Oligodeoxynucleotide Dendritic Cells to the TLR9 Agonist Cpg Distinct

Distinct Responses of Lung and Spleen Dendritic Cells to the TLR9 Agonist CpG Oligodeoxynucleotide

This information is current as Li Chen, Meenakshi Arora, Manohar Yarlagadda, Timothy of September 24, 2021. B. Oriss, Nandini Krishnamoorthy, Anuradha Ray and Prabir Ray J Immunol 2006; 177:2373-2383; ; doi: 10.4049/jimmunol.177.4.2373

http://www.jimmunol.org/content/177/4/2373 Downloaded from

References This article cites 62 articles, 32 of which you can access for free at: http://www.jimmunol.org/content/177/4/2373.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 24, 2021

*average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Distinct Responses of Lung and Spleen Dendritic Cells to the TLR9 Agonist CpG Oligodeoxynucleotide1

Li Chen,* Meenakshi Arora,* Manohar Yarlagadda,* Timothy B. Oriss,* Nandini Krishnamoorthy,* Anuradha Ray,*† and Prabir Ray2*†

Dendritic cells (DCs) sense various components of invading pathogens via pattern recognition receptors such as TLRs. CpG oligodeoxynucleotides (ODNs), which mimic bacterial DNA, inhibit allergic airways disease and promote responses in the spleen to bacterial components. Because many TLR agonists are currently being tested for potential therapeutic effects, it is important to characterize the expression and function of TLRs in different tissues. We show that both myeloid and plasmacytoid DCs in the spleen express TLR9, the receptor for CpG ODNs, but lung DCs show no detectable expression in either subset. TLR4 expression in contrast was detected on both lung and spleen DCs. LPS was superior to CpG ODN in increasing the allostimulatory potential of lung DCs and their expression of CD40. However, both agonists efficiently stimulated spleen DCs. CpG ODNs administered to Downloaded from mice efficiently inhibited Th2 cytokine production both in the lung draining lymph node and in the spleen. Surprisingly, inhibition of Th2 cytokine production was evident despite high levels of expression of GATA-3 and additional transcription factors that regulate Th2 responses. Although in the spleen CpG ODNs induced IL-6, a key cytokine induced via TLR9-MyD88 signaling, no IL-6 was detectable in lung LN cells. These studies show for the first time that lung DCs lack TLR9 expression, but, despite this deficiency, CpG ODNs induce potent inhibitory effects on Th2 cytokine production in the lung without inducing expression of the proinflammatory cytokine, IL-6, which has been linked to chronic diseases in the lung and the gut. The Journal of Immunology, http://www.jimmunol.org/ 2006, 177: 2373–2383.

endritic cells (DCs)3 play an important role in Ag pre- tegic distribution, and their ability to capture and process Ag, and sentation in different tissues (1, 2), including the lung (3, present them to T cells in the LNs make DCs the key APCs in the D 4). Research since the late 1980s has shown an impor- lung and in other mucosal surfaces. tant role for DCs in pulmonary immune responses (3, 5–13). DCs Lung DCs are constantly exposed to a myriad of inhaled agents. are present in high density in the proximal airways and are well However, the lung maintains homeostasis in the face of constant positioned to capture Ags. Overall, pulmonary DCs are immature provocation by these multiple stimuli. Interestingly, although LPS by guest on September 24, 2021 in that they express low levels of costimulatory molecules (CD80, and CpG oligodeoxynucleotides (ODNs) are both bacterial com- CD86, and CD40). During inflammation, increased numbers of ponents, LPS induces severe inflammation in the lung precipitating DCs are rapidly recruited from bone marrow progenitors to the sepsis (20, 21), while CpG ODNs suppress allergic airway inflam- airway epithelium, where these immature DCs efficiently capture mation (22–25) and also growth of Mycobacterium tuberculosis in Ags due to their high phagocytic ability (6, 12, 14). Subsequent to the lungs (25). Like the lung, the spleen also encounters pathogens Ag capture, the DCs undergo maturation, which is influenced by being the principal filter unit for pathogens that enter the blood- the microenvironment, and the mature DCs traffic to local draining stream. Administration of CpG ODNs to mice has been shown to lymph nodes (LNs). The transition from immature to mature state cause transient splenomegaly (26), and CpG ODNs potentiate ef- is accompanied by the production of various cytokines and che- fects of TLR agonists in the spleen (27, 28). Bacterial CpG motifs mokines by the DCs that regulate their ability to interact with naive have been shown to promote chronic intestinal inflammation via T cells to direct T cell differentiation (1, 2, 6, 15–19). Their stra- secretion of proinflammatory cytokines such as IL-6 and IFN-␥ (29). CpG ODNs are increasingly being used in various clinical trials (30). Therefore, it is important to better understand the effects *Department of Medicine, Pulmonary, Allergy, and Critical Care Division, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213; and †Department of Immu- of CpG ODNs in different tissues and the expression characteris- nology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213 tics of TLR9, which is the best-studied receptor for CpG ODNs. Received for publication November 17, 2005. Accepted for publication May Because little is known about TLR expression in lung DCs, which 26, 2006. are the key APCs in the lung, we investigated whether CpG ODNs The costs of publication of this article were defrayed in part by the payment of page exert similar or differential effects on DCs from the lung vs the charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. spleen, which serve distinct functions in the body. 1 This work was supported by Grants HL 60207 (to P.R.) and HL 77430 (to A.R.) We show that while LPS had similar effects on lung and spleen from the National Institutes of Health. DCs and the DCs in turn stimulated CD4ϩ T cells in a comparable 2 Address correspondence and reprint requests to Dr. Prabir Ray, Department of Med- fashion, CpG ODNs exerted differential effects on lung and spleen icine, Pulmonary, Allergy, and Critical Care Medicine, University of Pittsburgh DCs both in in vitro assays as well as when administered to mice. School of Medicine, 3459 Fifth Avenue, MUH 628 NW, Pittsburgh, PA 15213. E- mail address: [email protected] Studies of the expression profile of the two TLRs at the two lo- 3 Abbreviations used in this paper: DC, dendritic cell; alum, aluminum hydroxide; cations showed that while TLR4 expression is comparable be- cRPMI, complete RPMI; CT, cholera toxin; GUS, ␤-glucuronidase; IDO, indoleam- tween lung and spleen DCs, only the latter express TLR9 with no ine 2,3-dioxygenese; i.n., intranasal; LN, lymph node; pDC, plasmacytoid DC; mDC, myeloid DC; ODN, oligodeoxynucleotide; TRIF, Toll/IL-1 receptor domain-contain- detectable expression in lung DCs. Despite the lack of TLR9 ex- ing adapter-inducing IFN-␤; wt, wild type. pression in lung DCs, CpG ODN administration caused potent

