Journal of Leukocyte Biology eBioscience full spectrum cell analysis
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published online as doi:10.1189/jlb.0107036 on March 14, 2007

Published online before print March 14, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.0107036v1
81/6/1577    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schroder, K.
Right arrow Articles by Sweet, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schroder, K.
Right arrow Articles by Sweet, M. J.
(Journal of Leukocyte Biology. 2007;81:1577-1590.)
© 2007 by Society for Leukocyte Biology

PU.1 and ICSBP control constitutive and IFN-{gamma}-regulated Tlr9 gene expression in mouse macrophages

Kate Schroder*,{dagger}, Monika Lichtinger{ddagger}, Katharine M. Irvine*, Kristian Brion*,{dagger}, Angela Trieu*,{dagger}, Ian L. Ross*,{dagger}, Timothy Ravasi§, Katryn J. Stacey*,{dagger}, Michael Rehli{ddagger}, David A. Hume*,{dagger} and Matthew J. Sweet*,{dagger},||,1

* Special Research Centre for Functional and Applied Genomics, Institute for Molecular Bioscience, and
|| School of Molecular and Microbial Sciences, University of Queensland, Brisbane, Queensland, Australia;
{dagger} Cooperative Research Centre for Chronic Inflammatory Diseases, Australia;
{ddagger} Abt. für Hämatologie und Internistische Onkologie, Klinikum der Universität Regensburg, Germany; and
§ Scripps NeuroAIDS Preclinical Studies Centre and Department of Bioengineering, Jacobs School of Engineering, University of California San Diego, La Jolla, California, USA

1 Correspondence: Institute for Molecular Bioscience, University of Queensland, St. Lucia, Brisbane 4072, Australia. E-mail: m.sweet{at}imb.uq.edu.au


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Macrophages are activated by unmethylated CpG-containing DNA (CpG DNA) via TLR9. IFN-{gamma} and LPS can synergize with CpG DNA to enhance proinflammatory responses in murine macrophages. Here, we show that LPS and IFN-{gamma} up-regulated Tlr9 mRNA in murine bone marrow-derived macrophages (BMM). The ability of LPS and IFN-{gamma} to induce Tlr9 mRNA expression in BMM was dependent on the presence of the growth factor, CSF-1, which is constitutively present in vivo. However, there were clear differences in mechanisms of Tlr9 mRNA induction. LPS stimulation rapidly removed the CSF-1 receptor (CSF-1R) from the cell surface, thereby blocking CSF-1-mediated transcriptional repression and indirectly inducing Tlr9 mRNA expression. By contrast, IFN-{gamma} activated the Tlr9 promoter directly and only marginally affected cell surface CSF-1R expression. An ~100-bp proximal promoter of the murine Tlr9 gene was sufficient to confer basal and IFN-{gamma}-inducible expression in RAW264.7 cells. A composite IFN regulatory factor (IRF)/PU.1 site upon the major transcription start site was identified. Mutation of the binding sites for PU.1 or IRF impaired basal promoter activity, but only the IRF-binding site was required for IFN-{gamma} induction. The mRNA expression of the IRF family member IFN consensus-binding protein [(ICSBP)/IRF8] was coregulated with Tlr9 in macrophages, and constitutive and IFN-{gamma}-inducible Tlr9 mRNA expression was reduced in ICSBP-deficient BMM. This study therefore characterizes the regulation of mouse Tlr9 expression and defines a molecular mechanism by which IFN-{gamma} amplifies mouse macrophage responses to CpG DNA.

Key Words: colony-stimulating factor-1 • inflammation • lipopolysaccharide • interferon • IRF8 • Toll-like receptor


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The vertebrate immune system has evolved to recognize invariant pathogen-associated molecular patterns (PAMPs) for the early detection and subsequent clearance of microbes. One such PAMP is unmethylated CpG-containing DNA (CpG DNA), found in bacterial and viral genomes [1 ]. Unmethylated CpG motifs within specific nucleotide contexts are recognized by a TLR family member, TLR9 [2 ]. To date, 13 mammalian TLRs have been identified (TLR1–13), many of which have demonstrated functions in innate immunity. TLR9 is located in the endoplasmic reticulum in resting cells but upon cellular uptake of CpG-containing sequences, relocates to the lysosomal compartment to activate signaling [3 4 5 ]. The signal is relayed through the MyD88 adaptor protein, members of the IL-1 receptor-associated kinase family, and TNF receptor-associated factor-6 (TRAF6) to promote I{kappa}B degradation and activate JNK and p38 MAPK. This ultimately results in activation of a number of transcription factors, including NF-{kappa}B and AP-1 and transcriptional control over genes directing immune function (reviewed in refs. [6 7 8 9 ]).

In mice, Tlr9 is expressed by macrophages, B cells, myeloid dendritic cells (DCs), and plasmacytoid DCs (PDCs; reviewed in ref. [10 ]). A number of studies demonstrated an absence of TLR9 expression in the monocyte/macrophage lineage in humans [11 12 13 ], suggesting divergent regulation among the species. However, recent reports have claimed that TLR9 expression was inducible by bacterial challenge [14 ] and IFN-{gamma} [15 ] in human myeloid cells. TLR9 stimulation of mouse B cells triggers proliferation and IL-6 secretion and skews class switch recombination to "Th1-like" Igs [16 ]. In human and mouse PDCs, TLR9 ligation results in cellular differentiation and activation to secrete Th1-polarizing cytokines such as Type I IFN, IL-12, and TNF [17 , 18 ]. Mouse monocyte/macrophages are activated by CpG DNA to release cytokines such as IL-12, IL-6, and TNF [2 ]. In the context of bacterial pathogen recognition, innate immune cells such as macrophages are likely todetect CpG DNA in concert with other bacterial products such as LPS or proinflammatory cytokines such as IFN-{gamma}. LPS and IFN-{gamma} can amplify responses to CpG DNA [2 , 19 , 20 ]. Bacterial CpG DNA action can be mimicked by synthetic phosphodiester or phosphorothioate oligonucleotides [2 ], although phosphorothioate oligonucleotides can also have CpG DNA-independent effects on cellular function [21 22 23 24 25 ]. CpG DNA, in the form of the phosphorothioate-stabilized oligodinucleotide, has potential for clinical exploitation as a vaccine adjuvant and in the treatment of cancer and infectious diseases.

Mice with targeted deficits in IFN-{gamma} signaling display compromised innate and acquired immunity during CpG DNA stimulation [26 ] and are highly resistant to LPS-induced toxicity [27 ]. This highlights the physiological significance of IFN-{gamma} priming in vivo, as it suggests that IFN-{gamma} is normally produced during the response to these TLR agonists and functions to amplify the TLR-induced cellular responses. Many mechanisms have been elucidated for IFN-{gamma} priming of macrophage responses to LPS (reviewed in refs. [28 , 29 ]) and involve the reciprocal cross-regulation of signaling molecules between the IFN-{gamma} and LPS signaling pathways. These mechanisms include induction of TLR4 and inhibition of LPS-induced TLR4 down-regulation, induction of MyD88 and the myeloid differentiation protein 2 adaptor molecule, potentiation of NF-{kappa}B activity, and transcription factor synergy at the promoters of target genes. However, it is less clear how IFN-{gamma} primes macrophage responses to CpG DNA (reviewed in ref. [29 ]).

Although CpG DNA-regulated signaling events are well-characterized, information regarding regulation of the Tlr9 gene remains limited. One report characterizing the human TLR9 promoter [30 ] identified four promoter elements [cAMP-response element (CRE), 5' PU.1, 3' PU.1, and c/EBP sites], which conferred gene regulation in B cells. Another recent study identified some of the elements necessary for constitutive and IFN-ß-inducible regulation of the Tlr9 promoter in mouse macrophages [31 ]. This study found that a proximal promoter AP-1 site conferred constitutive expression, and a distal regulatory region containing two IFN-stimulated response element/IFN regulatory factor (ISRE/IRF) sites, ~2.4 kb upstream of the transcription start site, contributed to basal and IFN-ß-inducible Tlr9 expression. In the present report, we set out to define the mechanisms by which IFN-{gamma} and LPS prime responses to CpG DNA. We show that LPS and IFN-{gamma} induced Tlr9 mRNA in mouse macrophages through divergent mechanisms. The induction of Tlr9 by IFN-{gamma} occurred at the level of the proximal promoter, and we identified regulators of basal and IFN-{gamma}-inducible Tlr9 expression. Definition of the key elements of the mouse Tlr9 promoter provides insight into expression differences arising from evolutionary divergence of the promoter in mammals.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
General reagents
Recombinant mouse (rm)IFN-{gamma} (R&D Systems, Minneapolis, MN, USA) was used at a final concentration of 500 pg/ml, and rmIFN-ß (PBL Biomedical Laboratories, Piscataway, NJ, USA) was used at 100 U/ml. LPS (from Salmonella minnesota, Sigma Aldrich, St. Louis, MO, USA) was used at a final concentration of 10 ng/ml. Recombinant human (rh)CSF-1 (a gift from Chiron, Emeryville, CA, USA) was used at a final concentration of 1 x 104 U/ml (100 ng/ml).

