Cellular Signalling 27 (2015) 61–68

Contents lists available at ScienceDirect

Cellular Signalling

j ourna l homepage: www.e lsev ie r .com/ locate /ce l l s ig

miR-206 modulates lipopolysaccharide-mediated inflammatorycytokine production in human astrocytes

Xiaodong Duan a,b, Ali Zohaib a,b, Yunchun Li a,b, Bibo Zhu a,b, Jing Ye a,b, Shengfeng Wan a,b, Qiuping Xu a,b,Yunfeng Song a,b, Huanchun Chen a,b, Shengbo Cao a,b,⁎a State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan, Hubei 430070, PR Chinab Laboratory of Animal Virology, College of Veterinary Medicine, Huazhong Agricultural University, Wuhan, Hubei 430070, PR China

Abbreviations: LPS, lipopolysaccharide; CCL5, chemoklength; SOCS-1, suppressor of cytokine signalling-1; MKkinasephosphatase1;MRE,miRNArecognitionelements;Bregulated MAP kinase; MEK, MAP kinse-ERK kinase; Umutation; GAPDH, glyceraldehyde-3-phosphatedehydroNF-κB, nuclear factor-kappaB; NR4A2, nuclear receptor suJNK, c-JunN-terminalkinase.⁎ Corresponding author at: State Key Laboratory

Huazhong Agricultural University, Wuhan, Hubei 430087282608.

E-mail address: sbcao@mail.hzau.edu.cn (S. Cao).

http://dx.doi.org/10.1016/j.cellsig.2014.10.0060898-6568/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 10 October 2014Accepted 16 October 2014Available online 23 October 2014

Keywords:LPSmiR-206InflammationNR4A2Astrocyte

Astrocyte-derived inflammation is a common component of acute or chronic injury in the central nervous system.MicroRNAs (miRNAs) are small non-coding RNAs that play important regulatory roles in the inflammatory re-sponse. In this study, we found thatmiR-206 is inducedupon stimulationwith lipopolysaccharide. Overexpressionof miR-206 in astrocytes led to increased expression of inflammatory cytokines (interleukin-6, interleukin-1β,CCL5) upon exposure to lipopolysaccharide, whereas knockdown of miR-206 had completely opposite effects.We used a combination of bioinformatics and experimental techniques to demonstrate that NR4A2,which belongsto the nuclear receptor (NR) 4 family of orphan nuclear receptors, is a direct target of miR-206. Overexpression ofmiR-206 mimics decreased the activity of a luciferase reporter containing the NR4A2 3′-untranslated region andled to decreased NR4A2 mRNA and protein levels. In contrast, ectopic expression of an miR-206 inhibitor led toelevated NR4A2 expression. We also found that miR-206 modulated the lipopolysaccharide-induced proinflam-matory response by targeting NR4A2 and activating nuclear factor-kappa B activity. Finally, we demonstratedthat the transcription factor AP-1 plays a critical role in lipopolysaccharide-induced expression of miR-206and that the extracellular signal-regulated kinase signaling pathway contributes to the regulation of miR-206level in astrocytes. These data demonstrate that miR-206 positively regulates the lipopolysaccharide-inducedinflammatory response in human astrocytes.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Inflammation of the central nervous system is a fundamental pro-cess involved in the progression of many neurological diseases [1,2].Increasing evidence suggests that astrocyte-derived inflammatoryresponses play a critical role in neurological diseases [3–5]. The complexprocess of neuroinflammation is initiated by innate immune systemreceptors, which recognize a diverse range of pathogens or damage sig-nals and tightly regulate protection and repair mechanisms.

The lipopolysaccharide (LPS)-induced inflammatory response in thebrain is mediated by the toll-like receptor (TLR)-4 signaling pathway.

ine (C-Cmotif) ligand 5; FL, fullP-1, mitogen-activated proteinS,bindingsite;ERK,extracellularTR, untranslated region; MUT,genase; TLR, toll-like receptor;bfamily 4, group A, member 2;

of Agricultural Microbiology,70, PR China. Tel./fax: +86 27

Activation of TLR4 results in activation of nuclear factor-kappa B(NF-κB) and mitogen-activated protein kinase signaling pathways andleads to the production of inflammatory cytokines, chemokines, and in-terferons, which directly contribute to neuronal damage [6,22]. Hence,proper regulation of inflammation will be an effective method forcontrolling/reducing brain damage.

MicroRNAs (miRNAs) are small, non-coding RNAs that play crucialroles in post-transcriptional gene regulation by targeting the 3′-untranslated region (3′-UTR) of mRNAs, resulting in translationalrepression or degradation of mRNA [7]. A critical role for miRNAs inregulating both innate and adaptive immune responses has also beenreported. Accumulating evidence also suggests a decisive role formiRNAs in TLR4 signaling. For example, miR-155 and miR-221 regulatecytokine production in human dendritic cells by targeting SOCS-1 fol-lowing LPS stimulation of these cells [8]. Another miRNA, miR-101,also positively regulates inflammatory cytokineproduction by negative-ly regulating MKP-1 expression in LPS-stimulated macrophages [9]. Onthe other hand, miR-147, miR-98, and miR-34a act as negative regula-tors of inflammation in LPS-induced macrophages [10–12]. Recently,miR-146a, miR-200b, and miR-200c were reported to regulate proin-flammatory cytokine production via TLR4 signaling [13,14]. miR-146and miR-181 also regulate neuroinflammation mediated by astrocytes

62 X. Duan et al. / Cellular Signalling 27 (2015) 61–68

[15,16]. Thus, a number of miRNAs regulate LPS-induced immune re-sponses during neuroinflammation. However, the patterns of miRNA-mediated regulation of neuroinflammation remain to be fully elucidated.