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 2374 DISTINCT RESPONSES OF LUNG AND SPLEEN DCs TO CpG ODN inhibition of allergen-induced Th2 cytokine production in both the Flow cytometry spleen and the lung LN. However, only in the spleen, production Cells were first incubated for 5 min with Fc block (10 ␮g/ml; BD Bio- of IL-6, a key cytokine downstream of TLR9 (31–33) and MyD88 sciences) to minimize nonspecific Ab binding and were then incubated (34), was noted. Interestingly, this inhibition of cytokine produc- with saturating concentrations of appropriate Abs for 5 min on ice in 2% tion was evident in the presence of high levels of expression of FBS/PBS, after which the cells were washed in the same buffer. Before GATA-3, which we and others (35, 36) previously demonstrated to be intracellular TLR9 staining, DCs were stained with dye-conjugated Abs against the cell surface molecules CD11c, MHC class II, and B220. The the master regulator of Th2 differentiation. With the recent identifi- cells were washed, fixed for 30 min using freshly prepared Fix/Perm so- cation of TLR9-independent mechanisms of CpG action (37, 38), our lution (eBioscience), and were washed in PBS and 1ϫ permeabilization data suggest that CpG ODNs can exert both TLR9-dependent (IL-6) buffer (eBioscience) successively to permeabilize the cells. The cells were and -independent effects to regulate immune responses. incubated with biotinylated mAb against murine TLR9 (clone 5G5) or with isotype control (mouse IgG2a), as described (41). Streptavidin-PerCP (BD Biosciences) was used for detection. TLR9 staining was also investigated Materials and Methods using another anti-TLR9 Ab IMG-431 (Imgenex). All samples were ana- Mice lyzed using a FACSCalibur flow cytometer and CellQuest software (BD Biosciences). Dead cells were excluded using forward and side light scatter Ϫ/Ϫ BALB/cByJ, CD40 mice (backcrossed to BALB/c background), and properties. Cell sorting was conducted using a FACSAria cell sorter and C57BL/6 male mice, obtained from The Jackson Laboratory, and DO11.10 FACS Diva software (BD Biosciences). TCR transgenic mice, originally provided by K. Murphy (Washington Uni- versity School of Medicine, St. Louis, MO), were housed under pathogen- Microscopy free conditions, were generally used at 6–8 wk of age, and were used under appropriate institutional guidelines. For scanning electron microscopy, freshly isolated DCs were plated onto 12-mm glass coverslips in 24-well plates and allowed to attach for 15 min Downloaded from Reagents at 37°C. The cells were then fixed with 2.5% glutaraldehyde and prepared for electron microscopy using standard techniques. Purified preparations of LPS from Escherichia coli (O55:B5) were pur- For light microscopy, DCs were cytospun onto glass slides (800 rpm, 10 Ј chased from List Biological Laboratories. The CpG ODN 1826 (5 -TC min) and air dried. The cells were stained with Hema-3 reagent (Fisher Ј CATGACGTTCCTGACGTT-3 ), which is known to be optimal for stim- Scientific), according to the manufacturer’s recommendations. ulation of murine cells, and the control ODN 1911 (5Ј-TCCAGGAC TTTCCTCAGGTT-3Ј) were used. The LPS level in the ODN preparations MLRs http://www.jimmunol.org/ was very low (Ͻ0.1 ng/mg DNA). The following mAbs were purchased from BD Pharmingen: PE-labeled Ab against CD11c (clone HL3), allo- DCs were gamma irradiated at 2000 rad, washed with fresh cRPMI, and phycocyanin-labeled Ab against CD11c (clone HL3), biotinylated Ab seeded in triplicate in round-bottom 96-well plates for use as stimulator ϩ against CD40 (clone 3/23), PE-conjugated anti-CD11b (clone M1/70), anti- cells at 1.2–20 ϫ 103 cells/well. Allogeneic responder CD4 T cells were CD19 (clone 1D3), anti-Gr-1 (clone RB6-8C5), anti-CD80 (clone16- obtained from freshly isolated splenocytes from C57BL/6 mice by mag- ϩ 10A1), anti-CD86 (clone GL1), and PerCP-labeled anti-CD3e (clone 145- netic bead positive selection methods. Purified CD4 T cells (1 ϫ 105 2C11). PE-labeled anti-MHC class II (clone NIMR-4) was purchased from cells/well) were added to the DCs in a total volume of 200 ␮l of cRPMI, Southern Biotechnology Associates. Biotinylated anti-DEC 205 was a gift and then cultured for 96 h. For the final 18 h of culture, the cells were from R. Hendricks (University of Pittsburgh, Pittsburgh, PA). Biotinylated pulsed with 1 ␮Ci of [3H]thymidine/well (catalog no. NET-027; Abs against TLR4 (clone MTS510) and TLR9 (clone 5G5) were obtained PerkinElmer). Each sample was assayed in triplicate. Cells were harvested from HyCult Biotechnology. Streptavidin-conjugated allophycocyanin was using a cell harvester, and incorporation of radioactivity was assessed by by guest on September 24, 2021 obtained from Caltag Laboratories. The appropriate isotype controls used liquid scintillation counting. Results are expressed as mean cpm. were: hamster IgG1 PE (clone G235-2356), hamster IgG1 allophycocyanin (clone G235-2356), hamster IgG1 PerCP (clone A19-3), rat IgG2a PE RT-PCR (clone R35-95), rat Ig2a biotin (clone R35-95), mouse IgG2a biotin (clone Total RNA was extracted from freshly