genes were performed using SYBR Green PCR Master Mix (Roche Diagnostics), following the manufacturer’s instructions, in a LightCycler 480 System (Roche Diagnostics). Data analyses were done using the LightCycler 480 relative quantification software with data normalized to the mRNA level of the cyclophilin A housekeeping gene, and efficiencies were corrected of target and housekeeping genes.

Ethanol administration and brain immunohistochemistry

Mice (wild-type and knockout) were administered with an i.p. injection of saline or 25% (v/v) ethanol (4 g/kg) in isotonic saline daily for 3 consecutive days. After 24 h of ethanol administration, mice were sacrificed and brains were rapidly removed and frozen. To visualize CD11b⁺ microglia-expressing cells, 40-μm-thick cryostat coronal brain sections (Leica Microsystems) were mounted onto microscope slides. Tissue sections were rinsed in TBS and fixed with 4% paraformaldehyde in 0.1 M phosphate buffer at pH 7.4 for 15 min. After blocking with 5% bovine serum for 60 min, sections were incubated with an anti-CD11b Ab (Abcam) overnight at 4°C. Afterward the tissue was rinsed in TBST and incubated with a biotinylated anti-rat IgG Ab (Vector Laboratories). Finally, the avidin-biotin peroxidase method, using 3,3′-diaminobenzidine as a substrate (Sigma-Aldrich), was performed according to the manufacturer’s protocol. As a negative control, slides were processed without a primary Ab. The results were digitally recorded with a camera (Leica Microsystems, model DFC-480) mounted onto a Leica microscope (Leica Microsystems, model DM-6000B). The CD11b immunoreactivity was quantified using the Meta Imaging Series 7.0 analysis software. Quantification was assessed by measuring the percentage of threshold area occupied by CD11b⁺ staining relative to the whole area of tissue in cerebral cortex sections of nontreated (control) and ethanol-treated animals. Results were expressed as fold increase over control values. Approximately 18–20 fields of the cerebral cortex from three to four coronal brain sections of three to four animals per group were analyzed.

Ethanol changes the morphology of microglial cells

Microglial cells, the resident macrophage population of the brain, are activated in response to different stimuli, which lead to morphological and functional changes. Therefore, to investigate whether ethanol activates the morphology of microglia, cells were incubated with ethanol (50 mM) for either 7 or 24 h. Fig. 1⇓A shows that ethanol treatment induced an activated state presenting a rod-like bipolar morphology with elongated cell bodies. Additionally, ethanol-treated cells also showed bushy morphology characterized by multiple processes radiating from the central cell body (arrows) and ameboid or macrophage-like morphology (arrowheads).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Ethanol activates the morphology and phagocytic activity of microglial cells. A, Ethanol treatment for 7 and 24 h changed the morphology of microglial cells. Arrows and arrowheads show examples of bushy and ameboid morphology, respectively. Scale bar, 10 μm. B, Microglial cells treated with or without ethanol for 24 and 48 h were exposed to fluorescein-labeled latex beads for 30 min to internalize them, which reveals the phagocytic activity of these cells. Microglia were labeled with tomato lectin. Scale bar, 20 μm. A representative photomicrograph from five different experiments is shown.

To assess whether the changes in morphology in the ethanol-treated microglia were associated with changes in cell proliferation, we incorporated BrdU into the microglial cells treated with or without 50 mM ethanol for 12, 24, and 48 h. The results revealed that, in the absence of serum, very low levels in the BrdU incorporation were detected in either cells treated with or without ethanol for 12, 24, and 48 h, and no differences were noted between control and ethanol-treated cells (data not shown). These results indicate that the changes in morphology were not associated with changes in cell proliferation.

To establish the phagocytic capacity of ethanol-treated microglia, cells were exposed to fluorescein-labeled latex beads for 30 min. Fig. 1⇑B shows that ethanol-treated cells exhibited a more robust phagocytic response after 48 h than after 24 h of treatment, indicating that ethanol stimulates the phagocytic machinery of microglia cells, which were capable of ingesting latex beads.