Here, we found that the expression of miR-206 was upregulated inLPS-treated U251 cells, a human astrocytoma cell line, and that miR-206 upregulated the production of proinflammatory cytokines follow-ing LPS stimulation in these cells. Our studies thus identify a previouslyunrecognized role for miR-206 as a positive regulator of neuroinflam-mation in LPS-treated astrocytes.

2. Materials and methods

2.1. Cell culture and treatment

U251 cells (human astrocytoma cell line) and H293T cells werecultured and maintained in Dulbecco's Modified Eagle's Medium sup-plemented with 10% FBS, 100 U/ml penicillin, and 100 mg/ml strepto-mycin sulfate at 37 °C in 5% CO2. U251 cells were plated in 12-wellplates (2 × 105 cells/well), grown to 80% confluency, and transfectedwith plasmids or RNAs using Lipofectamine 2000 (Invitrogen). After24 h, the cell medium was removed, and cells were treated with orwithout LPS for another 24 h. When needed, cells were treated withvarious pharmacological inhibitors together with LPS as recommendedby the different suppliers.

2.2. Reagents

Pharmacological inhibitors of MEK (PD98059), ERK (U0126), andJNK (SP600125) were obtained from Calbiochem-Merck and dissolvedin dimethylsulfoxide (Sigma). Antibodies against NF-κB were obtainedfrom Cell Signaling Technology. Antibodies against NR4A2, GAPDH,and lamin Awere purchased from Proteintech (ABclonal TechnologyTM,Wuhan, China). Horseradish peroxidase–conjugated anti-mouse/rabbitsecondary antibodies (Boster, China) were used. Hsa-miR-206 dsRNAmimic, ssRNA inhibitor, and control oligonucleotides were synthesizedby Genepharma (Shanghai, China).

2.3. Constructs and plasmids

The psiCheck-2 dual-luciferase reporter vector (Promega) harboringthe 3′-UTR of NR4A2 was inserted into the Xho I and Not I restrictionsites 3′ to the end of the Renilla luciferase gene and used to check theeffect of miR-206 on Renilla luciferase activity. The 3′-UTR of NR4A2and its coding regions were amplified from cDNA derived from U251cells and cloned into psiCheck-2 and pCDNA4 (Life Technologies). Theluciferase reporter plasmids pre-2904, pre-2349, pre-1787, pre-1149,and pre-544 were constructed with PCR amplification using genomicDNA from U251 cells as template and subsequently cloned into PGL3-basic (Promega). To construct expression vectors for the transcriptionfactors c-Jun and c-fos, the coding regions were amplified from cDNAfrom U251 cells and cloned into pCDNA4.0. The mutated 3′-UTR ofNR4A2 and miR-206 promoter constructs containing deletions or site-specific mutations at transcription factor binding sites were generatedwith overlap-extension PCR. Primers are listed in Table S1. All con-structs were verified with sequencing.

2.4. Luciferase reporter assays

H239T cells were co-transfected with 50 ng psiCheck-2 with inser-tion of the wild-type or mutated 3′-UTR of NR4A2, along with miR-206 mimics, inhibitors, or controls (final concentration, 50 nM). After24 h, the cells were collected for use in the Dual-Luciferase ReporterSystem (Promega). U251 cells were co-transfected with 100 ng full-length, truncated, or mutant promoter firefly luciferase reporter con-structs and 10 ng Renilla luciferase vector (pRL-TK). Luciferase activitieswere determined with the Dual-Luciferase Reporter Assay System and

were expressed as relative luciferase activity by normalizing fireflyluciferase activity to Renilla luciferase activity.

2.5. RNA extraction and quantitative real-time PCR

Total RNA was extracted from treated cells with TRIzol (Invitrogen),and 1 μg RNA was used to synthesize cDNA using a first-strand cDNAsynthesis kit (TOYOBO). Quantitative real-time PCR analysis was per-formed using a 7500 Real-time PCR System (Applied Biosystems) andSYBR Green PCR master mix (TOYOBO). Data were normalized to thelevel of β-actin expression in each sample. To quantify mature miRNAexpression, a commercial Bulge-Loop™ miRNA quantitative reversetranscription (RT)-PCR detection method was used. Briefly, 1 μg totalRNA was used as the template and reverse transcribed using an miR-206–specific RT primer. The resulting cDNA was used for quantitativereal-time PCR with a universal reverse primer and a specific forwardprimer. Amplification was performed for 2 min at 50 °C and 10 min at95 °C, followed by 40 cycles of 95 °C for 10 s, 60 °C for 10 s, and 72 °Cfor 30 s. The relative expression of miRNAs was normalized to that ofinternal control U6 snRNA within each sample using the 2−ΔΔCt

method. Expression was then standardized to the miRNA levels inmock-infected or control miRNA-treated cells. Primers used are listedin Table S1. All constructs were verified with sequencing.