isolated lung and spleen DCs with G155-178) (BD Pharmingen), and rat IgG2b PE (clone KLH/G2b-1-2) TRIzol reagent and RNeasy Mini Kit (Qiagen). (Southern Biotechnology Associates). Isolation and purification of DCs from lungs of mice Semiquantitative RT-PCR Lung DCs were isolated by a modification of previously published methods First strand cDNA was synthetized using 300 ng of total RNA using oli- (24, 39). Briefly, BALB/cByJ mice were anesthetized with ketamine/xy- go(dT) (catalog no. Y01212; Invitrogen Life Technologies) and Super- script II reverse transcriptase (Invitrogen Life Technologies). PCR was lazine mixture. After exsanguination via the abdominal aorta, the pulmo- Ј Ј nary vasculature was perfused with sterile 10 U of heparin/ml PBS to performed with the primers 5 -AGTGGGTCAAGGAACAGAAGCA-3 and 5Ј-CTTTACCAGCTCATTTCTCACC-3Ј for TLR4, 5Ј-CCAGACG remove peripheral blood cells. The perfused lungs were removed, cut into Ј Ј Ј small pieces, and incubated in an enzyme solution containing 0.7 mg/ml CTCTTCGAGAACC-3 and 5 -GTTATAGAAGTGGCGGTTGT-3 for ␮ TLR9, and 5Ј-GTTGGATACAGGCCAGACTTTGTTG-3Ј and 5Ј-GAG collagenase A (Boehringer Mannheim) and 30 g/ml type IV bovine pan- Ј creatic DNase I (Sigma-Aldrich) in serum-free RPMI 1640 for 90 min at GGTAGGCTGGCCTATAGGCT-3 for hypoxanthine phosphoribosyl- 37°C. Digested lung tissue was ground on a cell strainer (70 ␮m), partic- transferase. The PCR products were separated by electrophoresis in 2% ulate matter was removed by rapid filtration through a new cell strainer, agarose gels and were visualized by ethidium bromide staining. and the filtered cells were washed in complete RPMI (cRPMI) (RPMI 1640 Ϫ Quantitative PCR supplemented with 10% FBS, 5 ϫ 10 5 M 2-ME, 1 mM sodium pyruvate, 100 U/ml penicillin, 100 ␮g/ml streptomycin, and 20 ␮g/ml gentamicin). Predesigned gene-specific Taqman probe and primer sets were used for After washing with 0.5% BSA, 2 mM EDTA in PBS, lung cells were quantitative RT-PCR of TLR-4, TLR-9, and ␤-glucuronidase (GUS). First resuspended in 2% FBS in PBS, overlaid on the same volume of Nycodenz strand cDNA was synthesized using High Capacity cDNA Archive kit (density 1.068 g/ml), and centrifuged at 600 ϫ g for 15 min at room (Applied Biosystems). Samples were then subjected to real-time PCR anal- temperature. After centrifugation, cells at the interphase were collected and ysis using the ABI PRISM 7700 Sequence System (Applied Biosystems). washed with 0.5% BSA, 2 mM EDTA in PBS. Lung DCs were further Relative mRNA abundance of each transcript was normalized against purified by positive selection using CD11c microbeads (Miltenyi Biotec). GUS, calculated as 2(Ct[GUS] Ϫ Ct[gene]), where Ct represents the threshold cycle for each transcript and the resulting numbers were multiplied by a Isolation and purification of spleen DCs factor of 103. Spleen DCs were isolated by minor modifications of a previously published method (40). Briefly, spleen cell suspensions were prepared by collagenase Mouse immunizations and culture of isolated cells A and DNase I digestion, and single cells were cultured for2hat37°C in Intranasal (i.n.) immunization and isolation of lung-draining LNs. Mice plastic culture plates, followed by removal of nonadherent cells. Adherent were lightly anesthetized by isoflurane inhalation, and then they received cells were collected and washed with PBS containing 2% FBS three times. by i.n. route 5 ␮g of CpG ODN, 100 ␮g of OVA plus 1 ␮g of cholera toxin CD11cϩ DCs were purified from these cells with anti-CD11c magnetic (CT), or 5 ␮g of CpG ODN first and followed by 3 h later with 100 ␮gof beads (Miltenyi Biotec). OVA plus 1 ␮g of CT for 3 consecutive days. The mice were sacrificed on The Journal of Immunology 2375 day 5 post last immunization. The lung-draining LNs were harvested, and single-cell suspensions were used, unless otherwise stated. Intraperitoneal immunization and isolation of spleens. Mice were immunized by i.p. injection of 25 ␮g of CpG ODN alone or 10 ␮gofOVA plus 2 mg of aluminum hydroxide (alum), or 25 ␮g of CpG ODN and followed by 3 h later with 10 ␮g of OVA plus 2 mg of alum. Following 5 days of rest, mice were given 1 boost (i.p.) with the same amount of Ag and adjuvants. Mice were sacrificed 24 h later, and spleens were harvested and single splenocytes were isolated. RBC were lysed by using ammonium chloride lysing reagent (BD Biosciences). Lung LN or spleen cells from each group were cultured in cRPMI (2.5 ϫ 106 cells/well) in the presence of OVA protein (100 ␮g/ml) for 5 days. Supernatants were collected for cytokine profile analysis. Cytokine assays Cytokine assays were performed by ELISA (R&D Diagnostic) or using a multiplex system (Luminex; Bio-Rad) using commercially available kits, according to the manufacturer’s instructions. Preparation of nuclear extracts and Western blotting Lung DCs were incubated for 40 h in cRPMI with or without LPS, control ODN 1911, or CpG ODN 1826. The cells were washed with fresh medium Downloaded from and cultured for another 5 days with CD4ϩ T cells from spleens of DO11.10 TCR transgenic mice, in the presence