Ethanol activates microglial cells to release inflammatory mediators, which induces neuronal cell death

As one of the characteristic of activated microglia is to secrete proinflammatory mediators, we determined the production of TNF-α, IL-1β, and NO in the medium of microglial cells treated with ethanol, at different concentrations (10, 50, and 100 mM), for 7, 24, and 48 h. As a positive control, we used microglial cells stimulated with LPS (50 ng/ml) for 24 h. Fig. 2⇓A shows that both LPS and ethanol, at either low or high concentrations, stimulated microglia to produce cytokines and NO. In general, ethanol stimulated the release of TNF-α, IL-1β, and NO in microglial cells from 7 h onward at all the concentrations used (10, 50, and 100 mM), although the maximum increase was observed at 24 h. At this time, the production of IL-1β was similar in cells treated with either ethanol or LPS. Nonetheless, the levels of TNF-α and NO were ∼2-fold higher in LPS (TNF-α, 188 ± 10 pg/ml; NO, 7.72 ± 1.3 μM) than in the ethanol-treated microglia (TNF-α, 65–84 pg/ml; NO, 2.4–3.2 μM). The cytokine and NO levels were in the previously described range for LPS-activated microglia (33, 34).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Ethanol increases the production of TNF-α, IL-1β, and NO in microglia and causes the apoptosis of cortical neurons. A, Microglial cells were treated with ethanol (10, 50, and 100 mM) for 7, 24, and 48 h or LPS (50 ng/ml) for 24 h, and the levels of TNF-α, IL-1β, and NO were measured in the culture media. B, For apoptosis analysis of cortical neurons by flow cytometry, cells were incubated for 1 and 2 days with microglia-conditioned medium obtained from ethanol-treated microglial cells (50 mM) for 24 and 48 h. Data represent means ± SEM; n = 5–6 independent experiments. ∗, p < 0.05 and ∗∗, p < 0.01 vs control group, according to the one-way ANOVA followed by Newman-Keuls post hoc test.

We next addressed whether the production of cytokines and other inflammatory mediators secreted by activated microglia can induce neuronal damage. For these experiments cortical neurons (at 5 days in vitro) were cultured with microglia-conditioned medium obtained from those cells treated with 50 mM ethanol for either 24 or 48 h, in which maximal release of inflammatory mediators was observed (Fig. 2⇑A). Neurons were cultured with the microglia-conditioned medium for 1 and 2 days and neuronal apoptosis was measured by flow cytometry. Furthermore, since the microglia-conditionrd medium contained some alcohol, we used neurons incubated with a medium in which ethanol was added at the same concentrations found in the conditioned medium of ethanol-treated microglia as control neurons. The ethanol concentrations found were 4.8 ± 1.1 and 2.5 ± 0.7 mM for the supernatant of the microglia treated with ethanol for 1 or 2 days, respectively.

The results in Fig. 2⇑B show a significant apoptosis in the neurons cultured for 2 days with the supernatant of microglia treated with 50 mM ethanol for 24 or 48 h. However, higher levels of apoptosis were observed in neurons incubated with conditioned medium of 48-h ethanol-treated microglia than in those neurons exposed to the conditioned-medium of microglia treated for 24 h with ethanol. The results also revealed that the addition of 2.5 or 4.8 mM ethanol to the culture medium did not significantly affect the natural apoptosis observed in those neurons cultured without alcohol (data not shown). These results indicate that microglia produced inflammatory mediators in response to ethanol and that the mediators accumulated in the medium were capable of inducing neuronal death.