2.6. Western blotting

Whole-cell lysates were generated in radioimmunoprecipitationassay (RIPA) buffer containing proteinase inhibitors (Roche). Cytosolicand nuclear extractswere prepared usingNE-PEPNuclear and Cytoplas-mic Extraction Reagents (ThemmoScienntific). Protein concentrationswere measured with the BCA Protein Assay kit (Thermo Scientific).Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was per-formed, followed by protein transfer to polyvinylidene fluoride mem-branes using Mini Trans-Blot Cell (Bio-Rad). Blots were probed withrelevant antibodies, and proteinswere detected using enhanced chemi-luminescence reagents (Thermo Scientific).

2.7. ELISA

Culture supernatants were collected from treated U251 cells atvarious times and stored at −80 °C. Levels of tumor necrosis factor(TNF)-α, interleukin (IL)-1β, IL-6, and CCL5 were measured withELISA kits (eBioscience).

2.8. Statistical analysis

All experiments were repeated at least three times with similarresults. Analyses were conducted using Prism (5.0) (GraphPad Soft-ware, San Diego, CA). Results are expressed as the mean ± SD. Datawere compared with either a two-way ANOVA, with subsequent ttests using Bonferroni post-tests for multiple comparisons, or with theStudent's t test. For all tests, P b 0.05 was considered significant.

3. Results

3.1. miR-206 expression is upregulated in LPS-treated astrocytes

LPS induces several miRNAs that regulate LPS-mediated inflamma-tion [10,11,17,18]. miR-206 is highly upregulated in ischemic brainspecimens [19–21]. To test whether miR-206 is also involved in LPS-mediated immune responses in astrocytes, the level ofmiR-206was de-termined in LPS-stimulated U251 cells (human astrocytoma cell line).miR-206 was expressed constitutively in these cells, and upon stimula-tion with LPS the level increased significantly in a time-dependent(Fig. 1A) and dose-dependent manner (Fig. 1B), suggesting possible

Fig. 1.miR-206 is upregulated in LPS-treated astrocytes. (A) U251 cells were incubated with 1 μg/ml LPS for 0, 2, 4, 6, 12, or 24 h. (B) U251 cells were also incubated with LPS at 0 ng/ml,100 ng/ml, 500 ng/ml, or 1000 ng/ml for 24 h. Total RNA was extracted, and the miR-206 level was detected with quantitative RT-PCR. Results are representative of three independentexperiments performed in triplicate. (*P b 0.1,**P b 0.01 and ***P b 0.001).

63X. Duan et al. / Cellular Signalling 27 (2015) 61–68

involvement of miR-206 in LPS-induced immune responses in brainastrocytes.

3.2. miR-206 modulates the production of inflammatory cytokines

LPS-mediated activation of NF-κB signaling pathways leads to theproduction of proinflammatory cytokines such as TNF-α, IL-1β, IL-6,and CCL5 [22]. To investigate the involvement of miR-206 in cytokineproduction, miR-206 was overexpressed in U251 cells for 12 h, andthen the cells were treatedwith LPS for an additional 24 h. ELISA resultsrevealed that the production of IL-1β, IL-6, and CCL5 increased signifi-cantly, whereas no obvious change was observed in TNF-α expression(Fig. 2A–D). To assess the involvement of miR-206 in modulating cyto-kine production, U251 cells were transfected with miR-206–specificinhibitors and stimulated with LPS for 24 h. In agreementwith previousresults, inhibition of miR-206 expression resulted in significant down-regulation of proinflammatory cytokine production (Fig. 3A–D). Wenext determined the impact of miR-206 on NF-κB activation in LPS-treated astrocytes. Nuclear translocation of NF-κB was detected withimmunoblotting. Transfection with miR-206 mimics increased thetranslocation of NF-κB from the cytoplasm to the nucleus (Fig. 2E). In

Fig. 2. miR-206 modulates LPS-induced inflammatory cytokine production. (A–D) U251 cellincubated with LPS (1 μg/ml) for 24 h. Levels of IL-1β, IL-6, TNF-α, and CCL5 were detected(final concentration, 50 nM) for 24h. One group of cellswas treatedwith LPS (1 μg/ml) for anotblotting to analyze the activation of NF-κB. Protein levels were quantified with immunoblot scadard deviation of results from three independent assays (ns = not significant; *P b 0.05; **P b

contrast, treatment of cells with miR-206 inhibitors significantlyinhibited the nuclear translocation of NF-κB in LPS-stimulated U251cells (Fig. 3E). Taken together, these findings demonstrated thatmiR-206 appears to regulate proinflammatory cytokine production byenhancing the activation of NF-κB signaling pathways in LPS-treatedastrocytes.

