IL-10-Dependent S100A8 Induction in Monocytes/Macrophages by Double-Stranded RNA

This information is current as Yasumi Endoh, Yuen Ming Chung, Ian A. Clark, Carolyn L. of September 27, 2021. Geczy and Kenneth Hsu J Immunol 2009; 182:2258-2268; ; doi: 10.4049/jimmunol.0802683 http://www.jimmunol.org/content/182/4/2258 Downloaded from

References This article cites 73 articles, 26 of which you can access for free at: http://www.jimmunol.org/content/182/4/2258.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 27, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

IL-10-Dependent S100A8 Gene Induction in Monocytes/Macrophages by Double-Stranded RNA1

Yasumi Endoh,* Yuen Ming Chung,* Ian A. Clark,† Carolyn L. Geczy,* and Kenneth Hsu2*

The S100 calcium-binding S100A8 and are elevated systemically in patients with viral infections. The S100A8- S100A9 complex facilitated viral replication in human CD4؉ T lymphocytes latently infected with HIV-1- and S100A8-induced HIV-1 transcriptional activity. Mechanisms inducing the S100 and the potential source of these proteins following viral activation are unknown. In this study, we show that S100A8 was induced in murine macrophages, and S100A8 and S100A9 in human monocytes and macrophages, by polyinosinic:polycytidylic acid, a dsRNA mimetic. Induction was at the transcriptional level and was IL-10 dependent. Similar to LPS-induced S100A8, induction by dsRNA was dependent on p38 and ERK MAPK. kinase R (PKR) mediates antiviral defense and participates in MyD88-dependent/independent signaling triggered by

TLR4 or TLR3. Like IL-10, S100 induction by polyinosinic:polycytidylic acid and by LPS was inhibited by the specific PKR Downloaded from inhibitor 2-aminopurine, indicating a novel IL-10, PKR-dependent pathway. Other mediators such as IFN-␤, which synergized with dsRNA, may also be involved. C/EBP␤ bound the defined promoter region in response to dsRNA. S100A8 was expressed in lungs of mice infected with influenza virus and was maximal at day 8 with strong immunoreactivity in epithelial cells lining the airways and in mononuclear cells and declined early in the recovery phase, implying down-regulation by mediator(s) up-regulated during resolution of the infection. IL-10 is implicated in viral persistence. Since S100A8/S100A9 levels are likely to be maintained

in conditions where IL-10 is raised, these proteins may contribute to viral persistence in patients infected by some RNA http://www.jimmunol.org/ viruses. The Journal of Immunology, 2009, 182: 2258–2268.

alcium-binding proteins S100A8 and S100A9 are ex- S100A9 negated this activity and S100A9-null mice are protected pressed constitutively in large amounts in neutrophils from endotoxic shock (9). S100A8 and S100A8/S100A9 were re- C and are inducible in monocytes and macrophages. cently shown to induce MMP-2, 3, 9, and particularly 13, in mu- S100A8/S100A9-positive macrophages are found in numerous au- rine bone marrow-derived macrophages (10), suggesting that the toimmune diseases (1) and in atherosclerotic plaque (2). They are S100 proteins mediate macrophage migration and matrix degrada- also inducible in other cell types such as epithelial cells (3) and tion via MMP production. Intracellularly, the complex is impli- fibroblasts (4) by appropriate stimulants and their expression is cated in arachidonic acid transport (11) assembly of the NADPH by guest on September 27, 2021 associated with several tumors. S100A8 and S100A9 have intra- oxidase complex (12) and may regulate cytoskeletal rearrangement cellular and extracellular functions and although the S100A8- of phagocyte membranes during migration by integrating MAPK S100A9 complex is considered the major functional form, separate and calcium-dependent signals that influence tubulin polymeriza- functions for the individual proteins are reported (reviewed in tion (13, 14). Refs. 5 and 6). These proteins are proposed to function as novel Elevated systemic levels of S100A8/S100A9 are associated with damage-associated molecular pattern molecules (7). The complex viral infections such as human papillomavirus (15), and microarray is antimicrobial, apoptotic for some cells, and inhibits matrix met- of PBMC from patients with severe acute respiratory syndrome, alloproteinase (MMP)3 activities (8) by virtue of its ability to che- caused by a coronavirus, identified ϳ60-fold increase in the late zinc. Recently, S100A8 was reported to activate cytokine pro- S100A9 gene (16). S100A8/A9 levels in patients infected with a duction from murine bone marrow cells via TLR4 ligation; lentivirus, HIV-1, correlate with disease progression and low CD4ϩ counts (17), and in HIV-1-seropositive patients with ad- *Centre for Infection and Inflammation Research, School of Medical Sciences, Uni- vanced immunodeficiency (18, 19) correlate with the onset of and versity New South Wales, Sydney, Australia; and †School of Biochemistry and Mo- lecular Biology, Australian National University, Canberra, Australia with opportunistic infections (20–22). The 27E10 Ag, (detects the Received for publication August 14, 2008. Accepted for publication December S100A8/A9 heterocomplex) was suggested to be a potential 1, 2008. marker for different stages of HIV (19). S100A8 was isolated from The costs of publication of this article were defrayed in part by the payment of page cervico-vaginal secretions from women at high risk of HIV infec- charges. This article must therefore be hereby marked advertisement in accordance tion that contained HIV-inducing activity for an HIV-infected with 18 U.S.C. Section 1734 solely to indicate this fact. monocyte cell line, and recombinant S100A8 mimicked this ac- 1 This work was supported by grants from the National Health and Medical Research Council of Australia. Y.E. had an International Postgraduate Award. tivity (23). The source of this protein was suggested as being de- 2 Address correspondence and reprint requests to Dr. Kenneth Hsu, Centre for Infec- rived from granulocytes, macrophages, and epithelial cells present tion and Inflammation Research, School of Medical Sciences, University New South as a consequence of localized inflammation of the genital mucosa. Wales, Sydney, NSW 2052. E-mail address: [email protected] In another study, S100A8, S100A9, and S100A8/S100A9 facili- 3 Abbreviations used in this paper: MMP, matrix metalloproteinase; poly(I:C), tated viral replication in human CD4ϩ T lymphocytes latently in- polyinosinic:polycytidylic acid; PKR, protein kinase R; HPRT, hypoxanthine gua- nine phosphoribosyltransferase; COX-2, cyclooxygenase 2; 2-AP, 2-aminopurine; fected with HIV-1 and S100A8-induced HIV-1 transcriptional ac- siRNA, small interfering RNA; iNOS, inducible NO synthase; ChIP, chromatin tivity via an NF-␬B-dependent pathway (24). It was proposed that immunoprecipitation. these S100 proteins may particularly contribute to HIV-1 progres- Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 sion in seropositive patients with opportunistic infections and/or www.jimmunol.org/cgi/doi/10.4049/jimmunol.0802683 The Journal of Immunology 2259

Table I. Primers and conditions used for real-time RT-PCR amplification and ChIP

Gene Forward Primer (5Ј33Ј) Reverse Primer (5Ј33Ј)

Mouse COX-2 TGAGCAACTATTCCAAACCAGC GCACGTAGTCTTCGATCACTATC HPRT GCAGTACAGCCCCAAAATGG AACAAAGTCTGGCCTGTATCCAA IL-10 ACCTGCTCCACTGCCTTGCT GGTTGCCAAGCCTTATCGGA iNOS GTTCTCAGCCCAACAATACAAGA GTGGACGGGTCGATGTCAC S100A8 CCGTCTTCAAGACATCGTTTGA GTAGAGGGCATGGTGATTTCCT S100A9 ATACTCTAGGAAGGAAGGACACC TCCATGATGTCATTTATGAGGGC S100A8 PRO AGAGCGTTGTCTCCATAGCC CGGAGCTGACAAAGGATATGT S100A8 EXON2 CCTTGAGCAACCTCATTGATG CCTCCTCACCTGCACAAACT

Human S100A8 GGGATGACCTGAAGAAATTGCTA TGTTGATATCCAACTCTTTGAACCA S100A9 GTGCGAAAAGATCTGCAAAATTT GGTCCTCCATGATGTGTTCTATGA S100A12 CTGCTTACAAAGGAGCTTGCAA GGCCTTGGAATATTTCATCAATG ␤-actin CATGTACGTTGCTATCCAGGC CTCCTTAATGTCACGCACGAT

inflammatory conditions because such conditions would likely cells and promoted resolution of a chronic viral infection (re- Downloaded from have elevated systemic levels. viewed in Ref. 27). IL-10 inhibits adaptive T cell-mediated responses, primarily by Our earlier studies showed that S100A8 expression in murine mac- anti-inflammatory functions through effects on Ag-presenting den- rophages induced by LPS is a late-phase response that is neutralized dritic cells and in the generation of regulatory T cells. It down- by anti-IL-10 Abs, although IL-10 does not directly induce the gene regulates proinflammatory cytokine production in macrophages. (28). Moreover, corticosteroids potentiate S100A8 induction by LPS On the other hand, IL-10 can stimulate B lymphocyte proliferation in an IL-10-dependent manner (29), suggesting that S100A8 may http://www.jimmunol.org/ and activation of NK cells and CD8ϩ CTLs (reviewed in Ref. 25). have anti-inflammatory properties. IL-10 down-regulates release of Recent studies implicate IL-10 in the pathogenesis of viral persis- reactive oxygen and nitrogen species in macrophages (30) and this tence, possibly because it induces a generalized immunosuppres- may be facilitated by S100A8, which we showed to be a potent scav- sion. Clinical observations indicate that a diminished capacity to enger of peroxide and hypochlorite generated by the myeloperoxidase produce IL-10 is associated with improved viral clearance and an- system (31). These effects are reminiscent of induction of heme ox- tiviral immunity can be promoted by targeting the IL-10 signaling ygenase 1, the rate-limiting enzyme in heme catabolism by IL-10, that pathway or by neutralizing IL-10 (reviewed in Ref. 26). Blockade generates carbon monoxide and other metabolites that contribute to of IL-10 receptor prevented functional exhaustion of memory T the anti-inflammatory responses attributed to IL-10 (32). by guest on September 27, 2021