Cell culture
Specific, homogenous bone marrow-derived macrophages (BMM) were obtained by ex vivo differentiation from mouse BM progenitors in the presence of rCSF-1 as described previously [32 , 33 ]. All mice were C57Bl/6, specific pathogen-free and 6–8 weeks old at the time of BM collection. In experiments with IFN consensus-binding protein (ICSBP)–/– versus wild-type (WT) BMM, matched numbers of female and male mice were used for BM collection. Male mice were used in all other experiments. BM cells were cultured in RPMI 1640 (Invitrogen, Carlsbad, CA, USA) supplemented with 10% FBS (heat-inactivated and endotoxin-tested; JRH Biosciences, Lenexa, KS, USA), 20 U/ml penicillin and 20 µg/ml streptomycin (Invitrogen), 2 mM L-glutamine (Glutamax-1, Invitrogen), and CSF-1 on bacteriological plastic (Sterilin, Staffordshire, UK). CSF-1-replete media was replaced on Day 5, and cells were re-seeded at Day 6 onto tissue-culture plastic (Iwaki, Tokyo, Japan) in the presence or absence of CSF-1. Cells were stimulated (e.g., with IFN-{gamma} or LPS) early on Day 7. The macrophage-like cell line RAW 264.7 was maintained on bacteriological plates in RPMI 1640 supplemented with 5% FBS, 20 U/ml penicillin, 20 µg/ml streptomycin, and 2 mM L-glutamine. Prior to experiments, cells were seeded onto tissue-culture plastic. All cells were maintained in a 37°C incubator venting 5% CO2.

RNA purification and analysis
RNA extraction from mammalian cells or tissues was performed using RNeasy kits (Qiagen, Valencia, CA, USA), genomic DNA was removed from RNA preparations using DNA-Free (Ambion, Austin, TX, USA), and cDNA was synthesized using Superscript III (Invitrogen). These procedures were performed according to the manufacturer’s instructions. Negative control reactions were performed for cDNA synthesis in the absence of the Superscript III enzyme.

Where possible, gene-specific PCR primer pairs were designed, such that one of the primers overlapped an exon junction and produced an amplicon of ~100 bp. The amplification of a single product was ensured by melt-curve analysis for each primer pair. The sequences of the primer pairs used for cDNA quantitation are as follows: Tlr9 forward (F), 5'-AGGCTGTCAATGGCTCTCAGTT-3', reverse (R), 5'-TGAACGATTTCCAGTGGTACAAGT-3'; hypoxanthine guanine phosphoribosyl transferase (Hprt) F, 5'-GCAGTACAGCCCCAAAATGG-3', R, 5'-AACAAAGTCTGGCCTGTATCCAA-3'; Icsbp F, 5'-ACGAGGTTACGCTGTGCTCTG-3', R, 5'-CACGCCCAGCTTGCATTT-3'; Ifnß F, 5'-CCACAGCCCTCTCCATCAAC-3', R, 5'-TGAAGTCCGCCCTGTAGGTG-3'; Irf4 F, 5'-AGCTCATCACAGCTCATGTGGA-3', R, 5'-GCGGTGGTAATCTGGAGTGGT-3'.

Gene expression was quantitated relative to Hprt using the Platinum SybrGreen quantitative PCR (qPCR) system (Invitrogen) according to the manufacturer’s instructions, using an ABI Prism 7000 sequence detection system (Applied Biosystems, Foster City, CA, USA). Each cDNA sample was detected in experimental triplicate. Negative control reactions were cycled alongside test samples to ensure the absence of contaminating genomic DNA. Data were analyzed using the ABI Prism software, and expression was determined relative to Hprt mRNA. Transcript abundance (gene/Hprt) and SD were calculated as recommended by Applied Biosystems. Unless otherwise stated, error bars in figures indicate the SD of triplicate cDNA quantitations in the same thermal cycler run.

Surface CSF-1 receptor (CSF-1R) quantitation
Cells for flow cytometric quantification of surface CSF-1R expression were prepared using prechilled solutions, and all centrifugation steps were carried out at 4°C. Cells were harvested in lift buffer (1 mM EDTA, 0.1% azide in PBS), washed thoroughly, pelleted, and incubated in 50 µl 2.4G2 hybridoma medium (rat anti-FcR mAb) for 20 min. Cells were pelleted and resuspended in 100 µl 0.1% BSA/0.1% azide in PBS (PBA) containing 4 µg/ml rat antimouse CSF-1R (IgG1) antibody conjugated to PE or the rat IgG1-PE isotype control (Serotec, Oxford, UK). Samples were incubated for 45 min in the dark before addition of 1 ml PBA. Cells were pelleted, washed again with 1 ml PBA, and resuspended in 500 µl PBA, and surface CSF-1R expression was quantitated using a FACSCalibur (BD Biosciences, San Jose, CA, USA) and CellQuest software (BD Biosciences).

5' RACE
The RACE PCR was performed as specified by the manufacturer (Invitrogen). Briefly, 5 µg total spleen RNA was treated with DNase I and reverse-transcribed. The 5' end of mouse Tlr9 was PCR-amplified using the GeneRacer forward primer and a Tlr9-specific reverse primer (5'-AGCCAGGAAGGTTCTGGGCTCAA-3', which binds to the second exon of the Tlr9 gene). Extension products were gel-purified and topoisomerase (TOPO)-cloned into pCR4-TOPO vector (Invitrogen). Clones were sequenced by automated dsDNA sequencing using the ABI PRISM BigDye sequencing kit and the ABI PRISM 7000 sequence detection system (Applied Biosystems) at the Australian Genome Research Facility (University of Queensland, St. Lucia).

DNA manipulation
The Tlr9 promoter was PCR-amplified from genomic DNA (primers: F, 5'-AACTCAACTCTGGGTAAAAGATGG-3', R, 5'-GGTTCCATGAGTGTCCCAAG-3'), and the 2158-bp PCR product was partially digested with BsaI, the DNA overhangs were blunted, and the fragment was digested further with XbaI. The blunt end of the restriction fragment was ligated into the SmaI site of pGL2B, and the cohesive, XbaI-digested site was ligated into the compatible NheI-digested site of pGL2B to form the plasmid pGL2B_Tlr9p_2 kb. The 100-bp, proximal Tlr9 promoter inserts were PCR-amplified from the pGL2B_Tlr9p_2-kb plasmid using the same forward primer and WT or nonhomologous reverse primers to introduce mutations (primers: F, 5'-GCTAGCGACTATGCAAATGATGTGTGACTCAT-3'; WT Tlr9 promoter R, 5'-CAGAGAATTGAGGAAGTGACACTTTCAC-3'; mutant IRF site Tlr9 promoter R, 5'-CAGAGAATTGAGGAAGTGTCTCTTTCAC-3'; mutant PU.1 site Tlr9 promoter R, 5'-CAGAGAATTGAGGTTGTGACACTTTCAC-3'). PCR products were blunted, digested with NheI, and inserted into the NheI/blunted BglII site of pGL2B. All constructs were sequenced before use.