3.3. NR4A2 is a molecular target of miR-206

To assess specific miR-206 targets with potential relevance to theregulation of inflammatory cytokine production, several miRNA publictarget-prediction algorithms were used (Pictar, TargetScan, miRanda,PIT, and RNAhybird) [23–26]. NR4A2 is an important negative regulatorof LPS-induced inflammation and protects dopaminergic neurons byreducing inflammatory gene expression in astrocytes and microglia.NR4A2 displayed potential seed matches for miR-206 and was furtherinvestigated [27]. In addition, as shown in Fig. 4A, the miR-206 bindingsite was also broadly conserved among different species. To elucidatewhether NR4A2 is a target ofmiR-206,we generated a luciferase report-er carrying a putative binding site along with a mutant constructcontaining an miR-206 target sequence with base pair mutations in

s were transfected with miR-206 mimics or the negative control (nc) for 24 h and thenwith ELISA. (E) U251 cells were transfected with miR-206 mimics or the mimic controlher 24 h. Nuclear extractswere separated fromwhole-cell lysates and subjected towesternnning and normalized to the amount of LaminA expression. Error bars represent the stan-0.01).

Fig. 3. Knockdown of miR-206 attenuates LPS-triggered production of inflammatory cytokines. (A–D) U251 cells were transfected with inhibitor or inhibitor control for 24 h and thenincubated with LPS (1 μg/ml) for 24 h. Levels of IL-1β, IL-6, TNF-α, and CCL5 were detected with ELISA. (E) U251 cells were transfected with the miR-206 inhibitor or control inhibitor(final concentration, 50 nM) for 24h, and one group of cellswas treatedwith LPS (1 μg) for another 24 h. Nuclear extractswere separated fromwhole-cell lysates and subjected towesternblotting to assess the activation of NF-κB. Protein levelswere quantifiedwith immunoblot scanning and normalized to the amount of LaminA expression. Error bars represent the standarddeviation of results from three independent assays (*P b 0.05; **P b 0.01).

Fig. 4.NR4A2 is a direct target ofmiR-206 in astrocytes. (A) Alignment of theNR4A23′-UTR showed that theMRE (blue)was conserved among different species. Bioinformatics analyses ofmiR-206 potential targets and sequence alignment ofmiR-206 and its binding sites in the 3′-UTRs of NR4A2 as predictedwith software and a schematic representation ofmutant reportersofNR4A2 3′-UTR are shown. The blue indicates the regionwith thepointmutations (red). (B) U251 cellswere co-transfectedwith psicheck-NR4A2 3′-UTR (100 ng) or psicheck-NR4A23′-UTR mutant (100 ng) together with miR-206 mimics or the mimic control, miR-206 inhibitor, negative control (nc) and control inhibitor (final concentration, 50 nM) as indicated. After24 h, firefly luciferase activity was determined and normalized to Renilla luciferase activity. Similar results were obtained in three independent experiments. (C, D) U251 cells weretransfected with miR-206 mimics or the mimic control, miR-206 inhibitor, or control inhibitor (final concentration, 50 nM) for 24 h, and then NR4A2 mRNA and protein levels weremeasured with quantitative RT-PCR and immunoblotting, respectively. (*P b 0.05; **P b 0.01).

64 X. Duan et al. / Cellular Signalling 27 (2015) 61–68

65X. Duan et al. / Cellular Signalling 27 (2015) 61–68

the seed region (Fig. 4A). A marked reduction in luciferase activitywas observed in cells transfected with miR-206 mimics and NR4A2wild-type 3′-UTR, whereas significantly increased luciferase levelswere observed following application of miR-206 inhibitors. Moreover,mutation of the miR-206 target sequence completely abrogated thenegative effects of miR-206 on the NR4A2 3′-UTR reporter construct(Fig. 4B). These data indicated that miR-206 directly targets the 3′-UTR of NR4A2. To confirm that miR-206 targets NR4A2, the expressionof endogenous NR4A2 was determined in U251 cells treated withmiR-206mimics or inhibitors. As shown in Fig. 4C–D, ectopic expressionof miR-206 resulted in a significant reduction in NR4A2 mRNA andprotein levels, whereas application of miR-206 inhibitors restored theexpression of NR4A2.

3.4. NR4A2 abrogates the miR-206–modulated LPS-induced inflammatoryresponse

NR4A2plays an important anti-inflammatory role in activated astro-cytes and microglial cells [28]. To further investigate the link betweenmiR-206–mediated regulation of NR4A2 and inflammation, U251 cellswere transiently transfected with miR-206 mimics and an NR4A2expression plasmid. ELISA was used to quantify IL-1β, IL-6, TNF-α, andCCL5. As expected, overexpression of the miR-206 mimic togetherwith NR4A2 abrogated the miR-206–modulated LPS-induced proin-flammatory response (Fig. 5A–D). These results indicated that NR4A2is involved in the process bywhichmiR-206 regulates LPS-induced pro-duction of proinflammatory cytokines.

