SEARCH

SEARCH BY CITATION

Keywords:

  • apoptosis;
  • ethanol;
  • immunopathology;
  • inflammation;
  • lipid rafts;
  • TLRs

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Alcohol consumption can induce brain damage, demyelination, and neuronal death, although the mechanisms are poorly understood. Toll-like receptors are sensors of the innate immune system and their activation induces inflammatory processes. We have reported that ethanol activates and recruits Toll-like receptor (TLR)4 receptors within the lipid rafts of glial cells, triggering the production of inflammatory mediators and causing neuroinflammation. Since TLR2 can also participate in the glial response and in the neuroinflammation, we investigate the effects of ethanol on TLR4/TLR2 responses. Here, we demonstrate that ethanol up-regulates TLR4 and TLR2 expression in microglial cells, inducing the production of inflammatory mediators which triggers reactive oxygen species generation and neuronal apoptosis. Ethanol also promotes TLR4/TLR2 recruitment into lipid rafts-caveolae, mimicking their activation by their ligands, lipopolysaccharide, and lipoteichoic acid (LTA). Immunoprecipitation and confocal microscopy studies reveal that ethanol induces a physical association between TLR2 and TLR4 receptors, suggesting the formation of heterodimers. Using microglia from either TLR2 or TLR4 knockout mice, we show that TLR2 potentiates the effects of ethanol on the TLR4 response reflected by the activation of MAPKs and inducible NO synthase. In summary, we provide evidence for a mechanism by which ethanol triggers TLR4/TLR2 association contributing to the neuroinflammation and neurodegeneration associated with alcohol abuse.

Abbreviations used
CNS

central nervous system

FBS

fetal bovine serum

LR

lipid rafts

LTA

lipoteichoic acid

Microglial cells are considered the resident macrophage-like population within the CNS and they are the prime component of the brain immune system (Streit and Xue 2009). Microglia detects danger signals through various receptors, including Toll-like receptors (TLRs), which play a crucial role in innate immunity by recognizing microbial pathogens and ligands from injured cells (Lehnardt 2010). Activation of TLRs stimulates glial cells to prevent infections or tissue damage. However, a heightened inflammatory state can lead to unnecessary damage and neuroinflammation (Rivest 2009; Amor et al. 2010). Activation of TLR2 and TLR4 is involved in neurodegeneration. Mice deficient in these TLRs exhibit reduced levels of pro-inflammatory cytokines and milder clinical disease following traumatic brain injury (Koedel et al. 2007; Ziegler et al. 2011). The levels of TLR4 and TLR2 have been reported to increase in Parkinson's disease, stroke, and amyotrophic lateral sclerosis (Letiembre et al. 2009; Okun et al. 2009).

Alcohol is a neurotoxic compound and its abuse can cause brain damage (Pfefferbaum 2004; Harper and Matsumoto 2005). Nevertheless, the mechanism by which ethanol induces neurodegeneration remains largely elusive. Previous work done in our laboratory demonstrates the critical role of TLR4 in alcohol-induced microglia (Fernandez-Lizarbe et al. 2009) and astroglial activation (Alfonso-Loeches et al. 2010), demyelination, and neuronal damage (Fernandez-Lizarbe et al. 2009; Alfonso-Loeches et al. 2012), indicating that activation of the TLR4 response by ethanol can be an important mechanism of ethanol-induced neuroinflammation and neurodegeneration. We also demonstrate that ethanol is capable of activating TLR4 signaling by promoting translocation and the clustering of TLR4 and signaling molecules (IL-1 receptor-associated kinase, myeloid differentiation primary response protein 88, extracellular-signal-regulated kinase) into the membrane microdomains lipid rafts (Blanco et al. 2008; Fernandez-Lizarbe et al. 2008). Accordingly, different studies have demonstrated that, upon activation, TLRs are recruited into lipid rafts microdomains, acting as signaling platforms for several TLRs and leading to innate immune activation (Triantafilou et al. 2011).

Among TLRs, TLR2 seems to be the most promiscuous TLR receptor capable of recognizing the most diverse set of pathogens. TLR2 complexes with TLR1 or TLR6 are involved in the recognition of bacterial lipoproteins (Akira and Takeda 2004; Gay and Gangloff 2007). TLR2 can also interact with other molecules such as CD36 (Triantafilou et al. 2006) or CD14 (Yang et al. 1999; Flo et al. 2002) and can induce multimerization in response to different microbial ligands (Triantafilou et al. 2006).

We propose that ethanol, through its interaction with lipid rafts, can recruit several receptors like TLR4 and TLR2, triggering innate immune activation and leading to neuroinflammation. Here, we report that ethanol not only up-regulates TLR4 and TLR2 receptors in microglial cells but also promotes the recruitment and association of both receptors into rafts-caveolae, leading to TLRs signaling production and cytokine release. These events might contribute to the neuroinflammation and neuronal cell death associated with alcohol abuse.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals

Female Wistar rats (Harlan Ibérica, Barcelona, Spain), C57BL/6 wild-type mice (Harlan Ibérica), TLR4−/− and TLR2−/− knockout mice (C57BL/6 background, kindly provided by Dr. S. Akira) were used for cell isolation. After mating, dams were placed into separate cages during the gestation period. Food and water were provided ad libitum. All the animal experiments were approved by the local Research and Ethics Committee of the Príncipe Felipe Research Center (CIPF). All the experimental procedures were carried out in accordance with the guidelines approved by the EC Council Directive (86/609/ECC) and by Spanish Royal Decree 1201/2005.

