JOURNAL OF INTERFERON & CYTOKINE RESEARCH 24:699–710 (2004)© Mary Ann Liebert, Inc.

Toll-like Receptor 2 Stimulation Decreases IFN-� ReceptorExpression in Mouse RAW264.7 Macrophages

HEATHER CURRY,1 GAIL R. ALVAREZ,1 BRUCE S. ZWILLING,2 and WILLIAM P. LAFUSE1

ABSTRACT

Interferon-� (IFN-�) is a key cytokine in the immune defense against mycobacteria. IFN-� activates macro-phages to resist the growth of mycobacteria and induces expression of MHC class II molecules required forantigen presentation. Macrophages infected with mycobacteria or stimulated by the interaction of mycobac-terial products with toll-like receptor 2 (TLR2) have reduced responses to IFN-�. Previous research has shownthat infection of mouse macrophages with Mycobacterium avium causes decreased expression of the IFN-� re-ceptor (IFNGR). In the present study, we show that TLR2 stimulation of RAW264.7 macrophages with a syn-thetic lipoprotein, Pam3CSK4, also causes rapid decrease in expression of IFNGR-1 protein, with little changein IFNGR-2 protein levels. The decrease in IFNGR-2 expression in TLR2-stimulated cells required receptorinternalization and proteasomal degradation. The level of IFNGR-1 mRNA also decreased in TLR2-stimu-lated RAW264.7 cells and M. avium-infected cells. The decrease in IFNGR-1 mRNA was shown to be due todecreased transcription. In spite of the decrease in IFNGR-2 receptor expression, activation of Stat1 activa-tion by an optimal dose of IFN-� was identical between control and TLR2-stimulated RAW264.7 cells. How-ever, at low suboptimal doses of IFN-�, Stat1 activation was decreased in TLR2-stimulated cells.

699

INTRODUCTION

HUMAN AND MOUSE STUDIES have shown that interferon-�(IFN-�), released in an immune response by T lympho-

cytes and natural killer (NK) cells, plays an important role inthe host defense against mycobacteria.(1–5) IFN-� acts on mac-rophages to increase the ability of the macrophage to resist thegrowth of intracellular pathogens by enhancing production ofantimicrobial effector molecules, such as reactive oxygenspecies (ROS) and nitric oxide (NO).(6–8) The receptor forIFN-� is expressed on lymphoid cells, including monocytes,macrophages, T cells, B cells, NK cells, and nonlymphoid cells,such as fibroblasts and endothelial cells.(9) The receptor con-sists of a ligand-binding subunit, termed either IFNGR-1, andan accessory chain, IFNGR-2, required for signaling.(10–13)

Binding of IFN-� to the receptor activates the Jak-Stat signal-ing pathway,(14–16) in which receptor-associated Jak1 and Jak2tyrosine kinases phosphorylate the IFNGR-1 chain. The tyro-sine-phosphorylated receptor recruits signal transducer and ac-tivator of transcription 1 (Stat1) to the receptor, resulting in thetyrosine phosphorylation of Stat1. The phosphorylated Stat1

forms dimers and translocates to the nucleus, where it regulatesthe induction of IFN-�-responsive genes.

Human and mouse macrophages infected with mycobacteriahave reduced response to IFN-�, resulting in decreased in-duction of MHC class II genes and other IFN-�-inducedgenes.(17–19) Previous research from this laboratory has shownthat prior infection of mouse macrophages with Mycobacteriumavium caused decreased Stat1 activation by IFN-�.(17) This de-crease was shown to correlate with a downregulation in the pro-tein and mRNA expression of IFNGR-1 and IFNGR-2. Infec-tion of human mouse macrophages with Mycobacteriumtuberculosis also has been shown to inhibit IFN-�-induced geneexpression, but this effect was observed without inhibition ofStat1 activation.(18,19)

Macrophages respond to mycobacteria through the interac-tion of mycobacterial products with toll-like receptors(TLR).(20–22) M. avium stimulates macrophages through TLR2,whereas M. tuberculosis acts through both TLR2 andTLR4.(23–25) Mycobacterial products that interact with TLR2include lipoarbinomannan, phosphatidylinositolmannan, and a19-kDa lipoprotein from M. tuberculosis.(23,26–28) Studies have

1Department of Molecular Virology, Immunology, and Medical Genetics, and 2Department of Microbiology, The Ohio State University, Colum-bus, OH 43210.

shown that exposure of mouse macrophages to the 19-kDalipoprotien inhibits IFN-� induced MHC class II expression, re-sulting in decreased antigen presentation to T lympho-cytes.(29–31) We have also observed a similar effect on IFN-�-induced gene expression using the 19-kDa lipoprotein as wellas TLR2 agonists, lipoarbinomannan and Pam3CSK4.(32) Wealso showed that exposure of mouse RAW264.7 macrophagesto TLR2 agonists does not inhibit the DNA binding of Stat1and the tyrosine phosphorylation of Stat1�. Instead, we showedthat TLR2 stimulation increases the phosphorylation and ex-pression of Stat1�, which is transcriptionally inactivated andacts as a dominant negative. This increase in Stat1� expressionwas shown to result from increased stability of Stat1� mRNAin TLR2-stimulated cells.

In the present study, we investigated the effect of TLR2 stim-ulation on the expression of the IFNGR. We show that stimu-lation of RAW264.7 cells with Pam3CSK4 rapidly reduces theprotein and mRNA expression of IFNGR-1. We further showthat this decrease in IFNGR expression causes a dose-depen-dent effect on Stat1 activation. At high doses of IFN-�, thereis no effect of Pam3CSK4 treatment on the level of Stat1 acti-vation. However, at low doses of IFN-�, Stat1 activation is de-creased in Pam3CSK4 treated RAW264.7 cells compared with

control cells. Thus, in vivo, where IFN-� concentrations wouldbe expected to be limiting, a decrease in IFNGR expression me-diated by stimulation of TLR by mycobacteria would be ex-pected to be a contributing factor to the decreased responsive-ness of mycobacteria-infected macrophages to IFN-�.

