of April 6, 2018.This information is current as

B PathwayκHBx/ERK/NF-with Chronic Hepatitis B Is Mediated by the Upregulation of IL-23 Expression in Patients

Fan and Kaichun WuWang, Yongguo Zhang, Hao Hu, Yongzhan Nie, Daiming Limin Xia, Dean Tian, Wenjie Huang, Hongwu Zhu, Jing

http://www.jimmunol.org/content/188/2/753doi: 10.4049/jimmunol.1101652December 2011;

2012; 188:753-764; Prepublished online 14J Immunol 

Referenceshttp://www.jimmunol.org/content/188/2/753.full#ref-list-1

, 17 of which you can access for free at: cites 46 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2012 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Upregulation of IL-23 Expression in Patients with ChronicHepatitis B Is Mediated by the HBx/ERK/NF-kB Pathway

Limin Xia,*,† Dean Tian,† Wenjie Huang,‡ Hongwu Zhu,* Jing Wang,* Yongguo Zhang,*

Hao Hu,* Yongzhan Nie,* Daiming Fan,* and Kaichun Wu*

IL-23 is a newly discovered proinflammatory cytokine that contributes to the maintenance and expansion of Th17 cells. IL-23 has

recently been identified as playing a critical role in a number of chronic inflammatory diseases. However, the regulatory mechanism

of IL-23 in chronic hepatitis B (CHB) remains largely unknown. The aims of this study were to detect the expression of IL-23 in CHB

patients and to explore the molecular mechanism of hepatitis B virus (HBV)-induced IL-23 expression. Serum levels and hepatic

expression of IL-23 were significantly upregulated in CHB patients. A positive correlation was found between IL-23 expression and

the histological activity index score, HBV DNA load, and serum alanine aminotransferase and aspartate aminotransferase levels.

HBx protein increased IL-23 expression in a dose-dependent manner. It also aided in the nuclear translocation of NF-kB, which

directly bound to the promoters of IL-23 subunits p19 and p40 to facilitate their transcription. NF-kB inhibitors blocked the effect

of HBx on IL-23 induction, and NF-kB subunits p65 and p50 increased the augmented IL-23 expression. Inhibition of ERK1/2

activation and transfection with ERK dominant-negative plasmid significantly blocked the HBx-induced IL-23 expression. Fur-

thermore, PI3K and Ras–MEK–MAPK inhibitors significantly decreased the ERK1/2 activation and IL-23 expression. Thus, we

report a new molecular mechanism for HBV-induced IL-23 expression, which involves the activation of the ERK/NF-kB pathway

by HBx, leading to the transactivation of the IL-23 p19 and p40 promoters. The Journal of Immunology, 2012, 188: 753–764.

Hepatitis B virus (HBV) infection is one of the leadingcauses of chronic liver disease, which affects .300million people worldwide. Annually, HBV infection

accounts for 1 million deaths worldwide due to cirrhosis, liverfailure, and hepatocellular carcinoma (1, 2). Although there is nodirect cytopathic effect on the hepatocytes, HBV infection inducesimmune cell infiltration, which leads to the formation of nec-roinflammatory foci and mediates the disease processes (3, 4). Therecruitment and differentiation of immune cells to the inflamma-tory site is triggered and enhanced by proinflammatory cytokines,which play important roles in inducing liver injury (5, 6).HBV encodes a small genome consisting of partially double-

stranded, circular DNA that is encapsulated within an envelopedparticle. The HBV genome encodes four overlapping open readingframes (S, C, P, and X) and expresses seven viral proteins including

HBx, HBs, preS2, preS1, HBc, HBe, and HBp (2). Among these,HBx is considered one of the most important determinants of viralreplication and viral-mediated pathogenesis (7, 8). The HBx proteinis a small protein composed of 154 aa, and it is distributed in thecytoplasm and, to some extent, the nucleus. Distribution in the cy-toplasm may activate the signal transduction cascade, but distribu-tion in the nucleus activates specific transcription factors. HBx doesnot bind DNA directly but transactivates multiple transcriptionfactors, including AP-1, AP-2, NF-kB, activating transcription factor(CREB/ATF), and HIF-1 (9–11). NF-kB has been evaluated becauseof its pivotal role in regulating inflammatory-related genes (12).IL-23 is a heterodimeric cytokine belonging to the IL-12 cy-

tokine family. It consists of two subunits: p40, which is shared withIL-12, and p19, which shares sequence homology with the IL-12p35 subunit but is distinct and leads to unique biological effects(13). IL-23 is predominantly produced by macrophages and den-dritic cells (DCs) that are activated by bacterial products, theSendai virus, and proinflammatory cytokines (14). The importanceof IL-23 in inflammatory responses has been demonstrated inexperimental autoimmune diseases. IL-23 selectively promotesIL-17 cytokine production by maintaining and expanding Th17cell expansion/maintenance and promoting the production of otherproinflammatory mediators, such as IL-8, IL-6, MCP-1, and G-CSF (15). IL-23 p19 knockout mice are resistant to experimen-tal autoimmune encephalomyelitis and collagen-induced arthritis,which are associated with defects in IL-17 production (16, 17). Inaddition, multiorgan inflammation and premature death are ob-served in transgenic mice overexpressing the IL-23 p19 subunit(18). Previous studies have found that IL-12 family members, suchas IL-12 and IL-27, are involved in chronic inflammation and liverdamage from HBV and hepatitis C virus infections (19–22).However, the regulatory mechanism of IL-23 in chronic HBVinfection remains largely unknown.Recent studies have found that Th17 cells, which are expanded

by IL-23, increase the severity of liver damage in patients with

*State Key Laboratory of Cancer Biology, Xijing Hospital of Digestive Diseases,Fourth Military Medical University, Xi’an 710032, Shaanxi Province, People’s Re-public of China; †Division of Gastroenterology, Tongji Hospital of Tongji MedicalCollege, Huazhong University of Science and Technology, Wuhan 430030, HubeiProvince, People’s Republic of China; and ‡State Key Laboratory of Virology, WuhanInstitute of Virology, Chinese Academy of Sciences, Wuhan 430071, Hubei Province,People’s Republic of China

Received for publication June 9, 2011. Accepted for publication November 4, 2011.

This work was supported by grants from the National Natural Science Foundation ofChina (81000864, 81090270, and 81090273), the National Key and Basic ResearchDevelopment Program of China (2010CB529302), the National Municipal Scienceand Technology Project (2009ZX09103-667 and 2009ZX09301-009-RC06), and theChinese Postdoctoral Science Foundation (20100471776).

Address correspondence and reprint requests to Prof. Kaichun Wu, State Key Lab-oratory of Cancer Biology, Xijing Hospital of Digestive Diseases, Fourth MilitaryMedical University, Xi’an 710032, Shaanxi Province, People’s Republic of China.E-mail address: wu_kaichun@yahoo.cn

Abbreviations used in this article: ALT, alanine aminotransferase; AST, aspartateaminotransferase; CHB, chronic hepatitis B; ChIP, chromatin immunoprecipitation;DC, dendritic cell; DN-ERK, dominant-negative mutant of ERK1; HAI, histologicalactivity index; HBV, hepatitis B virus; HC, healthy control.

Copyright� 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1101652

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

HBV (23–25), indicating that IL-23 may play an important role inpromoting inflammatory injury in chronic hepatitis B (CHB). Inthis study, we evaluated the serum levels and hepatic expression of

IL-23 in CHB patients and analyzed its correlation with HBV-mediated inflammatory damage. We also explored the possiblemolecular mechanism of HBV-induced IL-23 expression in hep-atocytes.

Materials and MethodsPatients

CHB patients (n = 192) were recruited for this study. All patients werediagnosed based on described criteria (26) and were not taking antiviraltherapy or immunosuppressive drugs within 6 mo of blood sampling.Individuals with concurrent hepatitis C virus, hepatitis D virus, HIV, au-toimmune liver disease, or alcoholic liver disease were excluded. Sixtyage-matched and sex-matched healthy individuals were enrolled as healthycontrols (HCs). All patients and HC individuals were recruited from theXijing Hospital at the Fourth Military Medical University (Shanxi, Peo-ple’s Republic of China) between 2008 and 2009. Ethical approval wasobtained from the Research Ethics Committee of Xijing Hospital, andwritten and informed consent was obtained from each patient. The basiccharacteristics of the enrolled subjects are listed in Table I.