FIGURE 1. Induction of S100A8 mRNA by poly(I: C), poly(C), and poly(I). A, RAW 264.7 macrophages were incubated with 2.0, 10, 50, or 100 ␮g/ml poly(I:C), poly(C), or poly(I). Controls include untreated or 20 ng/ml LPS. Cells were harvested 24 h poststimulation and S100A8 mRNA was quantitated by real-time RT- PCR; HPRT was endogenous control. Data represent means (relative to HPRT mRNA levels) Ϯ SD of du- plicate measurements from at least three independent p Ͻ 0.05 compared with relative ,ء .experiments S100A8:HPRT mRNA ratio of untreated cells. B, S100A8 in RAW cell supernatants quantitated by ELISA. RAW cells were stimulated with poly(I:C), poly(C), poly(I)(10 or 50 ␮g/ml), or LPS (20 or 100 p Ͻ 0.01 compared with untreated ,ءء .ng/ml) for 36 h cells. C, RAW cells were stimulated with 10 ␮g/ml poly(I:C) Ϯ 20 ng/ml LPS for 24 h. mRNA levels rel- ative to HPRT were quantitated. Data represent means Ϯ SD of duplicate measurements from at least three independent experiments. Ctr, Control. 2260 IL-10 MEDIATES S100A8 INDUCTION BY dsRNA

In this study, we demonstrate induction of S100A8 in murine macrophages and of S100A8 and S100A9 in human monocytes by poly(I:C), a dsRNA mimetic. Induction was via a novel IL-10 and PKR-dependent pathway and required p38 and ERK MAPK. S100A8 was expressed in lungs of mice infected with influenza virus and declined early in the recovery phase, implying down- regulation by mediator(s) up-regulated during resolution. Since S100A8/S100A9 levels are likely to be maintained in conditions where IL-10 is raised, these proteins may contribute to viral per- sistence in patients infected by some RNA viruses. Materials and Methods Reagents LPS (Escherichia coli 0111:B4), poly(I:C), poly(I), and poly(C) (all po- tassium salts) were from Sigma-Aldrich. Pathway inhibitors U0126 and SP600125 were from Calbiochem; all others were from Sigma-Aldrich. TRIzol reagent, glycogen, and a SuperScript III First-Strand Synthesis sys- tem were from Invitrogen; other RNA extraction reagents were from Sigma- Aldrich and all reagents for real-time RT-PCR were from Invitrogen, ex- cept for TURBO-DNase which was from Ambion). Affinity-purified rabbit Downloaded from anti-murine IL-10, anti-human IL-10 mAb, and mouse IgG2b isotope con- trol were from R&D Systems. Monocyte and macrophage preparation FIGURE 2. S100A8 induction is IL-10 dependent. RAW cells were Tissue culture medium for all experiments was filtered through 0.22-␮m treated with 20 ␮g/ml poly(I:C) for the indicated times. A, S100A8 or Zetapore filters (Cuno) to remove traces of endotoxin. Medium and medi- ators were only used if endotoxin levels were Ͻ20 pg/ml (chromogenic IL-10 mRNAs were quantitated as given in Fig. 1. Maximum mRNA in- Limulus amebocyte assay; Associates of Cape Cod) (4, 29). Bovine calf duction by poly(I:C) was denoted as 100% maximal response. B, RAW http://www.jimmunol.org/ serum (HyClone) had been pretested before purchase. cells were untreated or stimulated with poly(I:C) (50 ␮g/ml); cells were Murine RAW 264.7 macrophages were cultured as previously described coincubated with IL-10 (10 ng/ml) or anti-IL-10 mAb (10 ng/ml) and (28) and activated 24 h after seeding. Animal experiments were done with mRNA levels relative to HPRT were quantitated. Data represent S100A8 the approval of the Animal Care and Ethics Committees (University New mRNA relative to poly(I:C)-stimulated samples without IL-10; means Ϯ Ϫ/Ϫ South Wales, Australia). C57BL/6 mice or C57BL/6-IL-10 mice (33), SD of three separate experiments. C, S100A8 in supernatants of RAW cells 6–8 wk old, were maintained under specific-pathogen free conditions. Res- cotreated with IL-10 (10 ng/ml) with or without poly(I:C) (50 ␮g/ml) for ident peritoneal cells were lavaged with cold RPMI 1640, washed, and 36 h. Data represent S100A8 protein levels relative to poly(I:C)-stimulated 0.5 ϫ 106 cells in 500 ␮l of RPMI 1640 containing 2% bovine calf serum, samples; means Ϯ SD of duplicate measurements from at least three in- 2mML-glutamine, and penicillin/streptomycin dispensed into 24-well Ͻ ء o plates (Nunc). Plates were incubated for2hat37Cin5%CO in air, dependent experiments. , p 0.05 compared with poly(I:C) alone. D, by guest on September 27, 2021 2 Ϫ Ϫ washed three times with warm PBS to remove nonadherent cells, and equil- Peritoneal macrophages from wild-type C57BL/6 mice (Ⅺ) or IL-10 / ibrated in the same medium overnight. Medium was replenished before mice (f) untreated or stimulated with LPS (20 ng/ml) or poly(I:C) (50 activation. Populations contained ϳ98% macrophages (ϳ98% viable by ␮g/ml) for 24 h and mRNA levels relative to HPRT were quantitated. Data trypan blue exclusion) and ϳ0.3% neutrophils by differential staining. represent S100A8 mRNA relative to poly(I:C)-stimulated samples; p Ͻ ,ءء p Ͻ 0.05 and ,ء .Human monocytes isolated from peripheral blood of healthy subjects as means Ϯ SD of three separate experiments given previously described (29) were analyzed using a Beckman Coulter counter. 0.01 compared with wild-type mice. Preparations generally contained ϳ10% monocytes, ϳ90% lymphocytes, and Ͻ1.5% granulocytes. PBMC (2.5 ϫ 106/well) were dispensed into 24-well plates (Costar), then incubated in RPMI 1640 containing 10% pared by performing reverse transcription reactions in the absence of re- heated (56°C, 30 min) autologous serum, 2 mM L-glutamine, and penicil- verse transcriptase. PCR amplification was performed with Platinum lin/streptomycin at 37°C in 5% CO in air for 4 h. After rinsing with warm 2 SYBR Green qPCR SuperMix UDG. Reactions were performed in dupli- medium to remove nonadherent lymphocytes, monocytes were cultured cate, and contained 2ϫ SYBR Green qPCR SuperMix, 1 ␮l of template overnight in the same medium. To generate macrophages, monocyte cul- cDNA or control, and 100 nM primers (Table I) in a final volume of 25 ␮l ture in 24-well Costar plates was continued for 7 days; cells were replen- and analyzed in 96-well optical reaction plates (Applied Biosystems). Re- ished with fresh medium with 10% autologous serum on day 4. Monocytes/ actions were amplified and quantified using an Applied Biosystems 7700 macrophages were activated with particular stimulants after replenishing sequence detector with standard cycle conditions and the Applied Biosys- with fresh culture medium. For experiments using pathway inhibitors, cells tems software. Relative quantities of mRNA in duplicate samples were were generally pretreated with inhibitors for 30 min, then untreated or obtained using the ⌬⌬C method and normalized against murine hypox- stimulated for 24 h, unless specifically detailed. T anthine-guanine phosphoribosyltransferase (HPRT) or human ␤-actin as Human lung epithelial A549 cell culture endogenous controls. Human A549 pulmonary type II epithelial cells, used to study gene induc- Dual luciferase reporter assay tion by viral infection (34), were obtained from Prof. P. Thomas (Depart- ment of Respiratory Medicine, Prince of Wales Hospital, Sydney, Austra- Construction of the murine S100A8 promoter-luciferase-fused reporter lia). Cells were cultured in F12 medium (Invitrogen) containing 10% FCS, plasmids was described previously (28). RAW 264.7 cells were tran- ␮ 4 100 IU penicillin/ml, and 100 g/ml streptomycin at 37°C in 5% CO2 in siently transfected as described elsewhere(35) using 3.0 ϫ 10 cells/500 air. Cells (2 ϫ 105) were seeded into 12-well plates (Nunc) in fresh me- ␮l seeded into 24-well plates (Nunc) 72 h before transfection, and then dium containing 2% FCS and stimulated with the indicated dose of 0.5 ␮g of luciferase reporter plasmid or 0.05 ␮g of reference plasmid poly(I:C) or LPS for 4 or 24 h. (pRL-TK) was transfected in the presence of DEAE-dextran (300 ␮g/ RNA preparation and analysis ml; Sigma-Aldrich). After 24 h, cells were stimulated for 20 h with poly(I:C), and firefly and Renilla luciferase activities were assayed with Adherent macrophages grown as monolayers in 24-well Costar plates were 15 ␮l of extract using the Luciferase Assay System (Promega) accord- lysed directly after stimulation with TRIzol reagent and RNA prepared as ing to the manufacturer’s instructions and measured using a TD-20/20 described elsewhere (29). Total RNA (1 ␮g) was treated with TURBO Luminometer. Results are expressed as mean Ϯ SD of luciferase ac- DNase and reverse transcribed using random hexamers and the Superscript tivity normalized to the reference Renilla luciferase from three separate III First-Strand Synthesis System for RT-PCR according to the manufac- experiments. Database searching (TFSERCH; http://www.cbrc.jp/ turer’s instructions. Negative controls (no first-strand synthesis) were pre- research/db/TFSEARCH.html) was used to search putative consensus The Journal of Immunology 2261 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 3. Poly(I:C) induces S100 mRNA in human monocytes and macrophages and IL-10 is involved. A, Monocytes were untreated or stimulated p Ͻ 0.05, compared with relative ,ء .with LPS (500 ng/ml) or poly(I:C) (50 ␮g/ml); means Ϯ SD of duplicate measurements of cells from five donors S100:␤-actin mRNA ratios of control. B, Monocytes were untreated or stimulated with poly(I:C) (50 ␮g/ml) with or without IL-10 (2 or 10 ng/ml). Data p Ͻ 0.05, compared with ,ء .represent S100 mRNA relative to poly(I:C)-stimulated samples without IL-10; means Ϯ SD of three separate experiments relative S100:␤-actin mRNA ratios of poly(I:C) alone. Human macrophages (C and D) or monocytes (E and F) were untreated or stimulated with LPS (500 ng/ml) or poly(I:C) (50 ␮g/ml) with or without anti-IL-10 mAb (5 ␮g/ml) or IgG isotype control for 24 h, or pretreated with 2-AP (4 mM) for 4 h. S100A8 (C and E) and S100A9 (D and F) mRNA levels relative to ␤-actin mRNA were quantitated. Data represent means Ϯ SD of duplicate measurements and are representative of two to three independent experiments. IL-10 mRNA levels relative to ␤-actin mRNA were also quantitated. Ctr, Control.