MATERIALS AND METHODS

Reagents

Fetal bovine serum (FBS) was obtained from Harlan Bio-products for Science (Indianapolis, IN). Mouse IFN-� was ob-tained from Roche Molecular Biochemicals (Indianapolis, IN).Actinomycin D and filipin were purchased from Sigma (SaintLouis, MO). Proteasome inhibitor I and chlorpromazine were ob-tained from Calbiochem (San Diego, CA), and Pam3CSK4 wasobtained from EMC Microcollections (Tübingen, Germany).

Mycobacteria

M. avium (ATCC 35713) was passed once through mice andcultured in Middlebrook 7H9 broth supplemented with oleicacid-albumin-dextrose-catalase (ODAC) (Difco, Detroit, MI).

CURRY ET AL.700

FIG. 1. TLR2 stimulation of RAW264.7 macrophages causes rapid decrease in the expression of IFNGR-1. RAW264.7 cellswere stimulated with 500 ng/ml Pam3CSK4 or infected with M. avium at 10:1. Cell lysates were prepared, and IFNGR expres-sion was determined by immunoblotting using anti-IFNGR-1 antibody (A) and anti-IFNGR-2 antibody (B). Data shown are rep-resentative Western blots and the densitometric analysis of data pooled from three separate experiments. The decrease in IFNGR-1 in Pam3CSK4-stimulated and M. avium-infected cells was significant by one-way ANOVA and Tukey’s test (p � 0.001).

Mycobacteria were stored in 1 ml aliguots in 10% glycerol at�70°C until used in experiments. Gamma-irradiated M. tuber-culosis H37Rv (Colorado State University, Fort Collins, CO;National Institutes of Health, National Institute of Allergy andInfectious Diseases Contract N01 AI-75320) was diluted inMiddlebrook 7H9 broth and centrifuged at 800 rpm for 10 minto eliminate clumped bacteria. The supernatant was stored inaliquots at �80°C. The protein concentration in the supernatantwas determined by the Bradford protein assay (Bio-Rad Labo-ratories, Hercules, CA).

Cell lines

The mouse macrophage cell line, RAW264.7, was obtainedfrom the American Type Culture Collection (Manassas, VA).RAW264.7 cells expressing dominant negative TLR2 were ob-tained by transfecting RAW264.7 cells with a Flag-taggedTLR2 �14 pcDNA3.1 plasmid (gift from Dr. M. Fenton,). Astable clone was obtained by selection with hygromycin andscreening for TLR2 expression by Western blotting using anti-Flag antibodies. RAW264.7 and RAW-TLR2 DN cells werecultured in DMEM supplemented with 10% FBS, 5 mM so-

dium pyruvate, 100 U/ml penicillin, and 100 �g/ml strepto-mycin. For experiments, cells were seeded into 6-well plates at5 � 106 cells/well and allowed to adhere for 2 h before the ad-dition of Pam3CSK4 or M. avium.

Western blot analysis

Cell lysates were obtained by lysis of cells in buffer con-taining 1% Triton as previously described.(32) Cell debris wasremoved by centrifugation at 4°C at 14,000 � g for 10 min.Protein concentrations were determined by Bio-Rad protein as-say. Proteins (40 �g) were resolved by SDS PAGE using 10%Tris-glycine gels (Invitrogen, Carlsbad, CA), transferred to PDFmembranes (Hybond P) (Amersham, Arlington Heights, IL),and probed with specific antibodies overnight at 4°C. The de-tection step was performed with peroxidase-coupled antimouseand antirabbit IgG antibodies (Abs) (diluted 1:5000) (Genotech,St. Louis, MO). Blots were developed with the femtoLucent de-tection system (Genotech). Primary antibodies used were rab-bit antimouse IFNGR-1 (Santa Cruz Biotechnology, SantaCruz, CA) (diluted 1:750), rabbit antimouse IFNGR-2 (SantaCruz) (diluted 1:1000), monoclonal phospho-Stat1 (Zymed,

TLR2 STIMULATION DECREASES IFNGR-1 EXPRESSION 701

FIG. 2. The decrease in IFNGR-1 expression in TLR2-stimulated RAW264.7 macrophages is concentration dependent.RAW264.7 macrophages were stimulated with 0–500 ng/ml Pam3CSK4, and cell lysates were prepared after 20 h. (A) Westernblots probed with anti-IFNGR-1 and anti-IFNGR-2 antibodies. (B) Densitometeric analysis of data pooled from three indepen-dent experiments. The decrease in IFNGR-1 expression was significant by one-way ANOVA and Tukey’s test (p � 0.001).

South San Francisco, CA) (diluted 1:2000), and monoclonalanti-Flag (Sigma) (1:1000).

Northern blot hybridization

RNA was isolated by the acid guandinium isothiocyanate-phenol-chloroform extraction method.(33) RNA (30 �g) wasfractionated in 1% formaldehyde agarose gels and transferredto Hybond-N� membranes (Amersham). Northern blot hy-bridization was performed as previously described.(34) Theprobes used were cDNA of IFNGR-1, IFNGR-2, and GAPDH.The IFNGR-1 and IFNGR-2 probes were obtained by RT-PCR.(17) The GAPDH was previously isolated from a mousemacrophage cDNA library.(34)

Measurement of mRNA transcript stability

To measure the effects of M. avium and TLR2 stimulationon IFNGR mRNA, RAW264.7 cells were cultured overnight in6-well plates. The cells were then infected with M. avium at

20:1 bacteria/macrophage and treated with 0.50 �g/mlPam3CSK4, or left untreated. After 1 h, actinomycin D (2�g/ml) was added to inhibit transcription. RNA was isolated at0, 1, 2, 4, 6, and 8 h. IFNGR-1 and IFNGR-2 mRNA was de-tected by Northern blot hybridization. Blots were stripped andreprobed with a GAPDH cDNA probe. Autoradiographs werequantified by densitometric analysis using SigmaScan Pro 4software (SPSS, Chicago, IL). The intensity of IFNGR-1 andIFNGR-2 mRNA was normalized to GAPDH. Linear regres-sion analysis of semilogarithmic plots of percent IFNGR mRNAremaining vs. time after addition of actinomycin D was used todetermine mRNA t1/2.