of the specific OVA peptide (pOVA323–339)at5␮g/ml and IL-2 (10 U/ml) in fresh medium. Nuclear extracts were prepared from CD4ϩ T cells following culture with differentially treated lung DCs, as described previously (35, 42). Nu- FIGURE 1. Purity and morphology of lung DCs. A, CD11cϩ cells were clear protein content was determined by protein assay (Bio-Rad). Equal collected from the lung of mice using a combination of density-gradient amounts of nuclear protein were subjected to 4.5–15% gradient SDS- ϩ http://www.jimmunol.org/ centrifugation and positive selection methods. The purity of the CD11c PAGE, and the resolved proteins were transferred to polyvinylidene diflu- cells was Ͼ97%. B, Scanning electron microscopy and light microscopy of oride membrane. The blots were probed with anti-GATA3 mAb, anti-T-bet ϩ polyclonal Ab, or anti-CREB-1 Ab (Santa Cruz Biotechnology), followed freshly isolated CD11c DCs. C, Lack of CD3-, CD19-, and Gr-1-express- by HRP-conjugated secondary Ab. Bands were visualized by ECL ing cells in the purified DCs. D, Expression of CD11b, MHC II, and (Amersham). DEC205 on CD11cϩ DCs after brief incubation with GM-CSF (10 ng/ml). Light lines indicate staining with isotype control Ab. Shown is a represen- Statistical analysis tative of three independent experiments. Results are presented as mean Ϯ SD. Statistical comparison between two different groups was performed using Student’s t test. Statistical significance was defined as p Ͻ 0.05. DCs (pDCs) and myeloid DCs (mDCs) express TLR9, and pDCs by guest on September 24, 2021 typically reside in LNs draining respective tissues unless recruited Results ϩ into the tissues due to specific infections. Because it is not feasible Isolation of lung CD11c DCs and characterization to isolate enough DCs from lung-draining LNs of naive mice, we A combination of density-gradient centrifugation and CD11c-pos- immunized mice with Ag ϩ adjuvant to examine TLR9 expression itive selection was used to recover lung and spleen DCs (24, 39). in DCs isolated from lung-draining LNs and spleens. CpG ODN, The density-gradient-positive selection combination approach con- known to promote Th1 responses in different tissues including the sistently resulted in the recovery of a population of cells that was lung, and the best known TLR9 agonist, was used with Ag (OVA) Ͼ97% CD11cϩ with the characteristics of DCs as observed by and administered either i.n. to induce immune responses in the scanning electron microscopy and light microscopy (Fig. 1, A and lung (43), or i.p. to stimulate responses in the spleen (44, 45). DCs B). Also, this population was largely devoid of CD3ϩ, CD19ϩ,or were purified from lung-draining LNs or spleens of mice, and Gr-1ϩ cells (Fig. 1C). This approach resulted in a typical recovery TLR9 expression was examined by flow cytometry. As shown in of 1.0ϳ1.2 ϫ 105 CD11cϩ DCs cells per mouse lung. The cells Fig. 2C, TLR9 expression was readily detected in both mDCs and were briefly cultured in GM-CSF and analyzed for expression of pDCs isolated from spleens with greater expression in mDCs. Sur- relevant cell surface molecules. As shown in Fig. 1D, these cells prisingly, TLR9 protein expression could not be appreciated in expressed CD11b, MHC class II, and DEC205. lung DCs in either subset. We have also observed a high level of TLR9 expression in thioglycolate-elicited peritoneal macrophages Differential expression of TLR9 in lung and spleen DCs (data not shown). Additionally, we have confirmed this differential To study the response of lung DCs to TLR4 and TLR9 agonists, TLR9 expression profile between lung and spleen DCs using a we first examined expression of the TLRs in the purified lung DCs different anti-TLR9 Ab (data not shown). and compared with expression in spleen DCs. The expression of both TLRs was examined by semiquantitative and quantitative RT- Allostimulatory potential of stimulated DCs PCR. Using both techniques, lung DCs were found to express We tested the allostimulatory capacity of DCs isolated from lung slightly higher levels of TLR4 compared with spleen DCs, while tissue and spleen after treating with either LPS, a TLR4 agonist, or the converse was true for TLR9 expression (Fig. 2A). At the pro- CpG ODN, a TLR9 agonist, to assess the functional consequence tein level, as determined by flow cytometry, TLR4 expression was of differential expression of TLRs on lung and spleen DCs. We clearly evident on the cell surface of both lung and spleen DCs first used spleen DCs to establish a dose response to the agonists. (Fig. 2B). Because the RT-PCR data showed low expression of Based on the results of the dose-response study, as shown in Fig. TLR9 in lung tissue DCs, we investigated TLR9 expression in DCs 3A,20␮g/ml CpG ODN and 10 ␮g/ml LPS were used to stimulate isolated from lung-draining LNs and spleens. This approach was DCs from the two tissues. Lung DCs do not fare well in low serum undertaken based on the knowledge that in mice both plasmacytoid medium or with other supplements, and therefore we have used 2376 DISTINCT RESPONSES OF LUNG AND SPLEEN DCs TO CpG ODN