Promoter activity analysis
RAW 264.7 cells were transfected with endotoxin-free plasmid preparations (plasmid maxi kit, Qiagen) by electroporation in 0.4 cm cuvettes. RAW 264.7 cells (5x106) were electroporated (280 V/1 mFd, Gene-Pulser, Bio-Rad, Hercules, CA, USA) with 10 µg plasmid in 250 µl complete media supplemented with 10 mM HEPES (Thermotrace, San Diego, CA, USA). For experiments in which the promoter activity of multiple plasmids with different sizes was compared, separate transfections included 10 µg of the largest plasmid and molar equivalents of the remaining plasmids. Cells were electroporated immediately after the addition of plasmid DNA and washed thoroughly without delay after transfection to limit RAW 264.7 activation by the plasmid DNA [34 ]. After electroporation, the cells were divided into two to four wells of a six-well tissue-culture plate (Iwaki). Transfected cells were cultured for 24 h before cellular harvest or treatment with IFN-{gamma} or LPS. Cellular harvest and luciferase activity quantitation were performed using the high-sensitivity luciferase reporter gene assay kit (Roche, Basel, Switzerland) according to the manufacturer’s instructions. Luciferase activity was normalized to the total protein concentration to give relative light units (RLU) per µg. The total protein concentration of transfected cell lysates was determined by the Bradford assay kit (Bio-Rad).

EMSA
Nuclear extracts for EMSA were prepared by hypotonic lysis at 4°C. All buffers contained phosphatase inhibitors (1 mM sodium vanadate, 1 mM sodium pyrophosphate, 1 mM sodium molybdate, 10 mM sodium fluoride) and a 1x protease inhibitor cocktail (complete, mini, EDTA-free, Roche) in all steps of nuclear extract preparation. Cells were harvested and washed twice in PBS. The pellet was resuspended in 1 ml per 10 x 106 cells Buffer A (10 mM HEPES, pH 7.6, 10 mM NaCl, 0.5 mM EDTA, 0.5 mM EGTA, 0.5 mM DTT, 0.5 mM PMSF). Buffer B [0.5 ml per 10x106 cells; 25 mM HEPES, pH 7.6, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.6% Nonidet P-40 (NP-40), 2 mM DTT, 1 mM PMSF] was added, and the solution was vortexed immediately at low speed for 30 s. The nuclei were pelleted and resuspended in 100 µl per 10 x 106 cells Buffer C (25 mM HEPES, pH 7.6, 350 mM NaCl, 1 mM EDTA, 1 mM EGTA, 20% glycerol, 1 mM DTT, 0.5 mM PMSF) and shaken vigorously for 20 min. Cellular debris was removed by centrifugation, and nuclear extracts were aliquoted and stored at –80°C. Total protein in each extract was quantitated and normalized before the binding reaction.

Oligonucleotide end-labeling was performed according to the manufacturer’s instructions for T4 polynucleotide kinase (New England Biosciences, Beverly, MA, USA). MgCl2 was added to a final concentration of 5 mM, and strands were annealed by heating the sample to 80°C for 2 min and allowing it to cool down slowly to room temperature. Probes were purified using a Nick Sephadex column (Amersham Biosciences, Piscataway, NJ, USA). Complementary oligonucleotides used for ds probe preparation were as follows (bold depicts mutated nucleotides): WT Tlr9 promoter, 5'-GTAGTGAAAGTGTCACTTCCTCAATTCTCTGAGA-3', 5'-TCTCAGAGAATTGAGGAAGTGACACTTTCACTAC-3'; mutant IRF site Tlr9 promoter, 5'-GTAGTGAAAGAGACACTTCCTCAATTCTCTGAGA-3', 5'-TCTCAGAGAATTGAGGAAGTGTCTCTTTCACTAC-3'; mutant PU.1 site Tlr9 promoter, 5'-GTAGTGAAAGTGTCACAACCTCAATTCTCTGAGA-3', 5'-TCTCAGAGAATTGAGGTTGTGACACTTTCACTAC-3'.

Nuclear extract proteins were bound to probe in a 10 µl reaction containing 20 mM HEPES, pH 7.9, 0.5 mM DTT, 2 mM EDTA, 40 mM KCl, 12% glycerol, and 1 µg polyinosinic-polycytidylic (Sigma Aldrich), 0.04 pmol purified probe, and 2 µg nuclear extract. Binding reactions were incubated for 30–60 min at room temperature. Reactions in which competitor probes were added included ten-, 50-, or 100-fold molar excess of unlabeled, ds competitor probe (0.4, 2, or 4 pmol, respectively). The supershift reaction included 2 µg rabbit polyclonal antimouse PU.1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA).

The Tris-glycine-EDTA gel system was used for size separation. Acrylamide (8%; 29:1) gels were cast by 0.1% w/v ammonium persulfate and 0.2% N,N,N',N'-tetramethylethylenediamine in 0.15 M Tris-HCl, pH 8.8. Gels were electrophoresed in 1x 25 mM Tris-HCl, pH 8.3, 192 mM glycine, 0.2 mM EDTA (TGE buffer) at 100 V, fixed in 10% acetic acid for 15 min, dried, and exposed to X-ray film (Fuji, Tokyo, Japan).

Chromatin immunoprecipitation (ChIP) assay
Cells (20x106) in media were cross-linked by incubation for 10 min with formaldehyde to a final concentration of 1%. The reaction was quenched by addition of glycine to a final concentration of 0.125 M. Unless otherwise stated, all further steps of ChIP were performed at 4°C with prechilled solutions. After washing twice with PBS, the cells were harvested, resuspended in 1 ml lysis buffer (85 mM KCl, 1 mM EDTA, 1% NP-40, 10 mM HEPES, pH 7.9, 1 mM PMSF, 1x Roche complete protease inhibitor), and incubated on ice for 15 min. The nuclei were pelleted, resuspended in homogenization buffer (50 mM Tris, pH 8.1, 1% SDS, 0.5% Empigen BB, 10 mM EDTA, 1 mM PMSF, 1x Roche complete protease inhibitor), and sonicated for 14 pulses of 45 s duration. The lysate was cleared by centrifugation for 10 min at 16,000 g, and part of the supernatant was kept as 10% input for normalization. Lysate was diluted 2:3 with buffer (20 mM Tris, pH 8.1, 100 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 1x Roche complete protease inhibitor) and precleared with 25 µl salmon sperm DNA/Protein A agarose-50% slurry (Upstate Biotechnology, Lake Placid, NY, USA) per 200 µl-diluted lysate by rotation for 30 min. Protein A agarose was removed by exclusion from a Millipore Ultrafree MC spin column. Precleared lysate (200 µl) was incubated with 3 µg antibody (rabbit polyclonal antimouse PU.1 antibody and IgG rabbit isotype control, Santa Cruz Biotechnology) and rotated overnight to allow binding. Complexes were recovered by adding 50 µl salmon sperm DNA/Protein A agarose-50% slurry and rotating for 2 h. Complexes were washed in Millipore Ultrafree MC spin columns by sequential application of wash buffer 1 (20 mM Tris/HCl, pH 8.1, 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 2 mM EDTA), Wash Buffer 2 (10 mM Tris/HCl, pH 8.1, 250 mM LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA), Wash Buffer 3 (10 mM Tris/HCl, pH 8.1, 250 mM LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA), and two applications of 10 mM Tris-HCl, pH 8.0, 1 mM EDTA. The complexes were recovered from protein A agarose by eluting with 200 µl elution buffer (0.1 M NaHCO3, 1% SDS). NaCl was added to eluates and 10% input samples to a final concentration of 190 mM, and the cross-links were reversed by incubation at 65°C for 5 h. Protein was digested by incubation with 20 µg proteinase K for a further 30 min at 65°C. DNA was purified using a Qiaquick PCR purification column, and DNA was analyzed by qPCR as described above, using primer pairs that amplified the IRF/PU.1 site (F, 5'-TGCTCTTTCAGGGTAGGGACA-3'; R 5'-GGCGGCAGAGAATGATGTTC-3') or the upstream PU.1 site (F, 5'-CATTGACCTTGGGCTGGAGT-3'; R, 5'-GGCAGACAGATAACCCCCTCA-3') of the Tlr9 promoter.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
IFN-{gamma} and LPS induce Tlr9 mRNA in mouse macrophages through divergent mechanisms
During pathogenic challenge, macrophages are likely to detect CpG DNA in combination with bacterial products such as LPS and proinflammatory cytokines such as IFN-{gamma}. As LPS and IFN-{gamma} can synergize with CpG DNA to activate proinflammatory gene expression in macrophages, we assessed whether these factors could influence Tlr9 mRNA expression in macrophages. A mechanism for priming TLR9-dependent responses by IFN-{gamma} might be an up-regulation of TLR9 itself, as reported for human TLR4 [35 , 36 ]. Indeed, IFN-{gamma} induced Tlr9 expression approximately tenfold in the presence of CSF-1 in BMM (Fig. 1A ). In the absence of CSF-1, basal expression of Tlr9 mRNA was elevated as reported previously [37 , 38 ] and was not significantly induced further by IFN-{gamma}. We have shown recently that LPS induced a group of coregulated genes including Tlr9 in BMM only in the presence of CSF-1 [38 ], and we confirmed that finding here (Fig. 1B) . In the case of LPS, Tlr9 mRNA induction was mediated by LPS-induced down-regula tion of cell surface CSF-1R, which alleviates the repressive effect of CSF-1 on Tlr9 gene expression [38 ]. To determine whether a similar mechanism accounted for induction by IFN-{gamma}, we compared the effects of LPS and IFN-{gamma} on cell surface CSF-1R expression in BMM. IFN-{gamma} decreased surface CSF-1R expression only marginally (Fig. 1C 1D 1E 1F) , whereas LPS triggered a rapid and sustained down-regulation of cell surface CSF-1R (Fig. 1G 1H 1I 1J) as expected. Such data suggest that an alternative mechanism accounts for Tlr9 mRNA induction by IFN-{gamma}. In keeping with this hypothesis, Figure 2 shows that IFN-{gamma} but not LPS induced Tlr9 mRNA in the murine macrophage cell line RAW264.7, which expresses low levels of cell surface CSF-1R [39 ]. By Northern blotting, two Tlr9 mRNA transcripts can be detected [2 ], and we found that both transcripts were coregulated by CSF-1, LPS, and IFN-{gamma} (data not shown). It is interesting that in comparison with Tlr9, Tlr4 induction by IFN-{gamma} in mouse BMM was modest (data not shown), despite published data suggesting this as a major mechanism for IFN-{gamma} priming of LPS responses in human monocytes and macrophages [35 , 36 ].