binding sequences of transcription factors in the murine S100A8 8.1), 167 mM NaCl, 1.2 mM EDTA, 0.01% SDS, and 1.1%Triton promoter. X-100) supplemented with Complete protease inhibitor mixture (Roche), and immunoprecipitated overnight at 4°C using the Ab-coated Chromatin immunoprecipitation (ChIP) assays beads. Magnetic-separated beads were washed five times with high-salt RAW 264.7 cells (5 ϫ 106) were untreated or stimulated with 50 ␮g/ml radioimmunoprecipitation assay wash buffer (50 mM HEPES (pH 7.9), poly(I:C) and harvested between 30 min and 8 h. Chromatin was cross- 500 mM LiCl, 1 mM EDTA, 1% Nonidet P-40, and 0.7% sodium de- lined with 1% formaldehyde for 10 min at room temperature. Cells were oxycholate) and once with Tris-EDTA buffer. Extraction of DNA and ␮ washed twice with PBS, then lysed (5 mM PIPES (pH 8.0), 85 mM KCl, reversal of cross-linking were performed in 150 l of elution buffer (50 and 0.5% Nonidet P-40) for 10 min on a rocking platform. Nuclei were mM Tris-HCl (pH 8.0), 10 mM EDTA, and 1% SDS) supplemented ␮ obtained by centrifugation at 1000 ϫ g for 5 min, then resuspended in with proteinase K (50 g/ml) in a hybridization oven with rotation at nuclei lysis buffer (50 mM Tris-HCl (pH 8.1), 10 mM EDTA, and 1% 65°C for 2 h. The DNA was purified using a PCR purification kit SDS) and sonicated six times with 15-s pulses followed by 45-s recov- (Qiagen). ery periods at 60% output (Knotes Microultrosonicator). Protein G Enriched DNA was detected by real-time quantitative PCR (LightCycle Dynabeads (Invitrogen) were washed with block solution (0.5% BSA in 480; Roche) and normalized to 10% of each input as previously described (36). 1ϫ PBS) and incubated with 2 ␮g against C/EBP-␤ Ab (Santa Cruz The following regions (relative to the transcription start site determined by Biotechnology) or normal rabbit IgG (-Aldrich) for 6 h with rotation, aligning reference RNA and genomic sequences in GenBank) were amplified then washed three times with block solution. The purified chromatin using specific primer pairs (sequences listed in Table I): S100A8 PRO (ϩ14 to was diluted 10-fold with ChIP dilution buffer (16.7 mM Tris-HCl (pH Ϫ82) and S100A8 exon 2 (ϩ493 to ϩ621). 2262 IL-10 MEDIATES S100A8 INDUCTION BY dsRNA Downloaded from http://www.jimmunol.org/ FIGURE 4. PKR is implicated in S100A8 and IL-10 up-regulation. A, RAW cells were untreated or pretreated with 2-AP (4 mM) and then untreated or stimulated with poly(I:C) (50 ␮g/ml) for 8 h (IL-10 mRNA; f) or 24 h (S100A8 mRNA; Ⅺ). B, S100A8 in supernatants of RAW cell pretreated with 2-AP, then stimulated with poly(I:C) for 36 h. Data represent the S100A8 protein level relative to poly(I:C)-stimulated samples; means Ϯ SD of duplicate p Ͻ 0.05 compared with poly(I:C) alone. C, RAW cell pretreated with 2-AP were untreated ,ء .measurements from at least three independent experiments or stimulated with poly(I:C) in the presence (f) or absence (Ⅺ) of IL-10 (10 ng/ml) for 24 h. mRNA levels relative to HPRT were quantitated. For A and p Ͻ 0.05 ,ء .C, data represent mRNA levels relative to poly(I:C)-stimulated samples without 2-AP; means Ϯ SD of three separate experiments are given compared with poly(I:C) alone. D, Human monocytes were pretreated with 2-AP (4 or 10 mM), then untreated or stimulated with poly(I:C) (50 ␮g/ml). p Ͻ ,ءء p Ͻ 0.05 and ,ء .Data represent mRNA levels relative to poly(I:C)-stimulated samples without 2-AP; means Ϯ SD of three separate experiments 0.01 compared with relative S100A8:␤-actin mRNA ratios of poly(I:C) alone. E, Human macrophages were untreated or stimulated with LPS (500 ng/ml) by guest on September 27, 2021 or poly(I:C) (50 ␮g/ml) with or without 2-AP (4 mM) for 4 h. IL-10 mRNA levels relative to ␤-actin mRNA were quantitated.

ELISA Statistical analysis A double sandwich ELISA was used to detect murine S100A8 as described Values in figures are expressed as means Ϯ the SD of the indicated number elsewhere (35, 37) using recombinant S100A8 as standard (38). of observations. Statistical analyses were performed using the Student t test. Influenza virus infection of BALB/c mice and localization of S100A8 in lung Results Viral stock and infection procedures were as described previously (39). Synthetic dsRNA induces S100A8, but not S100A9, in murine Briefly, stocks of influenza virus A/Japan/305/57 (A/Jap, H2N2), a nega- tive ssRNA virus, were grown in embryonated eggs. Virus-containing al- RAW cells lantoic fluid was harvested and stored in aliquots at Ϫ70°C. Virus content This study was initiated because our initial attempts to silence was determined by hemagglutination assay using erythrocytes from Gallus induction of the S100A8 gene in murine RAW 264.7 macrophages domesticus. Influenza virus infection was established by inoculating 2,2,2- tribromoethanol (Avertin)-anesthetized BALB/c female mice, 6–8 wk old, with small interfering RNAs (siRNAs) were unsuccessful. In fact, intranasally with 50 hemagglutination units of virus. Mice were weighed transfection of enzymatically synthesized S100A8 siRNAs (four before infection and then daily. Survival was monitored for 30 days. constructs) potentiated S100A8 mRNA levels induced by LPS 21- Procedures for immunohistochemistry were as described elsewhere to 180-fold (data not shown). In the absence of stimulation, trans- (40). Briefly, formalin-fixed tissue samples from murine lung embedded in fection of enzymatically synthesized GAPDH siRNA suppressed paraffin were sectioned onto poly-L-lysine-coated slides and stained with H&E for routine morphology. Rabbit anti-murine S100A8 IgG and control GAPDH mRNA by 62%, compared with untransfected RAW IgG were used; in another control, the primary Ab was omitted. The Abs cells. However, enzymatically synthesized GAPDH siRNA caused to murine S100A8 do not cross-react with S100A9 (38). Ag retrieval was 28-fold induction of S100A8 mRNA (data not shown), indicating performed by immersion in 0.01 M citrate buffer (pH 6.0) in a water bath broad induction of the gene by enzymatically synthesized siRNA. at 95°C for 20 min and cooled to room temperature while still immersed in We next tested whether S100 proteins were up-regulated by buffer (41). After quenching with 3% H2O2 and treating with primary Ab (dilution of stock solution, 1/500 to 1/4000, v/v) at room temperature for stimulation of murine macrophages with synthetic dsRNA. 1 h, biotin-conjugated secondary Ab and streptavidin-conjugated HRP Poly(I:C) at 2 ␮g/ml did not induce ; 10 ␮g/ml ϩ from an LSAB kit (DakoCytomation) were applied to sections for 20 min increased S100A8 mRNA levels ϳ 31-fold and ϳ466-fold with 50 Ј at room temperature to amplify the Ag signal for subsequent 3,3 -diami- ␮ ␮ nobenzidine staining. Known positive controls (murine bone marrow cells) g/ml (Fig. 1A). Although 100 g/ml poly(I:C) increased mRNA were stained in each run and runs were duplicated on different days to further, this dose can cause some toxicity, and 50 ␮g/ml was used confirm repeatability. Sections were counterstained with hematoxylin. in subsequent experiments. S100A8 in supernatants (0.48 Ϯ The Journal of Immunology 2263 Downloaded from http://www.jimmunol.org/