Transient transfection experiments

A luciferase reporter containing the mouse IFNGR-1 promoterwas constructed (GenBank accession number U05970).(35) TheDNA fragment, consisting of the �460 to �1 sequence, was syn-thesized by PCR using BALB/c genomic DNA. The PCR prod-ucts were digested with Kpn I and Bgl II, and the digested frag-

CURRY ET AL.702

FIG. 3. TLR2 stimulation decreases Stat1 activation at low concentrations of IFN-�. RAW264.7 cells were stimulated with 500ng/ml Pam3CSK4 or left untreated (Control) for 20 h. The macrophages were then stimulated with IFN-� (0–50 U/ml). After 45min, cell lysates were prepared, and activation of Stat1 was determined by immunblotting with antiphosphotyrosine-Stat1. (A)Representative Western blots. (B) Densitometeric analysis of data from three independent experiments. Results are shown as %of the maximal response at 50 U/ml.

ments were ligated into pGL3 basic luciferase vector (Promega,Madison, WI). The forward primer containing the Kpn I site andthe reverse primer containing the Bgl II site were: sense, 5�-CGCGGGTACCCTAGTGAAACTTCT CCGCCCACG-3�; an-tisense, 5�-GCGAAGATCTTCTGCGCGAGGGACACC TCT-GACAG-3�. Plasmid DNA was isolated using an Endofreeplasmid kit (Qiagen, Valencia, CA). RAW264.7 cells were platedin 24-well plates at 5 � 105 cells/ well overnight. The next day,the cells were transfected with 0.50 �g of the IFNGR-1-pGL3plasmid using LipofectAMINE PLUS (Invitrogen). After 3 h, thecells were stimulated with 0–500 ng/ml Pam3CSK4 or gamma-irradiated M. tuberculosis (0–750 (�g protein/ml). After 16 h,the cells were lysed, and luciferase activity was measured usingthe Promega luciferase assay system. The protein concentrationof the lysate was determined using the Bio-Rad protein assay,

and the results of the experiment are expressed as relative lightunits (RLU)/�g protein.

RESULTS

TLR2 stimulation decreases IFNGR-1 protein expression

In order to determine the effect of TLR2 stimulation on theexpression of the IFNGR, RAW264.7 macrophages were stim-ulated with Pam3CSK4; a synthetic lipoprotein that acts as aTLR2 agonist.(20) RAW264.7 cells were also infected with M.avium. Protein expression of IFNGR-1 and IFNGR-2 was de-termined by Western blotting. As shown in Figure 1A, TLR2stimulation caused a rapid decrease in IFNGR-1 expression,

TLR2 STIMULATION DECREASES IFNGR-1 EXPRESSION 703

FIG. 4. Proteasome inhibition and inhibition of receptor endoctyosis blocks the decrease in IFNGR-1 expression in TLR2-stim-ulated cells. (A) RAW264.7 macrophages were stimulated with 500 ng/ml Pam3CSK4 in the presence or absence of 1 �M (Con-trol) proteasome-inhibitor I (Prot In). After 1–4 h, cell lysates were prepared, and IFNGR-1 expression was determined by im-munobloting. The effect of proteasome-inhibitor I was significant at 3 and 4 h by one-way ANOVA (p � 0.020). (B) Receptorendocytosis was inhibited by treating RAW264.7 cells for 0–4 h with 500 ng/ml Pam3CSK4 in the presence of 25 �M chlor-promazine to block endocytosis by clathrin-coated pits and 5 �g/ml filipin to block receptor endocytosis by caveolae. Expres-sion of IFNGR-1 was determined by immunoblotting. The effect of chlorpromazine was significant at 3 h (p � 0.050) and 4 h(p � 0.020) by one-way ANOVA. The effect of filpin was significant at 3 h (p � 0.001) and 4 h (p � 0.010) by one-way ANOVA.Data shown in A and B are representative Western blots and the densitometric analysis of data pooled from three separate ex-periments.

which was equivalent to the decrease observed in M. avium-in-fected cells. In contrast, TLR2 stimulation had no effect onIFNGR-2 expression, and M. avium infection caused decreasedIFNGR-2 expression only at 16 h after infection. The effect ofPam3CSK4 on IFNGR-1 expression was concentration depen-dent (Fig. 2).

TLR2 stimulation alters the kinetics of Stat1 phosphorylation

As TLR2 stimulation reduces the level of IFNGR-1 expres-sion to only 25% of control RAW264.7 cells (Fig. 1A), we ex-amined the effect of TLR2 stimulation on activation of the Jak-Stat1 pathway. Control and Pam3CSK4-treated RAW264.7 cellswere stimulated with increasing concentrations of IFN-�. Ac-tivation of the Jak-Stat1 signaling pathway was determined byWestern blotting for tyrosine-phosphorylated Stat1. Maximalphosphorylation of Stat1 was observed at 50 U/ml IFN-�. Thelevel of Stat1 phosphorylation at this concentration was equiv-

alent in control RAW264.7 cells and Pam3CSK4-treated cells(Fig. 3). However, at lower concentrations of IFN-�, Pam3CSK4

treatment altered the kinetics of the response such that higherconcentrations of IFN-� were required to reach 50% of the max-imal response in TLR2-stimulated cells compared with controlcells. These results indicate that although TLR2 stimulationdoes not inhibit activation of the Jak-Stat1 pathway by IFN-�,the decrease in IFNGR-1 expression inhibits the ability of themacrophage to respond to low doses of IFN-�.