3.5. The transcription factor AP-1 is essential for transcription of miR-206

After confirming thatmiR-206 indeed regulates LPS-induced inflam-mation in astrocytes, we investigated the mechanism underlying miR-206 regulation. Genomic information and chromosomal location of

Fig. 5.NR4A2 abrogates themiR-206–modulated LPS-induced proinflammatory response. (A–DNR4A2 (1 μg) for 24 h and then treated with 1 μg/ml LPS for another 24 h. The concentrationrepresentative of three independent experiments performed in triplicate. (*P b 0.05; **P b 0.01

miR-206 were obtained from miRBase [24]. To evaluate the transcrip-tional regulation of miR-206, its potential promoter was analyzed.A schematic representation of the miR-206 promoter is shown inFig. 6A. To define the boundaries of the minimal promoter region, anapproximately 3-kb DNA fragment containing non-coding sequencesupstream of the first nucleotide site of pre–miR-206 was selected forpromotermapping. For the deletionmutation assay, luciferase reporterswith different lengths of the miR-206 promoter were transfected intoU251 cells to determine the basal and LPS-induced promoter activity.The highest luciferase activity was produced by reporter pre-1787(Fig. 6A), suggesting that pre-1787 contained the intact promoterregion. Transfection with pre-2904, pre-2349, pre-1149, or pre-544 re-porter plasmids resulted in relatively lower basal promoter activity.These results indicated that the region from pre-1787 to pre-1149 is re-quired for promoter activity. This regionwas chosen for further analysis.And the region from pre-2904 to pre-1787 or pre-1149 to pre-544 maycontain a negative-acting or incomplete regulatory element(s). Bioin-formatic analyses of the selected promoter region using JASPAR andCONSITE software predicted AP-1 (contains two subunits: c-jun andc-fos) as a crucial potential transcription factor. The four binding sitesof AP-1 on the selected promoter region are shown in Fig. 6B. To deter-mine if AP-1 contributes to the expression ofmiR-206 inU251 cells afterLPS stimulation, a series of truncated promoter constructs includingfull-length (i.e., pre-1787), deletion Δ1 (BS1), Δ2 (BS2 and BS3), andΔ3 (BS4) were generated (Fig. 6C). Compared with the full-lengthpromoter construct, deletion of each AP-1 binding site decreased theluciferase activity (Fig. 6D).

To confirm these findings, a panel of five different constructs (withspecific point mutations) includingmutation 1–4 (Mut1–4) was gener-ated in the context of the full-length promoter to assess the functionalrole of the specific predicted elements (Fig. 6E). Transient transfectionassays showed that mutations of AP-1 sites 1, 2, and 4 dramaticallydecreased luciferase activity compared with the full-length promoter,

)U251 cellswere transfectedwithmiR-206 or negative control (nc) alongwith pCDNA4.0-of cytokines including IL-1β, IL-6, TNF-α, and CCL5 was measured with ELISA. Results are).

66 X. Duan et al. / Cellular Signalling 27 (2015) 61–68

Fig. 7. The ERK pathway regulatesmiR-206 transcription. (A) U251 cells were transfectedwith pre-1787 (100 ng) and pRL-TK plasmid (10 ng). After 24 h, cells were treatedwith U0126,PD98059, or SP600125 for another 24 h, and the promoter activities weremeasured with luciferase analysis. (B) U251 cells were pretreated with U0126, PD98059, or SP600125 for 24 h.miR-206 expression was analyzedwith quantitative RT-PCR. Results are representative of three independent experiments performed in triplicate. ns, not significant; *P b 0.05; **P b 0.01.

67X. Duan et al. / Cellular Signalling 27 (2015) 61–68

and mutations of AP-1 site 3 did not further decrease luciferase activity(Fig. 6F). All these data indicated that sites 1, 2, and 4 for AP-1 binding tothe miR-206 promoter appeared to be potent sites for the contributionof AP-1 to the expression of miR-206. In addition, AP-1 (c-jun/c-fos) ex-pression was significantly increased in LPS-treated astrocytes, andforced expression of AP-1 enhanced the transcriptional activity of themiR-206 promoter as well as the expression of endogenous miR-206(Fig. 6G–I).Taken together, these results indicated that LPS-induced up-regulation of AP-1 contributes to transcriptional regulation of miR-206in astrocytes.

3.6. Involvement of ERK signaling pathways in miR-206 transcription

To examine which pathway regulates the transcription of miR-206,different kinase inhibitors were used.When U251 cells were pretreatedwith U0126 (ERK inhibitor), the promoter activity of miR-206 was sig-nificantly decreased, whereas pretreatment with PD98059 (MEK inhib-itor) or SP600125 (JNK inhibitor) had no appreciable inhibitory effecton the transcription of miR-206 (Fig. 7A). To confirm these results,quantitative RT-PCR was also performed for miR-206. The miR-206level was significantly reduced in cells pretreated with U0126 (Fig. 7B).These results suggested that ERK signaling pathways are involved inthe expression of miR-206.