FIGURE 5. Pathways involved in poly(I:C)-activated S100A8 induction. A, RAW cells were untreated or stimulated with poly(I:C) (50 ␮g/ml) with or without anti-IFN-␤ mAb (4000 neutralization units/ml), One neutralization unit is the amount of antiserum required to neutralize1Uofmouse IFN-␤. Eight hours poststimulation, iNOS mRNA (Ⅺ) or S100A8 mRNA (f) levels relative to HPRT were quantitated. B, To determine the effects of IFN-␤, RAW cells ␤ were incubated with the indicated dose of IFN- (U/ml) with or without poly(I:C). Data represent S100A8 mRNA relative to poly(I:C)-stimulated samples; by guest on September 27, 2021 ,(p Ͻ 0.05 compared with poly(I:C) alone. C, RAW cells were pretreated with DMSO (vehicle control ,ء .means Ϯ SD of three separate experiments given NS389 (10 (M), U0126 (2.5 ␮M), SB202190 (2.5 ␮M), and SP600125 (10 ␮M), then untreated or stimulated with poly(I:C) (50 ␮g/ml); mRNA levels relative to HPRT were quantitated. Data represent S100A8 mRNA relative to poly(I:C)-stimulated samples without inhibitors; means Ϯ SD of three separate p Ͻ 0.05 compared with relative S100A8:HPRT mRNA ratios of poly(I:C) stimulation alone. D, RAW cells were pretreated with ,ء .experiments given SP600125 (10 ␮M), then untreated or stimulated with poly(I:C) (50 ␮g/ml) for 4 h; mRNA levels relative to HPRT were quantitated. Data represent COX-2 (Ⅺ) or IL-10 (f) mRNA relative to poly(I:C)-stimulated samples without inhibitors; means Ϯ SD of three separate experiments given.

0.68 ␮g/106 cells with 10 ␮g/ml poly(I:C) and 1.98 Ϯ 0.72 ␮g mRNA increases ϳ30-fold above control levels, then declined with 50 ␮g/ml poly (I:C)) reflected this increase; the latter were over 16 h (Fig. 2A). IL-10 significantly enhanced the poly(I:C)- 3-fold higher than protein levels in supernatants and reflected activated response ϳ5- fold at the mRNA (Fig. 2B) and ϳ3-fold at mRNA levels from cells stimulated with 20 ng/ml LPS (Fig. the protein level (Fig. 2C). An optimized dose (28) of a neutral- 1B), a control, positive activator (29). Poly(I:C) markedly am- izing anti-IL-10 mAb suppressed relative S100A8 mRNA levels plified S100A8 mRNA when cultured together with LPS (Fig. by ϳ70% (Fig. 2B). To confirm the role of IL-10, peritoneal mac- 1C). S100A8 mRNA levels in RAW cells stimulated with 10 rophages from IL-10Ϫ/Ϫ mice were tested. Fig. 2D shows signif- ␮g/ml poly(I:C) alone or 20 ng/ml LPS alone were 4- and 49- icant S100A8 mRNA induction by LPS and poly(I:C) in macro- fold above control levels, respectively; when combined, induc- phages from wild-type mice. In marked contrast, induction by tion increased ϳ447-fold. poly(I:C) or LPS in IL-10 Ϫ/Ϫ macrophages was significantly less Fig. 1A shows that synthetic ssRNA at 50 ␮g/ml induced only than that in wild type ( p Ͻ 0.01 compared with wild-type mac- low amounts of S100A8; poly(C) and poly(I) increased mRNA rophages). S100A8 mRNA levels induced by these agents in IL- levels up to ϳ9- and ϳ29-fold above control, respectively, whereas 10Ϫ/Ϫ macrophages were elevated ϳ2-fold, but not significantly poly(I:C) promoted a ϳ470-fold increase. S100A9 mRNA was not different to those in unstimulated samples. This confirmed that induced in RAW cells by poly(I:C) at any time point or in any ex- S100A8 induction by LPS and poly(I:C) was IL-10 dependent. periment (data not shown). Mechanisms of gene induction of these S100s in human mono- cytes/macrophages have not been as carefully characterized as in mu- S100A8 induction by poly(I:C) is IL-10 dependent rine macrophages. S100A8 mRNA was induced with poly(I:C) and IL-10 does not directly induce S100A8 mRNA but is essential for, by LPS in human monocytes (Fig. 3A), but unlike RAW cells (Fig. and increases, the LPS-activated response (28). S100A8 mRNA 1C), the combination of poly(I:C) and LPS was not synergistic (data induction by poly(I:C) in RAW cells was evident after 8 h, max- not shown). Poly(I:C) also increased S100A9 and S100A12 mRNA; imal at 24 h, and gradually declined over 48 h, whereas IL-10 poly(I:C)- or LPS-induced S100A12 mRNA levels were 20- to 40- mRNA induction was evident within 2 h and maximal at 8 h, with fold less than those of S100A8 or S100A9 (Fig. 3A). Ratios of 2264 IL-10 MEDIATES S100A8 INDUCTION BY dsRNA

FIGURE 6. Poly(I:C)-responsive regions in the murine S100A8 promoter. A, RAW cells were transiently cotransfected with pRT-TK-luciferase and full-length (Ϫ917/465) or a series of 5Ј deletion S100A8 promoter-luciferase reporters. pGL-basic plasmids (promoterless) were transfected as controls. The medium from transfected cells was replaced and left untreated (Ⅺ) or stimulated with poly(I:C) (25 ␮g/ml; f) for 16 h. Firefly luciferase activities Downloaded from in cell extracts were analyzed against Renilla luciferase. Data represent means Ϯ SD of duplicates and are representative of three experiments. B, Time course of ChIP assays indicated binding of C/EBP to the S100A8 promoter (f) in RAW cells after stimulation with 50 ␮g/ml poly(I:C). The same region was not enriched by IgG control (Ⅺ); amplification of a region located in exon 2 (u) was used as specificity control. Results are presented as means Ϯ SD of duplicate measurements and are representative of two to three independent experiments.

␤

S100A9 mRNA levels relative to S1008 in monocytes stimulated with cells were incubated with poly(I:C) in the presence of anti-IFN- http://www.jimmunol.org/ LPS or poly(I:C) varied among individuals (from 0.2 to 5; n ϭ 5). mAb; the dose was initially optimized by measuring effects on IL-10 enhanced poly(I:C)-activated S100A8 and S100A9 mRNA in- inducible NO synthase (iNOS) mRNA induction by poly(I:C). Fig. duction 2- to 3-fold (Fig. 3B). In this system, IL-10 directly induced 5A shows that anti-IFN-␤ suppressed iNOS mRNA by 81.1%, low levels of S100A9 mRNA, although these were not significantly whereas S100A8 mRNA levels were reduced by only 37%, al- more than levels in unstimulated monocytes. though this was statistically less ( p Ͻ 0.05) than with poly(I:C) Fig. 3, C and D, shows that induction of S100A8 and S100A9 alone. S100A8 mRNA levels were potentiated by 20 or 100 U/ml mRNAs in human macrophages by LPS and by poly(I:C) was IFN-␤, increasing 1.9- or 2.6-fold (Fig. 5B). Thus, IFN-␤ may suppressed by anti-IL-10. Similarly, poly(I:C) induced both genes contribute to the poly(I:C)-mediated induction of S100A8 and can in human monocytes and anti-IL-10 reduced mRNA to control synergize with dsRNA and LPS. by guest on September 27, 2021 levels (Fig. 3, E and F), strongly supporting the findings with LPS induction of S100A8 in murine macrophages is also PGE2/ murine macrophages that induction was via an IL-10-dependent cAMP dependent (28). We next tested whether the poly(I:C) response mechanism. was dependent on cyclooxygenase-2 (COX-2) using the specific in- hibitor NS398, which reduces the LPS-stimulated response by 60% PKR mediates poly(I:C)-induced IL-10 and S100A8 mRNA (28); however, changes in mRNA levels were not significant (Fig. Poly(I:C) can also signal through a TLR3-independent pathway 5C). Among the MAPK inhibitors, the JNK inhibitor SP600125, via PKR (reviewed in Ref. 42). To test its involvement, 2-amin- which partially blocked COX-2 and IL-10 mRNA induction (Fig. opurine (2-AP), which competes for ATP at the ATP binding site 5D), did not suppress poly(I:C)-induced S100A8 mRNA but of PKR and thereby inhibits its autophosphorylation (43), was caused some enhancement (Fig. 5C). As reported for induction by used. S100A8 mRNA levels induced by poly(I:C) in RAW cells LPS (29), Fig. 5C shows that the p38 (SB202190) and ERK in- were significantly inhibited by 2-AP by 90% (Fig. 4A) and protein hibitors (U0126) almost abolished S100A8 induction by poly(I:C). levels by 90% (Fig. 4B). IL-10 mRNA was almost abolished by These results indicate that S100A8 mRNA induction by poly(I:C)