Primary microglial and astrocytes cell cultures

Primary rat and mouse cultures of cortical mixed glial cells were prepared as previously described from 1- to 3-day-old neonatal rats or mice (Fernandez-Lizarbe et al. 2009). Cells were seeded at a density of either 250 000 cells/mL or 125 000 cells/mL for rat and mouse cultures, respectively. Cells were maintained at 37°C in humidified 5% CO2-95% air. The medium was replaced every 4–5 days, and cultures were used between 12 and 20 days in vitro. Then, confluent mixed glial cultures were subjected to mild trypsinization for microglia purification as described (Fernandez-Lizarbe et al. 2009). Microglial cells were treated with 50 mM ethanol (Merck Sharp and Dohme, Madrid, Spain) or lipopolysaccharide (LPS) (50 ng/mL, Sigma-Aldrich, Madrid, Spain) or LTA (1 μg/mL purified lipoteichoic acid from Staphylococcus aureus, InvivoGen, San Diego, CA, USA) for different times. To prevent ethanol evaporation, culture plates were placed in sealed containers. In some experiments, microglia was treated with 50 mM ethanol for 24 h. Then, the microglia-conditional medium, with or without 50 mM ethanol, was collected, frozen, and stored at −80°C until use. Alcohol determinations in the medium were assessed with a spectrophotometric assay kit (Sigma-Aldrich).

Primary cultures of mice cortical astrocytes were prepared as previously described (Guerri et al. 1990). Cells were used after 14 days in culture.

Western blot analysis

After cell stimulation with ethanol or LPS or LTA, microglial cells were dissolved in lysis buffer (1% Nonidet P-40, 20 mmol/L Tris-HCl pH8, 4 mmol/L sodium chloride, 40 mmol/L sodium fluoride, and protease inhibitors) for 30 min on ice. An equal amount (40–50 μg) of cell lysate of each sample was loaded onto sodium dodecyl sulfate–polyacrylamide gel electrophoresis, was blotted onto polyvinylidene fluoride membranes, and then was blocked with 50 g/L dried milk in tris buffered saline and tween-20. Membranes were incubated overnight with the following antibodies: anti-glyceraldehyde 3-phosphate dehydrogenase, anti-iNOS, anti-TLR4, anti-phosphorylated extracellular-signal-regulated kinase, anti-ERK, anti-PJNK, anti-JNK, anti-Pp38, anti-p38, anti-flotillin-1, anti-MyD88 and anti-TRAF6 (Santa Cruz Biotechnology, Madrid, Spain), anti-TLR2, (Imgenex, San Diego, CA, USA), and anti-caveolin-1 (Sigma-Aldrich) and next with the appropriate secondary antibody. Proteins were visualized with either the enhanced chemiluminescence system (ECL Plus, Amersham Pharmacia Biotech., Madrid, Spain) or alkaline phosphatase conjugate (Sigma-Aldrich). Band densities were quantified using the AlphaImager 2200 software (Alpha Innotech Corporation, San Leandro, CA, USA).

Quantitive Real-Time RT-PCR

Cells were lysed in 1 mL of Tri-Reagent solution (Sigma, St Louis, MO, USA) and RNA was isolated according to the manufacturer's instructions. The amount of purified RNA and its purity was assessed and RNA 2 μg of each sample was subjected to DNAase treatment (Invitrogen, Carlsbad, CA, USA) and reverse transcribed using the Transcriptor First Strand cDNA Synthesis kit (Roche, Madrid, Spain). cDNA was amplified in a rapid thermal cycler (LightCycler Instrument; Roche Diagnostics) in 10 μL of LightCycler 480 SYBR Green I Master (Roche) using 0.5 μM of each oligonucleotide. Primers were designed using Primer Blast program (NCBI) covering at least one intron. Samples underwent 45× cycles of RT-PCR (2 min at 95°C; 10 s at 62°C Melting Temperature and 18 s at 72°C). For each primer pair, melting temperature was optimized choosing 57°C for peptidylprolyl isomerase A (PPIA) and 62°C for TLR2 and TLR4. The sequence of primer pairs used in this study was as follows: PPIA, 5′- GCG TCT GCT TCG AGC TGT TTG C-3′ (forward) and 5′- ACC ACA TGC TTG CCA TCC AGC-3′ (reverse); TLR2, 5′-GAG GTC TCC AGG TCA AAT CTC AG-3′(forward) and 5′-AAT GGC CTT CCC TTG AGA GG-3′ (reverse); TLR4 5′-TGC CTC TCT TGC ATC TGG CTG G-3′(forward) and 5′-CTG TCA GTA CCA AGG TTG AGA GCT GG-3′(reverse). The mRNA levels of the housekeeping genes PPIA was used as an internal control for normalization. Relative quantification of the PCR products was made using the LightCycler 480 software (Roche).

