Background
Microglia are the resident immune cells in the central nervous system (CNS). They serve as the neuron-pathological sensor under various conditions such as inflammation [
1,
2]. The sensing of microglia to pathological stimuli leads to activation of microglia, which then produce trophic factors that are important for neuronal recovery [
3]. However, uncontrolled activation of microglia triggers neurotoxicity by overproducing cytokines such as nitric oxide (NO), TNF-α, and IL-6 [
4‐
6]. NO is synthesized by a family of NO synthase (NOS) consisting of three isoforms: endothelial NOS (eNOS), neuronal NOS (nNOS), and inducible NOS (iNOS) [
7]. NO generated from eNOS and nNOS primarily participates in cardiovascular regulation, neuronal signaling transduction, and neuronal protection [
8‐
10]. NO produced from iNOS [
11], however, promotes the development of neurodegenerative disorders associated with inflammation such as Parkinson’s disease and multiple sclerosis [
12,
13]. Similarly, TNF-α and IL-6 also have dual functions. For example, TNF-α preconditioning was found to protect neurons from Aβ-mediated cell toxicity [
14]. IL-6 treatment ameliorates trimethyltin-induced injury in neurons [
15]. Both TNF-α and IL-6 were confirmed to enhance neurotoxicity [
5,
6]. Collectively, control of induction of cellular cytokines in microglia might be important for regulation of numerous physiological or pathological processes.
An important feature of pro-inflammatory factors is their absence in quiescent inflammatory cells and induction by inflammatory inducers. Lipopolysaccharide (LPS) is an extensively characterized inducer of pro-inflammatory factors [
16]. It stimulates gene transcription of pro-inflammatory factors through the classical inhibitor of κB kinase (IKK)-inhibitor of κB α (IκB-α)-nuclear factor κB (NF-κB) signaling pathway. LPS binds with the Toll-like receptors leading to IκB-α degradation through the ubiquitin-proteasome system [
17]. The removal of IκB-α liberates transcriptional factor NF-κB. The active NF-κB is then free for translocation to the nucleus, where it initiates gene transcription [
17]. Besides IκB-α-NF-κB signals, mitogen-activated protein kinase (MAPK) including ERK1/2, p38, and JNK is also involved in induction of pro-inflammatory factors [
18‐
20]. Interfering with the MAPK signals is beneficial to coping with inflammation.
Plenty of small molecules extracted from traditional Chinese herbal medicines have been reported to regulate induction of pro-inflammatory factors through the classical IκB-α-NF-κB pathway. For instance, oregonin was found to inhibit iNOS gene transcription by reducing the nuclear translocation of NF-κB in LPS-stimulated microglia [
21]; curcumin suppresses the expression of NF-κB-dependent genes in rats [
22]. A monomer of stilbene from a traditional Chinese herbal medicine polygonummultiflorum, 2,3,4’ ,5-tetrahydroxystilbene 2-O-β-D-glucoside (TSG), has also been found to attenuate inflammatory responses [
23]. Its anti-inflammatory function was supported by the following evidence: 1) TSG suppresses COX-2 expression in a carrageenin-induced rat paw edema model [
24]; 2) TSG reduces NO levels in serum and the aorta in atherosclerotic rats [
25]; and 3) TSG decreases iNOS expression and infarct volume in the ischemic brain [
26]. In the third case, the effect of TSG was explained by the suppression of NF-κB nuclear translocation in neurons but not in microglia [
26]. In the present study, in view of the great importance of microglia in CNS disorders associated with inflammation [
27,
28], we explored the effect and mechanism of TSG on LPS-mediated inflammatory response in microglia. We found that TSG reduces iNOS expression and NO, TNF-α, and IL-6 release in microglia in a way that is independent of MAPK-IκB-α-NF-κB activation but likely represses NF-κB binding activity.