2-AP by 8 h poststimulation (Fig. 4A). Suppression of S100A8 was likely to be PGE2 independent but dependent on a p38- and mRNA induction in RAW cells by 2-AP was restored to levels ERK-mediated pathway; the JNK pathway was unlikely to be induced by poly(I:C) and were ϳ43% of the levels induced by involved. poly(I:C) plus exogenous IL-10 (Fig. 4C). Fig. 4D shows the sig- nificant inhibition of S100A8 and S100A9 induction by poly(I:C) Identification of poly(I:C)-responsive regions in the S100A8 in human monocytes by 2-AP by 82 and 69%, respectively; 2-AP promoter alone did not directly alter mRNA levels. Similarly, IL-10 levels To examine mechanisms of transcriptional regulation of the induced by poly(I:C) or LPS were totally suppressed by 2-AP S100A8 gene by poly(I:C), 5Ј-flanking sequences upstream of the treatment of these cells (Fig. 4E). Interpretation of suppression of transcription initiation site, untranslated exon 1, intron 1, and se- S100A12 mRNA was difficult because of its low expression levels. quences upstream of the translation start site on exon 2 were used These experiments strongly indicate that PKR may contribute to to evaluate activities of deletion constructs after transient transfec- S100A8/S100A9 gene up-regulation via its effects on IL-10 rather tion into RAW cells. Fig. 6 shows that levels of luciferase activity than through direct activation by dsRNA. after poly(I:C) stimulation were similar for all positive constructs, with 5- to 11-fold increases compared with unstimulated cells. The Other signaling pathways region Ϫ94 to Ϫ34 bp contained the essential promoter because its Because dsRNA activates production of type 1 IFNs, we tested deletion completely abrogated luciferase activity. The region their potential involvement in S100A8 expression. IFN-␤ en- Ϫ178 to Ϫ94 bp was responsible for luciferase activity in hanced S100A8 mRNA induction by LPS (data not shown). RAW poly(I:C) enhancement because deletion strongly reduced activity The Journal of Immunology 2265

FIGURE 7. S100A8 expression in lung tissues from mice infected with influenza A virus and human alveolar A549 epithelial cells. A–E, Lungs were harvested at the day of infection (A) or 5 days (B); arrows indicate pos- itive neutrophil staining) 6 days (C), 8 days (D), or 12 days (E) postinfec- tion. Samples were stained with anti- murine S100A8 IgG. Immunostaining in tissues treated with a negative con- trol IgG were similar to samples

shown in A (data not shown). Original Downloaded from magnification, ϫ1000. F, Human A549 epithelial cells were incubated with 2.5, 10, 25, or 50 ␮g/ml poly(I: C). Controls were untreated (Ctr) or treated with 20 ng/ml LPS. Cells were harvested 24 h poststimulation and

S100A8 mRNA was quantitated by http://www.jimmunol.org/ real-time RT-PCR; ␤-actin mRNA was endogenous control. Data are means Ϯ SD of duplicate measure- ments and are representative of three independent experiments. by guest on September 27, 2021

and lost enhancement. Consensus motifs for a number of transcrip- phase (days 6 and 7), patchy staining of S100A8ϩ interstitial cells tion factors, including C/EBP and E26 transformation specific, are was seen, in particular within interstitial lesions (Fig. 7C). At day located within this region. Constructs not containing the first exon 8, peribronchiolar lesions were strongly S100A8ϩ, particularly and intron (Ϫ178/ϩ1 bp) generated positive, although somewhat around the rims of the bronchioles (Fig. 7D). Destruction of the weak luciferase activities and may contain elements essential for epithelial lining was obvious at this time (data not shown). At day gene induction by poly(I:C). 12, when mice had recovered, no cells strongly expressing S100A8 Because C/EBP␣ and -␤ bind a similar region in the human were seen (Fig. 7E). This time course suggests that S100A8 ex- S100A9 promoter (44, 45) and to further confirm functional ac- pression may reflect the severity of influenza infection. tivity of this region, ChIP assays were performed. Fig. 6B indicates Airway epithelial cells express TLR3 (46) and because epithe- specific time-dependent binding of C/EBP␤ to the Ϫ178/ϩ1 re- lial cells lining the airways of infected mice expressed S100A8, we gion in response to poly(I:C), whereas there was little enrichment tested its induction in lung epithelial A459 cells. Induction was of DNA in exon 2, located approximately ϩ493 bp downstream of low after 6 h (data not shown) and maximal 24 h poststimulation the promoter region and used as a specificity control. with 10 ␮g/ml poly(I:C) but decreased with higher amounts (Fig. 7F). However, maximal induction in epithelial cells was ϳ104 Influenza A virus induced S100A8 in epithelial cells in murine times lower than responses induced in monocyte or macrophages lung in vivo when normalized to ␤- actin (cf with Fig. 3, C and E). To assess whether S100A8 was up-regulated by a viral infection in vivo, we infected mice with influenza A virus (an RNA virus) and Discussion examined lungs histologically over a time course. Anti-S100A8 Serum levels of S100A8/S100A9 are elevated in patients with viral did not react with normal lung epithelial cells or interstitial cells infections (17–20), but mechanisms regulating this are unknown. (Fig. 7A) but, as expected, neutrophils within blood vessels were Induction of S100A8 through activation of TLR4 by LPS is well strongly reactive (Fig. 7B). Increased numbers of neutrophils and studied in murine macrophages (28, 35). This study provides the S100A8ϩ mononuclear cells were seen in interstitial tissue (Fig. first evidence that S100 proteins can be induced in monocytes/ 7B); the cytoplasm of epithelial cells reacted positively and was macrophages by RNAs. In this study, we show that murine and independent of the localization of neutrophils. During the acute human monocytes and macrophages stimulated with dsRNA 2266 IL-10 MEDIATES S100A8 INDUCTION BY dsRNA strongly expressed S100A8 mRNA. Synthetic ssRNAs also in- invasion (56). Similar to LPS-activated RAW cells (35), IFN-␤ duced S100A8 mRNA, although this was comparatively weak. Ac- synergized with dsRNA to enhance S100A8 mRNA induction tivation was reflected by the elevated secreted levels of S100A8 (Fig. 5B). PKR is an IFN-inducible gene but its expression is cell from activated RAW cells. As reported for several activators of the type dependent (57). RAW cells constitutively express PKR, gene in murine macrophages (28, 35), S100A9 was not coex- whereas poly(I:C) or IFN-␤ induce it in numerous other cells pressed in murine macrophages but there was strong induction in (57, 58). the human cells, again confirming a divergence of S100A9 gene Poly(I:C) stimulates rapid activation of ERK, JNK, and p38 in regulation in rodents and humans. Similarly, glucocorticoids only RAW cells (59). Similar to LPS (35), dsRNA-induced S100A8 amplify LPS-induced S100A8 in murine macrophages, whereas was dependent on p38 and ERK MAPK (Fig. 5C). Moreover, the S100A8 and S100A9 are both directly up-regulated in human PKR inhibitor blocks numerous antiviral responses (60) including monocytes and macrophages (29). S100A8 mRNA induced by reduction of MEK6 and p38 MAPK phosphorylation in RAW dsRNA increased in synergy with LPS in murine macrophages, cells, since the p38 MAPK activator MEK6 has increased affinity whereas no synergy was seen with human monocytes, possibly due for PKR, forming a catalytic complex following exposure of cells to different expression levels of TLR3 and TLR4 (and/or CD14) to dsRNA (61). Although there are other possible mechanisms, between these cell types. such as inhibition of other signaling molecules, this mechanism Kinetic studies showed that dsRNA-induced S100A8 mRNA in could be involved in induction of IL-10 and S100A8, because the RAW cells was maximal 24 h poststimulation, similar to the time p38 pathway was crucial for both (ref. 62 and Fig. 5C). On the of optimal expression following LPS stimulation (29), suggesting other hand, pharmacological inhibition of JNK enhanced dsRNA- a secondary event. Induction of S100A8 in murine macrophages induced S100A8 mRNA (Fig. 5C), suggesting that JNK has a reg- Downloaded from by LPS is IL-10 and COX-2 dependent (28). IL-10 mRNA was ulatory role. Interestingly, JNK inhibition increases TNF-␣ levels also induced by dsRNA, peaking 8 h poststimulation (Fig. 2A). following dsRNA treatment in epithelial cells and enhances p38 IL-10 alone only slightly induced S100A8 mRNA (2.5-fold), but MAPK phosphorylation, suggesting that activation of JNK by synergistically enhanced dsRNA-induced S100A8 mRNA and dsRNA negatively regulates TNF-␣ induction (63). Its negative protein (Fig. 2, B and C). Moreover, inhibition of endogenous effect on S100A8 regulation is indirectly supported by the