Isolation of detergent-insoluble fraction

For lipid rafts isolation, a protocol previously described for this cell type was used (Kim et al. 2006). Microglial cells were stimulated with LPS, LTA, or ethanol for 30 min. Cells were then lysed gradually to obtain the soluble and insoluble fractions. Each fraction was analyzed by western blotting. In some experiments, filipin, a lipid rafts-disrupting drug (5 μg/mL, Sigma-Aldrich) was used, 10 min before and during ethanol treatment.

Immunofluorescence and confocal microscopy

Microglial cells plated on glass coverslips were fixed with 3.7% paraformaldehyde for 20 min, blocked with 3% bovine serum albumin for 30 min. Cells were then cell-incubated overnight at 4°C with primary antibodies: anti-TLR4, anti-TLR2 antibodies (both at 1 : 50; Santa Cruz Biotechnology), anti-caveolin-1 (1 : 200, Abcam, Cambridge, UK), or anti-flotillin-1 (1 : 100, BD Transduction Laboratories, Madrid, Spain). After several washings in phosphate-buffered saline, cells were incubated for 90 min at 37°C with the corresponding FITC (1 : 100, Jackson ImmunoResearch, West Grove, PA, USA) AlexaFluor 647 or AlexaFluor 405 (1 : 500; Invitrogen). Coverslips were mounted in FA mounting fluid (Difco, Madrid, Spain). Fluorescence images were quantified in single cells using a Leica confocal microscope (model TCS-SP2-AOBS, Mannheim, Germany). Fluorescence was quantified with the Meta Imaging Series 7.0 analysis software (Molecular Devices, Sunnyvale, CA, USA).

Co-immunoprecipitation assay

Microglial cells were lysed for 30 min on ice-modified ristocetin-induced platelet agglutination buffer (0.05 mol/L Tris-HCl, 10 mL/L Nonidet P-40, 2.5 g/L sodium deoxycholate, 0.15 mol/L NaCl, 1 mmol/L EDTA, 1 mmol/L phenylmethylsulfonyl fluoride, 1 μg/mL leupeptin, 1 μg/mL pepstatin, 1 mmol/L NaVO4). To reduce the background, lysates were incubated with protein A/G-agarose beads (Santa Cruz Biotechnology). Lysates weighing 150–250 μg were then incubated with 2 μg of anti-TLR4 or anti-TLR2 antibody at 4°C overnight and precipitated with protein A/G-agarose beads. Pre-immune serum was used as a control to discard non-specific interactions. Immunoprecipitated proteins were washed in lysis buffer, subjected to denaturing conditions by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and incubated with several antibodies as indicated in the text and figure legends, following the previously described immunoblotting analysis.

Detection of FRET by confocal microscopy

Förster Resonance energy transfer (FRET) is used to study the interactions between TLR4 and TLR2. This technique measures the energy transfer between a donor chromophore (FITC label) on one protein and an acceptor chromophore (TRITC label) on the other protein. Transfer of energy can occur only if the two chromophores (and hence the two proteins) are closer than 10 nm. Cells cultured on poly-d-lysin-coated coverslips were fixed with 37 mL/L paraformaldehyde in phosphate-buffered saline for 20 min, blocked with 30 g/L bovine serum albumin for 30 min, and incubated with the goat anti-TLR4 antibody and the rabbit anti-TLR2 antibody (both at 1 : 50; Santa Cruz Biotechnology) overnight at 4°C. Cells were then incubated with the corresponding FITC or TRITC-conjugated antibodies (1 : 50, Jackson Immunoresearch) for 2 h and were mounted. The Leica (Leica Microsistemas, Barcelona, Spain) confocal microscope hardware/software package allows the measurement of fluorescence of the donor, the acceptor, and then the FRET image. The results are expressed as the% of positive ROIs (regions of interest) with FRET efficiency greater than the negative control and FRET efficiency values.

Neuronal culture and apoptotic quantification by flow cytometry

The primary cultures of the mice cortical neurons from WT mice were prepared as previously described (Fernandez-Lizarbe et al. 2009). On day 5 in vitro, the neuronal medium was removed and replaced with a conditioned medium from the microglial cells from the WT, TLR4−/− and TLR2−/− mice treated with or without ethanol (50 mM) for 24 h and neurons were incubated with microglia-conditioned medium for 1 day.

To analyze apoptosis, cells were harvested by trypsinization for 5 min and incubated with PE-conjugated annexin-V (Molecular Probes, Eugene, OR, USA) in combination with the cell-impermeant DNA fluorophore 7-amino-actinomycin D (Molecular Probes), following the manufacturer's protocol. Double staining was used to differentiate between necrosis and apoptosis. Finally, samples were analyzed with CXP software in a Cytomics FC500 flow cytometer (Beckman Coulter, CA, USA).