Proteasome inhibition and inhibition of receptorendocytosis blocks decrease in IFNGR-1 expression in TLR2-stimulated cells

To determine if IFNGR-1 is being degraded by the protea-some pathway in TLR2-stimulated cells, RAW264.7 were stim-ulated by Pam3CSK4 in the presence of proteasome-inhibitor I,which is a reversible inhibitor of the chymotrypsin-like activ-ity of the proteasome complex. As shown in Figure 4A, inhi-

CURRY ET AL.704

FIG. 5. TLR2 stimulation decreases IFNGR-1 mRNA expression. RAW264.7 macrophages were stimulated with 500 ng/mlPam3CSK4 (A) or infected with M. avium at a 10:1 bacteria/macrophage ratio (B). RNA was isolated at the indicated times, andNorthern blots were hybridized with IFNGR-1, IFNGR-2, and GAPDH cDNA probes. Data shown are representative Northernblots and densitometeric analysis of pooled data from three experiments. Expression levels of IFNGR-1 and IFNGR-2 were nor-malized to GAPDH levels and then expressed as % change relative to the control 0 h time point.

bition of proteasome activity decreased the degradation ofIFNGR-1. Previous studies have shown that the IFNGR is in-ternalized by caveolae and clathrin-coated vesicles.(36) In orderto determine if receptor endocytosis by caveolae is involved inthe decreased IFNGR-1 expression, caveolae were disrupted bytreatment with filipin, which selectively binds cholesterol,forming complexes in the plasma membrane that dispersescaveolae.(37,38) Receptor internalization by clathrin-coated pitswas inhibited by treatment with chlorpromazine, which is acationic amphiphilic drug that inhibits receptor internalizationby causing disassembly of plasma membrane clathrin-coatedpits.(39) Both filipin and chlorpromazine inhibited the decreasein IFNGR-1 protein after Pam3CSK4 stimulation (Fig. 4B).These studies indicate that the decrease in IFNGR-1 involvesreceptor internalization and degradation by the proteasome.

TLR2 stimulation decreases IFNGR-1 mRNA expression

We have shown previously that M. avium infection decreasesexpression of IFNGR mRNA.(17) To determine if TLR2 stim-ulation also decreases IFNGR mRNA, RAW264.7 macro-phages were stimulated with Pam3CSK4, and expression of re-ceptor mRNA was determined by Northern blot hybridization.

As shown in Figure 5A, TLR2 stimulation decreases IFNGR-1mRNA levels beginning 3 h after stimulation. The expressionof IFNGR-1 mRNA was decreased by 40% at 8 and 16 h afterstimulation. This decrease in IFNGR-1 mRNA in TLR2-stim-ulated cells was comparable to the decrease in expression ob-served in M. avium-infected cells (Fig. 5B). In contrast, ex-pression of IFNGR-2 mRNA increased by 30%–50% inTLR2-stimulated cells at 3–8 h after stimulation (Fig. 5A) andincreased slightly at 3–5 h in M. avium-infected cells and thendeclined (Fig. 5B).

In order to determine if the decrease in IFNGR-1 mRNA inM. avium-infected cells was due to TLR2 stimulation, we usedRAW264.7 cells stably transfected with dominant negativeTLR2, which lacks 14 amino acids in the cytoplasmic domainrequired for signaling.(26) These cells have decreased proin-flammatory gene responses after infection as shown (Fig. 6A)by the decreased induction of tumor necrosis factor-� (TNF-�)mRNA by infection with M. avium. The TLR2-dominant neg-ative cell line without infection consistently expressed slightlyless IFNGR-1 mRNA than RAW264.7. However, M. avium in-fection did not decrease IFNGR-1 mRNA expression any fur-ther in the TLR2 dominant negative RAW264.7 cells (Fig.6B,C), indicating that TLR2 stimulation is required for the de-crease in IFNGR-1 mRNA in M. avium-infected cells. The ex-

TLR2 STIMULATION DECREASES IFNGR-1 EXPRESSION 705

FIG. 6. Effect of M. avium infection on IFNGR and TNF-� mRNA levels in RAW264.7 cells stably transfected with dominantnegative (DN) TLR2. (A) RAW264.7 TLR2 dominant negative cells were infected with M. avium at 1:1, 5:1, and 10:1 bacteria/mac-rophage ratios. After 6 h, RNA was isolated, and TNF-� mRNA levels were determined by northern blotting. (B) RAW264.7 cellsand RAW264.7– TLR2 dominant negative cells were infected with M. avium (10:1). RNA was isolated at the indicated times, andexpression levels of IFNGR-1 and IFNGR-2 mRNA were determined by Northern blotting. (C) Densitometric analysis of IFNGR-1 mRNA levels. (D) Densitometric analysis of IFNGR-2 mRNA levels. Data shown in C and D were pooled data from three sep-arate experiments and are expressed as % of control 0 time point determined after normalization to GAPDH.

pression of IFNGR-2 mRNA was also less in unstimulatedRAW264.7-TLR2 dominant negative cells than in RAW264.7cells. The effect of M. avium infection was mostly similar (Fig.6B,D), although at 5 h after infection, there was a greater in-crease in expression in the RAW-TLR2 dominant negative thanin the control RAW264.7 cells.

Decreased IFNGR-1 mRNA expression is not due todecreased mRNA stability

We determined the stability of IFNGR-1 and IFNGR-2mRNA in TLR2-stimulated RAW264.7 macrophages and M. avium-infected cells. As shown in Figure 7, TLR2 stimula-

tion and M. avium infection did not affect the stability ofIFNGR-1 mRNA. In contrast, the stability of IFNGR-2 mRNAwas decreased by one half compared with control, unstimulatedcells.