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		Figure 1. IFN-{gamma} and LPS induced Tlr9 expression through different mechanisms. BMM were differentiated and maintained overnight in the presence or absence of CSF-1 before treatment with (A, C–F) IFN-{gamma} or (B, G–J) LPS. (A, B) Tlr9 expression relative to Hprt was measured by qPCR in experimental triplicate. Data points represent the average of triplicate quantitations + SD. Profiles are representative of at least three independent experiments. (C–J) Surface CSF-1R was stained using flow cytometry. Histogram overlays represent antibody (Ab) isotype-control staining (dashed lines) and CSF-1R staining before (unfilled histograms) or after (filled histograms) treatment with IFN-{gamma} or LPS. Profiles are indicative of two independent experiments.

 

Figure 2
View larger version (14K):
[in this window]
[in a new window]

  		Figure 2. Tlr9 mRNA regulation in RAW 264.7 replicated BMM Tlr9 regulation by IFN-{gamma} but not LPS. The RAW 264.7 cell line was treated with (A) IFN-{gamma} or (B) LPS. Tlr9 expression relative to Hprt was measured by qPCR in experimental triplicate. Data points represent the average of triplicate quantitations + SD. Profiles are representative of at least three independent experiments.

 
IFN-{gamma} regulates Tlr9 expression at the promoter level in murine macrophages
To determine whether IFN-{gamma} regulates Tlr9 expression at the promoter level, we isolated the murine Tlr9 promoter. We first mapped the transcription start sites by 5' RACE (Fig. 3A ). The most 5' start site (designated +1, eight out of 18 clones) was 88 bp upstream of the translation start ATG, and a second major transcription start site was located at +29 (six out of 18 clones). A minor transcription start was located at +32 (two of 18 clones). 5' RACE analysis was verified independently using cap analysis gene expression (data not shown) [40 ]. In transient transfection analysis, an ~100-bp proximal promoter (located –94 to +9 bp) displayed activity ~300-fold greater than the pGL2B empty vector and had a similar activity to a 5'-extended, ~2 kb promoter construct (–1934 to +9 bp; Fig. 3B ). As with expression of the endogenous gene, the Tlr9 promoter was responsive to IFN-{gamma} and was induced threefold after 24 h IFN-{gamma} treatment (Fig. 3C) . The IFN-{gamma} responsiveness was apparent in the minimal 100-bp promoter construct, and a range of intermediate deletions between 100 bp and 2 kb had no significant impact on constitutive or IFN-{gamma}-induced activity (data not shown). The 100-bp proximal promoter is highly homologous in sequence to the human TLR9 promoter sequence (Fig. 3A) .


Figure 3
View larger version (47K):
[in this window]
[in a new window]

  		Figure 3. Characterization of the mouse Tlr9 promoter. (A) Clustal W alignment of the human TLR9 promoter with the putative mouse Tlr9 promoter. Functional transcription factor-binding sites, TATA box and transcription (C+1), and translation (*ATG) start sites demonstrated for the human gene [30 ] are included. Potential transcription factor-binding sites conserved between human and mouse are shown in dark gray. This includes the AP-1 site (–76 to –69) identified by others to contribute to basal activity of the mouse promoter [31 ]. Evolutionary, conserved regions (as identified by the Evolutionary Conserved Region browser, http://ecrbrowser.dcode.org) are shown in light gray. The major (+1, +29) and minor (+32) transcription start sites determined by 5' RACE for the mouse Tlr9 gene are annotated with arrows. Human and mouse Tlr9 gene sequences are numbered relative to the most 5' (+1) transcription start site for that species. (B and C) A 2-kb genomic sequence (–1937 to +9 bp) and a 100-bp 5' deletion (–94 to +9 bp) were compared for promoter activity by transfection analysis of promoter-luciferase constructs. RAW 264.7 cells were transfected, and each transfection was split into multiple wells. One well was harvested at 24 h post-transfection for calculation of promoter basal activity. Cells in the other wells were left untreated or stimulated for a further 24 h with IFN-{gamma} before harvest at 48 h post-transfection. Each construct was electroporated in triplicate transfections within a single experiment. Luciferase activity was assayed to calculate RLU per µg total protein. (B) Comparison of basal activities of the 100-bp and 2-kb Tlr9 promoters against the pGL2B control plasmid. The mean and SE were calculated from four independent experiments. (C) Comparison of the inducible activities of the 100-bp and 2-kb Tlr9 promoters. Activity is expressed as average + SD of RLU/µg-fold induction by IFN-{gamma} over unstimulated cells from the same transfection. Data shown are representative of three independent experiments.

 
A composite IRF/PU.1 site is necessary for regulated Tlr9 expression in murine macrophages
A recent study of the human TLR9 promoter identified PU.1, C/EBP, and CRE elements, which regulate gene expression [30 ]. Both of the PU.1-binding sites are completely conserved between mice and humans, and one PU.1 site overlaps the major transcription start site in the mouse (Fig. 3A) . This site is adjacent to a consensus IRF half-site. Composite PU.1/IRF sites have been implicated in IFN-{gamma}-inducible gene expression for a number of genes (reviewed in ref. [41 ]), and the PU.1/IRF site overlapping the mouse Tlr9 transcription start site has high sequence homology with such sites (Fig. 4A ). Mutations in the PU.1 and IRF sites in the context of the 100-bp proximal promoter were generated (Fig. 4B) , and the effects of these mutations on basal and IFN-{gamma}-induced promoter activity were analyzed (Fig. 4C and 4D) . Mutation of the PU.1 site reduced basal promoter activity dramatically but did not affect induction by IFN-{gamma} (Fig. 4C and 4D) . EMSA analysis confirmed that PU.1 bound strongly to this site in the Tlr9 promoter and that the corresponding oligonucleotide with a mutation in the PU.1-binding region did not bind PU.1 (Fig. 5A and 5B ). Mutation of the IRF site did not affect PU.1 binding, suggesting that PU.1 binding was independent of occupancy at the IRF site (Fig. 5B) . ChIP analysis demonstrated further that PU.1 bound this site in BMM in vivo (Fig. 5C) and also bound the other conserved PU.1 site (–130 to –125) further upstream (Fig. 5D) . Mutation of the IRF site also decreased basal promoter activity (Fig. 4C) and in contrast to mutation of the PU.1 site, attenuated early (6 h) induction of the Tlr9 promoter by IFN-{gamma} (Fig. 4D) . Delayed induction (24 h) did not require the IRF site. Despite the contribution of the IRF site for basal and IFN-{gamma}-inducible promoter activity, EMSA analysis using the PU.1/IRF site did not identify IFN-{gamma}-regulated binding proteins (Fig. 5A) . We suggest that the EMSA conditions used did not mimic the in vivo situation, which occurs during IFN-{gamma} induction of Tlr9 transcription. The IRF half-site, which contributes to IFN-{gamma} induction of the murine Tlr9 promoter, is not conserved in the human TLR9 promoter (Fig. 3A) .