Ethanol activates IRAK and MAPK pathways in microglia

Our previous results demonstrate that ethanol activates macrophages by stimulating TLR4 response, initiating an inflammatory processes (24). We therefore determined whether acute ethanol treatment could stimulate TLRs in microglial cells, by analyzing the activation of IRAK, ERK, p38, and JNK (35) in astroglia treated with 50 mM ethanol over different times. To asses IRAK phosphorylation, this kinase was immunoprecipitated and then detected with anti-P-IRAK1 (Ser³⁷⁶) by Western blot. Fig. 3⇓A shows that the phosphorylation of IRAK occurred rapidly at 10 min, increased at 30 min, and then remained high up to 3 h. Similarly, phosphorylation of ERK, p38, and JNK also occurred as early as 10 min, although maximal activation was noted between 30 and 60 min and decreased thereafter (Fig. 3⇓A). Furthermore, when cells were treated with different ethanol concentrations for 30 min, we observed a dose-dependent increase in the activation of ERK, JNK, and p38, although no significant differences in phosphorylation of these kinases were noted between 50 and 100 mM ethanol (Fig. 3⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Ethanol induces the activation of the MAPK and IRAK pathways in cortical microglial cells. A, Immunoblot analysis and quantification of P-IRAK, P-ERK, P-JNK, and P-p38 in cell extracts of ethanol (50 mM)-treated cells at different time periods (10 and 30 min and 1, 3, 7, and 24 h). PS, pre-immune serum. B, Immunoblot analysis and quantification of P-ERK, P-JNK, and P-p38 from microglia extracts incubated with or without different concentrations of ethanol (10, 50, and 100 mM) for 30 min. Values represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 and ∗∗, p < 0.01 compared with the control value, according to the one-way ANOVA followed by Dunnett’s post hoc test. Blots were stripped, and the total quantities of ERK, JNK, and p38 were also assessed. A representative immunoblot of each protein is shown.

Ethanol treatment activates the NF-κB pathway and increases iNOS and COX-2 in microglial cells

Activation of TLRs leads to a rapid downstream stimulation of NF-κB signaling (35), which triggers the transcriptional induction of some inflammatory mediators, including COX-2 and iNOS (36, 37). We therefore investigated the potential nuclear translocation of NF-κB following the stimulation of microglia with ethanol. For these experiments, microglial cells were treated with either ethanol (50 mM) or LPS (50 ng/ml) for 30 min, and the p65 subunit of NF-κB in the nuclear fraction was assessed. As shown in Fig. 4⇓A, either ethanol or LPS was capable of activating NF-κB, as demonstrated by the increased levels of the NF-κB subunit, p65, in the nucleus. We then examined whether microglia expressed the inflammatory mediators iNOS and COX-2 in response to ethanol treatment. Fig. 4⇓B shows that while the iNOS levels significantly increased at 1 and 3 h, and then decreased at 7 and 24 h of ethanol treatment, the COX-2 expression peaked 3 and 7 h of treatment, although these levels were still high after 24 h of the ethanol incubation. These results indicate that ethanol is capable of inducing a rapid response of NF-κB in microglia, triggering the expression of cytokines (e.g., TNF-α) and inflammatory mediators such as iNOS and COX-2.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Ethanol induces the activation of the NF-κB pathway and increases iNOS and COX-2 in cortical microglial cells. A, Immunoblot analysis and quantification of p65 in nuclear extracts from microglia treated with either ethanol (50 mM) or LPS (50 ng/ml) for 30 min. Values represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 compared with the control value, according to t test. B, Immunoblot analysis and quantification of COX-2 and iNOS in microglia extracts treated with ethanol (50 mM) for different times (10 and 30 min and 1, 3, 7, and 24 h). Values represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 and ∗∗, p < 0.01 compared with the control value, according to a one-way ANOVA followed by Dunnett’s post hoc test. A representative immunoblot of each protein is shown. C, Immunoblot analysis and quatification of P-ERK and COX-2 in microglial cells treated with ethanol (50 mM) or LPS (50 ng/ml) in the presence or absence of PMBS (10 μM) during 30 min and 3 h, respectively. Data represent means ± SEM of three independent experiments. ∗, p < 0.05 compared with the control value and #, p < 0.05 compared with the LPS value, according to a one-way ANOVA followed by a Newman-Keuls post hoc test. Blots were stripped, and the total quantities of ERK, GAPDH, and lamin A/C were also assessed. A representative immunoblot of each protein is shown.