Methods
Chemicals and reagents
DMEM/F12 was obtained from Gibco Invitrogen Corporation (Carlsbad, CA, USA). Heat-inactivated FBS was purchased from Hyclone (Logan, UT, USA). TSG was the product of the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). LPS, poly-L-lysine, and Hoechst 33258 were purchased from Sigma (Saint Louis, MO, USA). Antibodies against iNOS, IκB-α, p-NF-κB, NF-κB, Histone H2A, ERK1/2, phospho-ERK1/2, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were purchased from Cell Signaling Technology (Beverly, MA, USA). Protein A/G PLUS-Agarose and antibodies against caspase-3, p38, phospho-p38, JNK, and phospho-JNK were the products of Santa Cruz Biotechnology (Santa Cruz, CA, USA). Other related agents were purchased from commercial suppliers. All drugs were prepared as stock solutions, and stock solutions were stored at -20°C.
Cell preparation
BV-2 cells were grown in DMEM/F12 with 10% FBS. The use of mice was approved by the University Animal Ethics Committee of Nantong University (Permit Number: 2110836). Mouse primary cultured brain cells were prepared as described previously with some modifications [
29]. Briefly, newborn (day 0 to 1) C57/BL6 mice were decapitated, hippocampus were then removed and digested with 0.125% trypsin for 15 minutes at 37°C. Followed by trituration and centrifugation at 118
g for 6 minutes, cells were re-suspended and plated on poly-L-lysine (1 mg/mL)-coated culture flasks. For preparation of hippocampal neurons, the single-cell suspension was cultured in DMEM/F12 supplement with 2% B
27 and 1% penicillin-streptomycin (100 U/mL), and the medium was replaced every 3 days. For preparation of primary microglia, the individual cell suspensions were cultured in DMEM/F12 supplement with 10% FBS and 1% penicillin-streptomycin (100 U/mL). This medium was replaced every 3 days. After 12 days, mixed cells were shaken gently overnight (37°C, 18 h), and the supernatants were collected and plated on the new poly-L-lysine-coated culture flasks. All cells were maintained in a 37°C incubator containing 95% air and 5% CO
2.
Cell viability assay
Cell viability was measured using MTT Cell Proliferation and Cytotoxicity Assay Kit (Bi Yuntian Biological Technology Institution, Shanghai, China). Briefly, methylthiazolyldiphenyl-tetrazolium bromide (5 mg/mL) was dissolved in prepared MTT-dissolved solutions and kept at -20°C. After washing with PBS, the cells in plates were added 20 μL of MTT solutions and kept at 37°C for 4 h. The blue crystals were dissolved in formazan-dissolved solutions. The absorbance was read at 570 nm.
Western blot
To extract the total proteins, cells were lysed on ice for 30 minutes in lyses buffer (50 mM Tris–HCl, pH 7.4, 1 mM EDTA, 100 mM NaCl, 20 mM NaF, 3 mM Na3VO4, 1 mM PMSF with 1% (v/v) Nonidet P-40, and protease inhibitor cocktail). The lysates were centrifuged at 12,000 g for 16 minutes, and the supernatants were harvested. After denaturation, 30 μg of protein was separated on 10% SDS/PAGE gels and then transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA, USA). After blocking with 5% nonfat dried milk powder/Tris-buffered saline Tween-20 (TBST) for 1 h, membranes were probed with 1:500 primary antibodies against iNOS, caspase-3, IκB-α, p-NF-κB, NF-κB, ERK1/2, p-ERK1/2, JNK, p-JNK, p38, p-p38, and Histone H2A or 1:10,000 primary antibody against GAPDH overnight at 4°C. Primary antibodies were then removed by washing the membranes three times in TBST. Membranes were further incubated for 2 h at room temperature with IRDye 680-labeled secondary antibodies (1:3,000 to 1:5,000). Finally immunoblots were visualized by scanning using the Odyssey CLx western blot detection system. Isolated cytoplasmic and nuclear proteins were normalized to GAPDH and Histone-H2A respectively. The band density was quantified using Image J software.