serum-containing medium in all of our experiments. Using different end points in different experiments, the DCs were found not to undergo maturation when maintained in serum containing medium unless speciﬁc stimuli were added. As shown in Fig. 3B, between the two TLR agonists, LPS and CpG ODN, the former induced stronger allostimulatory potential in the lung DCs. However, the overall level of T cell proliferation was not particularly high. In the case of spleen DCs in contrast, pretreatment with either LPS or CpG ODN promoted comparable allostimulatory potential of the DCs. Furthermore, T cell proliferation achieved with splenic DCs was greater compared with that obtained with lung DCs (Fig. 3B). Similarly, LPS- or CpG ODN-treated bone marrow-derived DCs also induced vigorous T cell proliferation (data not shown). We also used an alternate approach of purifying lung DCs based on high CD11c expression and low autoﬂuorescence, as previously reported (46) (Fig. 3C). The sorted cells were left in medium or treated with LPS or CpG ODN and used in MLR. Using these sorted cells expressing high levels of CD11c and exhibiting low

autoﬂuorescence, again greater proliferation of T cells was ob- Downloaded from served whether the DCs were previously exposed to LPS rather than CpG ODN (Fig. 3C). Phenotypic maturation of DCs in response to chemokines and TLR agonists

The results of the MLRs led us to examine the ability of the TLR http://www.jimmunol.org/ agonists LPS and CpG ODN to cause maturation of lung and spleen DCs (Fig. 4A). With the lung DCs, CpG ODN 1826, but not the control ODN 1911, induced a small increase in CD40- and CD86-expressing cells. LPS, in contrast, increased a higher per- centage of CD40-expressing cells (Fig. 4A). Consistently, simply FIGURE 2. TLR4 and TLR9 expression by lung and spleen DCs. A, culturing the lung DCs, whether in medium alone or with the two Expression of TLR4 and TLR9 mRNA in CD11cϩ lung and spleen DCs. agonists, resulted in down-regulation of CD80 expression on the mRNA was isolated from freshly isolated lung or spleen CD11cϩ DCs. lung DCs. TLR4 and TLR9 expression was determined by both semiquantitative and Compared with lung DCs, freshly isolated spleen DCs displayed by guest on September 24, 2021 quantitative RT-PCR. Hypoxanthine phosphoribosyltransferase expression a higher level of expression of MHC class II and costimulatory was assessed as a control for loading. The results for quantitative RT-PCR molecules, except for CD80 (Fig. 4B). Also, unlike what was ob- Ϯ are expressed as relative TLR4 or TLR9 mRNA expression (mean ratio served with lung DCs, both LPS and the CpG ODN induced high SD) normalized to expression of GUS. B, Cell surface TLR4 expression in levels of MHC class II, CD40, and CD86 expression on spleen freshly isolated CD11cϩ lung and spleen DCs as assessed by flow cytom- DCs in terms of mean fluorescence intensity (Fig. 4B). Addition- etry. The open histograms show staining by the specific Abs, and the filled histograms are staining by isotype controls. C, TLR9 is expressed in ally, unlike their effects on lung DCs, the TLR agonists also in- spleen, but not lung-draining LN DCs. Mice were immunized with OVA creased the level of CD80 expression on the spleen DCs. Collec- plus adjuvant i.n. (i.n. OVA plus CpG) or systemically (i.p. OVA plus tively, these results indicated that DCs from the lung and the CpG) to activate DCs in the lung-draining LNs and spleen, respectively. spleen responded differentially to TLR4 and TLR9 agonists. DCs were partially purified by density-gradient centrifugation, then were Spleen DCs responded to both LPS and CpG ODN to undergo surface stained to identify pDCs and mDCs using directly labeled anti- maturation. However, lung DCs appeared to be more responsive to CD11c FITC, anti-MHC class II PE, and anti-B220 allophycocyanin LPS compared with CpG ODN. mAbs. Appropriate isotype control Abs for each of these markers were used to establish the level of background staining for each. Expression of LPS-stimulated lung DCs selectively induce Ag-specific intracellular TLR9 was then determined by incubation of fixed, permeabil- responses in CD4ϩ T cells, resulting in GATA-3 and T-bet ized cells with either biotinylated anti-TLR9 Ab or relevant biotinylated activation isotype control Ab, followed by incubation with PerCP-labeled streptavi- ϩ din. Flow cytometric analysis