Figure 4
View larger version (31K):
[in this window]
[in a new window]

  		Figure 4. The Tlr9 promoter contains a consensus IRF/PU.1 motif. (A) The mouse Tlr9 promoter contains a sequence (reverse-complemented in this figure) homologous to composite PU.1/IRF sites functional in the regulation of a number of genes [41 , 42 ]. (B) The PU.1 and IRF sites of the predicted IRF/PU.1 motif were mutated within the context of the 100-bp promoter for use in transfection analysis. (C and D) RAW 264.7 cells were transfected with promoter mutants, and each transfection was split into three wells. One well was left untreated, and the other two wells were treated with IFN-{gamma} for 6 and 24 h. All wells were harvested at 48 h post-transfection. Each construct was electroporated in triplicate transfections within a single experiment. Luciferase activity was measured and normalized to total protein to calculate RLU per µg. Data points represent the average of triplicate quantitations + SD. Profiles are representative of at least three independent experiments. (C) Basal promoter activities of Tlr9 promoter mutants depicted in B were determined at 48 h post-transfection. (D) IFN-{gamma}-inducible promoter activities of Tlr9 promoter mutants depicted in B were determined at 48 h post-transfection.

 

Figure 5
View larger version (42K):
[in this window]
[in a new window]

  		Figure 5. The composite IRF/PU.1 site bound PU.1 in EMSA and ChIP. Nuclear extracts were prepared from IFN-{gamma}-stimulated RAW 264.7 and BMM maintained in CSF-1 or starved of CSF-1 overnight. Nuclear extracts were incubated with a P32-labeled probe containing the Tlr9 promoter IRF/PU.1 site, and the sample was size-fractionated on 8% polyacrylamide TGE gels and exposed to X-ray film. (A) Nuclear extracts from RAW 264.7 and CSF-1-replete or -starved BMM bound to the native IRF/PU.1 site in the Tlr9 promoter. Addition of a PU.1 antibody to the binding reaction supershifted (SS) the protein:probe complex. (B) EMSA studies with unlabeled competitor probes confirmed PU.1 bound to the consensus PU.1 site. Nuclear extracts from RAW 264.7 cells stimulated with IFN-{gamma} for 10 h were bound to probe cocktails containing unvarying amounts of labeled probe but zero-, ten-, 50-, or 100-fold molar excess of unlabeled competitor probes. The WT competitor probe sequence was identical to the labeled probe. The mutant probes varied from the labeled probe in that the IRF or PU.1 half-sites were mutated (as in Fig. 4B ). (A and B) Profiles are representative of six independent experiments using four independent preparations of nuclear extracts for each cell type. ChIP assay using an anti-PU.1 antibody to detect PU.1 binding to (C) the PU.1/IRF site (–12 to –1) or (D) the upstream PU.1 site (–130 to –125) of the Tlr9 promoter. Data points correspond to the average + SD of triplicate qRT-PCR quantitations within a single experiment. Data are representative of two independent experiments comparing anti-PU.1 antibody to a matched isotype control antibody.

 
ICSBP regulates basal and IFN-{gamma}-inducible Tlr9 expression in murine macrophages
Of the IRFs, only IRF4 (PU.1 interaction partner/lymphocyte-specific IRF/IFN consensus sequence-binding activated T cell) and ICSBP are known to interact with PU.1 to bind composite IRF/PU.1 sites. The expression patterns of IRF4 and ICSBP across IFN-{gamma} treatment of BMM and RAW 264.7 were investigated to determine if either was coregulated with Tlr9 mRNA. qPCR analysis showed a strong correlation between the expression of Icsbp and Tlr9 (Fig. 6A and B ). Like Tlr9, basal Icsbp mRNA expression was greatest in CSF-1-starved BMM, least in RAW 264.7, and intermediate in CSF-1-replete BMM. Similarly, Icsbp induction by IFN-{gamma} was greatest in RAW264.7 and CSF-1-replete BMM and weak in CSF-1-starved BMM. By contrast, Irf4 was expressed poorly in these cells, and its expression was not induced by IFN-{gamma} in CSF-1-replete BMM or RAW264.7 (Fig. 6C) , consistent with published data [43 ]. Because of the apparent coregulation of Tlr9 and Icsbp expression, we next investigated whether ICSBP contributes to basal and/or IFN-{gamma}-inducible Tlr9 mRNA expression in murine macrophages using ICSBP-deficient mice. Basal levels of Tlr9 mRNA were reduced 18-fold in ICSBP-deficient macrophages (Fig. 7A ). In addition, early but not late induction of Tlr9 mRNA by IFN-{gamma} was impaired in ICSBP-deficient macrophages (Fig. 7B) . These results are consistent with the effect of mutation of the IRF site in the Tlr9 promoter (Fig. 4D) .


Figure 6
View larger version (15K):
[in this window]
[in a new window]

  		Figure 6. Tlr9 and Icsbp were coregulated across CSF-1 and IFN-{gamma} stimulation. BMM were prepared and maintained overnight in the presence or absence of CSF-1. RAW 264.7 cells and CSF-1-replete or -starved BMM were treated with IFN-{gamma} for 7 or 24 h. mRNAs for (A) Tlr9, (B) Icsbp, and (C) Irf4 were quantified by qPCR from the same cDNA time series in experimental triplicate. Data shown are the average + SD of triplicate quantitations within a single experiment. Profiles are representative of at least two experiments using independent RNA samples.

 

Figure 7
View larger version (16K):
[in this window]
[in a new window]

  		Figure 7. Basal and IFN-{gamma}-inducible Tlr9 expression was compromised in ICSBP-deficient BMM. BMM from WT and ICSBP-deficient (ICSBP–/–) mice were prepared and stimulated with IFN-{gamma} in the presence of CSF-1. Tlr9 mRNA was measured by qPCR in experimental triplicate. The profiles depicted represent the mean and SE calculated from quantitation of three independent RNA samples. (A) Tlr9 expression relative to Hprt in unstimulated BMM; (B) Tlr9 expression relative to Hprt in IFN-{gamma}-stimulated BMM, expressed as fold induction over basal expression within the genotype at that time-point.

 
Type I IFNs induce Tlr9 expression in BMM
The observation that IRF site mutation or ICSBP deletion affected early but not delayed Tlr9 mRNA induction by IFN-{gamma} indicated that a separate mechanism was responsible for the late-stage induction. Given the time course, we hypothesized that an autocrine factor would be involved. Type I IFNs are major autocrine regulators of gene expression during macrophage activation [44 ], and Figure 8A shows that IFN-{gamma} induced Ifnß expression in CSF-1-replete BMM and RAW264.7 cells. We therefore assessed whether Type I IFN induced Tlr9 expression in BMM. Figure 8B shows that IFN-ß potently induced Tlr9 expression and that this effect was, at least in part, independent of ICSBP. Thus, Type I IFNs may contribute to the sustained induction of Tlr9 by IFN-{gamma}.