Peripheral blood samples were collected from all enrolled subjects. Liverbiopsy samples were collected from 60 CHB patients, and 16 healthy livertissue samples were obtained from healthy donors whose livers were used

Table I. Clinical characteristics of the populations enrolled in the study

Parameter HCs CHB Patients

n 60 192Sex, n (M/F) 36/24 138/54Age, y (range) 31 (19–47) 32 (18–56)ALT, U/l (range) 26 (8–41) 240 (24–970)AST, U/l (range) 27 (11–42) 233 (26–1,084)HBV DNA, copies/ml

(range)ND 849,682,247

(15,000–8,100,000,000)HBsAg positive, n 0 192HBeAg positive, n 0 192HBeAb positive, n 0 0HBcAb positive, n 0 192

HBcAb IgM positive, n 0 0

Data are shown as median and range.ND, not determined.

FIGURE 1. Serum levels and the hepatic expression of IL-23 are increased in CHB patients. A, Serum IL-23 levels in CHB patients (n = 192) and HCs

(n = 60) were analyzed by ELISA. B and C, Hepatic IL-23 expression in CHB patients (n = 60) and HCs (n = 16) was investigated with real-time PCR. D,

Immunohistochemical analysis of IL-23 p19 expression in CHB liver and normal liver tissues. The IL-23 p19 protein in the tissue sections was detected

with SP kit. The nuclei were then counterstained with hematoxylin. Original magnification 340 (upper panels), 3200 (middle panels), and 3400 (lower

panels). E, Statistical analysis of IL-23 p19 expression in CHB liver and normal liver tissues. *p , 0.001.

754 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

for transplantation. The degree of hepatic inflammation was graded usingthe modified histological activity index (HAI) described by Scheuer (27).The basic clinical data for the CHB patients with available liver biopsysamples are presented in Table II.

Cell culture

Human hepatocellular carcinoma cells (HepG2 and Huh-7), normal he-patocyte cells (Chang liver and HL-7702), and HepG2.2.15 cells werecultured in DMEM at 37˚C in a 5% CO2 incubator. The medium wassupplemented with 10% FBS, 100 mg/ml penicillin, and 100 mg/mlstreptomycin.

Real-time PCR

Total RNA was extracted using TRIzol Reagent (Invitrogen), and reversetranscription was performed using an Advantage RT-for-PCR Kit (Takara)according to the manufacturer’s instructions. For the real-time PCR analysis,aliquots of double-stranded cDNAwere amplified using a SYBR Green PCRKit (Applied Biosystems). The cycling parameters were 95˚C for 15 s, 55˚Cfor 15 s, and 72˚C for 15 s for 45 cycles. A melting curve analysis was thenperformed. Ct was measured during the exponential amplification phase, andthe amplification plots were analyzed using SDS 1.9.1 software (AppliedBiosystems). For the cell lines, the relative expression level (defined as foldchange) of the target gene was determined by the equation 2–DDCt (DCt =DCttarget – DCtGAPDH; DDCt = DCtexpressing vector – DCtcontrol vector). Theexpression level was normalized to the fold change detected in the corre-sponding control cells, which was defined as 1.0. For clinical CHB andnormal liver tissue samples, the fold change of the target gene was deter-mined by the equation 2–DDCt (DDCt = DCttumor – DCtnontumor), and it wasnormalized to the average fold change in the normal liver tissues, which wasdefined as 1.0. All reactions were performed in duplicate. Primer sequencesare listed in Table III.

Immunohistochemistry

In brief, after deparaffinization and hydration, the slides were treated withendogenous peroxidase in 0.3% H2O2 for 30 min. Then, the sections wereblocked for 2 h at room temperature with a 1.5% blocking serum. Sectionswere incubated with an anti–IL-23 Ab (ab45420; Abcam) at 4˚C overnight.Labeled HRP was applied for 30 min at room temperature, and then,a diaminobenzidine solution was applied until the color developed. Theslides were counterstained with hematoxylin, and the results were observedunder a light microscope.

Plasmid constructions

All primers are shown in Table III. A 1348-bp IL-23 p19 promoter con-struct [(21269/+79) p19] was generated from human genomic DNA,

which corresponded to the sequence from 21269 to +79 (relative to thetranscriptional start site) of the 59-flanking region of the human IL-23 p19gene. This construct was generated using F1 and R1 as the forward andreverse primers incorporating SacI and XhoI sites at the 59 and 39 ends,respectively. The PCR product was cloned into the SacI and XhoI sites ofthe pGL3-Basic vector (Promega). The construct was confirmed by DNAsequencing. The 59-flanking deletion constructs of the IL-23 p19 promoter[(2565/+79) p19 and (277/+79) p19] were similarly generated with the(21269/+79) p19 construct as a template and the forward primers F2 andF3. Two NF-kB sites in the IL-23 p19 promoter were mutated witha QuikChange II site-directed mutagenesis kit (Stratagene). Truncated andmutated IL-23 p40 promoter constructs were cloned in the same way. Theexpression vectors containing the seven individual, viral structural genes ofthe HBVand NF-kB subunits (p65 and p50) were a generous gift from Dr.Wu Jianguo (Wuhan University, Wuhan, People’s Republic of China).

Transient transfections

The cells were plated at a density of 1 3 105 cells/well in a 24-well plate.After 12–24 h, the cells were cotransfected with 0.6 mg of the expressionvector plasmids, 0.18 mg of the promoter reporter plasmids, and 0.02 mg ofthe pRL-TK plasmids using Lipofectamine 2000 (Invitrogen) according tothe manufacturer’s instructions. After 5 h of transfection, the cells werewashed and allowed to recover overnight in fresh medium supplementedwith 1% FBS for 48 h. Serum-starved cells were used for the assay.

Luciferase reporter assay

The luciferase activity was detected with the Dual Luciferase Assay(Promega) according to the manufacturer’s instructions. The transfectedcells were lysed in the culture dishes with a lysis buffer, and the lysateswere centrifuged at maximum speed for 1 min in an Eppendorf micro-centrifuge. The relative luciferase activity was determined by a ModulusTD20/20 Luminometer (Turner Biosystems), and the transfection effi-ciency was normalized by Renilla activity.

ELISA

The culture supernatants were assayed for IL-23 protein according to themanufacturer’s instructions (cat. no. D2300B; R&D Systems) at roomtemperature. The cells were transfected with 0.8 mg of the relative plas-mids for 48 h. The supernatants were then collected and assayed for IL-23using ELISA. The concentration of IL-23 was determined using a standardcurve obtained with recombinant IL-23.

Western blot analysis

In brief, proteins from lysed cells were fractionated by SDS-PAGE andtransferred to nitrocellulose membranes. Nonspecific binding sites were

FIGURE 2. Serum levels and the hepatic expression of IL-23 are positively correlated with liver injury in CHB patients. A and B, Correlation analyses of

serum IL-23 levels with (A) HBV DNA loads and (B) serum ALT and AST levels. C, Compared with patients with lower HAI scores, CHB patients with

higher HAI scores had higher expression of hepatic IL-23 p19 mRNA. D and E, Correlation analyses of hepatic IL-23 p19 mRNA levels with (D) HBV

DNA loads and (E) serum ALT and AST levels. Correlation coefficients and levels of statistical significance are indicated.

The Journal of Immunology 755

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

blocked with 5%milk in TBST [120 mMTris-HCl (pH 7.4), 150 mMNaCl,and 0.05% Tween 20] for 1 h at room temperature. Blots were incubatedwith a specific Ab overnight at 4˚C. The membranes were then washed withPBS three times and incubated with an HRP-conjugated secondary Ab.Proteins were visualized using a Dura SuperSignal Substrate (Pierce).