4. Discussion

Neuroinflammation is characterized by the activation of glial cells,increased cytokine production, and infiltration of peripheral immunecells [29,30]. Astrocytes are the most abundant cell type in the centralnervous system,where they play important roles in promoting localizedinflammation [3]. MultiplemiRNAs regulate human innate immune sig-naling pathways and result in inflammatory responses [31,32], but theroles of miRNAs in astrocyte-derived inflammatory responses are notfully understood. In this study, we showed that astrocytes expresshigh levels of miR-206 in LPS-treated U251 cells. We also demonstrated

Fig. 6. Analysis of the miR-206 promoter and potential transcription factors. (A) pRL-TK waslengths of the miR-206 promoter into U251 cells. The first nucleotide of the mature pre-miR-2and luciferase activity was measured. Luciferase activity of the mock-transfected empty vectohuman chromosome 6. The promoter region from pre-1878 to pre-1149 was used for analysshown as diamonds. The sequences ofmutated sites are shown under the locus diagram. (C) Sch(D) The above promoter reporter constructs (100 ng) together with PRL-TK (10 ng) were co-trfor another 24 h. Sampleswere collected and analyzed for dual luciferase activity. Results are plas the percentage of LPS-inducible full-length promoter activity (100%). (E) Schematic represdisrupt transcription factor binding were introduced into the promoter construct and are ind(Mut 1 to Mut 4) and PRL-TK as described in (D). Fold increase after LPS stimulation relative1000 ng/ml LPS for 0, 6, 12, or 24 h. RNA samples were used for quantitative RT-PCR to det(100 ng), pRL-TK plasmid (10 ng), and pCNDA4.0-c-Jun/c-Fos (100 ng). After 24 h, their prowith the transcription factors c-Jun, c-Fos, or AP-1 (200, 500, or 1000 ng). After 24 h, the leveindependent experiments performed in triplicate. (*P b 0.05; **P b 0.01; ***P b 0.001).

for thefirst time thatmiR-206 targets the 3′-UTR of NR4A2 andpositive-ly regulates LPS-induced proinflammatory cytokine production throughactivation of NF-κB signaling pathways.

miR-206 contributes to the pathogenesis of Alzheimer's disease bytargeting brain-derived neurotrophic factor and promotes the apoptoticactivator of cell death by inhibiting notch3[33,34]. miR-206 also playsan important role in muscle differentiation andmaintenance and repairof neuromuscular junctions [36,37,39]. In the present study, we provideevidence thatmiR-206 is expressed at high levels in astrocytes stimulat-ed with LPS. These data are consistent with previous studies showingthat miR-206 expression is elevated in the degenerating brain [33,35,37]. In our study, we found that miR-206 has potent proinflammatoryproperties because it exacerbated the production of inflammatory cyto-kines in human astrocytes treated with LPS by enhancing the activationof NF-κB signalingpathways. These data indicate thatmiR-206mayplaya very important role in brain diseases.

Wehypothesized thatmiR-206may target a suppressor of inflamma-tion in astrocytes.We found thatmiR-206 targets 3′-UTR of NR4A2 in as-trocytes. NR4A2 belongs to the nuclear receptor (NR)4 family of orphannuclear receptors and functions as an anti-inflammatory mediator pri-marily via NF-κB signaling pathways [27,38,39]. The anti-inflammatoryactivity of NR4A2 is mediated by the NR4A2/CoREST transrepressionpathway that results in clearance and transcriptional repression of theNF-κB subunit, p65, in astrocytes [28]. Recently, the anti-inflammatoryeffects of NR4A2 have also been reported in human macrophages [40].Consequently, the direct effect ofmiR-206 on the production of its target,namely NR4A2, suggests that both of these factors are involved ininflammation-associated brain diseases. We found that NR4A2 overex-pression significantly reduced cytokine production, and co-transfectionof miR-206 mimics and an NR4A2 expression plasmid attenuated themiR-206–modulated LPS-induced proinflammatory response. Intrigu-ingly, our findings demonstrate that miR-206 enhances the activationof NF-κB pathways in LPS-treated astrocytes. Taken together, these re-sults indicate that LPS-induced miR-206 regulates the production of in-flammatory cytokines by enhancing NF-κB activity.

co-transfected with an miR-206 promoter reporter plasmid (100 ng) containing various06 is assigned +1. After 24 h, U251 cells were treated with 1 μg/ml LPS for another 24 h,r pGL3 Basic group was regarded as 1. (B) Schematic diagram of miR-206 genomic locionis of potential transcription factor binding sites. Four putative binding sites for AP-1 areematic representation of deletionmutants△1 to△3 in the full-lengthmiR-206 promoter.ansfected into U251 cells. At 24 h post-transfection, cells were treated with or without LPSotted as firefly luciferase activity normalized to Renilla luciferase activity and are expressedentation of point mutations (Mut 1 to Mut 4) in the wild-type promoter. Mutations thaticated by empty diamonds.(F) U251 cells were co-transfected with mutated constructsto the basal condition is indicated for each construct. (G) U251 cells were treated withect the mRNA level of c-Fos and c-Jun. (H) U251 cells were transfected with pre-1787moter activities were measured with luciferase analysis. (I) U251 cells were transfectedl of miR-206 was detected with quantitative RT-PCR. Results are representative of three