TLR2 stimulation decreases IFNGR-1 promoter activity

In order to determine if TLR2 stimulation inhibits transcrip-tion of the IFNGR-1 alpha gene, we cloned the promoter of theIFNGR-1 gene(35) into the pGL3-luciferase vector. The IFNGR-1-pGL3 plasmid was transiently transfected into RAW264.7 cells,and the effect of Pam3CSK4 and irradiated M. tuberculosis stim-

CURRY ET AL.706

FIG. 7. Effect of M. avium and TLR2 stimulation on IFNGR-1 and IFNGR-2 mRNA stability. RAW264.7 cells were stimu-lated with Pam3CSK4 (500 ng/ml), M. avium (20:1 bacteria/macrophage), or left untreated (Control). After 1 h, transcription wasstopped with actinomcyin D and RNA was isolated at the indicated times. Northern blots were hybridized with (A) IFNGR-1and (B) IFNGR-2 cDNA probes. (C) Half-lives were determined by linear regression analysis of semilogarithmic plots of per-cent IFNGR mRNA remaining vs. time after addition of actinomycin D. The data represent the mean half-life � SD of three sep-arate experiments. IFNGR-1 mRNA stability was not significantly different. IFNGR-2 mRNA was significantly greater in con-trol cells than in Pam3CSK4 stimulated and in M. avium-stimulated cells by one-way ANOVA and Tukey’s test (p � 0.050).

ulation on promoter activity was determined. As shown in Figure8A, TLR2 stimulation with Pam3CSK4 inhibited IFNGR-1 pro-moter activity. A similar inhibition of IFNGR-1 promoter activ-ity was observed with irradiated M. tuberculosis (Fig. 8B), indi-cating that mycobacteria decrease promoter activity.

DISCUSSION

We have shown previously that prior stimulation ofRAW264.7 macrophages with TLR2 agonists inhibited geneexpression induced by IFN-�.(31) Here, we report that TLR2stimulation causes a rapid decrease in the protein expression of

IFNGR-1, whereas the expression of IFNGR-2 remains un-changed. This decrease in IFNGR-1 expression in TLR2-stim-ulated cells is identical to the decrease observed in M. avium-infected cells. Similar effects on IFNGR receptor expressionhave been demonstrated previously with two other pathogens.Studies by Ray et al.(40) showed that infection of phorbol myris-tate acetate (PMA)-differentiated U937 cells with Leishmaniadonovani decreased IFNGR-1 protein levels after infection.Kierszenbaum et al.(41) showed that infection of B lymphocyteswith Trypansoma cruzi decreased both cell surface and intra-cellular expression of IFNGR. In our studies, the expression ofIFNGR-1 was reduced to 20%–40% by 4–8 h after treatmentwith Pam3CSK4 or infection with M. avium. By using specificinhibitors, we also determined that the decrease in IFNGR-1

TLR2 STIMULATION DECREASES IFNGR-1 EXPRESSION 707

FIG. 8. TLR2 stimulation decreases IFNGR-1 promoter activity. RAW264.7 cells were transiently transfected with a lucifer-ase reporter construct containing �460 to �1 of the mouse IFNGR-1 promoter. The transfected cells were stimulated with (A)Pam3CSK4 or (B) irradiated M. tuberculosis for 16 h, and the luciferase activity was measured. Each condition was performedin quadruple, and the RLUs were normalized to the protein concentration of the lysate. The data represent mean � SD of a rep-resentative experiment of three independent experiments. The decreases in luciferase activity resulting from Pam3CSK4 and ir-radiated M. tb were significant by one-way ANOVA with Tukey’s test (p � 0.001).

chain induced by TLR2 stimulation required receptor endocy-tosis and degradation by the proteasome.

The IFNGR-1 chain exists both on the cell surface and in alarge intracellular pool.(42–44) Upon ligand binding, the recep-tor is internalized by receptor endocyotosis via caveolae andclathrin-coated pits.(36,45–47) In our studies, disruption of eithercavolae or clathrin-coated pits inhibited the decrease in IFNGR-1 in TLR2-stimulated RAW264.7 macrophages. Afterinternalization, the receptor complex dissociates in an acidifiedendosome; free IFN-� traffics to the lysosome and is degraded,and the IFNGR-1 chain enters the intracellular pool and even-tually recycles back to the cell surface. The rate of intracellu-lar degradation of IFN-� was found by Celada and Schreiber(43)

to be independent of the level of receptor occupancy. This sug-gests that the IFNGR, like other recycling receptors, continu-ously recycles, even in the absence of ligand. At present, wedo not know how TLR2 stimulation causes the loss of IFNGR-1 chains. The observation that the decrease in IFNGR-1 ex-pression is inhibited by a proteasome inhibitor suggests that theIFNGR in TLR2-stimulated cells, instead of recycling to thecell surface, is being degraded by a proteasome pathway. Onepossible reason is that receptor is being targeted for degrada-tion by interaction with a protein induced by TLR2 stimulation.One candidate is suppressor of cytokine signaling 3 (SOCS3),which contains the conserved SOCS box motif that has beenshown to bind elongins B and C, which couple proteins to theproteasomal protein degradation pathway.(48) We have foundexpression of SOCS3 mRNA to be upregulated through TLR2stimulation in M. avium-infected macrophages (unpublished ob-servations). Pai et al.(30) also have reported that expression ofSOCS1 and SOCS3 is upregulated by the M. tuberculosis 19-kDa lipoprotein.

In this study, we also show that TLR2 stimulation decreasesIFNGR-1 mRNA expression, which is comparable to the de-crease in mRNA expression in M. avium-infected RAW264.7macrophages. Using RAW264.7 macrophages transfected withdominant negative TLR2, which lacks amino acids in the cy-toplasmic region for signaling but is able to bind ligand, wewere able to determine that the decrease in IFNGR-1 mRNAin M. avium-infected macrophage requires TLR2 activation. Itis possible that the decrease in IFNGR-1 protein expression re-sults from this decrease in mRNA level. However, the decreasein protein expression is greater than the decrease in mRNA,suggesting that the decreased mRNA levels alone may not ac-count for the loss of IFNGR-1 protein.