Figure 8
View larger version (18K):
[in this window]
[in a new window]

  		Figure 8. IFN-ß-dependent induction of Tlr9 was partially ICSBP-independent. (A) RAW 264.7 and CSF-1-replete BMM were left untreated or stimulated with IFN-{gamma} for 7 and 21 h. Expression of Ifnß mRNA relative to Hprt was measured by qPCR. (B) BMM from WT and ICSBP-deficient (ICSBP–/–) mice were prepared and stimulated with IFN-ß in the presence of CSF-1. Tlr9 mRNA was measured by qPCR and expressed as fold induction over basal expression within the cell type or treatment. Data points represent the mean and SE calculated from three experiments using independent RNA samples.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, we set out to define pathways by which IFN-{gamma} and LPS prime macrophage responses to CpG DNA and in doing so, have comprehensively analyzed the regulation of the murine Tlr9 gene. We found that transcriptional regulation of the mouse Tlr9 gene is geared toward promoting macrophage recognition and response to bacteria when in the presence of extracellular "danger" signals provided by the host (IFN-{gamma} and IFN-ß) and the invading pathogen (LPS). Our findings also provide a molecular mechanism for the well-documented ability of IFN-{gamma} to prime macrophage responses to CpG DNA [19 ]. Although IFN-{gamma} and LPS induced Tlr9 mRNA, there were fundamental differences in the mechanisms used. LPS primarily induced Tlr9 mRNA by down-regulation of surface CSF-1R and blockade of CSF-1 signaling (Fig. 1G 1H 1I 1J ; ref. [38 ]), although LPS-induced IFN-ß is also likely to contribute to regulation (Fig. 8B) . Conversely IFN-{gamma} did not greatly regulate cell surface CSF-1R (Fig. 1C 1D 1E 1F) but up-regulated activity of the Tlr9 promoter (Fig. 3C) . Indeed, IFN-{gamma} but not LPS induced Tlr9 mRNA in the RAW264.7 cell line (Fig. 2) , which expresses low levels of CSF-1R [39 ]. The dramatic (~65-fold at 21 h) up-regulation of Tlr9 expression by IFN-{gamma} in RAW264.7 cells presented here contrasts with another report [31 ], which showed only slight induction (approximately threefold at 24 h) and may reflect differences in cell culture protocols.

A number of studies have highlighted differences between mice and humans in the recognition of CpG motifs by TLR9. Human and mouse TLR9 have differences in the optimal CpG-containing sequence, which they recognize [13 , 45 ]. However, we showed recently that this difference is restricted to phosphorothioate-modified oligonucleotides and does not occur with natural phosphodiester DNA [46 ]. Hence, human and mouse TLR9 have not evolved to recognize different DNA sequences during pathogen challenge. The expression patterns of TLR9 between mice and humans have also been reported to be divergent. In particular, a number of studies suggest that TLR9 is expressed in myeloid cells in mice but not humans [2 , 11 12 13 ]. This is still an area of debate, as the human/mouse comparisons are somewhat confounded by a lack of directly comparable data; the human myeloid cells used for these investigations are generally human monocytes, and the mouse myeloid cells often studied are myeloid DCs, peritoneal macrophages, BMM, or cell lines such as RAW264.7. We have reported previously that CSF-1 down-regulates Tlr9 expression in murine macrophages [37 ], implying that mature murine tissue macrophages would be unlikely to express TLR9 and would be unresponsive to CpG DNA. Indeed, we (unpublished observations) and others [47 ] have shown that resident or thioglycollate-elicited peritoneal macrophages are poorly responsive to CpG DNA and express low levels of TLR9. This is in contrast to other publications and may reflect differential levels of circulating Type I and Type II IFNs in mice from different colonies or mouse strains; we show here that IFN-{gamma} and IFN-ß potently induced Tlr9 mRNA in mouse macrophages (Figs. 1A and 8B) , and the effect of IFN-{gamma} is also apparent in thioglycollate-elicited peritoneal macrophages (unpublished observations).

Divergence in regulation of Tlr promoters between mice and humans is well known. For example, although TLR4 is expressed predominantly in myeloid cells and B cells in humans and mice, basal tissue expression between the species is quite distinct. LPS induces TLR4 expression in human monocytes/macrophages but represses TLR4 expression in mouse macrophages (reviewed in ref. [48 ]). This is despite highly homologous promoter regions between the species. The human and mouse Tlr4 promoters resemble typical myeloid promoters in that they are short, TATA-less, harbor multiple PU.1 sites, and contain clustered transcription start sites [42 , 48 ]. A recent characterization of the human TLR9 promoter showed that PU.1, C/EBP, and CRE elements were important for expression in the human B cell line, RPMI 8226. Our study analyzed the murine promoter in a murine macrophage cell line. As with the Tlr4 promoter, there are likely to be differences in Tlr9 gene regulation between humans and mice, as well as differences between B cells and macrophages. Consequently, we found that many of the elements required for activity of the human promoter in B cells were dispensable for maximal promoter activity in murine macrophages (Fig. 3) . This included the 5' PU.1 box (–130 to –125), which although it bound PU.1 in BMM (Fig. 5D) , did not contribute significantly to basal promoter activity as seen by transfection analysis of 5' promoter deletions in RAW264.7 (Fig. 3B) . We did find that a conserved PU.1 site overlapping the murine transcription start site that was important for basal activity in human B cells and was also required for optimal basal activity in murine macrophages (Fig. 4C) . It remains possible that the 5' and 3' PU.1 sites are functionally redundant, and occupancy at only one site is required for maximal basal expression. We also identified a functional IRF half-site adjacent to the 3' PU.1 site in the murine promoter, which was not conserved in the human promoter (Fig. 3A) . The IRF/PU.1 site and the upstream PU.1 site identified in the present study were not identified in a previous report [31 ], which demonstrated only that an AP-1 site in the proximal promoter contributed to basal Tlr9 expression in mouse macrophages. The binding of AP-1 to the proximal Tlr9 promoter was also confirmed in the present study (data not shown). Apart from the nonconserved IRF site, the mouse and human Tlr9 promoters differ further; the mouse promoter, unlike its human counterpart, lacks a TATA box and is structurally typical of myeloid promoters. Like many other macrophage-specific promoters such as c-fms and FC{gamma}R1B [49 , 50 ], the transcription start sites of Tlr9 cluster around the PU.1 site. The transcription initiation of TATA-less promoters is not well-characterized; however, there is evidence that at least some TATA-less promoters form the preinitiation complex upon pyrimidine-rich initiator (Inr) motifs of consensus C/TC/TA(+1)NT/AC/TC/T [51 52 53 ]. PU.1 has been shown to interact in vitro with transcription factor IID [54 ], a general transcription factor necessary for Inr-mediated transcription in the absence of a TATA box. The sequence flanking the mouse Tlr9 A + 1 conforms to the Inr consensus, and the contiguous upstream sequence is a functional PU.1 site. This putative Inr motif, perhaps in concert with PU.1, may interact with an Inr-binding protein to tether the preinitiation complex to the DNA and drive transcription. Such differences in promoter architecture are likely to contribute to differential Tlr9 expression in the myeloid compartment in mouse versus humans.

We confirmed that the IRF site in the murine proximal promoter was required for basal activity and was partly required for induction of promoter activity by IFN-{gamma} using transfection analysis with a mutant promoter construct. The possibility that the effect of this mutation on basal activity was mediated by interference with PU.1 binding to the adjacent site is unlikely, as the mutated IRF site could still bind PU.1 in EMSA analysis (Fig. 5B) . The adjacent PU.1 site was dispensable for IFN-{gamma} induction, and in fact, mutation of this site modestly enhanced IFN-{gamma} induction (Fig. 4D) . It is possible that the Tlr9 promoter IRF/PU.1 site may be atypical in that rather than acting in synergy, PU.1 binding prevents optimal IRF interaction at this site. This IRF site is distinct from the ISRE/IRF sites identified by others in an upstream enhancer element [31 ], and it should be noted that none of the Tlr9 promoter constructs used in the present study contained these enhancer sites. This suggests that the effect of IFN-{gamma} is mediated via the proximal promoter IRF/PU.1 site, acting alone or in combination with the upstream enhancer. IRF4 and ICSBP are the only IRFs reported to be able to bind PU.1 at composite IRF/PU.1 sites. Although IRF4 and ICSBP are closely related, the phenotypes of the knockout mice demonstrated that they perform nonredundant functions in vivo [55 56 57 58 ]. They can both function as transcriptional activators or repressors, depending on cell type and sequence context. Typically, transcriptional repression occurs when heterodimeric complexes of IRF4 or ICSBP interacting with other IRFs bind to ISRE. However, IRF4 or ICSBP interaction with PU.1 and perhaps other IRFs upon composite IRF/PU.1 sites tend to result in promoter activation (reviewed in ref. [41 ]). The Icsbp, but not Irf4, mRNA expression pattern was coregulated with Tlr9 mRNA (Fig. 6) , and deletion of the Icsbp gene dramatically impaired basal Tlr9 expression as well as early Tlr9 mRNA induction by IFN-{gamma}, suggesting that ICSBP binds together with PU.1 at the composite IRF/PU.1 site. ICSBP is expressed primarily by B cells and macrophages [59 60 61 ], and as in this study, others have found that it is highly IFN-{gamma}-inducible at the mRNA and protein level in macrophages [59 , 62 63 64 ]. In fact, nuclear protein levels of ICSBP, but not IRF4, were induced 26-fold by IFN-{gamma} in RAW264.7, while nuclear levels of PU.1 remained unchanged [61 ]. Furthermore, we found expression levels of ICSBP to be at least tenfold higher than IRF4 in resting macrophages and at least 100-fold higher in IFN-{gamma}-stimulated macrophages at the mRNA level (Fig. 6) . Despite this evidence for ICSBP in regulating Tlr9 expression, we were unable to detect ICSBP interaction with the Tlr9 promoter or other reported targets by EMSA or ChIP assay in this study (data not shown). This is not entirely unexpected, as few reports demonstrate binding of an IRF/PU.1 complex to DNA by EMSA analysis using nuclear extracts from untransfected cells; instead, many used in vitro-translated protein [65 66 67 68 ] or nuclear extracts from transfected cells [69 ]. This suggests that native IRF/PU.1/DNA complexes are difficult to resolve for reasons of sensitivity or instability. The slight sequence deviation in the Tlr9 promoter IRF half-site from the consensus (TGACA instead of TGAAA; Fig. 4A ) may have weakened the interaction between ICSBP and DNA and added to difficulties in detecting the complex. Although our data suggest that IFN-{gamma} activates the Tlr9 promoter via ICSBP interaction with the IRF/PU.1 site, we cannot discount the possibilities that ICSBP regulates Tlr9 expression indirectly (i.e., through mechanisms other than occupation of the Tlr9 promoter) or that as found for other promoters [61 , 70 ], IRF4 may also bind PU.1 cooperatively at the IRF/PU.1 site. Further clarification of the roles of ICSBP and IRF4 in B cell-specific expression of Tlr9 is required.