EMSA

Biotin end-labeled oligonucleotides were synthesized and annealed to obtaindsDNA fragments. The oligonucleotide sequences are shown in Table II.EMSA was performed using the LightShift Chemiluminescent EMSA Kit(Pierce) according to the manufacturer’s protocol. DNA binding reactionswere performed in 20-ml samples containing 20 fmol biotin-labeled oli-gonucleotides, 4 mg nuclear extracts, 2 mg d(I-C), 2 ml 103 bindingbuffer, 0.1 mM EDTA, and 10% glycerol at room temperature for 20 min.For a competitor EMSA, 20-fold or 100-fold dilution of unlabeled oligo-nucleotides was added prior to the addition of the labeled probe. Thesamples were run on 6% nondenaturing polyacrylamide gels with 0.53TBE buffer and transferred to a nylon membrane (Amersham Biosciences,Pharmacia, Buckinghamshire, U.K.) at 380 mA (∼100 V) for 30 min. Achemiluminescent detection method using a luminal/enhancer solution andstable peroxide solution was used as described by the manufacturer, and themembranes were scanned by the Image Station 4000R (Kodak).

Chromatin immunoprecipitation assay

HepG2 cells transfected with relative plasmids were cross-linked using 1%formaldehyde at 37˚C for 10 min. After washing with PBS, the cells wereresuspended in 300 ml lysis buffer [50 mM Tris (pH 8.1), 10 mM EDTA,1% SDS, and 1 mM PMSF]. DNA was sheared to small fragments bysonication. The supernatants were precleared using a herring sperm DNA/protein G–Sepharose slurry (Sigma-Aldrich). The recovered supernatantswere incubated with anti-p50, p65, p52, c-Rel, RelB, or an isotype controlIgG for 2 h in the presence of herring sperm DNA and protein G–Sepharose beads. The immunoprecipitated DNA was retrieved from thebeads with 1% SDS and a 1.1 M NaHCO3 solution at 65˚C for 6 h. DNAwas then purified using a PCR purification kit (Qiagen), and real-time PCRwas performed on the extracted DNA using IL-23 p19 and p40 promoter-specific primers (Table III).

Statistical analyses

Data are expressed as the means 6 SD. Serum and hepatic IL-23 ex-pression levels were compared between groups using the Student t test.Correlation analyses were evaluated using the Spearman rank correlationtest. A p value ,0.05 was considered statistically significant. Statisticalanalysis data were analyzed using SPSS software (version 11.0).

ResultsSerum levels and hepatic expression of IL-23 are significantlyupregulated in CHB patients

To determine the potential association between IL-23 and HBVinfection, serum IL-23 levels from 192 CHB patients and 60 HCs(Table I) were evaluated by ELISA. Serum IL-23 levels weresignificantly increased in CHB patients compared with those ofthe HCs (Fig. 1A). Real-time PCR was used to detect hepatic IL-23 (subunits p19 and p40) mRNA expression in these patients andin HCs. Compared with the HCs, p19 and p40 mRNA levels weresignificantly increased in CHB patients (Fig. 1B, 1C). Further-more, immunohistochemistry was also performed to detect theexpression of IL-23 in the CHB patients. IL-23 was mainly vi-sualized in hepatocytes and infiltrating inflammatory cells in CHBliver tissues. However, IL-23 could not be detected in normal livertissues (Fig. 1D, 1E).

Increased IL-23 expression is positively correlated with liverinjury in CHB patients

To determine whether serum IL-23 expression is correlated withliver injury, we determined the correlations between serum IL-23levels and serum HBV DNA load, serum alanine aminotransferase(ALT) levels, or aspartate aminotransferase (AST) levels in CHBpatients. Significant, positive correlations were found betweenserum IL-23 levels and HBV DNA load (R = 0.544, p, 0.001) and

serum ALT (R = 0.508, p , 0.001) and AST (R = 0.442, p ,0.001) levels in CHB patients (Fig. 2A, 2B).We subsequently analyzed the association between hepatic IL-23

mRNA expression and liver injury scores in CHB patients ( Table II).Compared with patients with lower HAI scores, CHB patientswith higher HAI scores had higher expression of hepatic IL-23p19 mRNA (Fig. 2C). Further analysis found that hepatic IL-23p19 mRNA levels were positively correlated with HBV DNA load

Table II. Basic clinical data for CHB patients with available liverbiopsy samples

Case Sex Age (y)HBV DNA(copies/ml) ALT (U/l) AST (U/l) HAI Score

1 F 28 120,000,000 316 218 G32 F 29 16,000,000 154 269 G23 M 22 1,800,000,000 599 797 G34 M 44 1,200,000 186 311 G15 M 22 180,000,000 228 167 G26 M 31 1,700,000 130 227 G17 M 35 58,000,000 162 196 G28 M 25 12,000,000 226 216 G19 M 35 4,400,000 32 59 G210 M 33 570,000,000 229 144 G211 F 34 320,000,000 187 181 G312 M 35 500,000,000 108 124 G213 M 39 450,000 154 148 G114 M 25 560,000,000 321 116 G315 M 29 12,000,000 109 375 G216 M 22 330,000,000 118 74 G217 F 24 1,600,000 55 78 G118 F 38 730,000,000 173 154 G319 M 41 520,000,000 239 239 G120 F 37 710,000,000 162 140 G121 F 43 290,000,000 123 133 G322 M 34 11,000,000 144 124 G123 F 26 430,000,000 179 291 G124 M 27 37,000,000 214 217 G325 M 28 13,000,000 152 109 G226 M 34 39,000 144 140 G227 F 26 230,000 183 169 G128 F 31 350,000,000 249 159 G229 M 28 210,000,000 439 319 G230 F 24 920,000,000 101 379 G231 F 35 32,000,000 299 281 G232 M 25 75,000 101 196 G133 M 38 670,000,000 451 478 G234 M 22 150,000,000 141 108 G235 M 26 7,500,000 129 145 G136 M 44 35,000,000 142 176 G337 M 41 56,000 121 45 G138 M 36 160,000,000 223 230 G139 M 25 43,000 82 69 G140 M 26 22,000,000 270 103 G141 M 26 120,000,000 227 189 G242 M 21 140,000,000 100 300 G143 F 25 170,000 154 151 G144 M 23 380,000,000 153 262 G345 M 21 2,200,000 155 44 G146 F 42 21,000 96 131 G147 M 25 83,000,000 28 26 G248 F 39 180,000,000 231 237 G449 M 25 780,000 53 54 G150 M 33 670,000,000 222 226 G151 M 29 11,000,000 142 146 G152 M 33 8,100,000 333 133 G153 M 36 400,000 24 39 G154 M 23 24,000,000 128 158 G155 F 24 520,000 167 113 G156 M 31 34,000,000 98 47 G157 M 21 5,600,000 131 142 G158 F 36 6,300,000,000 283 352 G359 F 38 180,000 36 50 G160 F 38 280,000,000 178 189 G2

756 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

(R = 0.700, p, 0.001) and serum ALT (R = 0.519, p, 0.001) andAST (R = 0.377, p , 0.001) levels (Fig. 2D, 2E).

HBV directly induces IL-23 expression in hepatocytes

The previous experiments suggest that IL-23 expression is sig-nificantly increased in CHB patients and positively correlated withliver injury and that the hepatocytes were one of the main cells thatsecreted IL-23 in CHB patients. However, the mechanism ofIL-23 overexpression in hepatocytes remains unclear. To investi-gate whether HBV directly induces expression of IL-23, theplasmid pBlue-HBV, which contained the 1.3-fold HBV genome,

was transfected into hepatocytes (HepG2, Huh-7, Chang liver, andHL-7702). Forty-eight hours after transfection, the protein levels inthe supernatants and the mRNA levels in the cells were measured.Secretion of IL-23 significantly increased after transfection withpBlue-HBV (Fig. 3A). Accordingly, the mRNA levels of the p19and p40 subunits were also increased in cells transfected withpBlue-HBV (Fig. 3B). We also compared the expression of IL-23in HepG2 and HepG2.2.15 cells, which were stably transfectedwith the 1.3-fold HBV genome. IL-23 expression levels weremuch higher in the HepG2.2.15 cells than in the HepG2 cells (Fig.3C, 3D), as assessed by ELISA and real-time PCR.