We also report that TLR2 stimulation inhibits IFNGR-1 pro-moter activity to the same extent that the mRNA level is re-duced. TLR2 stimulation does not change the stability of theIFNGR-1 mRNA. Therefore, our data indicate that TLR2 stim-ulation decreases transcription of the IFNGR-1 gene. Additionalresearch will be needed to identify the regulatory element re-sponsible for the decrease in transcription. Regulatory motifspresent in the mouse IFNGR-1 gene include multiple SP1 andAP-2 sites and NF1 and CRE sites.(35) The IFNGR-1 gene isthought not to be regulated by external stimuli,(49) although arecent study showed that the human IFNGR-1 promoter is ac-tivated by Sp1 in response to phorbol ester treatment.(50) TLR2stimulation also affects the levels of IFNGR-2 mRNA. We haveobserved an increase in expression at 3–8 h after stimulation.The increase in mRNA expression occurred despite a 2-fold de-

crease in mRNA stability in TLR2-stimulated cells. This sug-gests that IFNGR-2 regulation by TLR2 is most likely both tran-scriptional and posttranscriptional.

In spite of the loss of IFNGR-1 protein, there was no dif-ference in the level of Stat1 phosphorylation in Pam3CSK4-stimulated cells when the cells were activated by saturating lev-els of IFN-�. This result is consistent with our report(32) thatTLR2 stimulation does not affect Stat1� phosphorylation or StatDNA binding in RAW264.7 cells stimulated with high levelsof IFN-� (200 U/ml). However, at lower doses of IFN-�, therewas reduced Stat1 phosphorylation in TLR2-stimulated cellscompared with control cells. This indicates that the decrease inIFNGR-1 does affect Stat1 activation when the concentrationof IFN-� is suboptimal. Thus, under limiting concentrations ofIFN-�, which would be expected to be encountered by macro-phages in vivo, the decrease in IFNGR-1 may contribute to re-duced expression of IFN-�-induced gene expression. However,decreased IFNGR-1 expression cannot account for the decreasein IFN-�-induced gene expression in TLR2-stimulated macro-phages that has been reported,(30,32) as this inhibition occurredat maximal levels of IFN-�. From these studies, we concludethat the major effect of TLR2 stimulation and mycobacterial in-fection is at the level of gene transcription, not at the level ofthe Jak-Stat1 signaling pathway. One mechanism by whichTLR2 stimulation decreases IFN-�-induced gene transcriptionis through upregulation of the dominant negative Stat1� iso-form, which lacks 38 amino acids at the C-terminus that are re-quired for transcriptional activation.(32,51) We previouslyshowed that TLR2 stimulation increases Stat1� expression byincreasing the stability of Stat1� mRNA.(32) The ability ofStat1� to act as a dominant negative was demonstrated by over-expression of Stat1� in transient transfection experiments withRAW264.7 cells. However, the inhibition of IFN-�-inducedgene expression by Stat1� overexpression was not as great asthat observed in RAW264.7 cells treated with TLR2 agonistsor cells infected with M. avium. This suggests that there areother mechanisms of gene silencing to be discovered.

In conclusion, mycobacteria inhibit IFN-�-induced gene ex-pression by multiple mechanisms. One of these pathways is adecrease in IFNGR-1 chain expression by TLR2 stimulation.Mycobacteria survive for long periods within macrophages ingranulomas. The mycobacteria are restricted to granulomasthrough the continuous activation of T lymphocytes. The abil-ity to inhibit antimicrobial pathways induced by IFN-� is a ma-jor factor in the survival of mycobacteria.

ACKNOWLEDGMENTS

This work was supported in part by National Institutes ofHealth grants AI-42901 and HL-59795.

REFERENCES

1. COOPER, A.M., DALTON, D.K., STEWART, T.A., GRIFFIN, J.P.,RUSSELL, D.G., and ORME, I.M. (1993). Disseminated tuberculo-sis in interferon-� disrupted mice. J. Exp. Med. 178, 2243–2247.

2. FLYNN, J. L., CHAN, J., TRIEBOLD, K.J., DALTON, D.K.,STEWART, T.A., and BLOOM, B.R. (1993). An essential role for

CURRY ET AL.708

interferon-� in resistance in Mycobacterium tuberculosis infection.J. Exp. Med. 178, 2249–2254.

3. NEWPORT, M.J., HUXLEY, C.M., HUSTON, S., HAWRY-LOWICZ, C.M., OSOSTRA, B.A., WILLIAMSON, R., andLEVIN, M. (1996). A mutation in the interferon-� receptor geneand susceptibility to mycobacterial infection. N. Engl. J. Med. 335,1941–1949.

4. DORMAN, S., and HOLLAND, S.M. (1998). Mutation in the sig-nal-transducing chain of the interferon-� receptor and susceptibil-ity to mycobacterial infection. J. Clin. Invest. 101, 2364–2369.

5. BELLAMY, R. (2003). Susceptibility to mycobacterial infections:the importance of host genetics. Genes Immun 4, 4–11.

6. CHAN, J., XING, Y., MARGLIOZZO, R.S., and BLOOM, B.R.(1992). Killing of virulent Mycobacterium tuberculosis by reactivenitrogen intermediates produced by activated murine macrophage.J. Exp. Med. 175, 1111–1122.

7. KAUFMANN, S.H.E. (1993). Immunity to intracellular bacteria.Annu. Rev. Immunol. 11, 129–163.

8. MACMICKING, J., XIE, Q.-W., and NATHAN, C. (1997). Nitricoxide and macrophage function. Annu. Rev. 15, 323–340.

9. VALENTE, G., OZMEN, L., NOVELLI, F., GENUNA, M.,PALESTRO, G., FORMI, G., and GAROTTA, G. (1992). Distri-bution of interferon-� receptor in human tissues. Eur. J. Immunol.22, 2403–2412.

10. AUGET, M., DEMBIC, Z., and MERLIN, G. (1988). Molecularcloning and expression of the human interferon-� receptor. Cell55, 273–280.

11. HEMMI, S., PEGHINI, P., METZLER, M., MERLIN, G., DEM-BIC, Z., and AUGET, M. (1989). Cloning of murine interferon-�receptor cDNA: expression in human cells mediates high affinitybinding but is not sufficient to confer sensitivity to murine inter-feron-�. Proc. Natl. Acad. Sci. USA 86, 9901–9905.