68 X. Duan et al. / Cellular Signalling 27 (2015) 61–68

Despite the important function of the miR-206–modulated proin-flammatory response in LPS-treated astrocytes, how the transcriptionof the gene encoding miR-206 is controlled remains unknown. There-fore, research has focused on the transcriptional regulation of miRNAs.In this study, we first found that the principal contributing componentsof the transcription factor AP-1 (c-Fos and c-Jun) were dramaticallyincreased in LPS-treated astrocytes. We further confirmed that LPS-induced upregulation of AP-1 contributed to transcriptional regulationof miR-206 in astrocytes. Consistent with our results, recent studiesalso demonstrated that AP-1 directly participates in transcription ofmiR-144 and that bone morphogenetic protein-6 inhibits miR-21expression through the AP-1 signaling pathway [41,46–47].Enterovirus-induced AP-1 contributes to expression of miR-146a [42].S100P/RAGE–mediated regulation of miR-155 is also AP-1 dependent[43]. Thus, AP-1 may regulate the processing of multiple miRNAs. Fur-ther studies are also needed todetermine if other potential transcriptionfactors and AP-1 may collaborate to contribute to the transcriptionalregulation of miR-206 in LPS-treated astrocytes.

As AP-1was essential for the upregulation ofmiR-206 in LPS-treatedastrocytes and widely participated in multiple signaling pathways, weinvestigated which signaling pathway was involved in the processingof miR-206. In this regard, we used different specific inhibitors ofkinases involved in multiple signaling pathways. Pretreatment with anERK inhibitor significantly inhibited miR-206 promoter activity andreduced the endogenous expression ofmiR-206 inU251 cells, indicatinga regulatory role for the ERK signaling pathway in the transcription ofmiR-206. MEK, p38, and JNK regulate the expression of other miRNAs.For example, JNK regulates miR-155, MEK/ERK regulates miR-768-3pexpression, and p38 regulates miR-34c and miR-133/1 induction byLPS [44,45]. In our study, however, no appreciable role for MEK or JNKwas observed in the regulation of miR-206.

In summary, this study provides evidence that miR-206 acts as apositive regulator of neuroinflammation in LPS-stimulated human as-trocytes by enhancing NF-κB signaling pathways, resulting in increasedproduction of IL-1β, IL-6, and CCL5. Our data demonstrate that NR4A2 isa novel target of miR-206. Finally, LPS-induced AP-1 was identified as apositive transcriptional regulator of miR-206. To the best of our knowl-edge, our data suggest for the first time that miR-206–mediated regula-tion of NR4A2 participates in inflammatory processes in astrocytes.However, given that miR-206 is multifunctional andmay have differenttargets in different circumstances, as well as our results with transfec-tion of miR-206 mimics and the fact that the effects of the NR4A2expression plasmid were transient, a role for miR-206 as a positive reg-ulator of inflammatory responses in astrocytes needs to be carefullyestablished in future studies.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cellsig.2014.10.006.

Conflict of interest

All authors report no potential conflict of interest.

Acknowledgments

This work was supported by the National Natural SciencesFoundation of China (31172325, 31472221), 973 Project of China(No. 2010CB530100), Special Fund for Agro-scientific Research in thePublic Interest (201203082), and Fundamental Research Funds for theCentral University (2011PY002, 2013PY051).

References

[1] D.N. Xanthos, J. Sandkühler, Nat. Rev. Neurosci. 15 (2014) 43–53.[2] E. Ellwardt, F. Zipp, Exp. Neurol. (n.d.). http://dx.doi.org/10.1016/j.expneurol.2014.

02.006[3] M. Aschner, Neurotoxicology 19 (1998) 269–281.[4] J.M. van Noort, Drug Discov. Today 11 (2006) 74–80.[5] S.S. Choi, H.J. Lee, I. Lim, J. Satoh, S.U. Kim, PLoS ONE 9 (2014) e92325.[6] T.-M. Yuan, Y. Sun, C.-Y. Zhan, H.-M. Yu, J. Neuroimmunol. 229 (2010) 16–25.[7] N. Bushati, S.M. Cohen, Annu. Rev. Cell Dev. Biol. 23 (2007) 175–205.[8] C. Lu, X. Huang, X. Zhang, K. Roensch, Q. Cao, K.I. Nakayama, et al., Blood 117 (2011)

4293–4303.[9] Q.-Y. Zhu, Q. Liu, J.-X. Chen, K. Lan, B.-X. Ge, J. Immunol. Baltim. Md 1950 (185)

(2010) 7435–7442.[10] Y. Liu, Q. Chen, Y. Song, L. Lai, J. Wang, H. Yu, et al., FEBS Lett. 585 (2011)

1963–1968.[11] G. Liu, A. Friggeri, Y. Yang, Y.-J. Park, Y. Tsuruta, E. Abraham, Proc. Natl. Acad. Sci. 106

(2009) 15819–15824.[12] P. Jiang, R. Liu, Y. Zheng, X. Liu, L. Chang, S. Xiong, et al., Exp. Cell Res. 318 (2012)

1175–1184.[13] E.B. Wendlandt, J.W. Graff, T.L. Gioannini, A.P. McCaffrey, M.E. Wilson, Innate