FIGURE 3. HBV induces IL-23 expression in hepatocytes. A, The hepatocytes (HL-7702, Chang liver, HepG2, and Huh-7) were transfected with pBlue-

HBV, which contains the 1.3-fold HBV genome, and serum IL-23 levels were measured by ELISA. B, After the cells were transfected with pBlue-HBV, IL-

23 mRNA levels (subunits p19 and p40) were detected by real-time PCR. C, Serum IL-23 levels from HepG2 and HepG2.2.15 cells were measured by

ELISA. D, IL-23 mRNA levels in HepG2 and HepG2.2.15 cells were detected by real-time PCR. E, The cells (HL-7702, Chang liver, HepG2, and Huh-7)

were transfected with HBV viral protein plasmids (pCMV-HBx, pCMV-HBs, pCMV-preS2, pCMV-preS1, pCMV-HBc, pCMV-HBe, or pCMV-HBp), and

IL-23 protein levels were measured by ELISA. F, HBx transactivates the promoters of IL-23 p19 and p40. HepG2 cells were cotransfected with IL-23 p19

or p40 promoter luciferase reporter vectors [(21269/+79) p19 or (2979/+106) p40] and seven HBV viral protein plasmids. Relative luciferase activities

were determined by standard procedures. G, HepG2 cells were cotransfected with an increasing amount of plasmid expressing the HBx protein (pCMV-

HBx) and reporter plasmids (21269/+79) p19 or (2979/+106) p40. The luciferase activity of the mock pCMV-Taq group was designated as 1.00. The

results are the mean 6 SD of three experiments performed in duplicate. H, HBx protein levels were determined by Western blot analysis with an anti-Flag

Ab. The results are the mean 6 SD of three experiments performed in duplicate. *p , 0.05.

The Journal of Immunology 757

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

HBx upregulates IL-23 expression and transactivates itspromoter activityTo confirm further the effect of HBV on IL-23 expression, plas-mids containing seven individual viral genes of HBV (pCMV-HBx,pCMV-HBs, pCMV-preS2, pCMV-preS1, pCMV-HBc, pCMV-HBe, and pCMV-HBp) or control plasmids (pCMV-Tag) weretransfected into four different hepatocytes, respectively. IL-23

protein levels increased in cells transfected with pCMV-HBx,but the other viral proteins (S, preS2, preS1, HBc, HBe, andHBp) had no significant effects on IL-23 expression (Fig. 3E).IL-23 is a heterodimeric cytokine composed of the p19 and p40

subunits. To determine whether HBx regulates the transcriptionalactivity of these subunits, a luciferase reporter plasmid containingthe promoter for p19 or p40 was cotransfected with the seven viral

FIGURE 4. The NF-kB sites in the IL-23 p19 and p40 promoters are involved in HBx-induced IL-23 expression. A and B, HepG2 cells were treated with

three different NF-kB inhibitors after transfection with pCMV-HBx, and IL-23 expression was determined by ELISA and real-time PCR. The p19 and p40

mRNA levels in the untreated group were designated as 1.00. The results are the mean 6 SD of three experiments performed in duplicate. C, HepG2 cells

were treated with NF-kB inhibitors after cotransfection with pCMV-HBx and the reporter vector (21269/+79) p19 or (2979/+106) p40. After 48 h,

luciferase activity was measured. D, Effect of mutated NF-kB sites on the activity of the IL-23 p19 promoter. HepG2 cells were cotransfected with pCMV-

HBx and NF-kB1 MUT-p19 or NF-kB2 MUT-p19. The left panel is a schematic representation of the reporter gene constructs. The bar graphs in the right

panel represent the relative levels of luciferase activity in each of the transfected samples. E, The effect of the HBx protein on the truncated IL-23 p19

promoter. HepG2 cells were cotransfected with truncated IL-23 p19 promoter constructs and pCMV-HBx, and the relative luciferase activity was deter-

mined. F and G, Effect of the HBx protein on the mutated and truncated p40 promoter. The luciferase activity of the mock pCMV-Taq group was designated

as 1.00. The results are the mean 6 SD of three experiments performed in duplicate. *p , 0.05.

758 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

protein plasmids into HepG2 cells. The transcriptional activities ofp19 and p40 were significantly increased by the HBx protein, butthe other viral proteins (S, preS2, preS1, HBc, HBe, and HBp) hadno significant effect on p19 and p40 promoter activities (Fig. 3F).We next investigated the dose dependence of HBx on p19 and

p40 promoter activities. Reporter plasmids containing the p19 orp40 promoter were cotransfected with an increasing concentrationof pCMV-HBx plasmid into HepG2 cells. p19 and p40 promoteractivities were increased as the concentration of the pCMV-HBxplasmid increased, which indicated that the transactivation ofthe p19 and p40 promoters by the HBx protein was concentrationdependent (Fig. 3G). HBx protein levels were measured byWestern blot using an anti-Flag Ab in the transfected HepG2 cells(Fig. 3H). Together, these data suggest that HBx is a criticalregulator of IL-23 expression during HBV infection.

NF-kB activity is required for HBx-induced IL-23 expression

Recent studies have demonstrated that the p19 and p40 genes aretranscriptionally regulated by NF-kB in murine macrophages andDCs (28–30). To determine whether the HBx-induced p19 and p40expression was regulated through NF-kB, HepG2 cells weretreated with three different NF-kB inhibitors (SN50, PDTC, andBAY) after transfection with pCMV-HBx. ELISA and real-timePCR showed that HBx-induced IL-23 expression was significantlyinhibited by blocking the activation of NF-kB (Fig. 4A, 4B). Theluciferase reporter assay also demonstrated that the transcriptionalactivities of p19 and p40 were gradually inhibited after NF-kBinhibition (Fig. 4C). These results suggest that NF-kB activationis required to activate the p19 and p40 promoters by the HBxprotein.

The p19 promoter contains two putative NF-kB sites located fromnt2884 to2874 (p19 NF-kB1) and from nt290 to280 (p19 NF-kB2). To investigate the roles of the two NF-kB binding sites inregulating p19 subunit expression via the HBx protein, these NF-kB sites were mutated (Table III). Mutating the NF-kB2 but not theNF-kB1 binding site effectively reduced the HBx-induced activa-tion of the p19 promoter (Fig. 4D). To confirm further the role ofthe NF-kB2 binding site in regulating the p19 promoter by the HBxprotein, a series of 59 deletions in the p19 promoter were con-structed and cotransfected with pCMV-HBx into HepG2 cells.Deletion from nt21269 to nt2565 did not affect the HBx-inducedp19 luciferase activity; however, deletion from nt 2565 to nt 277significantly decreased the HBx-induced p19 luciferase activity(Fig. 4E). These results suggest that the NF-kB2 binding site isrequired for the HBx-induced activation of the p19 promoter.The p40 promoter contains only one NF-kB site, which is located

from nt2131 to nt2122 (p40 NF-kB). Mutating the NF-kB bindingsite significantly reduced the HBx-induced transactivation of the p40promoter (Fig. 4F). Consistent with this result, a deletion from nt2979 to nt 2268 did not affect the HBx-induced p40 promoteractivity. A further deletion to nt 2121 significantly decreased theHBx-induced p40 promoter activity (Fig. 4G). These results indicatethat the NF-kB binding site located from nt 2131 to nt 2122 isrequired for the HBx-induced activation of the p40 promoter.