Real-time PCR
At the end of each treatment, total RNA was isolated from BV-2 cells using the RNeasy mini kit according to the manufacturer’s instructions (Qiagen, GmbH, Hilden, Germany). First-strand cDNA was generated by reverse transcription of total RNA using the RT system (Promega, Madison, WI, USA). Real-time PCR reactions were conducted with FaststartSYBR Green Master Mix (Roche Molecular Biochemicals, Shanghai, China). Briefly, 2 μL of diluted cDNA, 0.5 μM primers, 2 mM MgCl
2, and 1 × FastStartSYBR Green Master mix were employed. The primers are quoted as follows [
30,
31]: iNOS 5’-CTC ACT GGGACAGCA CAG AA-3’ (forward), 5’-TGG TCA AAC TCT TGG GGT TC-3’ (reverse); TNF-α 5’-CTG TGA AGG GAA TGG GTG TT-3’ (forward), 5’-GGT CAC TGT CCC AGC ATC TT-3’ (reverse); IL-6 5’-TTC CAT CCA GTT GCC TTC TT-3’ (forward), 5’-CAG AAT TGC CAT TGC ACA AC-3’ (reverse); 18S rRNA 5’-GTA ACCCGTTGAACC CCA TT-3’ (forward), 5’-CCA TCCAATCGG TAG TAG CG-3’ (reverse). PCR products were detected by monitoring the fluorescence increase of double-stranded DNA-binding dye SYBR Green during amplification. The expression levels of target genes were normalized to the housekeeping gene (18S rRNA). The fold changes in the target gene expression between experimental groups were expressed as a ratio. Relative gene expression was calculated by the comparative cycle threshold (Ct) method. Melt-curve analysis and agarose gel electrophoresis were used to examine the authenticity of the PCR products.
Detection of NO
Total nitrite levels in collected supernatants were measured with a Griess reagent kit (Invitrogen). The reaction consisted of 20 μL of Griess Reagent, 150 μL of supernatant, and 130 μL of de-ionized water. After incubation of the mixture for 30 minutes at room temperature, nitrite levels were measured at 548 nm using an M2 spectrophotometric microplate reader (Molecular Devices).
Measuring of lactate dehydrogenase (LDH)
For LDH testing, hippocampal neurons were seeded into 24-well plates. BV-2 cells were treated with LPS (2 μg/mL) for 24 h, and then the supernatants were collected to incubate hippocampal neurons. After additional 48 h, the supernatants from hippocampal neurons were collected to be prepared for further experiments. LDH levels were determined using LDH Cytotoxicity Assay Kit according to the supplier’s recommendation (Bi Yuntian Biological Technology Institution, Shanghai, China).
Hoechst 33258 staining
For Hoechst 33258 staining, hippocampal neurons in different groups were fixed with 2% paraformaldehyde in 0.01 M PBS (pH 7.4) for 20 minutes, and then rinsed three times with PBS for 10 minutes each. Cells were treated with Hoechst 33258 staining solution for 15 minutes at 4°C. The images of Hoechst 33258 staining were viewed with a Nikon Eclipse 800 microscope. Cells with condensed bright nuclei were regarded as apoptotic cells. The apoptosis rate was calculated by the ratio between the numbers of cells with condensed bright nuclei and total cell numbers. The numbers of apoptotic or total cells were counted from the resulting four phases for each point with the digital camera and microscope, and then averaged for each experimental condition. The data presented were generated from three separate assays.
NF-κB binding assays
The nuclei were extracted from BV-2 cells or primary microglia by first incubating them in hypotonic buffer (10 mM Tris–HCl, pH 7.5, 10 mM NaCl, 1.5 mM MgCl2) at 4°C for 15 minutes. After homogenization, cell homogenates were spun at 3,000 g for 5 minutes. The supernatants were collected for western blot analysis. The pellets were recovered, extensively washed, and re-suspended in the nuclear extraction buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 0.25% sodium deoxycholate, 10% glycerol, 50 mM NaF, 1 mM Na3VO4, 5 mM sodium pyrophosphate, protease inhibitors). The NF-κB binding activity of nuclear extracts was measured with the TransFactor NF-κB colorimetric kit (Clontech, Mountain View, CA, USA) according to the manufacturer’s instruction.