12. SOH, J., DONNELLY, R.J., KOTENKO, S., MARIANO, T.M.,COOK, J.R., WANG, N., EMANUEL, S., SCHWARTZ, B., MIKI,T., and PESTKA, S. (1994). Identification and sequence of an ac-cessory factor required for activation of the human interferongamma receptor. Cell 76, 793–802.

13. HEMMI, S., BOHNI, R., STARK, G., DI-MARCO, F., andAUGET, M. (1994). A novel member of the interferon receptorfamily complements functionality of the murine interferon gammareceptor in human cells. Cell 76, 803–810.

14. DARNELL, J.E., KERR, I.M., and STARK, G.R. (1994). Jak-Statpathway and transcriptional activation in response to IFNs andother extracellular signaling proteins. Science 264, 1415–1421.

15. SCHINDLER, C., and DARNELL, J.E. (1995). Transcriptional re-sponses to polypeptide ligands: the Jak-Stat pathway. Annu. Rev.Biochem. 64, 621–651.

16. IHLE, J.N. (1996). Stats: signal transducers and activators of tran-scription. Cell 84, 331–334.

17. HUSSAIN, S., ZWILLING, B.Z, and LAFUSE, W.P. (1999). My-cobacterium avium infection of mouse macrophages inhibits IFN-� Janus kinase-Stat signaling and gene induction by down-regulation of the IFN-� receptor. J. Immunol. 163, 2041–2048.

18. TING, L.M., KIM, A.C., CATTAMANCHI, A., and ERNST, J.D.(1999). Mycobacterium tuberculosis inhibits IFN-� transcriptionalresponses without inhibiting activation of Stat1. J. Immunol. 163,3898–3906.

19. KINCAID, E.Z., and ERNST, J.D. (2003). Mycobacterium tuber-culosis exerts gene-selective inhibition of transcriptional responsesto IFN-� without inhibiting Stat1 function. J. Immunol. 171,2041–2049.

20. ALIPRANTIS, A.O., YANG, R.-B., MARK, M.R., SUGGETT, S.,DEVAUX, B., RADLOF, J.D., KLIMPEL, G.R., GODOWSKI, P.,and ZYCHLINNSKY, A. (1999). Cell activation and apopotosisby bacterial lipoproteins through toll-like receptor-2. Science 285,736–739.

21. UNDERHILL, D.M., OZINSKY, A., SMITH, K.D., and ADR-EREM, A. (1999) Toll-like receptor-2 mediates mycobacteria-in-duced proinflammatory signaling in macrophages. Proc. Natl.Acad. Sci. USA 96, 14459–14463.

22. HELDWEIN, K.A., and FENTON, M.J. (2002). The role of toll-like receptors in immunity against mycobacterial infection. Mi-crobes Infect. 4, 937–944.

23. STENGER, S., and MODLIN, R.L. (2002). Control of Mycobac-terium tuberculosis through mammalian toll-like receptors. Curr.Opin. Immunol. 14, 452–457.

24. MEANS, T.K., WANG, S., LEIN, E., YOSHIMURA, A., GOLEN-BOCK, D.T., and FENTON, M.J. (1999). Human toll-like recep-tors mediate cellular activation by Mycobacterium tuberculosis. J.Immunol. 163, 3920–3927

25. LEIN, E., SELLATI, T.J., YOSHIMURA, A., FLO, T.H.,RAWADI, G., FINBEG, R.W., CARROLL, J.D., ESPEVIK, T.,INGALLS, R.R., RADOLF, J.D., and GOLENBACK, D.T. (1999).Toll-like receptor 2 functions as a pattern recognition receptor fordiverse bacterial products. J. Biol. Chem. 274, 33419–33425.

26. MEANS, T.K, LEIN, E., YOSHIMURA, A., WANG, S., GOLEN-BOCK, D.T., and FENTON, M.J. (1999). The CD14 ligandslipoarabinomannan and lipopolysaccharide differ in their require-ment for toll-like receptors. J. Immunol. 163, 6748–6755.

27. JONES, B.W., MEANS, T.K., HELSWEIN, K.A., KEEN, M.A.,HILL, P.J., BELISLE, J.T., and FENTON, M.J. (2001). Differenttoll-like receptor agonists induce distinct macrophage responses. J.Leukocyte Biol. 28, 1036–1044.

28. BRIGHTBILL, H.D., LIBRATY, D.H., KRUTZIK, S.R., YANG,R., BELISLE, J.T., BLEHARSKI, J.R., MAITLAND, M., NOR-GARD, M.V., PLEVY, S.E., SMALE, S.T., BRENNAN, P.J.,BLOOM, B.R., GODOWSKI, P.J., and MODLIN, R.L. (1999).Host defense mechanisms triggered by microbial lipoproteinsthrough toll-like receptors. Science 285, 732–736.

29. NOSS, E.H., PAI, R.K., SELLATI, T.J., RADOLF, J.D., BEIL-ISLE, J., GOLENBOCK, D.T., BOOM, W.H., and HARDING,C.V. (2001). Toll-like receptor 2-dependent inhibition of macro-phage class II MHC expression and antigen processing by 19–kDalipoprotein of Mycobacterium tuberculosis. J. Immunol. 167,910–918.

30. PAI, R.K., CONVERY, M., HAMILTON, T.A., BOOM, W.H.,and HARDING, C.V. (2003). Inhibition of IFN-�-induced class IItransactivator expression by a 19-kDa lipoprotein from Mycobac-terium tuberculosis: a potential mechanism for immune evasion. J.Immunol. 171, 175–184.

31. GEHRING, A.J., ROJAS, R.E., CANADAY, D.H., LAKEY, D.L.,HARDING, C.V., and BOOM, W.H. (2003). The Mycobacteriumtuberculosis 19-kilodalton lipoprotein inhibits gamma interferon-regulated HLA-DR and Fc�R1 on human macrophages throughtoll-like receptor 2. Infect Immun 71, 4487–4497.