The NF-kB subunit p65/p50 binds to the promoter of IL-23subunits p19 and p40

To determine whether NF-kB binds to the p19 promoter aftertransfection with pCMV-HBx, EMSAwas performed using biotin-labeled NF-kB consensus oligonucleotides in the p19 promoter

Table III. Primer sequences used in the study

Primer Name Primer Sequences

Primers for IL-23 p19 promoter sequential truncationsF1 (21269/+79)p19 sense 59-CGTTGAGCTCGTGTCATTGGGCAGTTTTGA-39F2 (2565/+79)p19 sense 59-TGTCGAGCTCTGAAACCAGGACCATCCAGT-39F3 (277/+79)p19 sense 59-TGTAGAGCTCTGCTGTGAGTCACCTGCTGG-39R1 (21269/+79)p19 antisense 59-TGATCTCGAGTGATTCTCTCTGTTCAGCGC-39

Primers for IL-23 p19 promoter site-directed mutagenesisF4 NF-kB1 MUT-p19 sense 59-GGTCCAAAGCTtaGACTGatCCGACCCCAGT-39R4 NF-kB1 MUT-p19 antisense 59-ACTGGGGTCGGatCAGTCtaAGCTTTGGACC-39F5 NF-kB2 MUT-p19 sense 59-GATGCAGGGAGtaGAATatCACCTGCTGTG-39R5 NF-kB2 MUT-p19 antisense 59-CACAGCAGGTGatATTCtaCTCCCTGCATC-39

Primers for IL-23 p40 promoter sequential truncationsF6 (2979/+106)p40 sense 59-TTCTGAGCTCCCTGTTCACGTGCAGACTGG-39F7 (2268/+106)p40 sense 59-TTCTGAGCTCTAGCATCTCCATCTCCTTCC-39F8 (2121/+106)p40 sense 59-TGTTGAGCTCCAGAAGGTTTTGAGAGTTGT-39R6 (2979/+106)p40 antisense 59-TCTTCTCGAGCTACACCTGGACAGCTGCAG-39

Primers for IL-23 p40 promoter site-directed mutagenesisF9 NF-kB MUT-p40 sense 59-GGAACTTCTTGcgATTCatCCAGAAGGTTT-39R9 NF-kB MUT-p40 antisense 59-AAACCTTCTGGatGAATcgCAAGAAGTTCC-39

Primers for probes in EMSANF-kB2 WT-p19 (290/280) 59-GATGCAGGGAGGGGAATCCCACCTGCTGTG-39NF-kB2 MUT-p19 (290/280) 59-GATGCAGGGAGtaGAATatCACCTGCTGTG-39NF-kB WT-p40 (2131/2122) 59-GGAACTTCTTGAAATTCCCCCAGAAGGTTT-39NF-kB MUT-p40 (2131/2122) 59-GGAACTTCTTGcgATTCatCCAGAAGGTTT-39

Primers used for ChIP assayP19 NF-kB2 sense 59-GCAACCACACTACTCATT-39P19 NF-kB2 antisense 59-TCTCCCTTTGTCTCTGAG-39P40 NF-kB sense 59-GATGTCTATGTTCCCTCCT-39P40 NF-kB antisense 59-TCTACTCCTTTCTGATGG-39

Primers for real-time PCRP19 sense 59-TGTTCCCCATATCCAGTGTGG-39P19 antisense 59-CTGGAGGCTGCGAAGGATTT-39P40 sense 59-AGAGGCTCTTCTGACCCCCAAG-39P40 antisense 59-CTCTTGCTCTTGCCCTGGACCTG-39GAPDH sense 59-GCACCGTCAAGGCTGAGAAC-39GAPDH antisense 59-TGGTGAAGACGCCAGTGGA-39

The Journal of Immunology 759

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

(290 to280) as a probe. The DNA binding activity of NF-kB wassignificantly increased in cells transfected with pCMV-HBx com-pared with cells transfected with the control plasmid. To determinethe specificity of the NF-kB binding activity, nuclear extracts wereincubated with the labeled NF-kB consensus oligonucleotides inthe presence of either an unlabeled, wild-type NF-kB bindingprobe or a mutated probe. The wild-type NF-kB consensus oli-gonucleotides but not the mutated oligonucleotides showed sig-nificantly abrogated NF-kB complexes (Fig. 5A). Consistently, theaddition of NF-kB inhibitors (SN50, PDTC, and BAY) inhibitedthe DNA binding activity of NF-kB. Similar results were found forthe p40 promoter (Fig. 5B). These data indicate that NF-kB bindsto the NF-kB binding site in the p19 and p40 promoters.The NF-kB family consists of five structurally related proteins

(p65, c-Rel, RelB, p50, and p52). We next performed a chromatinimmunoprecipitation (ChIP) assay to confirm whether NF-kBbinds to the p19 promoter in vivo. Chromatin fragments wereprepared from HepG2 cells transfected with pCMV-HBx andimmunoprecipitated with Abs against p65, c-Rel, RelB, p50, andp52 subunits, respectively. A 150-bp DNA fragment containingthe NF-kB2 binding site in the p19 promoter was amplified byreal-time PCR using specific primers. The NF-kB2 binding site atthe p19 promoter was immunoprecipitated from the cells trans-fected with pCMV-HBx in the presence of an anti-p50 or anti-p65

Ab; however, they could not be immunoprecipitated in the pres-ence of control plasmids and Abs against c-Rel, RelB, or p52 (Fig.5C). These data indicate that the NF-kB subunits p65/p50 bind tothe NF-kB2 binding site at the p19 promoter, and similar resultswere found for the p40 promoter.(Fig. 5D).

The NF-kB subunit p65/p50 increases the transcriptionalactivities of the IL-23 p19 and p40 promoters

To investigate the effects of the p65 and p50 subunits of NF-kBon the expression of IL-23, HepG2 cells were cotransfected withplasmids expressing p65 and p50 (pCMV-p65 and pCMV-p50).IL-23 protein levels were measured by ELISA. The p65 and p50proteins increased the expression of the IL-23 protein, andcotransfection with p65 and p50 further increased IL-23 proteinlevels (Fig. 6A). Furthermore, plasmids expressing p65 or p50were cotransfected with the IL-23 (21269/+79) p19 reporterplasmids into HepG2 cells, and the luciferase activity was mea-sured. The transcriptional activity of p19 was highest when cellswere transfected with both p65 and p50. However, mutating theNF-kB2 binding site in the IL-23 p19 promoter abolished theeffect of the p50 and/or p65 protein on the activation of the IL-23p19 promoter (Fig. 6B). Similar results were also found for the IL-23 p40 promoter (Fig. 6C). Therefore, these results indicate thatthe p65/p50 heterodimer induces IL-23 expression.

FIGURE 5. NF-kB subunits p65/p50 bind to the IL-23 p19 and p40 promoters directly. A and B, HepG2 cells were transfected with pCMV-HBx or the

control plasmid pCMV-Taq for 48 h. Nuclear extracts were subjected to EMSAwith the biotin-labeled (A) p19 NF-kB2 oligonucleotide or (B) NF-kB site of

the IL-23 p40 promoter in the absence and presence of the indicated unlabeled or unlabeled-mutated competitors. In addition, three NF-kB inhibitors were

involved in the indicated case. C and D, The ChIP assay showed the direct binding of p65/p50 to the IL-23 p19 and p40 promoters. Abs against p65, p50,

p52, c-Rel, and RelB or a control IgG Ab pulled down the DNA fragment containing the NF-kB2 binding site in the p19 promoter and the NF-kB site in the

p40 promoter. Real-time PCR was performed to detect the amount of immunoprecipitated products. The results are the mean 6 SD of three experiments

performed in duplicate. *p , 0.05.

760 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

HBx induces the translocation of the NF-kB subunit p65 intothe nucleus

The above results suggest the involvement of NF-kB subunits p65/p50 in HBx-induced IL-23 expression. Therefore, we determinedwhether HBx induces p65 translocation. Confocal microscopy andWestern blot were used to examine the effect of HBx on thetranslocation of p65 from the cytosol into the nucleus. Transfec-tion with pCMV-HBx but not with pCMV-Taq caused a remark-able translocation of p65 from the cytosol into the nucleus (Fig.6D). Western blot analysis showed that p65 protein levels de-creased in the cytosol and increased in the nucleus after trans-fection with pCMV-HBx for 48 h (Fig. 6E). However, thisphenomenon was not observed in cells transfected with pCMV-Taq. These results suggest that HBx induces the translocation ofp65 from the cytosol to the nucleus.