was performed using multiple gating tech- We next investigated Ag-specific responses in CD4 T cells when niques to identify pDCs and mDCs according to the parameters listed stimulated with lung or spleen DCs previously exposed to the TLR ϩ above. For spleen cells, the expression of MHC class II was similar (data agonists. Ag-specific response was examined using CD4 T cells not shown) and CD11c expression varied slightly such that pDCs were from DO11.10 TCR transgenic mice with specificity for an epitope CD11c low, MHC class IIϩ, B220ϩ, and mDCs were CD11c high, MHC (aa 323–339) in OVA. We investigated expression of the transcrip- ϩ Ϫ class II , B220 . For lung-draining LN cells, CD11c expression was sim- tion factors T-bet and GATA-3, which program CD4ϩ T cells to ilar and MHC class II varied (data not shown) such that pDCs were differentiate into the Th1 or Th2 lineage, respectively, as shown CD11cϩ, MHC class II low, B220ϩ, and mDCs were CD11cϩ, MHC class Ϫ previously by us and others (35, 36, 45, 47). LPS-treated lung DCs II high, B220 . TLR9 expression is shown relative to CD11c expression ϩ induced high levels of T-bet and GATA-3 expression in the CD4 for pDCs and mDCs in lung-draining LNs and spleens. The isotype control Ab staining for TLR9 is shown in the upper panels. The percentages listed T cells (Fig. 5A). However, neither T-bet nor GATA-3 expression represent the proportion of TLR9 expression on the multiple-gated cell was induced in the T cells by CpG ODN-stimulated lung DCs. The population shown. This is a representative of two independent experiments. ability of LPS-treated DCs to induce a mixed Th1/Th2 response in CD4ϩ T cells was also noted in previous studies (48). When T cells were exposed to spleen DCs, GATA-3 expression was found The Journal of Immunology 2377 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 3. The effect of TLR agonists on allostimulatory function of lung and spleen DCs. A, Dose response of TLR agonists on allostimulatory function of spleen DCs. Freshly isolated spleen DCs were cultured for 40 h with different concentrations of LPS or CpG ODN, as indicated, or ODN control 1911 (20 ␮g/ml). The ratio of DC:T cell was 1:5. A dose of 10 ␮g/ml LPS and 20 ␮g/ml CpG ODN was used to stimulate DCs in all subsequent experiments. B, Allostimulatory ability of lung and spleen DCs stimulated by LPS or CpG ODN. Freshly isolated lung and spleen DCs were maintained in medium alone or stimulated with the indicated agents for 40 h. DCs were then washed with cRPMI and cocultured with allogeneic CD4ϩ T cells for p Ͻ 0.05 LPS vs medium; and #, p Ͻ 0.05 CpG ,ء ;h. [3H]Thymidine was added during the final 18 h of culture. ϩ, p Ͻ 0.05 LPS vs CpG treatment 96 ODN vs control ODN control. Shown is a representative experiment of four. C, Sorting of CD11cϩ, low autofluorescent cells and allostimulatory function after stimulation by LPS or CpG ODN. CD11cϩ cells were isolated from lungs by positive selection methods (see Presort population), and the less abundant low autofluorescent CD11cϩ cells were enriched by sorting (Ͼ80% purity in the Postsort low autofluorescent population). The sorted cells were then used p Ͻ 0.05 LPS vs medium. The experiment was repeated twice with ,ء ;in MLRs. The data shown are mean Ϯ SD. ϩ, p Ͻ 0.05 LPS vs CpG treatment similar results. to be high irrespective of whether the DCs were previously incu- pression of costimulatory molecules and have poor T cell stimu- bated in medium alone or stimulated with LPS or CpG ODN. This latory ability. LPS stimulation of lung DCs in contrast promotes may be due to the relatively high level of basal CD86 expression CD40 expression and renders them competent to up-regulate both on spleen DCs since CD86 is known to promote Th2 differentia- T-bet and GATA-3 expression in T cells. Spleen DCs in contrast tion (49). T-bet expression in the T cells, however, increased if the are highly responsive to both TLR agonists, as reflected in in- cells were cocultured with LPS or CpG ODN-treated DCs as com- creased expression of costimulatory molecules on the DCs and pared with incubation with DCs cultured in medium alone. These their ability to activate T cells. We also examined cytokine pro- results showed that CpG-treated lung DCs have low levels of ex- duction by the cocultured CD4ϩ T cells, and the cytokine secretion 2378 DISTINCT RESPONSES OF LUNG AND SPLEEN DCs TO CpG ODN