Chromatin immunoprecipitation (ChIP)
The ChIP experiment was performed as described previously with some modifications [
32]. Briefly, BV-2 cells were treated with LPS (2 μg/mL) for 1 h in the presence and absence of TSG. 1% of formaldehyde was added to the culture medium, and after incubation on the rocker for 10 minutes at room temperature, cells were rinsed twice with PBS and lysed for 15 minutes at 4°C. After sonication, the lysate was used as DNA input control. The remaining lysates were diluted 10-fold with ChIP dilution buffer followed by incubation with NF-κB p65 antibody overnight at 4°C. Immunoprecipitated complexes were collected using protein A/G Plus-agarose beads. The precipitates were extensively washed and then incubated in the elution buffer containing 1% SDS and 0.1 M NaHCO
3 at room temperature for 20 minutes. Cross-linking of protein-DNA complexes was reversed at 65°C for 4 h. DNA was extracted with the QiagenPCR purification kit. For ChIP assays, we used the following primers (forward) 5’-CAAGCCAGGGTATGTGGTTT-3’ and (reverse) 5’-GCAGCAGCCATCAGGTATTT-3’ to get a fragment at 290-bp. The resulting product was separated by 2% agarose gel electrophoresis.
Statistical analysis
Data are expressed as means ± standard error (SE). One-way analysis of variance (ANOVA) followed by the post hoc test was used for the statistical analysis, employing SPSS 11.0 software. Differences were considered significant at P < 0.05.
Discussion
The endotoxin- or pathogen-mediated induction of pro-inflammatory factors in microglia are implicated in pathophysiological processes of neurotoxicity [
33‐
35]. Numerous anti-oxidative molecules have been shown to protect neurons from cell toxicity in inflammatory disorders by attenuating the production of pro-inflammatory factors such as NO, TNF-α, and IL-6 in microglia [
21,
22,
36]. TSG, an active component of the rhizome extract from
Polygonummultiflorum, exhibits its function through anti-inflammation [
24], anti-apoptosis [
26], and anti-oxidation [
37]. In this study, we found that TSG impairs LPS-mediated inflammatory response in microglia. This effect was exemplified by the decrease in the production of pro-inflammatory factors as well as the DNA binding activity of NF-κB in LPS/TSG-stimulated microglia. This finding underscores the importance of TSG in regulation of inflammation, and extends the role of TSG beyond a cardiovascular protective molecule to a modulator of microglia activation. Recently, interesting work reported that TSG prevents the overexpression of α-synuclein in APPV717I transgenic mice with Alzheimer’s disease (AD), strongly showing a potential role of TSG in prevention or treatment of AD [
38]. Together with the fact that microglia are believed to mediate the development of AD [
39,
40] and that the damage of neurons in neurodegenerative disorders is usually secondary to microglia activation [
41,
42], we believe that our data may provide evidence to clarify how TSG exerts its protective effect in AD.
In vivo experiments designed to investigate the role and mechanism of TSG in different neurodegenerative disorders in the CNS are in progress.
The functional consequence of microglia activation depends on the induction and release of pro-inflammatory factors. The appropriate amount of pro-inflammatory factor is indispensable for various physiological processes, such as neuronal protection [
14,
15] and synaptic plasticity [
31]. However, overwhelmingly generated pro-inflammatory factors can act as neurotoxins and cause neuronal injury [
5,
6,
43,
44]. Strategies to inhibit the excessive production of pro-inflammatory factors are crucial to attenuate the neurotoxicity induced by inflammatory events. In the present study, we showed that TSG reduces the content of NO, TNF-α, and IL-6, and this reduction protects the hippocampal neurons from microglia-conditioned medium-induced cell injury. This finding about the role of TSG in induction of pro-inflammatory factors in microglia and protection of hippocampal neurons from inflammatory stimulation provides a new insight into the pharmacological role of TSG in inflammatory disorders.