HBx-induced IL-23 expression is mediated by the ERK–p65/p50 signaling pathway

It is known that HBx can induce the phosphorylation of ERK1/2,JNK, and p38 MAPKs in various cell types (9). We sought todetermine whether any of the MAPKs were involved in the in-duction of IL-23 expression by HBx. U0126, SP600125, andSB202190 have been used specifically to inhibit the activation ofERK1/2, JNK, and p38 MAPKs, respectively. Pretreatment ofHepG2 cells with U0126 significantly reduced HBx-induced IL-23protein expression. However, pretreatment of cells with SP600125or SB202190 did not significantly affect HBx-induced IL-23 ex-pression (Fig. 7A).To determine whether HBx-induced IL-23 p19 and p40 promoter

activities were dependent on ERK1/2 activity, HepG2 cells werecotransfected with IL-23 (21269/+79) p19 or (2979/+106) p40reporter plasmids and pCMV-HBx and treated with U0126,

SP600125, and SB202190. The luciferase reporter assay showedthat treatment of the cells with the U0126 but not the SP600125or SB202190 significantly decreased the HBx-induced IL-23 p19and p40 promoter activities (Fig. 7B).We further determined the role of ERK1/2 in the HBx-induced

activation of IL-23 p19 and p40 promoters by introducing adominant-negative mutant of ERK1 (DN-ERK). This mutantblocks kinase activities by competing with endogenous kinases.HepG2 cells were cotransfected with pCMV-HBx (21269/+79)p19 or (2979/+106) p40 reporter plasmids, and different con-centrations of DN-ERK plasmids. The luciferase assay showedthat DN-ERK significantly inhibited the transactivation of the p19and p40 promoters by HBx (Fig. 7C).Next, we investigated the role of the HBx protein in the acti-

vation of signal transduction pathways. The protein levels ofphospho-ERK1/2, phospho-JNK, and phospho-p38 were measuredby Western blot analysis in the transfected HepG2 cells. HBxexpression significantly increased the phosphorylation level of theERK1/2 protein (Fig. 7D). The EMSA assay showed that the in-hibition of ERK1/2 phosphorylation by U0126 decreased thebinding of NF-kB to the p19 and p40 promoters and inhibited IL-23 expression (Fig. 7E, 7F). Taken together, these results suggestthat the HBx-induced IL-23 expression in hepatocytes might bemediated through the ERK–p65/p50 signaling pathway.

PI3K and Ras–MEK–MAPK pathway is involved inHBx-induced ERK1/2 activation and IL-23 expression

In the previous study, it was demonstrated that ERK1/2 activitycould be regulated by the PI3K and Ras–MEK–MAPK pathway(31–33). To investigate the role of the PI3K and Ras–MEK–MAPK pathway in HBx-induced ERK1/2 activation and IL-23expression, wortmannin and BAY43-9006 were used to inhibit

FIGURE 6. NF-kB subunits p65/p50 increase the expression of IL-23. A, HepG2 cells were transfected with the p65 and/or p50 expression vectors

(pCMV-p65 or pCMV-p50). IL-23 protein levels were measured in the supernatants with an ELISA kit. B, HepG2 cells were cotransfected with (21269/

+79) p19 or NF-kB2 MUT-p19 reporter constructs and p65 or p50 expression constructs. The luciferase activity of the cells cotransfected with pCMV-Taq

and the indicated reporter plasmids were set to 1.00. C, HepG2 cells were cotransfected with (2979/+106) p40 or NF-kB MUT-p40 reporter constructs and

p65 or p50 expression constructs. After 48 h, luciferase activity was measured. D, HBx induces the translocation of the NF-kB subunit p65 into the nucleus.

HepG2 cells were transfected with pCMV-HBx or the control plasmid pCMV-Taq for 48 h. The cells were stained with an anti-p65 Ab, and immuno-

fluorescence was monitored by confocal microscopy. Original magnification3400. E, Levels of cytosolic and nuclear p65 proteins were determined by Western

blot using an anti-p65 Ab after being transfected with pCMV-HBx. The results are the mean 6 SD of three experiments performed in duplicate. *p , 0.05.

The Journal of Immunology 761

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

the activation of PI3K and Raf-1, respectively. ELISA assayshowed that wortmannin and BAY43-9006 treatment significantlydecreased IL-23 protein expression induced by HBx (Fig. 8A).To determine whether HBx-induced IL-23 p19 and p40 promoter

activities were dependent on activation of the PI3K and Ras–MEK–MAPK pathway, HepG2 cells were cotransfected with IL-23(21269/+79) p19 or (2979/+106) p40 reporter plasmids andpCMV-HBx and treated with wortmannin and BAY43-9006, re-spectively. Luciferase reporter assay showed that wortmannin andBAY43-9006 treatment significantly decreased p19 and p40 pro-moter activity induced by HBx (Fig. 8B). Furthermore, the effectsof wortmannin and BAY43-9006 on the ERK1/2 phosphorylationwere also measured by Western blot. It was found that block of thePI3K and Ras–MEK–MAPK pathway significantly decreasedERK1/2 phosphorylation (Fig. 8C). These data suggest that HBx-

induced IL-23 expression and ERK1/2 activation is mediated bythe PI3K and Ras–MEK–MAPK pathway.

DiscussionIL-23 has recently been identified as playing a critical role ina number of chronic inflammatory diseases, including rheumatoidarthritis, inflammatory bowel disease, and multiple sclerosis (34).IL-23 is essential for the survival and activation of Th17 cells,which produce proinflammatory cytokines, such as TNF-a, IL-17A, and IL-6. However, IL-23 expression and its role in CHBare still unknown. Recent studies have demonstrated that Th17cells increase the severity of liver damage in CHB patients (23–25), indicating that IL-23 may play an important role in promotinginflammatory injury in CHB. In the current study, we found thatserum levels and hepatic IL-23 expression were positively corre-

FIGURE 7. HBx-induced IL-23 expression is mediated by the ERK1/2 signaling pathway. A, HepG2 cells were pretreated with inhibitors to ERK1/2,

JNK, and p38 (U0126, SP600125, and SB202190) before being transfected with pCMV-HBx or the control plasmid pCMV-Taq. IL-23 protein levels were

measured in the supernatant by ELISA. B, HepG2 cells were pretreated with inhibitors to ERK1/2, JNK, and p38 and cotransfected with pCMV-HBx and an

IL-23 (21269/+79) p19 or (2979/+106) p40 reporter plasmid. The luciferase activity of the mock pCMV-Taq group was designated as 1.00. C, HepG2

cells were cotransfected with pCMV-HBx, with increasing concentrations of DN-ERK and IL-23 (21269/+79) p19 or (2979/+106) p40 reporter plasmids,

and the luciferase activity was measured 48 h after transfection. The luciferase activity of the mock pCMV-Taq group was designated as 1.00. D, HepG2

cells were transfected with pCMV-HBx, and the phosphorylated and total ERK1/2, JNK, and p38 protein levels were analyzed by Western blot. E and F,

HepG2 cells were pretreated with inhibitors to ERK1/2, JNK, and p38, and the nuclear extracts were isolated. EMSAwas performed to measure the binding

of NF-kB to the (E) p19 and (F) p40 promoters. The results are the mean 6 SD of three experiments performed in duplicate. *p , 0.05.

762 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

lated with HBV DNA load and serum ALT and AST levels, whichoften serve as markers of liver injury. In addition, according to theliver biopsy diagnoses, patients with higher HAI scores had a higherexpression of hepatic IL-23 p19 than did patients with lower HAIscores. Taken together, these results indicate that IL-23 is closelyassociated with liver damage that is induced by HBV infection.IL-23 is predominantly produced by macrophages and DCs

in response to various stimuli. In this study, we found that IL-23was upregulated in CHB patients. Furthermore, the immunohisto-chemistry results demonstrated that, besides inflammatory cells,hepatocytes also produce a significant amount of IL-23. Theseresults are consistent with previous research because other col-leagues and our group have also demonstrated that hepatocytes arecapable of producing proinflammatory cytokines, such as TNF-a,IL-6, IL-8, IL-18, MIG, and IP-10 (35–39). To confirm further thedirect effect of HBVon IL-23 expression, the plasmid pBlue-HBVwas transfected into hepatocytes. The results from ELISA, real-time PCR, and the luciferase reporter assay demonstrated thatHBV could directly induce IL-23 expression. Furthermore, wefound that the viral protein HBx, which is encoded by HBV,transactivated the promoters of the IL-23 subunits p19 and p40.Synthesis of biologically active IL-23 heterodimers requires the

expression of both the p19 and p40 subunits. Several pieces ofevidence have indicated that expression of the p19 and p40 genesare regulated by the NF-kB transcription factor. Recent studieshave demonstrated that IL-23 p19 is mainly regulated by the NF-kB heterodimer (p65/c-Rel) that binds to the p19 promoter andthat it is necessary for the TLR-induced IL-23 p19 expression inmurine macrophages and DCs (28, 29). Similarly, c-Rel is re-quired for LPS-induced IL-23 p40 gene transcription in macro-phages (40). To determine the role of NF-kB in HBx-induced IL-23 p19 and p40 transcriptions, the NF-kB binding sites on the p19and p40 promoters were mutated. Our results showed that the NF-kB2 binding sites in the p19 and p40 promoters are essential forregulating the transcription of IL-23 p19 and p40 by HBx.NF-kB belongs to the Rel family of transcription factors, which