To assess whether the ethanol-induced inflammatory signaling was due to the direct effects of ethanol and not to contaminating endotoxin, microglial cells were incubated with the LPS antagonist PMBS (10 μg/ml) 15 min before and during ethanol (50 mM) or LPS (50 ng/ml) treatment. The levels of ERK phosphorylation and COX-2 expression were evaluated. We found that while PMBS completely abolished the LPS-induced ERK phosphorylation and COX-2 expressions, treatment with PMBS did not significantly change the effects of ethanol on these proteins (Fig. 4⇑C). These results confirm direct effects of ethanol on the activation of ERK and COX-2 expression, rather than to contaminating endotoxin.

Ethanol induces the activation of both IRF-3 and the STAT-1/IRF-1 pathways in cortical microglial cells

Recent evidence indicated that TLR4 and TLR3 responses trigger two distinct signaling pathways, MyD88-dependent and -independent (38). The second pathway utilized TRAM/TRIF adaptors and led to the activation of IRF-3 and to the induction of IFN-inducible genes (39, 40). Previous results in our laboratory demonstrated that ethanol promotes the activation of TLR4 signaling response in macrophages, since blocking these receptors abolishes the inflammatory response induced by ethanol (24). We therefore evaluated whether ethanol could stimulate a MyD88-independent pathway. Activation of IRF-3 by ethanol was assessed by detecting the formation of IRF-3 dimers, which results from the phosphorylation of IRF-3 at several residues in the C-terminal end (41). Fig. 5⇓A shows the IRF-3 activation in microglial cells treated with 50 mM ethanol. Ethanol significantly increased the formation of IRF-3 dimers and the phosphorylation of IRF-3 after 3 h of ethanol treatment. Furthermore, because activation of IRF-3 induces the production of IFN-β, which may in turn initiate STAT-1/IRF-1 signaling pathway in microglia (13), we assessed the effects of ethanol on STAT-1 phosphorylation (Fig. 5⇓B) and IRF-1 expression (Fig. 5⇓C). Ethanol stimulation increased both the STAT-1 phosphorylation and the IRF-1 expression at 1 and 3 h of ethanol treatment. These results suggest that ethanol is capable of activating a MyD88-independent pathway and the production of IFN-β, which might initiate the STAT-1/IRF-1 pathway.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Ethanol induces the activation of both IRF-3 and the STAT-1/IRF-1 pathways in cortical microglial cells. A, To detect the formation of IRF-3 dimers, the microglia cells treated with or without 50 mM ethanol for 1 or 3 h were harvested and lysates were run on native gels as described in Materials and Methods. The monomeric and dimeric IRF-3 forms in the blots were detected by using IRF-3 Ab. Then, the blots were reprobed with the P-IRF-3 (Ser³⁹⁶) Ab to detect the phosphorylation of IRF-3. B, For STAT-1, lysates from microglia treated with or without ethanol for 1 or 3 h were immunoprecipitated, subjected to Western blot, and P-STAT-1 was detected using anti-phosho-STAT-1 (Ser). PS, pre-immune serum. C, Immunoblot analysis and quantification of IRF-1 in cell extracts of microglia treated with or without 50 mM ethanol (50 mM) for different times (10 and 30 min and 1, 3, 7, and 24 h). Values represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 compared with the control value, according to a one-way ANOVA followed by Dunnett’s post hoc test. Blots were stripped, and the total quantities of GAPDH were also assessed. A representative immunoblot of each protein is shown.