Measuring of ROS production

2′-7′-Dichlorodihydrofluorescein diacetate enters cells passively and is de-acetylated by esterases to non-fluorescent DCFH. DCFH reacts with reactive oxygen species (ROS) to form DCF, the fluorescent product (Kim et al. 2011). To analyze the ROS formation, the medium was replaced with the conditioned microglial medium in neurons on day 5 in vitro. After 1 day of incubation, cells were exposed to 20 μmol/L 2′-7′-Dichlorodihydrofluorescein diacetate in Hank's buffered salt solution for 30 min at 37°C. Fluorescence was read immediately at the wavelengths of 485 nm for excitation and 530 nm for emission using a Victor-II spectrofluorimeter plate reader (Wallac, Gaithersburg, MD, USA).

Cytokine analysis

The TNF-α and IL-1β levels were determined in the cell culture media of the microglia from WT, TLR4−/− and TLR2−/− mice by the TNF-α ELISA Kits from Diaclone Research (Bionova Científica, Madrid, Spain) and the IL-1β ELISA kit from eBioscience (San Diego, CA, USA) following the manufacturer's protocols.

Statistical analysis

All statistical analyses were performed using GraphPad Software Inc., La Jolla, CA, USA. The results were represented as Mean ± SEM. Data were analyzed using a one-way anova followed by Dunnett's post hoc test. For the inhibitory experiments, data were analyzed using one-way anova followed by Newman–Keuls post hoc test. A Student's two-tailed t-test was used when two groups were compared and values of p < 0.05; p < 0.01; p < 0.001 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Ethanol increases the expression of TLR4 and TLR2 in microglial cells

To test whether alcohol induces the expression of TLR4 and TLR2 in microglia, cells were treated with an acute dose of ethanol (50 mM) and the expression of both receptors was evaluated by western blot and by qPCR at different ethanol treatment times. Ethanol increases the expression of both the receptors TLR4 and TLR2 (Fig. 1a). The up-regulation of TLR4 occurred rapidly at 5 min, although a significant increase was noted at 10 min and a maximal level was observed at 30 min, which progressively decreased after 1 and 3 h. Similarly, the TLR2 protein level significantly increased in the cells treated with ethanol after 30 min and 1 h of the ethanol treatment. Analysis by RT-PCR also demonstrates that ethanol significantly up-regulated the expression of TLR4 mRNA at 30 and 60 min upon ethanol treatment and then decreased at 3 h. Likewise, TLR2 mRNA levels significantly increased at 10 and 30 min upon ethanol treatment and then decreased at 3 h. (Fig. 1a). These results indicate that ethanol is capable of up-regulating the expression of TLR4 and TLR2 in microglia.

image

Figure 1. Ethanol increases toll-like receptor (TLR)4 and TLR2 levels and promotes their recruitment into the lipid rafts (LR) (a) Microglial cells were treated with 50 mM ethanol during different time periods: 5, 10, and 30 min; 1, 3, and 24 h. The immunoblot analysis and quantification of the expression of TLR4 and TLR2 in cell extracts were assessed. Blots were stripped, and GAPDH levels were also assessed. For LR isolation, microglial cells from rat were treated with LPS 50 ng/mL, Lipoteichoic acid 1 μg/mL (b), or ethanol 50 mM (c), for 5, 10, and 30 min. Insoluble fractions to detergent were isolated. A representative immunoblot of each protein is shown n = 6–8 *p < 0.05, **p < 0.01 compared with the control value. In some experiments, cells were pre-treated with 5 μg/mL of filipin 10 min before and during ethanol treatment (d). #p < 0.05 versus ethanol treatment.

Download figure to PowerPoint

Stimulation of microglial cells with ethanol produces the translocation of TLR4 and TLR2 into lipid rafts

Previous studies have demonstrated that TLR4 and TLR2 can be recruited into LR membrane microdomains upon stimulation by bacterial products (Triantafilou et al. 2004). To corroborate these results, we assessed whether TLR4 and TLR2 are recruited into lipid rafts (LR) upon ligand stimulation.

After isolation of these microdomains, we measured the protein expression of the LR markers, caveolin-1 and flotillin-1, in the detergent-soluble and insoluble fractions. As expected, both LR markers appear to be mainly located in the Triton X-100 insoluble fractions (Fig. 1b–d). Microglia stimulation of TLR4 and TLR2 with their ligands LPS and LTA leads to the recruitment of these receptors into LR-enriched fractions (Fig. 2b), thus supporting the involvement of these microdomains in the activation of TLR4 and TLR2. Our previous studies demonstrate that ethanol mimics the ligand-mediated activation of TLR4 receptors via LR in macrophages and astrocytes (Blanco et al. 2008; Fernandez-Lizarbe et al. 2008), so we evaluated whether ethanol promotes the recruitment of both TLR4 and TLR2 receptors into LR. Ethanol treatment triggers the translocation of TLR4 and TLR2 as well as signaling molecules (MyD88, TRAF6), from the soluble fraction to the LR detergent-resistant fractions (Fig. 1c). However, alcohol efficacy is lower than LPS or LTA in triggering the recruitment of TLR4 or TLR2 receptors into the LR-enriched fractions. Indeed, the proportion of the TLR4 and TLR2 expressions in the LR-insoluble fraction versus the soluble fraction was higher in the cells stimulated with either LPS or LTA than when cells were stimulated with ethanol (Fig. 1c).

image

Figure 2. Toll-like receptor (TLR)4 and TLR2 recruitment into lipid rafts after LPS and Lipoteichoic acid (LTA) treatment. Microglial cells were treated with LPS 50 ng/mL or LTA 1 μg/mL for 5 or 30 min, and were then fixed and immunostained for the expressions of caveolin-1, TLR4, and TLR2. (a) The co-localization of these proteins with caveolin-1, generated by the treatments, was quantified by a confocal microscope. (b, c) A representative photomicrograph of four different experiments is shown. The results are the mean ± SEM of data from six different fields per condition from four different cultures. *p < 0.05, **p < 0.01 versus Scale bar = 10 μm, amplifications = 22 μm.