IL-10 by anti-IL-10 mAb substantially reduced S100A8 mRNA strong induction of S100A8 and S100A9 seen in the epidermis http://www.jimmunol.org/ levels in murine and human macrophages (Figs. 2B and 3C). Im- of JunB/c-Jun double-knockout mice (64), because activated portantly, S100A8 mRNA levels induced by dsRNA or by LPS JNKs phosphorylate c-Jun to enhance its transcriptional activity were significantly less in macrophages from IL-10Ϫ/Ϫ mice than (65). In contrast to its effects on S100A8 however, the JNK those wild-type mice (Fig. 2D). Together, these results confirm inhibitor suppressed IL-10 mRNA (Fig. 5D), suggesting addi- that S100A8 is an IL-10-dependent secondary response gene prod- tional mechanisms. uct and indicate that IL-10 is a major enhancer, rather than an This study indicated divergence of pathways used by dsRNA inducer of the gene. and LPS to express S100A8. Unlike the LPS-provoked response Recognition of viral dsRNA and of poly(I:C) activates produc- (28), dsRNA-induced S100A8 was independent of COX-2 metab- tion of antiviral factors via multiple pathways including TLR3, olites because suppression of the COX-2 gene had no effect on by guest on September 27, 2021 PKR, and the helicase retinoic acid-induced protein I (46–48). S1008 mRNA levels. Inhibition of JNK attenuates dsRNA-stimu-

Although dsRNAs and LPS are recognized by different TLR re- lated COX-2 mRNA accumulation and PGE2 production by RAW ceptors, there is divergence and convergence in TLR4 and TLR3 cells (59). Because JNK inhibition amplified S100A8 expression signaling. TLR4 signaling can occur via MyD88-dependent and by dsRNA, this supports our finding that the COX-2 pathway is -independent pathways, whereas TLR3 signaling is generally unlikely to be involved. Thus, at least two pathways may indepen- MyD88 independent (49, 50) and the adaptor proteins more re- dently contribute to induction and/or enhancement of S100A8 in stricted in their TLR interactions. PKR can participate in MyD88- macrophages, depending on the stimulant. independent signaling triggered by TLR3 (51) and in MyD88-de- Experiments with a S100A8 luciferase reporter construct con- pendent and -independent pathways by TLR4 (52). Thus, PKR firmed that S100A8 induction by dsRNA is at the transcriptional activation can be common to dsRNA and LPS signaling. In fact, level. Although deletion constructs had various basal responses, LPS activation of murine alveolar macrophages induces rapid possibly due to activation by the transfection procedure, we nar- phosphorylation of PKR (53). The PKR inhibitor 2-AP almost to- rowed the region containing response elements to poly(I:C) to tally abolished IL-10 mRNA induction in murine macrophages Ϫ178 to Ϫ94 of the S100A8 promoter. This is also the reported (Fig. 4), consistent with the reported role for PKR in IL-10 induc- LPS/IL-10 response region (28), which contains consensus se- tion in human PBMC by HIV-Tat stimulation (54). Consequently, quences for c-E26 transformation specific, C/EBP, and atypical S100A8 mRNA and protein levels were also reduced (Fig. 4); STAT motifs. The critical role of IL-10 in dsRNA induction in- S100A9 expression was also reduced in human monocytes, pre- dicates a potential role for STAT binding; cross-talk between Jak- sumably because IL-10 production was suppressed. 2-AP-sup- STAT and TLR pathways in macrophage activation is reported pressed S100A8 induction was restored to just above levels in- (66). C/EBP␣ and -␤ bind the human S100A9 promoter in a my- duced by poly(I:C) alone and the potentiation seen when poly(I:C) eloid differentiation-dependent manner (44, 45) and we identified and IL-10 were added together was not significant (Fig. 4C), sug- several consensus C/EBP boxes that are conserved in the defined gesting that additional mediators such as IFN-␤ may contribute to promoter regions of human and murine S100A8 and S100A9. As full gene expression. Notwithstanding, the results reported here described here for S100A8, poly(I:C) induces iNOS in astroglia indicate a crucial and convergent role for PKR in S100A8 induc- through activation of PKR, which induces p38-MAPK-coupled ac- tion by LPS and by dsRNA. tivation of C/EBP␤ (67). Specific binding of C/EBP␤ to the dsRNA can activate two opposing antiviral strategies (55) and S100A8 promoter in RAW cells in response to dsRNA indicates its PKR plays a role in both. It contributes to self-elimination of in- critical role in gene induction by poly(I:C). IFN-␤ potentiated the fected cells via apoptosis by decreasing the rate of host cell protein poly(I:C)-induced response (Fig. 5B), although no IFN-stimulated synthesis to prevent viral replication and also regulates NF-␬B and response element motif is obvious within the essential promoter, MAPK signaling, leading to expression of genes to combat viral whereas a conserved motif is located within intron 1. The Journal of Immunology 2267