FIGURE 4. Effects of TLR agonists on phenotypic maturation of lung and spleen DCs. Freshly isolated lung (A) or spleen (B) DCs were cultured with the indicated stimuli for 40 h. Cells were stained with PE-conjugated CD80, CD86, or MHC II or Downloaded from biotin-conjugated CD40, followed by incubation with streptavidin-conjugated allophycocyanin. The percentages indicate positive staining relative to control Ab staining (ﬁlled histograms), and the numbers in parenthe- http://www.jimmunol.org/ ses indicate mean ﬂuorescence intensity. Similar results were obtained in two independent experiments. by guest on September 24, 2021

proﬁle also matched T-bet and GATA-3 expression in the T cells CD40Ϫ/Ϫ mice was the known involvement of both microbial and (Fig. 5B). Only LPS-stimulated lung DCs induced high levels of CD40-CD154 interaction in stimulation of IL-12 production by secretion of the cytokines IFN-␥, IL-4, IL-5, IL-13, and IL-10 DCs (50). Although our previous experiments suggested that lung secretion from the CD4ϩ T cells. In the case of spleen DCs, both DCs are refractory to CpG ODN, this was found to be not entirely LPS and CpG ODN treatments increased cytokine production from true when IL-12p40 production was examined from the treated the T cells. Despite comparable levels of IL-5 and IL-13 in the DCs. IL-12p40 secretion was found to be higher from CpG ODN- supernatants of cocultures of CD4ϩ T cells with lung DCs and stimulated lung DCs than from LPS-stimulated cells (Fig. 6). In- spleen DCs, there was a marked difference in IL-4 levels, with terestingly, the level was similar to that produced by CpG ODN- CD4ϩ T cells incubated with LPS-treated lung DCs secreting a treated spleen DCs (Fig. 6). 1L-12p70 production from the higher level of IL-4. stimulated DCs was not detected at a high level from either lung or spleen DCs. The levels detected were, however, comparable to IL-12p40 and IL-12p70 production by lung and spleen DCs those reported in other studies using spleen DCs (51). Although stimulated with LPS or CpG ODN and dependence on CD40 CpG ODN induced higher levels of IL-12p40 secretion from lung Because all of the above experiments showed an attenuated re- DCs compared with LPS, the opposite was true with respect to sponse of lung DCs to CpG ODN, we investigated whether this IL-12p70 with LPS stimulating a higher level of IL-12p70 pro- was also reﬂected in cytokine production by the stimulated DCs. duction. Also, consistent with the reported involvement of CD40 We investigated IL-12 production by the stimulated DCs because stimulation in IL-12 production by DCs, lung or spleen DCs iso- it is known to be induced by both LPS and CpG ODN. In these lated from CD40Ϫ/Ϫ mice produced lower levels of IL-12 (p40 and experiments, we also used DC-T cell cocultures using DCs isolated p70) in response to both agonists (Fig. 6). Interestingly, the re- from either wild-type (wt) or CD40Ϫ/Ϫ mice. The reason for the quirement for CD40 appeared to be more stringent for LPS-in- inclusion of DC-T cell cocultures and DCs from both wt and duced IL-12p40 production. Collectively, these results showed that The Journal of Immunology 2379