Unexpectedly that we found that CSF-1 down-regulates expression of Icsbp mRNA in BMM (Fig. 6B) . This is likely to provide a mechanism for CSF-1-mediated repression of Tlr9 expression and downstream, TLR9-dependent responses [37 ]. This hypothesis is supported further by LPS induction of Icsbp mRNA, specifically in CSF-1-replete BMM but not in CSF-1-starved BMM or the CSF-1-unresponsive cell lines RAW264.7 and WEHI 231 (unpublished observations). In this way, the ICSBP/PU.1 complex may orchestrate integration of incoming signals from the host and potential pathogen for appropriate sensitization to CpG motifs. In addition to its transcriptional regulator activities, a recent article suggested that ICSBP may participate more directly in TLR signaling, via interaction with TRAF6 [71 ]. In light of the regulation of ICSBP by IFN-{gamma} and CSF-1 presented in this report, ICSBP may also function to integrate signals from CSF-1 and IFN-{gamma} directly into the TLR signaling pathway by modulating TRAF6 function.

The role of ICSBP in regulating Tlr9 gene expression has important implications for functional studies of ICSBP-deficient mice. ICSBP–/– mice show increased susceptibility to a range of infections [55 , 72 73 74 ], presumably as a result of their compromised ability to respond to IFNs and execute macrophage effector functions such as induction of gp91phox, gp67phox, inducible NO synthase, IL-18, and IL-1ß [70 , 75 76 77 ]. It is possible that part of the susceptibility to infections is mediated by deficiencies in basal and inducible Tlr9 expression. In addition, a recent study by Tsujimura et al. [78 ] using ICSBP–/– mice implicated ICSBP as a signaling molecule, which relays TLR9-specific signaling in DCs. We cannot discount this possibility, but given the effect of ICSBP deletion on Tlr9 expression, it is more likely that ICSBP is required for Tlr9 expression, rather than for signaling. This requires further clarification.

It is somewhat surprising that induction of Tlr9 mRNA by IFN-{gamma} at later time-points was not affected by mutation of the IRF site (Fig. 4D) or ablation of the Icsbp gene (Fig. 7B) . The time course suggests that an autocrine factor is responsible for this effect, and we provide evidence for the involvement of IFN-ß (Fig. 8) , although this cannot be concluded definitively. An IFN-ß-responsive element was mapped by others to an enhancer element ~2.4 kb upstream of the transcription start site [31 ], which was not contained in our promoter-reporter constructs. This suggests that IFN-{gamma}-induced IFN-ß may positively regulate Tlr9 expression via the upstream enhancer as well as an uncharacterized promoter-proximal element. A proposed mechanism for the IFN-{gamma}-mediated induction of Tlr9 via ICSBP and autocrine IFN-ß is shown in Figure 9 .


Figure 9
View larger version (35K):
[in this window]
[in a new window]

  		Figure 9. Proposed model for Tlr9 promoter regulation. In the absence of signaling by Type I or II IFN, LPS, or CSF-1, the Tlr9 promoter is activated by ICSBP/PU.1, and Tlr9 mRNA is expressed. CSF-1 suppresses transcription of Icsbp mRNA, thereby suppressing Tlr9 mRNA expression. Macrophage stimulation with IFN-{gamma} in the presence of CSF-1 activates the Icsbp promoter directly via STAT1 interaction with an IFN-{alpha}-activated site [79 ], and Icsbp expression is induced. In CSF-1-starved macrophages, IFN-{gamma} is unable to induce Icsbp mRNA beyond its current maximal level. In addition to direct Icsbp promoter activation, it is possible that induction of Icsbp mRNA by IFN-{gamma} occurs via an additional mechanism, which involves blockade of CSF-1 action downstream of the CSF-1R. IFN-{gamma} induces the expression and subsequent autocrine activities of Type I IFN, which is able to further transactivate the Tlr9 promoter through ICSBP-independent mechanisms, possibly including activation of upstream enhancer elements [31 ]. IFNAR, IFN-{alpha} receptor; IFNGR, IFN-{gamma} receptor.

 
In summary, the induction of Tlr9 mRNA by IFN-{gamma} in mouse macrophages provides a mechanistic explanation for the priming of CpG DNA responses by IFN-{gamma}. In contrast to induction by LPS, IFN-{gamma} targets the proximal promoter to up-regulate Tlr9. Our analysis of the murine Tlr9 promoter has also provided further mechanistic knowledge of differential regulation of TLRs between mice and humans.


   ACKNOWLEDGEMENTS
 
M. J. S., K. J. S., and D. A. H. are supported by the National Health and Medical Research Council (NHMRC) Project, grant numbers 301211 and 301210. T. R. is supported by a research grant for the Scripps NeuroAIDS Preclinical Studies Center (SNAPS) from the National Institute of Mental Health (NIMH), grant number 2P30MH062261-07. M. L. and M. R. are supported by a grant from the Deutsche Forschungsgemeinschaft (Re1310/7-1). We thank I. Horak for providing the ICSBP–/– mice used in these experiments.