Most natural dsRNA activators of PKR are synthesized in virus- Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, en- infected cells as by-products of viral replication or transcription. dotoxin-induced shock. Nat. Med. 13: 1042–1049. 10. van Lent, P. L., L. Grevers, A. B. Blom, A. Sloetjes, J. S. Mort, T. Vogl, Particularly for RNA viruses, dsRNA replicative forms are oblig- W. Nacken, W. B. van den Berg, and J. Roth. 2007. Myeloid related proteins atory intermediates for the synthesis of new genomic RNA copies S100A8/S100A9 regulate joint inflammation and cartilage destruction during an- tigen-induced arthritis. Ann Rheum. Dis. 67: 1750–1758. and dsRNA may be critical for the outcome of influenza A infec- 11. Kerkhoff, C., M. Klempt, V. Kaever, and C. Sorg. 1999. The two calcium-binding tion. S100A8 was expressed in lungs of mice infected with influ- proteins, S100A8 and S100A9, are involved in the metabolism of arachidonic enza virus (Fig. 7), although it was not obvious on the day of acid in human neutrophils. J. Biol. Chem. 274: 32672–32679. 12. Kerkhoff, C., W. Nacken, M. Benedyk, M. C. Dagher, C. Sopalla, and infection. Our results strongly support in vivo induction of S100A8 J. Doussiere. 2005. The arachidonic acid-binding protein S100A8/A9 promotes in response to an RNA viral infection in mice, as well as related NADPH oxidase activation by interaction with p67phox and Rac-2. FASEB J. 19: S100 proteins in human infections with a papillomavirus (16) and 467–469. 13. Leukert, N., T. Vogl, K. Strupat, R. Reichelt, C. Sorg, and J. Roth. 2006. Cal- a lentivirus (18). Expression was observed in our mice early in cium-dependent tetramer formation of S100A8 and S100A9 is essential for bio- infection and was maximal at day 8 when strong immunoreactivity logical activity. J. Mol. Biol. 359: 961–972. in epithelial cells lining the airways and in mononuclear cells was 14. Vogl, T., S. Ludwig, M. Goebeler, A. Strey, I. S. Thorey, R. Reichelt, D. Foell, V. Gerke, M. P. Manitz, W. Nacken, et al. 2004. MRP8 and MRP14 control obvious. Pulmonary epithelial cells express TLR3 (46, 68) and microtubule reorganization during transendothelial migration of phagocytes. PKR (46, 68, 69). S100A8 is up-regulated in bronchial epithelial Blood 104: 4260–4268. 15. Tugizov, S., J. Berline, R. Herrera, M. E. Penaranda, M. Nakagawa, and cells by LPS (3) and S100A8/S100A9 is expressed in tracheal J. Palefsky. 2005. Inhibition of human papillomavirus type 16 E7 phosphoryla- epithelial cells from patients with cystic fibrosis (70). In this study, tion by the S100 MRP-8/14 protein complex. J. Virol. 79: 1099–1112. we show that poly(I:C) induced S100A8 mRNA in alveolar A549 16. Reghunathan, R., M. Jayapal, L. Y. Hsu, H. H. Chng, D. Tai, B. P. Leung, and A. J. Melendez. 2005. Expression profile of immune response genes in patients epithelial cells (Fig. 7F), although levels were low and other me- with severe acute respiratory syndrome. BMC Immunol. 6: 2. Downloaded from diators may be involved in vivo. Moreover, S100A8/A9 induces 17. Muller, F., S. S. Froland, P. Aukrust, and M. K. Fagerhol. 1994. Elevated serum IL-8 production by these cells, suggesting a mechanism for am- levels in HIV-infected patients: the calprotectin response during ZDV treatment is associated with clinical events. J. Acquir. Immune Defic. Syndr. plification of neutrophilic inflammation (71). S100A8 expression 7: 931–939. declined early in the recovery phase, implying that it may be 18. Strasser, F., P. L. Gowland, and C. Ruef. 1997. Elevated serum macrophage inhibitory factor-related protein (MRP) 8/14 levels in advanced HIV infection down-regulated by mediator(s) up-regulated in the resolution and during disease exacerbation. J. Acquir. Immune Defic. Syndr. Hum. Retro- phase of the infection. This is consistent with reports that serum virol. 16: 230–238. http://www.jimmunol.org/ levels correlate with the clinical course in patients with HIV in- 19. Lugering, N., R. Stoll, T. Kucharzik, G. Burmeister, C. Sorg, and W. Domschke. 1995. Serum 27E10 antigen: a new potential marker for staging HIV disease. fection (20–22). Clin. Exp. Immunol. 101: 249–253. Increased expression of S100A8 in HIV-1-infected patients 20. Dunlop, O., J. N. Bruun, B. Myrvang, and M. K. Fagerhol. 1991. Calprotectin in could be induced by opportunistic pathogens or by direct HIV cerebrospinal fluid of the HIV infected: a diagnostic marker of opportunistic central nervous system infection? Scand J. Infect. Dis. 23: 687–689. infection of monocytes. Alternatively, production could be subse- 21. Sweet, S. P., A. N. Denbury, and S. J. Challacombe. 2001. Salivary calprotectin quent to IL-10 secretion by HIV-infected cells. Recently, two levels are raised in patients with oral candidiasis or Sjo¨gren’s syndrome but groups independently identified an important role for the IL-10 decreased by HIV infection. Oral Microbiol. Immunol. 16: 119–123. 22. Myint, M., S. Steinsvoll, Z. N. Yuan, B. Johne, K. Helgeland, and K. Schenck. in viral persistence. Blockade of IL-10 signaling increased vi- 2002. Highly increased numbers of leukocytes in inflamed gingiva from patients rus-specific CD8ϩ and CD4ϩ T cells and enhanced their func- with HIV infection. AIDS 16: 235–243. by guest on September 27, 2021 tion, resulting in resolution (72, 73). Since S100A8 levels may 23. Hashemi, F. B., J. Mollenhauer, L. D. Madsen, B. E. Sha, W. Nacken, M. B. Moyer, C. Sorg, and G. T. Spear. 2001. Myeloid-related protein (MRP)-8 be maintained in conditions where IL-10 is raised and human from cervico-vaginal secretions activates HIV replication. AIDS 15: 441–449. S100A8 accentuates HIV-1 transcription and virus production 24. Ryckman, C., G. A. Robichaud, J. Roy, R. Cantin, M. J. Tremblay, and P. A. Tessier. 2002. HIV-1 transcription and virus production are both accentu- (23, 24), this amplification may reflect clinical manifestations in ated by the proinflammatory myeloid-related proteins in human CD4ϩ T lym- patients with HIV-1. phocytes. J. Immunol. 169: 3307–3313. 25. Vicari, A. P., and G. Trinchieri. 2004. Interleukin-10 in viral diseases and cancer: exiting the labyrinth? Immunol. Rev. 202: 223–236. Acknowledgments 26. Filippi, C. M., and M. G. von Herrath. 2008. IL-10 and the resolution of infec- We are grateful to Dr. Alison C. Budd for performing the tions. J. Pathol. 214: 224–230. 27. Blackburn, S. D., and E. J. Wherry. 2007. IL-10, T cell exhaustion and viral immunohistochemistry. persistence. Trends Microbiol. 15: 143–146. 28. Xu, K., T. Yen, and C. L. Geczy. 2001. IL-10 up-regulates macrophage expres- Disclosures sion of the S100A8. J. Immunol. 166: 6358–6366. 29. Hsu, K., R. J. Passey, Y. Endoh, F. Rahimi, P. Youssef, T. Yen, and C. L. Geczy. The authors have no financial conflict of interest. 2005. Regulation of S100A8 by glucocorticoids. J. Immunol. 174: 2318–2326. 30. Nathan, C., and M. U. Shiloh. 2000. Reactive oxygen and nitrogen intermediates in the relationship between mammalian hosts and microbial pathogens. Proc. References Natl. Acad. Sci. USA 97: 8841–8848. 1. Foell, D., and J. Roth. 2004. Proinflammatory S100 proteins in arthritis and 31. Harrison, C. A., M. J. Raftery, J. Walsh, P. Alewood, S. E. Iismaa, S. Thliveris, autoimmune disease. Arthritis Rheum. 50: 3762–3771. and C. L. Geczy. 1999. Oxidation regulates the inflammatory properties of the 2. McCormick, M. M., F. Rahimi, Y. V. Bobryshev, K. Gaus, H. Zreiqat, H. Cai, murine S100 protein S100A8. J. Biol. Chem. 274: 8561–8569. R. S. A. Lord, and C. L. Geczy. 2005. S100A8 and S100A9 in human arterial 32. Lee, T.-S., and L.-Y. Chau. 2002. Heme oxygenase-1 mediates the anti- wall: implications for atherogenesis. J. Biol. Chem. 280: 41521–41529. inflammatory effect of interleukin-10 in mice. Nat. Med. 8: 240–246. 3. Henke, M. O., A. Renner, B. K. Rubin, J. I. Gyves, E. Lorenz, and J. S. Koo. 33. Ku¨hn, R., J. Lo¨hler, D. Rennick, K. Rajewsky, and W. Mu¨ller. 1993. Interleukin- 2006. Up-regulation of S100A8 and S100A9 protein in bronchial epithelial cells 10-deficient mice develop chronic enterocolitis. Cell 75: 263–274. by lipopolysaccharide. Exp. Lung Res. 32: 331–347. 34. Garofalo, R., M. Sabry, M. Jamaluddin, R. K. Yu, A. Casola, P. L. Ogra, and 4. Rahimi, F., K. Hsu, Y. Endoh, and C. L. Geczy. 2005. FGF-2, IL-1␤; and TGF-␤ A. R. Brasier. 1996. Transcriptional activation of the interleukin-8 gene by re- regulate fibroblast expression of S100A8. FEBS J. 272: 2811–2827. spiratory syncytial virus infection in alveolar epithelial cells: nuclear transloca- 5. Donato, R. 2003. Intracellular and extracellular roles of S100 proteins. Microsc. tion of the RelA transcription factor as a mechanism producing airway mucosal Res. Tech. 60: 540–551. inflammation. J. Virol. 70: 8773–8781. 6. Passey, R. J., K. Xu, D. A. Hume, and C. L. Geczy. 1999. S100A8: emerging 35. Xu, K., and C. L. Geczy. 2000. IFN-␥ and TNF regulate macrophage expression functions and regulation. J. Leukocyte Biol. 66: 549–556. of the chemotactic S100 protein S100A8. J. Immunol. 164: 4916–4923. 7. Foell, D., H. Wittkowski, T. Vogl, and J. Roth. 2006. S100 proteins expressed in 36. Ramirez-Carrozzi, V. R., A. A. Nazarian, C. C. Li, S. L. Gore, R. Sridharan, phagocytes: a novel group of damage-associated molecular pattern molecules. A. N. Imbalzano, and S. T. Smale. 2006. Selective and antagonistic functions of J. Leukocyte Biol. 81: 1–9. SWI/SNF and Mi-2␤ nucleosome remodeling complexes during an inflammatory 8. Isaksen, B., and M. K. Fagerhol. 2001. Calprotectin inhibits matrix metallopro- response. Genes Dev. 20: 282–296. teinases by sequestration of zinc. Mol. Pathol. 54: 289–292. 37. Iismaa, S. E., S.-P. Hu, M. Kocher, M. Lackmann, C. A. Harrison, S. Thliveris, 9. Vogl, T., K. Tenbrock, S. Ludwig, N. Leukert, C. Ehrhardt, M. A. D. van Zoelen, and C. L. Geczy. 1994. Recombinant and cellular expression of the murine che- W. Nacken, D. Foell, T. van der Poll, C. Sorg, and J. Roth. 2007. Mrp8 and motactic protein, CP-10. DNA Cell Biol. 13: 183–192. 2268 IL-10 MEDIATES S100A8 INDUCTION BY dsRNA