TLR4 response is critical for ethanol-induced microglia activation

Because our previous studies suggest that ethanol induces inflammatory signaling by activating TLR4 receptors (24, 42), we assessed the potential role of TLR4 response in the ethanol-induced microglia activation by using microglial cells of wild-type and TLR4^−/− knockout mice. For these experiments, cells were stimulated with ethanol, and the production of cytokines and MAPK activation were assessed. Fig. 6⇓A shows that microglia of wild-type mice produced TNF-α in response to ethanol (50 mM), as demonstrated by the significant increase noted in the mRNA levels of TNF-α after 3 h of ethanol treatment. Conversely, no changes in the TNF-α levels in response to ethanol were observed in TLR4^−/− microglia. We then determined the activation of MAPKs induced by LPS or ethanol in microglial cells of wild-type and TLR4^−/− mice. According to the results with rat microglia (Fig. 3⇑), treatment with either LPS (50 ng/ml) or ethanol (10, 50, 100, and 200 mM) for 30 min induced the phosphorylation of ERK, JNK, and p38 (Fig. 6⇓, B and C) in microglia of wild-type mice. However, activation of the above kinases does not occur in LPS- or ethanol-stimulated microglia of TLR4^−/− mice. These results support the role of TLR4 response in the ethanol-induced activation of microglia cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Effects of ethanol on TLR4 signaling in microglia of TLR4^−/− and wild-type mice. A, Levels of TNF-α mRNA in microglial cells of wild-type and TLR4^−/− mice treated with or without ethanol (50 mM) for 1, 3, and 7 h. B, Immunoblot analysis and (C) quantification of P-ERK, P-JNK, and P-p38 in microglial extracts of wild-type and TLR4^−/− mice, treated either with ethanol (10, 50, 100, and 200 mM) or LPS (50 ng/ml)-treated cells of wild-type and TLR4^−/− mice for 30 min. D, Immunoblot analysis of P-ERK, P-JNK and P-p38 from wild-type (WT) or TLR4^−/− (KO) microglia treated with LTA (1 μg/ml) for 30 min and 2 h. A representative immunoblot from three different experiments is shown. E, To detect the formation of IRF-3 dimers, the microglia cells of TLR4^−/− and wild-type mice, treated with or without ethanol (50 mM) for 1 and 3 h, were harvested and run on native gels as described in Materials and Methods. The monomeric and dimeric IRF-3 forms in the blots were detected by IRF-3 Ab, and then blots were reprobed with the P-IRF-3 (Ser³⁹⁶) Ab. F, The IFN-β mRNA levels of microglial cells of wild-type and TLR4^−/− mice were treated with or without ethanol (50 mM) for 1, 3, and 7 h. G, Immunoblot analysis and quantification of IRF-1 in cell extracts of TLR4^−/− and wild-type microglia treated with ethanol (50 mM) for different times (10 and 30 min and 1, 3, 7, and 24 h). Data represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 and ∗∗, p < 0.01 compared with the control value, according to a one-way ANOVA followed by Dunnett’s post hoc test. Blots were stripped, and the total quantities of ERK, JNK, p38, and GAPDH were also assessed. A representative immunoblot of each protein is shown.

To investigate whether TLR4-deficient cells could respond to other TLRs ligands, microglial cells were treated with LTA, a TLR2 agonist, and the activation of ERK, JNK, and p38 phosphorylation was measured upon 30-min and 2-h LTA treatment. Fig. 6⇑D shows that LTA induced phosphorylation of ERK, JNK, and p38 upon 2 h of treatment in both TLR4 wild-type and TLR4-deficient microglia. These results reveal that TLR4-deficient cells are able to respond to TLR2 ligand.

Furthermore, to evaluate the role of the TLR4-mediate MyD88-independent pathway in ethanol-induced stimulation of microglia, we analyzed the activation of IRF-3, the IFN-β mRNA levels, and the IRF-1 expression in ethanol-treated microglia of wild-type and TLR4^−/− knockout mice. Fig. 6⇑E shows that while ethanol (50 mM) significantly increased both the formation of IRF-3 dimers and the phosphorylation of IRF-3 after 3 h of treatment in microglia of wild-type mice, no significant changes in the IRF-3 activation were observed in TLR4^−/− microglia. Likewise, ethanol significantly increased both the IFN-β mRNA levels and the IRF-1 expression at 3 and 7 h of treatment in wild-type microglia. In contrast, ethanol does not induce significant variations in either IFN-β mRNA levels or IRF-1 expression in microglia of TLR4^−/− mice (Fig. 6⇑, F and G). These results support the notion that ethanol-induced cytokine production, including IFN-β, is mediated by TLR4 activation via MyD88-dependent and -independent pathways.

A deficiency in the TLR4 function protects neurons from the effects of ethanol on microglia activation

We then assessed the relationship between the TLR4 function and the neurotoxicity associated with ethanol-induced microglia activation. To do this, we performed similar experiments as those described above, but in this case, the conditional medium from the wild-type or TLR4^−/− microglia treated with 50 mM ethanol for 24 and 48 h were added to mice cortical neurons at 5 days in vitro. The murine neurons were cultured with the microglia-conditioned medium for 1 day and the neuronal apoptosis was measured by flow cytometry.