Blockade of gene transcription in stimulated inflammatory cells is often due to one or multiple interruptions in the signaling transduction from the stimuli to the corresponding transcriptional cytokines. In LPS signals, the MAPKs-IKK-IκB-α-NF-κB pathway is downstream of LPS signaling transduction [
15,
34]. IκB-α is phosphorylated by IKK, which then leads to the degradation of IκB-α and the translocation of NF-κB from the cytoplasm to nucleus. However, we found that although TSG inhibited pro-inflammatory factor gene-transcription in microglia, it failed to affect LPS-induced IκB-α degradation and NF-κB phosphorylation or nuclear translocation. Neither did we observe any significant effect on the LPS-induced increase in ERK1/2, JNK, or the p38 phosphorylation levels. It indicates that TSG might affect the inflammatory response in microglia by a mechanism downstream of the nuclear translocation of NF-κB. In fact, our results showed that the increase in DNA binding activity of NF-κB in LPS-stimulated microglia was remarkably suppressed by TSG. Moreover, the reduction in active NF-κB from binding to the iNOS promoter in the ChIP assay further confirmed the function of TSG. In general, these data provide a plausible explanation why the gene transcription of pro-inflammatory factors is inhibited, despite the fact that activation of MAPK-IκB-α-NF-κB signals is not perturbed in TSG/LPS co-treated cells. However, as we did not check the influence of TSG in the binding of NF-κB to TNF-α and IL-6 promoters, to some extent these conclusions limit our observations for the role of TSG in inflammatory response in microglia. Some previous studies associated with the interaction between NF-κB and TNF-α and IL-6 promoters might give us evidence to offset the limitation. For example, a 120-bp TNF-α promoter was found to possess a binding site for NF-κB [
45], and the promoter region of the IL-6 gene was confirmed to have a putative NF-κB binding site [
46].
It is worth noting that our above observations in microglia are inconsistent with the
in vitro oxygen glucose deprivation/reperfusion (OGD/R)-stimulated neurons where the TSG-mediated reduction in iNOS expression and brain infarct volume is mediated by the inhibition of nuclear translocation of NF-κB [
26]. However, because there are large differences between
in vitro and
in vivo micro-environments and the damage of neurons in the
in vivo conditions usually occurs following excessive activation of microglia, we tend to suppose that the TSG-mediated reduction in infarct volume after brain ischemia might be mediated by the attenuation of inflammatory response in microglia with a mechanism that is different from that in neurons. That is, TSG may protect the brain tissues from ischemia by impairing the DNA binding activity of NF-κB. However, how this mechanism works in microglia is not clear. We notice that though resveratrol, a polyphenolicphytoalexin found in grapes, fruits, and root extracts of the weed
Polygonumcuspidatum[
47], has similar characteristics in structure to TSG [
48,
49], it exhibits distinct mechanisms in regulation of inflammatory response in microglia. For example, resveratrol was reported to decrease the production of pro-inflammatory factors primarily by impairing the phosphorylation and nuclear translocation of NF-κB [
49,
50]. This difference, combined with the highly similar structure of TSG and resveratrol, might help us to further clarify the exact mechanism of TSG in inflammatory response.
In summary, this study identifies a role for TSG in the induction of pro-inflammatory factors in microglia by a mechanism that is independent of MAPK activation, IκB-α degradation, and NF-κB phosphorylation/nuclear translocation, but probably relies on the repression of NF-κB binding activity.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CH participated in the design of this study. CH and YW performed the cell culture, western blot, statistical analysis in the whole research. YW carried out the real time-PCR assay. JW carried out the NO assay. WY participated in the design of this study and carried out the LDH and MTT assay. CH carried out Hoechst 33258 staining. XC and WY participated in the design of this study. YW carried out the NF-κB binding and ChIP assays. WZ participated in the design of this study and proofread the whole manuscript. All authors read and approved the final manuscript.