Download figure to PowerPoint

To obtain further insights of the role of LR in TLRs activation, we used the LR inhibitor filipin, which disrupts lipid microdomains by interacting with cholesterol (Simons and Toomre 2000). Cells pre-treated with 5 μg/mL filipin for 10 min were stimulated with 50 mM ethanol for 30 min. Filipin treatment abolishes the ethanol-induced recruitment of the TLR4 and TLR2 receptors into LR. This result confirms that an intact LR structure is required for ethanol-induced TLR4 and TLR2 recruitment in microglial cells (Fig. 1d).

Ethanol induces the translocation of TLR4 and TLR2 into lipid rafts-caveolae in microglial cells

Two types of LR have been proposed: planar (flotillin-rich domains) and caveolae. Caveolae are small invaginations that are characterized by the presence of caveolin, which appears to be responsible for stabilizing the invaginated structure (Pike 2003; Le Roy and Wrana 2005).

To study the role of LR microdomains in the TLR4 and TLR2 activation, we used confocal microscopy to assess the co-localization of TLR4 or TLR2 with either flotillin-1 or caveolin-1. In resting cells, neither TLR4 nor TLR2 displayed any co-localization with caveolae. Microglial cells were stimulated with 50 ng/mL LPS or 1 mg/mL LTA for 5, 10, and 30 min. As shown in Fig. 2, TLR4 co-localized with caveolae after 30 min of LPS stimulation, whereas the co-localization of TLR2 with caveolae significantly increased at 5 min of LTA stimulation. Conversely, neither TLR4 nor TLR2 co-localized with flotillin (data not shown). We next assessed the co-localization of TLR4 or TLR2 with caveolae upon the stimulation with ethanol for 5–30 min in microglia. When microglia was stimulated with ethanol 50 mM, TLR4 moved into the LR-caveolae, showing a co-localization of TLR4/caveolae at 5, 10, and 30 min of the ethanol treatment (Fig. 3). Co-localization of TLR2 with caveolin-1 was also observed after 30 min of ethanol treatment. These results strongly support the western blot findings, showing a recruitment of TLR4 and TLR2 into LR-caveolae after ethanol treatment.

image

Figure 3. Role of ethanol in the co-localization of toll-like receptor (TLR)4 and TLR2 in caveosomes. Microglial cells were treated with 50 mM ethanol for 5, 10, or 30 min, and were then fixed and immunostained for the expressions of caveolin-1, TLR4, and TLR2. (a) The co-localization of TLR4 and TLR2 with caveolin-1 (b) and the co-localization of TLR4 with TLR2 (c) generated by the treatments were quantified by a confocal microscope. A representative photomicrograph of four different experiments is shown. The results are the Mean ± SEM of the data from six different fields per condition from four different cultures. *p < 0.05, **p < 0.01 versus control. Scale bar = 10 μm, amplifications = 22 μm

Download figure to PowerPoint

Confocal microscopy also illustrated that while microglia treated with either LPS or LTA shows an activated rod-like bipolar morphology with elongated cell bodies (Nakamura et al. 1999), ethanol-treated microglia appeared as cells with an ameboid or macrophage-like morphology (Block et al. 2007; Napoli and Neumann 2009). These results confirm our previous studies demonstrating that ethanol is capable of changing the microglia morphology stimulating the phagocytic machinery (Fernandez-Lizarbe et al. 2009).

Figure 3c, also demonstrates that ethanol treatment, but not its ligands (Fig. 2c), promotes the association of TLR4 and TLR2, as demonstrated by the increase noted in the TLR4/TLR2 co-localization. These results suggest that while natural ligands LPS and LTA induce TLR4 and TLR2 recruitment into the LR, they do not promote the interaction between both receptors. Conversely, by acting indirectly through membrane lipid rafts, ethanol can induce TLR4/TLR2 recruitment and an association within LR-caveolae.