Fig. 7⇓A shows a significant apoptosis in the neurons cultured for 1 day with the supernatant of wild-type microglia treated with 50 mM ethanol for 48 h. Conversely, we did not observe significant changes in neuronal survival when neurons were cultured with the supernatant of ethanol-treated TLR4^−/− knockout cells. These findings support the notion that ethanol-induced neuronal death depends on the activated TLR4 pathway in microglial cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

A deficiency in TLR4 function protects neurons from the effects of ethanol on microglial activation. A, Apoptosis analysis of murine cortical neurons assessed by flow cytometry. Neurons were incubated for 1 day with wild-type (WT) or TLR4 knockout (KO) microglia-conditioned medium obtained from ethanol-treated microglial cells (50 mM) for 24 and 48 h. Data represent means ± SEM; n = 7 independent experiments. ∗, p < 0.05 compared with the control value, according to a one-way ANOVA followed by Dunnett’s post hoc test. B, Immunoblot analysis and quantification of P-ERK and iNOS in wild-type microglia incubated in presence or absence of PMBS (10 μM) for 30 min or 3 h with the supernatants of TLR4^−/− microglia either untreated or treated with 50 mM ethanol for 24 or 48 h. CM, conditioned medium. Data represent means ± SEM of three independent experiments. Values of p were not significant compared with the control value, according to a one-way ANOVA followed by a Newman-Keuls post hoc test. Blots were stripped, and the total quantities of ERK and GAPDH were also assessed. A representative immunoblot of each protein is shown.

Finally, as microglia can be activated by endogenous TLR ligands (43, 44), we evaluated the possibility of ethanol being able to release endogenous TLR4 ligands from microglia, which could, on TLR4 wild-type microglia, act in an autocrine- or a paracrine-like manner, leading to the activation of the TLR4 pathway. To investigate this possibility, the supernatants from TLR4^−/− microglia treated with or without 50 mM ethanol for 24 and 48 h were added to wild-type microglia plates in the presence or absence of PMBS, and P-ERK and iNOS levels were assessed after 30 min and 3 h of incubation, respectively. There were no significant differences in the levels of P-ERK and iNOS irrespective of the nature of the conditioned medium (control or ethanol-treated) or the presence or absence of PMBS (Fig. 7⇑B). These results suggest that the culture supernatant of ethanol-treated TLR4^−/− microglia does not contain endotoxin, endogenous TLR ligands, or other compounds that could activate wild-type microglia under our in vitro experimental conditions.

Role of TLR4 in ethanol-induced in vivo microglia activation

To provide more physiological relevance to our in vitro results on the role of TLR4 receptors in the ethanol-induced microglia activation, we performed in vivo experiments in which one i.p. injection of either saline or ethanol (4 g/kg) was administered to TLR4 wild-type and TLR4^−/− knockout animals for 3 consecutive days. The activated microglia in coronal brain sections were analyzed for CD11b expression by immunohistochemistry. Representative micrographs from the cerebral cortex of wild-type and TLR4^−/− knockout are shown in Fig. 8⇓. Morphometric analysis revealed that ethanol administration induced greater CD11b immunoreactivity in the cerebral cortex of wild-type mice than in TLR4^−/− knockout mice.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

CD11b immunoreactive microglia in cerebral cortex sections of wild-type (WT) and TLR4^−/− (KO) mice treated with one i.p. injection of either saline or ethanol (4 g/kg) for 3 consecutive days. Arrows represent greater CD11b immunoreactivity in cerebral cortex of wild-type mice treated with ethanol. Insets are ×40 images taken from the area outlined in lower magnification pictures. Scale bar of inset, 20 μm; scale bar of low magnification images (×5), 150 μm. Results are given as means ± SEM. ∗, p < 0.01 compared the ethanol-treated wild-type or TLR4^−/− knockout mice with their respective controls, according to t test.