Received January 18, 2007; accepted February 12, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Stacey, K. J., Young, G. R., Clark, F., Sester, D. P., Roberts, T. L., Naik, S., Sweet, M. J., Hume, D. A. (2003) The molecular basis for the lack of immunostimulatory activity of vertebrate DNA J. Immunol. 170,3614-3620[Abstract/Free Full Text]
  2. Hemmi, H., Takeuchi, O., Kawai, T., Kaisho, T., Sato, S., Sanjo, H., Matsumoto, M., Hoshino, K., Wagner, H., Takeda, K., Akira, S. (2000) A Toll-like receptor recognizes bacterial DNA Nature 408,740-745[CrossRef][Medline]
  3. Latz, E., Schoenemeyer, A., Visintin, A., Fitzgerald, K. A., Monks, B. G., Knetter, C. F., Lien, E., Nilsen, N. J., Espevik, T., Golenbock, D. T. (2004) TLR9 signals after translocating from the ER to CpG DNA in the lysosome Nat. Immunol. 5,190-198[CrossRef][Medline]
  4. Leifer, C. A., Kennedy, M. N., Mazzoni, A., Lee, C., Kruhlak, M. J., Segal, D. M. (2004) TLR9 is localized in the endoplasmic reticulum prior to stimulation J. Immunol. 173,1179-1183[Abstract/Free Full Text]
  5. Ahmad-Nejad, P., Hacker, H., Rutz, M., Bauer, S., Vabulas, R. M., Wagner, H. (2002) Bacterial CpG-DNA and lipopolysaccharides activate Toll-like receptors at distinct cellular compartments Eur. J. Immunol. 32,1958-1968[CrossRef][Medline]
  6. Sweet, M. J., Hume, D. A. (1996) Endotoxin signal transduction in macrophages J. Leukoc. Biol. 60,8-26[Abstract]
  7. Akira, S. (2000) Toll-like receptors: lessons from knockout mice Biochem. Soc. Trans. 28,551-556[Medline]
  8. Aderem, A., Ulevitch, R. J. (2000) Toll-like receptors in the induction of the innate immune response Nature 406,782-787[CrossRef][Medline]
  9. Dobrovolskaia, M. A., Vogel, S. N. (2002) Toll receptors, CD14, and macrophage activation and deactivation by LPS Microbes Infect. 4,903-914[CrossRef][Medline]
  10. Wagner, H. (2004) The immunobiology of the TLR9 subfamily Trends Immunol. 25,381-386[CrossRef][Medline]
  11. Krug, A., Towarowski, A., Britsch, S., Rothenfusser, S., Hornung, V., Bals, R., Giese, T., Engelmann, H., Endres, S., Krieg, A. M., Hartmann, G. (2001) Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amounts of IL-12 Eur. J. Immunol. 31,3026-3037[CrossRef][Medline]
  12. Kadowaki, N., Ho, S., Antonenko, S., Malefyt, R. W., Kastelein, R. A., Bazan, F., Liu, Y. J. (2001) Subsets of human dendritic cell precursors express different Toll-like receptors and respond to different microbial antigens J. Exp. Med. 194,863-869[Abstract/Free Full Text]
  13. Bauer, S., Kirschning, C. J., Hacker, H., Redecke, V., Hausmann, S., Akira, S., Wagner, H., Lipford, G. B. (2001) Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition Proc. Natl. Acad. Sci. USA 98,9237-9242[Abstract/Free Full Text]
  14. Saikh, K. U., Kissner, T. L., Sultana, A., Ruthel, G., Ulrich, R. G. (2004) Human monocytes infected with Yersinia pestis express cell surface TLR9 and differentiate into dendritic cells J. Immunol. 173,7426-7434[Abstract/Free Full Text]
  15. Takeshita, F., Leifer, C. A., Gursel, I., Ishii, K. J., Takeshita, S., Gursel, M., Klinman, D. M. (2001) Cutting edge: role of Toll-like receptor 9 in CpG DNA-induced activation of human cells J. Immunol. 167,3555-3558[Abstract/Free Full Text]
  16. Yi, A. K., Klinman, D. M., Martin, T. L., Matson, S., Krieg, A. M. (1996) Rapid immune activation by CpG motifs in bacterial DNA. Systemic induction of IL-6 transcription through an antioxidant-sensitive pathway J. Immunol. 157,5394-5402[Abstract]
  17. Hochrein, H., O’Keeffe, M., Wagner, H. (2002) Human and mouse plasmacytoid dendritic cells Hum. Immunol. 63,1103-1110[CrossRef][Medline]
  18. Wagner, H. (2002) Interactions between bacterial CpG-DNA and TLR9 bridge innate and adaptive immunity Curr. Opin. Microbiol. 5,62-69[CrossRef][Medline]
  19. Sweet, M. J., Stacey, K. J., Kakuda, D. K., Markovich, D., Hume, D. A. (1998) IFN-{gamma} primes macrophage responses to bacterial DNA J. Interferon Cytokine Res. 18,263-271[Medline]
  20. An, H., Yu, Y., Zhang, M., Xu, H., Qi, R., Yan, X., Liu, S., Wang, W., Guo, Z., Guo, J., et al (2002) Involvement of ERK, p38 and NF-{kappa}B signal transduction in regulation of TLR2, TLR4 and TLR9 gene expression induced by lipopolysaccharide in mouse dendritic cells Immunology 106,38-45[CrossRef][Medline]
  21. Pisetsky, D. S., Reich, C. (1993) Stimulation of in vitro proliferation of murine lymphocytes by synthetic oligodeoxynucleotides Mol. Biol. Rep. 18,217-221[CrossRef][Medline]
  22. Monteith, D. K., Henry, S. P., Howard, R. B., Flournoy, S., Levin, A. A., Bennett, C. F., Crooke, S. T. (1997) Immune stimulation—a class effect of phosphorothioate oligodeoxynucleotides in rodents Anticancer Drug Des. 12,421-432[Medline]
  23. Vollmer, J., Weeratna, R. D., Jurk, M., Samulowitz, U., McCluskie, M. J., Payette, P., Davis, H. L., Schetter, C., Krieg, A. M. (2004) Oligodeoxynucleotides lacking CpG dinucleotides mediate Toll-like receptor 9 dependent T helper type 2 biased immune stimulation Immunology 113,212-223[CrossRef][Medline]
  24. Baek, K. H., Ha, S. J., Sung, Y. C. (2001) A novel function of phosphorothioate oligodeoxynucleotides as chemoattractants for primary macrophages J. Immunol. 167,2847-2854[Abstract/Free Full Text]
  25. Sester, D. P., Brion, K., Trieu, A., Goodridge, H. S., Roberts, T. L., Dunn, J., Hume, D. A., Stacey, K. J., Sweet, M. J. (2006) CpG DNA activates survival in murine macrophages through TLR9 and the phosphatidylinositol 3-kinase-Akt pathway J. Immunol. 177,4473-4480[Abstract/Free Full Text]
  26. Van Uden, J. H., Tran, C. H., Carson, D. A., Raz, E. (2001) Type I interferon is required to mount an adaptive response to immunostimulatory DNA Eur. J. Immunol. 31,3281-3290[CrossRef][Medline]
  27. Car, B. D., Eng, V. M., Schnyder, B., Ozmen, L., Huang, S., Gallay, P., Heumann, D., Aguet, M., Ryffel, B. (1994) Interferon {gamma} receptor deficient mice are resistant to endotoxic shock J. Exp. Med. 179,1437-1444[Abstract/Free Full Text]
  28. Schroder, K., Hertzog, P. J., Ravasi, T., Hume, D. A. (2004) Interferon-{gamma}: an overview of signals, mechanisms and functions J. Leukoc. Biol. 75,163-189[Abstract/Free Full Text]
  29. Schroder, K., Sweet, M. J., Hume, D. A. (2006) Signal integration between IFN{gamma} and TLR signaling pathways in macrophages Immunobiology 211,511-524[CrossRef][Medline]
  30. Takeshita, F., Suzuki, K., Sasaki, S., Ishii, N., Klinman, D. M., Ishii, K. J. (2004) Transcriptional regulation of the human TLR9 gene J. Immunol. 173,2552-2561[Abstract/Free Full Text]
  31. Guo, Z., Garg, S., Hill, K. M., Jayashankar, L., Mooney, M. R., Hoelscher, M., Katz, J. M., Boss, J. M., Sambhara, S. (2005) A distal regulatory region is required for constitutive and IFN-ß-induced expression of murine TLR9 gene J. Immunol. 175,7407-7418[Abstract/Free Full Text]
  32. Tushinski, R. J., Oliver, I. T., Guilbert, L. J., Tynan, P. W., Warner, J. R., Stanley, E. R. (1982) Survival of mononuclear phagocytes depends on a lineage-specific growth factor that the differentiated cells selectively destroy Cell 28,71-81[CrossRef][Medline]
  33. Hume, D. A., Gordon, S. (1983) Optimal conditions for proliferation of bone marrow-derived mouse macrophages in culture: the roles of CSF-1, serum, Ca2+, and adherence J. Cell. Physiol. 117,189-194[CrossRef][Medline]
  34. Stacey, K. J., Ross, I. L., Hume, D. A. (1993) Electroporation and DNA-dependent cell death in murine macrophages Immunol. Cell Biol. 71,75-85
  35. Mita, Y., Dobashi, K., Shimizu, Y., Nakazawa, T., Mori, M. (2001) Toll-like receptor 2 and 4 surface expressions on human monocytes are modulated by interferon-{gamma} and macrophage colony-stimulating factor Immunol. Lett. 78,97-101[CrossRef][Medline]
  36. Bosisio, D., Polentarutti, N., Sironi, M., Bernasconi, S., Miyake, K., Webb,