Ethanol triggers a physical association between TLR4 and TLR2 in microglial cells

We further tested the possibility that TLR4 and TLR2 could interact upon ethanol treatment, so we performed co-immunoprecipitation experiments to study protein–protein interaction under in vivo conditions. As shown in Fig. 4a, TLR4 co-immunoprecipitated with TLR2 after ethanol treatment. Similar results were obtained when the cell extract was immunoprecipitated with anti-TLR4 and anti-TLR2 was used for the western blotting purposes. These results indicate that ethanol promotes TLR4 and TLR2 dimer or multimer in vivo.

image

Figure 4. Ethanol promotes toll-like receptor (TLR)4-TLR2 interaction. (a) Lysates from microglial cells from rat, untreated and treated with 50 mM ethanol for 30 min, were immunoprecipitated with anti-TLR4 or anti-TLR2, as a capture antibody, followed by a western blotting analysis. PS, pre-immune serum. A representative immunoblot from four different experiments is shown. (b) Microglial cells stimulated with 50 mM ethanol for 30 min, or non-stimulated, were dual labeled with anti-TLR4 plus anti-goat FITC and anti-TLR2, followed by anti-rabbit TRITC. Images of TLR4, TLR2, and Förster Resonance energy transfer (FRET) were collected and quantified with confocal microscope. The percentage of positive ROIs and FRET efficiency are the Mean ± SEM of the data from six different experiments. ***p < 0.001 compared with the control.

Download figure to PowerPoint

To confirm the TLR4/TLR2 interaction, we used FRET imaging microscopy. The increased emission of donor fluorescence is considered a measure of the interaction between the two proteins. Figure 4b shows that ethanol treatment (50 mM, 30 min) increases the FRET emission, measured as the% of positive ROIs (Regions of interest), and also increases FRET efficiency when comparing controls versus ethanol treatment. These results suggest that ethanol promotes a physical association of TLR4 and TLR2 in microglial cells.

Role of TLR4 and TLR2 in the ethanol-induced activation of TLR4 in glial cells

To study the role of TLR4 and TLR2 in ethanol-induced microglia activation, cells of wild-type or TLR4−/− or TLR2−/− mice were stimulated with ethanol, and the inducible NO synthase (iNOS) expression and MAPKs activation were measured. Figure 5a shows that ethanol treatment (50 mM) lasting 30 min and 3 h increases the phosphorylation of extracellular-signal-regulated kinase, p-38 and c-Jun N-terminal kinase, and the levels of iNOS in the WT microglia. Conversely, no changes in the levels of these proteins were observed in microglia of TLR4−/− and TLR2−/− mice in response to the ethanol treatment. Moreover, ethanol (25 mM or 50 mM) increased the P-ERK and iNOS levels in WT astrocytes, but these changes were abolished in TLR2−/− astrocytes (Fig. 5b). These findings support the role of both TLR4 and TLR2 signaling in ethanol-induced inflammatory mediators in glial cells. Alternatively by activating microglial cells through the TLR4 response, ethanol triggers the release of proinflammatory cytokines, and these cytokines and/or TLR2 signaling amplifies the response. Indeed, previous works have shown that TLR-cytokine signaling converges upon these pathways by amplifying responses since TNF-α−/− mice do not show any pathogen activation of microglial TLR2 agonist responses (Syed et al. 2007).

image

Figure 5. Role of toll-like receptor (TLR)4 and TLR2 in ethanol-induced glial activation. (a) Levels of iNOS and MAPKs (PERK, Pp38, and PJNK) of the microglial cells from WT, TLR4−/− and TLR2−/− mice treated with or without ethanol (50 mM) for 10, 30 min, 1 or 3 h. Data are mean ± SEM (n = 6 independent experiments), expressed as relative densitometric units with respect the values at 0 min, within each mice genotype (b) Levels of iNOS and P-ERK of the astroglial cells from WT and TLR2−/− mice treated with ethanol (25 mM or 50 mM) for 10 min. Blots were stripped, and the total quantities of GAPDH, ERK, JNK, and p38 were also measured. A representative immunoblot of each protein is shown. *p < 0.05, **p < 0.01 compared with the control value.

Download figure to PowerPoint

Deficiency in the TLR4 and TLR2 function protects neurons from the effects of ethanol on microglia activation

We have previously described how ethanol acts through TLR4 activating microglial cells to cause neuronal death (Fernandez-Lizarbe et al. 2009). To assess the potential involvement of TLR2 in neurotoxicity associated with ethanol-induced microglia activation, the conditioned medium from the WT, TLR4−/− or TLR2−/− microglia treated with 50 mM ethanol for 24 h was added to cortical neurons on day 5 in vitro. Then, neurons were cultured with the microglia-conditioned medium for 1 day and neuronal apoptosis was measured by flow cytometry (Fig. 6a) and ROS production was evaluated (Fig. 6b).

image

Figure 6. Microglial-conditioned medium treated with ethanol induces neuronal apoptosis through toll-like receptor (TLR)4 and TLR2 signaling. Cortical neurons from mice were incubated for 1 day with the microglial-conditioned medium obtained from WT, TLR4−/−, and TLR2−/− microglial cells treated with 50 mM ethanol for 24 h (a) Data of the apoptotic rate obtained by flow cytometry represent Mean ± SEM, n = 9 independent experiments. Representative histogram of each condition for 1 day is shown. (b) ROS generation was expressed in relation to each control from n = 9 independent experiments. (c) TNF-α and IL-1β levels were measured in the control and 50 mM ethanol 24 h conditioned medium from the WT, TLR4−/−, and TLR2−/− microglial cells (n = 4–6 independent experiments). We compared the ethanol versus control values within each mice genotype. *p < 0.05, **p < 0.01 versus control and #p < 0.05, ##p < 0.01 versus WT control.