regulate an exceptionally large number of genes, particularly those

involved in immune and inflammatory responses (41, 42). Themammalian NF-kB family consists of five members: c-Rel, RelA(p65), RelB, p50, and p52. NF-kB subunits are able to homodi-merize or heterodimerize to form transcription factors that bind tothe promoters of target genes and regulate the expression of manycytokines, such as TNF-a, IL-1b, IL-2, IL-6, IL-8, CXCL9,CXCL10, and IL-12 p35 (43). In the current study, the EMSAassay revealed that NF-kB binds to the p19 and p40 promoters andis involved in HBx-induced IL-23 expression. The ChIP assayfurther confirmed that the NF-kB heterodimer p60/p50 binds tothe IL-23 p19 and p40 promoters. Notably, the overexpression ofwild-type p50 and p65 stimulated IL-23 expression, and the in-hibition of NF-kB activation by the NF-kB inhibitor significantlysuppressed IL-23 expression. These data indicate that NF-kBheterodimer p60/p50 activation and binding to the IL-23 subunitgene promoters are essential for HBx-induced IL-23 expression.Previous studies have demonstrated that MAPK pathways are

important regulators in proinflammatory cytokine production (44,45). In mammals, three major MAPK pathways have been iden-tified: ERK, JNK, and p38 MAPK. Studies have found that HBxcan induce the phosphorylation of ERK1/2, JNK, and p38 MAPKsin various cell types. To determine whether MAPK pathways areinvolved in HBx-induced IL-23 expression, different inhibitors(U0126, SP600125, and SB202190) of the MAPK pathways wereused to block their activities. ELISA and the luciferase reporterassay showed that the U0126 significantly decreased HBx-inducedIL-23 expression. It was demonstrated that U0126 could bind toMEK1/2 and indirectly affects ERK1/2 function (46). In accor-dance with previous research, we found that U0126 significantlydecreased the phosphorylation of ERK1/2. Furthermore, EMSAassay showed that U0126 blocked the binding of NF-kB to the IL-23 p19 and p40 promoters. These results indicate that ERK1/2activation is required for HBx-induced IL-23 expression. Previ-ous studies have reported that ERK1/2 activity could be regulatedby PI3K and Ras–MEK–MAPK pathway (31–33). In the currentstudy, wortmannin and BAY43-9006 were used to inhibit PI3Kand Ras–MEK–MAPK pathway, respectively. Consistent with

FIGURE 8. PI3K and Ras–MEK–MAPK pathway is critical for HBx-induced IL-23 expression. A, HepG2 cells were treated with wortmannin and

BAY43-9006 after being transfected with pCMV-HBx or the control plasmid pCMV-Taq. IL-23 protein levels were measured in the supernatant by ELISA.

B, HepG2 cells were cotransfected with pCMV-HBx and an IL-23 (21269/+79) p19 or (2979/+106) p40 reporter plasmid before being pretreated with

wortmannin and BAY43-9006. The promoter activities of p19 and p40 were measured by luciferase reporter assay. C, Western blot was used to detect the

protein levels of HBx, phospho-ERK1/2, and total ERK1/2. *p , 0.001.

The Journal of Immunology 763

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

previous studies, wortmannin and BAY43-9006 significantly in-hibit ERK activation. Furthermore, we also found that wortmanninand BAY43-9006 decreased IL-23 expression induced by HBx.These data suggest that HBx-induced IL-23 expression and ERK1/2 activation is mediated by PI3K and Ras–MEK–MAPK pathway.In summary, our findings demonstrate that IL-23 is significantly

increased in CHB patients, and increased IL-23 expression ispositively correlated with liver injury. Furthermore, viral proteinHBx induces IL-23 expression in hepatocytes through the acti-vation of the ERK/NF-kB pathway.

AcknowledgmentsWe thank all CHB patients and healthy participants in this study.

DisclosuresThe authors have no financial conflicts of interest.

References1. Dienstag, J. L. 2008. Hepatitis B virus infection. N. Engl. J. Med. 359: 1486–

1500.2. Ganem, D., and A. M. Prince. 2004. Hepatitis B virus infection—natural history

and clinical consequences. N. Engl. J. Med. 350: 1118–1129.3. Rehermann, B., and M. Nascimbeni. 2005. Immunology of hepatitis B virus and

hepatitis C virus infection. Nat. Rev. Immunol. 5: 215–229.4. Chang, J. J., and S. R. Lewin. 2007. Immunopathogenesis of hepatitis B virus

infection. Immunol. Cell Biol. 85: 16–23.5. Bertoletti, A., and A. J. Gehring. 2006. The immune response during hepatitis B

virus infection. J. Gen. Virol. 87: 1439–1449.6. Rehermann, B. 2000. Intrahepatic T cells in hepatitis B: viral control versus liver

cell injury. J. Exp. Med. 191: 1263–1268.7. Zhang, X., H. Zhang, and L. Ye. 2006. Effects of hepatitis B virus X protein on

the development of liver cancer. J. Lab. Clin. Med. 147: 58–66.8. Arbuthnot, P., A. Capovilla, and M. Kew. 2000. Putative role of hepatitis B virus

X protein in hepatocarcinogenesis: effects on apoptosis, DNA repair, mitogen-activated protein kinase and JAK/STAT pathways. J. Gastroenterol. Hepatol. 15:357–368.

9. Bouchard, M. J., and R. J. Schneider. 2004. The enigmatic X gene of hepatitis Bvirus. J. Virol. 78: 12725–12734.

10. Diao, J., R. Garces, and C. D. Richardson. 2001. X protein of hepatitis B virusmodulates cytokine and growth factor related signal transduction pathwaysduring the course of viral infections and hepatocarcinogenesis. Cytokine GrowthFactor Rev. 12: 189–205.

11. Sung, W. K., Y. Lu, C. W. Lee, D. Zhang, M. Ronaghi, and C. G. Lee. 2009.Deregulated direct targets of the hepatitis B virus (HBV) protein, HBx, identifiedthrough chromatin immunoprecipitation and expression microarray profiling. J.Biol. Chem. 284: 21941–21954.

12. Ghosh, S., and M. S. Hayden. 2008. New regulators of NF-kappaB in inflam-mation. Nat. Rev. Immunol. 8: 837–848.

13. Hunter, C. A. 2005. New IL-12-family members: IL-23 and IL-27, cytokineswith divergent functions. Nat. Rev. Immunol. 5: 521–531.

14. Lyakh, L., G. Trinchieri, L. Provezza, G. Carra, and F. Gerosa. 2008. Regulationof interleukin-12/interleukin-23 production and the T-helper 17 response inhumans. Immunol. Rev. 226: 112–131.

15. Boniface, K., B. Blom, Y. J. Liu, and R. de Waal Malefyt. 2008. Frominterleukin-23 to T-helper 17 cells: human T-helper cell differentiation revisited.Immunol. Rev. 226: 132–146.

16. Murphy, C. A., C. L. Langrish, Y. Chen, W. Blumenschein, T. McClanahan,R. A. Kastelein, J. D. Sedgwick, and D. J. Cua. 2003. Divergent pro- andantiinflammatory roles for IL-23 and IL-12 in joint autoimmune inflammation. J.Exp. Med. 198: 1951–1957.

17. Cua, D. J., J. Sherlock, Y. Chen, C. A. Murphy, B. Joyce, B. Seymour, L. Lucian,W. To, S. Kwan, T. Churakova, et al. 2003. Interleukin-23 rather thaninterleukin-12 is the critical cytokine for autoimmune inflammation of the brain.Nature 421: 744–748.

18. Wiekowski, M. T., M. W. Leach, E. W. Evans, L. Sullivan, S. C. Chen,G. Vassileva, J. F. Bazan, D. M. Gorman, R. A. Kastelein, S. Narula, andS. A. Lira. 2001. Ubiquitous transgenic expression of the IL-23 subunit p19induces multiorgan inflammation, runting, infertility, and premature death. J.Immunol. 166: 7563–7570.