38. Kocher, M., P. A. Kenny, E. Farram, K. B. Abdul Majid, J. J. Finlay-Jones, and 56. Williams, B. R. 1999. PKR: a sentinel kinase for cellular stress. Oncogene 18: C. L. Geczy. 1996. Functional chemotactic factor CP-10 and MRP-14 are abun- 6112–6120. dant in murine abscesses. Infect. Immun. 64: 1342–1350. 57. Fremont, M., F. Vaeyens, C. V. Herst, K. L. De Meirleir, and P. Englebienne. 39. Budd, A., L. Alleva, M. Alsharifi, A. Koskinen, V. Smythe, A. Mullbacher, 2006. Double-stranded RNA-dependent protein kinase (PKR) is a stress-respon- J. Wood, and I. Clark. 2007. Increased survival after gemfibrozil treatment of sive kinase that induces NF␬B-mediated resistance against mercury cytotoxicity. severe mouse influenza. Antimicrob. Agents Chemother. 51: 2965–2968. Life Sci. 78: 1845–1856. 40. Clark, I., M. Awburn, R. Whitten, C. Harper, N. G. Liomba, M. Molyneux, and 58. Moran, J. M., M. A. Moxley, R. M. Buller, and J. A. Corbett. 2005. Encepha- T. Taylor. 2003. Tissue distribution of migration inhibitory factor and inducible lomyocarditis virus induces PKR-independent mitogen-activated protein kinase nitric oxide synthase in falciparum malaria and sepsis in African children. Ma- activation in macrophages. J. Virol. 79: 10226–10236. laria J. 2: 6. 59. Steer, S. A., J. M. Moran, B. S. Christmann, L. B. Maggi, Jr., and J. A. Corbett. 41. Clark, I. A., M. M. Awburn, C. G. Harper, N. G. Liomba, and M. E. Molyneux. 2006. Role of MAPK in the regulation of double-stranded RNA- and encepha- 2003. Induction of HO-1 in tissue macrophages and monocytes in fatal falcipa- lomyocarditis virus-induced cyclooxygenase-2 expression by macrophages. rum malaria and sepsis. Malaria J. 2: 41. J. Immunol 177:3413–3420. 42. Garcia, M. A., J. Gil, I. Ventoso, S. Guerra, E. Domingo, C. Rivas, and 60. Wathelet, M. G., I. M. Clauss, F. C. Paillard, and G. A. Huez. 1989. 2-Amin- M. Esteban. 2006. Impact of protein kinase PKR in cell biology: from antiviral opurine selectively blocks the transcriptional activation of cellular genes by virus, to antiproliferative action. Microbiol. Mol. Biol. Rev. 70: 1032–1060. double-stranded RNA and interferons in human cells. Eur J. Biochem. 184: 43. Hu, Y., and T. W. Conway. 1993. 2-Aminopurine inhibits the double-stranded 503–509. RNA-dependent protein kinase both in vitro and in vivo. J. Interferon Res. 13: 61. Silva, A. M., M. Whitmore, Z. Xu, Z. Jiang, X. Li, and B. R. Williams. 2004. 323–328. Protein kinase R (PKR) interacts with and activates mitogen-activated protein 44. Klempt, M., H. Melkonyan, H. A. Hofmann, I. Eue, and C. Sorg. 1998. The kinase kinase 6 (MKK6) in response to double-stranded RNA stimulation. J. Biol. ␣ transcription factors c-myb and C/EBP regulate the monocytic/myeloic gene Chem. 279: 37670–37676. MRP14. Immunobiology 199: 148–151. 62. Haddad, J. J., N. E. Saade, and B. Safieh-Garabedian. 2003. Interleukin-10 and 45. Kuruto-Niwa, R., M. Nakamura, K. Takeishi, and R. Nozawa. 1998. Transcrip- the regulation of mitogen-activated protein kinases: are these signalling modules ␣ ␤ tional regulation by C/EBP and - in the expression of the gene for the MRP14 targets for the anti-inflammatory action of this cytokine? Cell Signal 15: myeloid calcium binding protein. Cell Struct. Funct. 23: 109–118.

255–267. Downloaded from 46. Matsukura, S., F. Kokubu, M. Kurokawa, M. Kawaguchi, K. Ieki, H. Kuga, 63. Stewart, M. J., S. B. Kulkarni, T. R. Meusel, and F. Imani. 2006. c-Jun N- M. Odaka, S. Suzuki, S. Watanabe, T. Homma, et al. 2007. Role of RIG-I, terminal kinase negatively regulates dsRNA and RSV induction of tumor necrosis MDA-5, and PKR on the expression of inflammatory chemokines induced by factor-␣ transcription in human epithelial cells. J. Interferon Cytokine Res. 26: synthetic dsRNA in airway epithelial cells. Int. Arch. Allergy Immunol. 521–533. 143(Suppl.)1:80–83. 64. Zenz, R., R. Eferl, L. Kenner, L. Florin, L. Hummerich, D. Mehic, H. Scheuch, 47. Dong, L. W., X. N. Kong, H. X. Yan, L. X. Yu, L. Chen, W. Yang, Q. Liu, P. Angel, E. Tschachler, and E. F. Wagner. 2005. Psoriasis-like skin disease and D. D. Huang, M. C. Wu, and H. Y. Wang. 2008. Signal regulatory protein ␣ arthritis caused by inducible epidermal deletion of Jun proteins. Nature 437: negatively regulates both TLR3 and cytoplasmic pathways in type I interferon 369–375.

induction. Mol. Immunol. 45: 3025–3035. http://www.jimmunol.org/ 65. Karin, M. 1995. The regulation of AP-1 activity by mitogen-activated protein 48. Okahira, S., F. Nishikawa, S. Nishikawa, T. Akazawa, T. Seya, and kinases. J. Biol. Chem. 270: 16483–16486. M. Matsumoto. 2005. Interferon-␤ induction through Toll-like receptor 3 de- pends on double-stranded RNA structure. DNA Cell Biol. 24: 614–623. 66. Hu, X., J. Chen, L. Wang, and L. B. Ivashkiv. 2007. Crosstalk among Jak-STAT, 49. Yamamoto, M., S. Sato, H. Hemmi, K. Hoshino, T. Kaisho, H. Sanjo, Toll-like receptor, and ITAM-dependent pathways in macrophage activation. O. Takeuchi, M. Sugiyama, M. Okabe, K. Takeda, and S. Akira. 2003. Role of J. Leukocyte Biol. 82: 237–243. adaptor TRIF in the MyD88-independent Toll-like receptor signaling pathway. 67. Auch, C. J., R. N. Saha, F. G. Sheikh, X. Liu, B. L. Jacobs, and K. Pahan. 2004. Science 301: 640–643. Role of protein kinase R in double-stranded RNA-induced expression of nitric 50. Yamamoto, M., S. Sato, K. Mori, K. Hoshino, O. Takeuchi, K. Takeda, and oxide synthase in human astroglia. FEBS Lett. 563: 223–228. S. Akira. 2002. Cutting edge: a novel Toll/IL-1 receptor domain-containing 68. Groskreutz, D. J., M. M. Monick, L. S. Powers, T. O. Yarovinsky, D. C. Look, adapter that preferentially activates the IFN-␤ promoter in the Toll-like receptor and G. W. Hunninghake. 2006. Respiratory syncytial virus induces TLR3 protein signaling. J. Immunol. 169: 6668–6672. and protein kinase R, leading to increased double-stranded RNA responsiveness in airway epithelial cells. J. Immunol. 176: 1733–1740. 51. Jiang, Z., M. Zamanian-Daryoush, H. Nie, A. M. Silva, B. R. Williams, and by guest on September 27, 2021 X. Li. 2003. Poly(I-C)-induced Toll-like receptor 3 (TLR3)-mediated activation 69. Gern, J. E., D. A. French, K. A. Grindle, R. A. Brockman-Schneider, S.-I. Konno, of NF␬B and MAP kinase is through an interleukin-1 receptor-associated kinase and W. W. Busse. 2003. Double-stranded RNA induces the synthesis of specific (IRAK)-independent pathway employing the signaling components TLR3- chemokines by bronchial epithelial cells. Am. J. Respir. Cell Mol. Biol. 28: TRAF6-TAK1-TAB2-PKR. J. Biol. Chem. 278: 16713–16719. 731–737. 52. Horng, T., G. M. Barton, and R. Medzhitov. 2001. TIRAP: an adapter molecule 70. Renaud, W., M. Merten, and C. Figarella. 1994. Increased coexpression of CFTR in the toll signaling pathway. Nat. Immunol. 2: 835–841. and S100 calcium binding proteins MRP8 and MRP14 mRNAs in cystic fibrosis 53. Cabanski, M., M. Steinmuller, L. M. Marsh, E. Surdziel, W. Seeger, and human tracheal gland cells. Biochem. Biophys. Res. Commun. 201: 1518–1525. J. Lohmeyer. 2008. PKR regulates TLR2/TLR4-dependent signaling in murine 71. Ahmad, A., D. L. Bayley, S. He, and R. A. Stockley. 2003. Myeloid related alveolar macrophages. Am. J. Respir. Cell Mol. Biol. 38: 26–31. protein-8/14 stimulates interleukin-8 production in airway epithelial cells. 54. Li, J. C., D. C. Lee, B. K. Cheung, and A. S. Lau. 2005. Mechanisms for HIV Tat Am. J. Respir. Cell Mol. Biol. 29: 523–530. upregulation of IL-10 and other cytokine expression: kinase signaling and PKR- 72. Brooks, D. G., M. J. Trifilo, K. H. Edelmann, L. Teyton, D. B. McGavern, and mediated immune response. FEBS Lett. 579: 3055–3062. M. B. Oldstone. 2006. Interleukin-10 determines viral clearance or persistence in 55. Iordanov, M. S., J. Wong, J. C. Bell, and B. E. Magun. 2001. Activation of vivo. Nat. Med. 12: 1301–1309. NF-␬B by double-stranded RNA (dsRNA) in the absence of protein kinase R and 73. Ejrnaes, M., C. M. Filippi, M. M. Martinic, E. M. Ling, L. M. Togher, S. Crotty, RNase L demonstrates the existence of two separate dsRNA-triggered antiviral and M. G. von Herrath. 2006. Resolution of a chronic viral infection after inter- programs. Mol. Cell. Biol. 21: 61–72. leukin-10 receptor blockade. J. Exp. Med. 203: 2461–2472.