Download figure to PowerPoint

Neurons cultured for 1 day with the supernatant of the WT microglia treated with 50 mM ethanol for 24 h showed a significant increase in the apoptosis rate and ROS levels. No significant changes were observed in neuronal apoptosis when the supernatant of the ethanol-treated TLR4−/− cells were used. Interestingly, the natural neuronal apoptosis was significantly reduced with the conditioned medium derived from TLR2−/− microglia stimulated with ethanol when compared with the control-conditioned medium. We also used neurons incubated with a medium in which ethanol was added at the same concentrations (4.8 ± 1.1 mM) found in the conditional medium of ethanol-treated microglia as control neurons. In addition, this ethanol dose did not significantly affect the natural apoptosis observed in those neurons cultured without alcohol (data not shown). These findings support the notion that ethanol-induced neuronal death depends on activated TLR4 and TLR2 in microglial cells. In addition, no significant changes in ROS formation were noted when neurons were incubated with the TLR4−/− and TLR2−/− ethanol-treated microglia-conditioned medium. Our previous studies have also demonstrated that endogenous ligands are not mediated in the neuronal death induced by ethanol-conditioned medium (Fernandez-Lizarbe et al. 2009).

We also measured the levels of TNF-α and IL-1β in the microglia ethanol-treated conditioned medium. We show that, while the levels of TNF-α and IL-1β significantly increased in the conditioned medium of ethanol-treated microglia from WT mice, the levels of these cytokines were not high in the conditioned medium from ethanol-treated microglia of TLR4−/− and TLR2−/− mice (Fig. 6c). Indeed, the expression of IL-1β lowered in the activated microglia from TLR2−/− mice. No changes in the IL-10 levels were observed in either the conditioned medium from ethanol-treated WT or TLR4−/− or TLR2−/− (data not shown). We also noted that the basal levels of TNF-α and IL-1β were significantly lower in TLR4−/− and TLR2−/− mice than in WT mice, suggesting a deficient basal immune response.

Taken together, these findings suggest that ethanol can activate microglia through TLR4 and TLR2 and causes a release of cytokines and other inflammatory mediators, contributing to neuronal oxidative stress and apoptosis, which could participate in the neurodegenerative processes associated with ethanol consumption.

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Glial cells and TLRs are vital players in CNS immune responses. TLRs activation is a major inducer of neuroinflammation, contributing to both infectious and non-infectious CNS injury. Indeed, several studies have demonstrated the role of TLR2 and TLR4 in neurodegeneration (Lehnardt et al. 2008; Kawai and Akira 2010) and neuroinflammation (Babcock et al. 2006; Buchanan et al. 2010). We previously demonstrated that ethanol induces TLR4 activation in glial cells (Blanco et al. 2005; Fernandez-Lizarbe et al. 2009) causing neuroinflammation and brain damage (Alfonso-Loeches et al. 2010). This study extends previous findings by showing that ethanol, by interacting with LR membrane microdomains, can not only recruit TLR4 and TLR2 receptors within the LR-caveolae in microglial cells, but also can promote a physical association of these receptors by triggering TLRs signaling and leading to the induction of pro-inflammatory cytokine formation. These effects could contribute to ethanol-induced neuroinflammation and neuronal damage.

Different studies demonstrated that upon activation, TLRs are recruited with other receptors into membrane microdomains rich in sphingolipid and cholesterol, called lipid rafts (LR), and that these microdomains stabilize TLRs and form signaling platforms, which transduce signals leading to innate immune activation (Triantafilou et al. 2011). This combinational use of cell surface receptors determines the character of the immune response. Our previous studies demonstrate that while low/moderate ethanol concentrations (10-50 mM) induce clustering and the recruitment of TLR4 into LR-caveolae (Blanco et al. 2008; Fernandez-Lizarbe et al. 2008), high ethanol concentrations (100 mM) inhibit TLR4 signaling through the disruption of the LR and, consequently, receptor clustering (Fernandez-Lizarbe et al. 2008). Here, we further report that ethanol triggers the recruitment of TLR4 and TLR2 into LR-caveolae, leading to TLRs signaling, and acts as their natural ligands, such as LPS for TLR4 and LTA for TLR2. Significantly, we found that the recruitment of both receptors into the LR-caveolae promotes TLR4 and TLR2 interaction. This notion is supported by the results revealing that upon ethanol treatment, TLR4 and TLR2 co-localize within the rafts-caveolae and physically interact, as demonstrated by the co-immunoprecipitation experiments and the FRET studies. Disruption of LR with filipin abolishes ethanol-induced TLR4 and TLR2 recruitment.