19. Zhu, C., R. Zhang, L. Liu, S. T. Rasool, Y. Mu, W. Sun, Q. Hao, F. Liu, Y. Zhu,and J. Wu. 2009. Hepatitis B virus enhances interleukin-27 expression bothin vivo and in vitro. Clin. Immunol. 131: 92–97.

20. Zhang, Y., C. J. Ma, L. Ni, C. L. Zhang, X. Y. Wu, U. Kumaraguru, C. F. Li,J. P. Moorman, and Z. Q. Yao. 2011. Cross-talk between programmed death-1 andsuppressor of cytokine signaling-1 in inhibition of IL-12 production by monocytes/macrophages in hepatitis C virus infection. J. Immunol. 186: 3093–3103.

21. Zeuzem, S., and V. Carreno. 2001. Interleukin-12 in the treatment of chronichepatitis B and C. Antiviral Res. 52: 181–188.

22. Ma, C. J., L. Ni, Y. Zhang, C. L. Zhang, X. Y. Wu, A. N. Atia, P. Thayer,J. P. Moorman, and Z. Q. Yao. 2011. PD-1 negatively regulates interleukin-12expression by limiting STAT-1 phosphorylation in monocytes/macrophagesduring chronic hepatitis C virus infection. Immunology 132: 421–431.

23. Zhang, J. Y., Z. Zhang, F. Lin, Z. S. Zou, R. N. Xu, L. Jin, J. L. Fu, F. Shi,M. Shi, H. F. Wang, and F. S. Wang. 2010. Interleukin-17-producing CD4(+)T cells increase with severity of liver damage in patients with chronic hepatitisB. Hepatology 51: 81–91.

24. Wu, W., J. Li, F. Chen, H. Zhu, G. Peng, and Z. Chen. 2010. Circulating Th17cells frequency is associated with the disease progression in HBV infectedpatients. J. Gastroenterol. Hepatol. 25: 750–757.

25. Ge, J., K. Wang, Q. H. Meng, Z. X. Qi, F. L. Meng, and Y. C. Fan. 2010. Im-plication of Th17 and Th1 cells in patients with chronic active hepatitis B. J.Clin. Immunol. 30: 60–67.

26. Lok, A. S., and B. J. McMahon. 2007. Chronic hepatitis B. Hepatology 45: 507–539.

27. Scheuer, P. J. 1991. Classification of chronic viral hepatitis: a need for reas-sessment. J. Hepatol. 13: 372–374.

28. Mise-Omata, S., E. Kuroda, J. Niikura, U. Yamashita, Y. Obata, and T. S. Doi.2007. A proximal kappaB site in the IL-23 p19 promoter is responsible for RelA-and c-Rel-dependent transcription. J. Immunol. 179: 6596–6603.

29. Carmody, R. J., Q. Ruan, H. C. Liou, and Y. H. Chen. 2007. Essential roles of c-Rel in TLR-induced IL-23 p19 gene expression in dendritic cells. J. Immunol.178: 186–191.

30. Becker, C., S. Wirtz, X. Ma, M. Blessing, P. R. Galle, and M. F. Neurath. 2001.Regulation of IL-12 p40 promoter activity in primary human monocytes: roles ofNF-kappaB, CCAAT/enhancer-binding protein beta, and PU.1 and identificationof a novel repressor element (GA-12) that responds to IL-4 and prostaglandin E(2). J. Immunol. 167: 2608–2618.

31. Miller, K. A., N. Yeager, K. Baker, X. H. Liao, S. Refetoff, and A. Di Cristofano.2009. Oncogenic Kras requires simultaneous PI3K signaling to induce ERK ac-tivation and transform thyroid epithelial cells in vivo. Cancer Res. 69: 3689–3694.

32. Bates, M. E., J. B. Sedgwick, Y. Zhu, L. Y. Liu, R. G. Heuser, N. N. Jarjour,H. Kita, and P. J. Bertics. 2010. Human airway eosinophils respond to chemo-attractants with greater eosinophil-derived neurotoxin release, adherence to fi-bronectin, and activation of the Ras-ERK pathway when compared with bloodeosinophils. J. Immunol. 184: 7125–7133.

33. Morton, A. M., B. McManus, P. Garside, A. M. Mowat, and M. M. Harnett.2007. Inverse Rap1 and phospho-ERK expression discriminate the maintenancephase of tolerance and priming of antigen-specific CD4+ T cells in vitro andin vivo. J. Immunol. 179: 8026–8034.

34. Abraham, C., and J. H. Cho. 2009. IL-23 and autoimmunity: new insights intothe pathogenesis of inflammatory bowel disease. Annu. Rev. Med. 60: 97–110.

35. Lara-Pezzi, E., P. L. Majano, M. Gomez-Gonzalo, C. Garcıa-Monzon,R. Moreno-Otero, M. Levrero, and M. Lopez-Cabrera. 1998. The hepatitis Bvirus X protein up-regulates tumor necrosis factor alpha gene expression inhepatocytes. Hepatology 28: 1013–1021.

36. Lee, M. O., Y. H. Choi, E. C. Shin, H. J. Kang, Y.M. Kim, S. Y. Jeong, J. K. Seong,D. Y. Yu, H. Cho, J. H. Park, and S. J. Kim. 2002. Hepatitis B virus X proteininduced expression of interleukin 18 (IL-18): a potential mechanism for liverinjury caused by hepatitis B virus (HBV) infection. J. Hepatol. 37: 380–386.

37. Mahe, Y., N. Mukaida, K. Kuno, M. Akiyama, N. Ikeda, K. Matsushima, andS. Murakami. 1991. Hepatitis B virus X protein transactivates humaninterleukin-8 gene through acting on nuclear factor kB and CCAAT/enhancer-binding protein-like cis-elements. J. Biol. Chem. 266: 13759–13763.

38. Zhou, Y., S. Wang, J. W. Ma, Z. Lei, H. F. Zhu, P. Lei, Z. S. Yang, B. Zhang,X. X. Yao, C. Shi, et al. 2010. Hepatitis B virus protein X-induced expression ofthe CXC chemokine IP-10 is mediated through activation of NF-kappaB andincreases migration of leukocytes. J. Biol. Chem. 285: 12159–12168.

39. Xia, L. M., W. J. Huang, J. G. Wu, Y. B. Yang, Q. Zhang, Z. Z. Zhou, H. F. Zhu,P. Lei, G. X. Shen, and D. A. Tian. 2009. HBx protein induces expression ofMIG and increases migration of leukocytes through activation of NF-kappaB.Virology 385: 335–342.

40. Sanjabi, S., A. Hoffmann, H. C. Liou, D. Baltimore, and S. T. Smale. 2000.Selective requirement for c-Rel during IL-12 P40 gene induction in macro-phages. Proc. Natl. Acad. Sci. USA 97: 12705–12710.

41. Li, Q., and I. M. Verma. 2002. NF-kappaB regulation in the immune system. Nat.Rev. Immunol. 2: 725–734.

42. Sun, S. C., and S. C. Ley. 2008. New insights into NF-kappaB regulation andfunction. Trends Immunol. 29: 469–478.

43. Hayden, M. S., and S. Ghosh. 2004. Signaling to NF-kappaB. Genes Dev. 18:2195–2224.

44. Huang, G., L. Z. Shi, and H. Chi. 2009. Regulation of JNK and p38 MAPK in theimmune system: signal integration, propagation and termination. Cytokine 48:161–169.

45. Liu, Y., E. G. Shepherd, and L. D. Nelin. 2007. MAPK phosphatases—regulatingthe immune response. Nat. Rev. Immunol. 7: 202–212.

46. Favata, M. F., K. Y. Horiuchi, E. J. Manos, A. J. Daulerio, D. A. Stradley,W. S. Feeser, D. E. Van Dyk, W. J. Pitts, R. A. Earl, F. Hobbs, et al. 1998.Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J.Biol. Chem. 273: 18623–18632.

764 HBx UPREGULATES IL-23 EXPRESSION THROUGH ERK/NF-kB PATHWAY

by guest on April 6, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from