32. ALVAREZ, G.R., ZWILLING, B.S., and LAFUSE, W.P. (2003).Mycobacterium avium inhibition of IFN-� signaling in mouse mac-rophages: Toll-like receptor 2 stimulation increases expression ofdominant-negative Stat1� by mRNA stabilization. J. Immunol.171, 6766–6773.

33. CHOMCZYNSKI, P., and SACCHI, N. (1987). Single-step methodof RNA isolation by acid guanidinium thiocyanate-phenol-chloro-form extract. Ann. Biochem. 162, 156–159.

34. LAFUSE, W.P., BROWN, D., CASTLE, L., and ZWILLING, B.S.(1995). Cloning and characterization of a novel cDNA that is IFN-�-induced in mouse peritoneal macrophages and encodes a puta-tive GTP-binding protein. J. Leukocyte Biol. 57, 477–482.

35. RAVAL, P., OBICI, S., RUSSELL, S.A., and MURPHY, W.J.(1994). Characterization of the 5� flanking region and gene en-coding the mouse interferon-gamma receptor. Gene 154, 219–224.

36. SADIR, R., LAMBERT, A., LORTAT-JACOB. H., and MOREL,G. (2001). Caveolae and clathrin-coated vesicles: two possible in-

TLR2 STIMULATION DECREASES IFNGR-1 EXPRESSION 709

ternalization pathways for IFN-gamma and IFN-gamma receptor.Cytokine 16, 19–26.

37. MCGOOKEY, D.J., FAGERBERG, K., and ANDERSON, R.G.(1983). Filipin-cholesterol complexes form in uncoated vesiclemembrane derived from coated vesicles during receptor-mediatedendocytosis of low-density lipoprotein. J. Cell Biol. 96, 1273–1278.

38. BOLARD, J. (1986). How do the polyene macrolide antibiotics af-fect the cellular membrane properties? Biochem. Biophys. Acta864, 257–304.

39. WANG, L.-H., ROTHERG, K.G., and ANDERSON, R.G. (1993).Mis-asembly of clathrin lattices on endosomes reveals a regulatoryswitch for coated pit formation. J. Cell Biol. 123, 1107–1117.

40. RAY, M., GAM, A.A., BOYKINS, R.A., and KENNEY, R.T.(2000). Inhibition of interferon-� signaling by Leishmania dono-vani. J. Infect. Dis. 181, 1121–1128.

41. KIERSZENBAYM, F., LOPEX, H.M., TANNER, M.K., andSZTEIN, M.B. (1995). Trypanosoma cruzi-induced decrease in thelevel of interferon-� receptor expression by resting and activatedhuman blood lymphocytes. Parasite Immunol. 17, 207–214.

42. ANDERSON, P., YIP, Y.K., and VILCEK, J. (1983). Human in-terferon-gamma is internalized and degraded by cultured fibro-blasts. J. Biol. Chem. 258, 6497–6502.

43. CELADA, A., and SCHREIBER, R.D. (1987). Internalization anddegradation of receptor-bound interferon-� by murine macro-phages. Demonstration of receptor recycling. J. Biol. 139, 147–153.

44. FINBLOOM, D.S. (1988). Internalization and degradation of hu-man recombinant interferon-gamma in the human histocytic lym-phoma U937, relationship to Fc receptor enhancement and anti-proliferation. Clin. Immunol. Immunopathol. 47, 93–105.

45. SUBRAMANIAM, P.S., and JOHNSON, H.M. (2002). Lipid mi-crodomains are required sites for the selective endocytosis and nu-clear translocation of IFN-�, its receptor chain IFN-� receptor 1,and the phosphorylation and nuclear translocation of Stat1�. J. Im-munol. 169, 1959–1969.

46. WILEMAN, T., BOSHANS, R.L., SCHLESINGER, P., andSTAHL, P. (1984). Monensin inhibits recycling of macrophage

mannose-glycoprotein receptors and ligand delivery to lysosomes.Biochem. J. 220, 665–675.

47. WATTS, C. (1984). Rapid endocytosis of the transferrin receptorin the absence of bound transferrin. J. Cell Biol. 100, 633–637.

48. ZHANG, J.G., FARLEY, A., NICHOLSON, S.E, WILSON, T.A.,ZUGARO, L.M., SIMPSON, R.J., MORITZ, R.L., CARY, D.,RICHARDSON, R., HAUSMANN, G., KILE, B.J., KENT, S.B.,ALEXANDER, W.S., METCALF, D., HILTON, D.J., NICOLA,N.A., and BACA, M. (1999). The conserved SOCS box motif insuppressors of cytokine signaling binds to elongins B and C andmay couple bound proteins to proteasomal degradation. Proc. Natl.Acad. Sci. USA 96, 2071–2076.

49. BACH, E.A., AUGET, M.A., and SCHREIBER, R.D. (1997). TheIFN-� receptor: a paradigm for cytokine receptor signaling. Annu.Rev. Immunol. 15, 563–591.

50. SAKAMOTO, S., and TANIGUCHI, T. (2001). Identification ofa phorbol ester-responsive element in the interferon-� receptor 1chain gene. J. Biol. Chem. 276, 37237–37241.

51. ZHANG, J.J., VINKEMEIER, U., GU, W., CHAKRAVARTI, D.,HORVATH, C.M., and DARNELL, J.E., Jr. (1996). Two contactregions between Stat1 and CBP/p300 in interferon � signaling.Proc. Natl. Acad. Sci. USA 93, 15092–15096.

Address reprint requests to:Dr. William P. Lafuse

Department of Molecular Virology, Immunology, and Medical Genetics

The Ohio State University333 W. 10th Avenue

Columbus, OH 43210

Tel: (614) 292-0652Fax: (614) 292-9805

E-mail: lafuse.1@osu.edu

Received 8 June 2004/Accepted 16 August 2004

CURRY ET AL.710