Although the molecular mechanisms of the TLR4 and TLR2 interaction are not yet clear, several reports have shown that TLR2 can interact with other TLRs and that both TLR4 and TLR2 form clusters composed for several receptors (Triantafilou et al. 2006). For instance, bacterial lipopeptides through their acyl chains can directly crosslink TLR2 with TLR6 or TLR1 (Jin et al. 2007). TLR2 can also form complexes with various molecules, such as CD14 (Brightbill et al. 1999) or CD36 (Hoebe et al. 2005; Triantafilou et al. 2006). H pylori, P gingivalis, and LPS can induce receptor clusters comprising TLR2, TLR1, CD36, and CD11b/CD18 on vascular endothelial cells (Triantafilou et al. 2007), and CD14 has been implicated in facilitating TLR1/2-mediated responses to bacterial lipopeptides by enhancing the physical proximity of the ligand to TLR1/2 heterodimers (Manukyan et al. 2005; Nakata et al. 2006). An extensive hydrogen-bonding network, as well as hydrophobic interactions, between TLR2 and TLR4 could further stabilize the heterodimer, as demonstrated with the TLR2 and TLR1 receptors (Jin et al. 2007). Therefore, we propose that, by interacting with membrane lipid rafts, alcohol can induce TLR4/TLR2 recruitment within lipid rafts-caveolae, and can promote the formation of the TLR4/TLR2 heterodimer, causing the interaction of their toll-interleukin 1 receptor domains and initiating the cascade to endorse microglia activation.

Emerging evidence indicates the role of TLRs and other immune receptors in neural damage associated with various neurodegenerative diseases (Okun et al. 2009, 2011). Here, we demonstrate that ethanol induces microglia activation, leading to the release of pro-inflammatory cytokines, promotes ROS generation in neurons, which in turn causes oxidative stress and neuronal apoptosis, and that these effects are mediated by TLR4 and TLR2. These results extend previous findings demonstrating the potential role of TLR4 in ethanol-induced microglia activation and neuronal apoptosis (Fernandez-Lizarbe et al. 2009), and suggest that both receptors participate in ethanol-induced neuroinflammation and neural damage.

To support our hypothesis, ethanol abuse not only induces microglial markers and chemokine-MCP1 (He and Crews 2008) but also up-regulates the protein expression of the plasma membrane receptors TLR2 and TLR4 and the intracellular receptors TLR3 in post-mortem human alcoholic brain and in mice brain. Indeed, these effects have been correlated with lifetime alcohol consumption (Crews et al. 2013). These results suggest that chronic alcohol intake triggers brain neuroimmune activation through TLRs signaling (Crews et al. 2013). Alcohol consumption also increases the gene expression of TLR2 and TLR4 in the liver (Oliva et al. 2011), while it diminishes the ability of splenic macrophages to respond to TLR4 and TLR2 ligands (Goral and Kovacs 2005). Although the effects of ethanol on TLR4/TLR2 depend on the stimuli, ethanol concentration, and cell type or tissue analyzed, our results suggest that, in the brain, alcohol triggers neuroimmune activation by stimulating microglial cells and TLRs signaling response. Ethanol-induced neuroimmune signaling has also been linked with alcohol-induced drinking behavior, anxiety, and depression-like behaviors (Pascual et al. 2011; Crews et al. 2013).

The activation of TLR4 and TLR2 also contributes to the ischemic brain injury (Ziegler et al. 2007), experimental autoimmune encephalomyelitis (Zekki et al. 2002), or Alzheimer's disease (Liu et al. 2012) and the deficiency of these receptors reduces the expression of proinflammatory genes and decreased pain hypersensitivity after spinal nerve transection (Tanga et al. 2005; Kim et al. 2007). Elimination of TLR2 also reduces cell death induced by kainic acid administration (Hong et al. 2010), and attenuates leukocyte and microglial infiltration and neuronal death induced by focal ischemia (Ziegler et al. 2011). In agreement with these findings, our results demonstrate that both TLR4 and TLR2 participate in ethanol-induced microglia activation since ethanol not only activates several kinases (extracellular-signal-regulated kinase, p38, c-Jun N-terminal kinase) but also induces inflammatory cytokines (TNF-α, IL-1β) and mediators (iNOS) in WT microglia. These events were not observed in the ethanol-treated microglia from TLR4−/− and TLR2−/− mice. Therefore, by considering the role of both receptors in neuroinflammation, we propose that TLR4/TLR2 are important targets for ethanol-induced microglia activation, and that the elimination of one of the receptors might abolish the ethanol-induced recruitment and heterodimerization of TLR4/TLR2 by diminishing the action of TLRs signaling and neuroinflammation.

In summary, this study shows for the first time that ethanol, by triggering TLR4 and TLR2 recruitment into LR-caveolae, promotes TLR4 and TLR2 interactions and signaling in microglial cells, leading to the production of pro-inflammatory cytokines, which causes ROS generation and neuronal apoptosis. The results extend our previous findings (Fernandez-Lizarbe et al. 2009) and reveal the participation of TLR4 and TLR2 receptors in microglia activation, neuroinflammation, and neuronal apoptosis induced by ethanol.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank M. March and M. J. Morillo for their excellent technical assistance and Confocal Microscopy Service at the CIPF Center. We thank Dr. S. Akira, who provided us with the TLR4−/− and the TLR2−/− knockout mice. This study has been supported by grants from the Spanish Ministry of Science and Innovation (SAF-2009-07503; SAF2012-33747), the Spanish Ministry of Health: the Carlos III Health Institut (RTA-Network, RD12-0028-007), Plan Nacional sobre Drogas, and Generalitat Valenciana, PROMETEO/2009/072. The authors declare that there is no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References