Immunol. 18 (2012) 846–855.[14] K. Yang, Y.S. He, X.Q. Wang, L. Lu, Q.J. Chen, J. Liu, et al., FEBS Lett. 585 (2011)

854–860.[15] E.R. Hutchison, E.M. Kawamoto, D.D. Taub, A. Lal, K. Abdelmohsen, Y. Zhang, et al.,

Glia 61 (2013) 1018–1028.[16] A. Iyer, E. Zurolo, A. Prabowo, K. Fluiter, W.G.M. Spliet, P.C. van Rijen, et al., PLoS ONE

7 (2012) e44789.[17] Y. Xu, H. Jin, X. Yang, L. Wang, L. Su, K. Liu, et al., FEBS Lett. 588 (2014) 1692–1698.[18] A.L. Cardoso, J.R. Guedes, L. Pereira de Almeida, M.C. Pedroso de Lima, Immunology

135 (2012) 73–88.[19] M. Shioya, S. Obayashi, H. Tabunoki, K. Arima, Y. Saito, T. Ishida, et al., Neuropathol.

Appl. Neurobiol. 36 (2010) 320–330.[20] K. Jeyaseelan, K.Y. Lim, A. Armugam, Stroke J. Cereb. Circ. 39 (2008) 959–966.[21] B. Zhang, X. Pan, Environ. Health Perspect. 117 (2009) 231–240.[22] R. Gorina, M. Font-Nieves, L. Márquez-Kisinousky, T. Santalucia, A.M. Planas, Glia 59

(2011) 242–255.[23] B. John, A.J. Enright, A. Aravin, T. Tuschl, C. Sander, D.S. Marks, PLoS Biol. 2 (2004)

e363.[24] S. Griffiths-Jones, Methods Mol. Biol. Clifton NJ 342 (2006) 129–138.[25] B.P. Lewis, C.B. Burge, D.P. Bartel, Cell 120 (2005) 15–20.[26] M. Rehmsmeier, P. Steffen, M. Hochsmann, R. Giegerich, RNA N. Y. N. 10 (2004)

1507–1517.[27] S.J. Bensinger, P. Tontonoz, Cell 137 (2009) 26–28.[28] K. Saijo, B. Winner, C.T. Carson, J.G. Collier, L. Boyer, M.G. Rosenfeld, et al., Cell 137

(2009) 47–59.[29] M. Sospedra, R. Martin, Int. Rev. Immunol. 24 (2005) 393–413.[30] S.J. Pittock, C.F. Lucchinetti, Neurologist 13 (2007) 45–56.[31] A. Rebane, C.A. Akdis, J. Allergy Clin. Immunol. 132 (2013) 15–26.[32] F. Olivieri, A.D. Procopio, R.R. Montgomery, Curr. Opin. Immunol. 29C (2014) 29–37.[33] N. Tian, Z. Cao, Y. Zhang, Neurosci. Bull. 30 (2014) 191–197.[34] G. Song, Y. Zhang, L. Wang, J. Biol. Chem. 284 (2009) 31921–31927.[35] A.H. Williams, G. Valdez, V. Moresi, X. Qi, J. McAnally, J.L. Elliott, et al., Science 326

(2009) 1549–1554.[36] G. Valdez, M.P. Heyer, G. Feng, J.R. Sanes, PLoS ONE 9 (2014) e93140.[37] S.-T. Lee, K. Chu, K.-H. Jung, J.H. Kim, J.-Y. Huh, H. Yoon, et al., Ann. Neurol. 72 (2012)

269–277.[38] K. Sakurada, M. Ohshima-Sakurada, T.D. Palmer, F.H. Gage, Dev. Camb. Engl. 126

(1999) 4017–4026.[39] P.I. Bonta, Arterioscler. Thromb. Vasc. Biol. 26 (2006) 2288.[40] L. Pei, A. Castrillo, M. Chen, A. Hoffmann, P. Tontonoz, J. Biol. Chem. 280 (2005)

29256–29262.[41] J. Du, S. Yang, D. An, F. Hu, W. Yuan, C. Zhai, et al., Cell Res. 19 (2009) 487–496.[42] B.-C. Ho, I.-S. Yu, L.-F. Lu, A. Rudensky, H.-Y. Chen, C.-W. Tsai, et al., Nat. Commun.

5 (2014) 3344.[43] B.C. Onyeagucha, M.E. Mercado-Pimentel, J. Hutchison, E.K. Flemington, M.A. Nelson,

Exp. Cell Res. 319 (2013) 2081–2090.[44] D. Zhang, X. Li, C. Chen, Y. Li, L. Zhao, Y. Jing, et al., PLoS ONE 7 (2012) e41478.[45] I.G. Cannell, Y.W. Kong, S.J. Johnston, M.L. Chen, H.M. Collins, H.C. Dobbyn, et al.,

Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 5375–5380.[46] J. Du, S. Yang, D. An, F. Hu, W. Yuan, C. Zhai, et al., Cell Res. 19 (2009) 487–496.[47] C. Cheng, W. Li, Z. Zhang, S. Yoshimura, Q. Hao, C. Zhang, et al., J. Biol. Chem. 288

(2013) 13748–13761.