FIGURE 5. Effects of the TLR agonist-stimulated DCs on Ag-speciﬁc CD4ϩ T cell responses. Lung DCs were cultured with the indicated stimuli for 40 h. The culture supernatant was then removed, the cells were washed, and fresh medium was added containing CD4ϩ T cells from DO11.10

TCR transgenic mice, the OVA peptide, pOVA Downloaded from (323–339), at 5 ␮g/ml, and IL-2 (10 U/ml). The cells were cocultured for another 5 days, after which nuclear extracts were made and the culture supernatants were saved for cytokine estimation. A, Expression of transcription factors T-bet and GATA-3 in the CD4ϩ T cells stimulated by lung or spleen DCs as assessed by Western blot analysis of http://www.jimmunol.org/ the nuclear extracts. CREB-1 expression was assessed as a control for protein loading. B, Cytokine levels in the culture supernatants were determined by multiplex assay. The data shown are mean concentrations Ϯ SD of triplicate wells. The experiment was repeated three times with similar results. by guest on September 24, 2021

lung DCs are not completely unresponsive to the TLR9 agonist alone induced IL-12p40 production from both lung LN and spleen CpG ODN. cells. A combination of OVA ϩ CT induced appreciable levels of IL-4, IL-5, IL-10, IL-13, and IFN-␥ from lung LN cells. Similar or CpG ODN administration in vivo induces IL-6 production from slightly lower levels of Th2 cytokines (except for IL-5) were noted spleen, but not lung cells from splenic cells when animals were immunized with OVA ϩ Having failed to detect TLR9 expression in lung DCs in contrast to alum i.p. Surprisingly, IFN-␥ production from spleen cells was readily detectable expression in spleen DCs, and yet detecting a markedly lower than what was observed from lung LN cells. The CpG ODN-induced IL-12p40 response in the lung DCs, we were most notable difference was low/undetectable IL-6 production curious to determine the response of lung and spleen cells to CpG from lung LN cells compared with modest levels (ϳ350–400 pg/ ODN in vivo. Toward this end, we used three groups of mice each ml) from splenic cells. The lack of IL-6 production from lung LN for the analysis of lung and spleen responses. Mice received either cells parallels the recent results of Sanjuan et al. (38), in which CpG ODN alone, OVA ϩ the Th2-skewing mucosal adjuvant CT CpG ODN induced tyrosine phosphorylation of target proteins in to study the response in lung LNs (43), or OVA ϩ alum to study macrophages in the absence of TLR9 or MyD88, but failed to response in the spleen (44, 45), or a combination of OVA ϩ CT/ induce IL-6 secretion in the absence of MyD88. When a combi- alum ϩ CpG ODN. Although CpG ODN has been shown to inhibit nation of CpG ODN and CT/alum was used, the production of Th2 allergic airway inﬂammation in mice (22–25), the response of lung cytokines was drastically reduced. Although IL-4 and IL-5 pro- LN cells to CpG ODN has not been reported to date. It was also duction was almost abolished, the levels of IL-10 and IL-13 were important to check whether CpG ODN triggered IL-12p40 in the reduced by half in the lung. However, there was no inhibitory lung LNs as observed in vitro. As shown in Fig. 7, CpG ODN effect on IFN-␥ production, and if anything, IFN-␥ levels increased 2380 DISTINCT RESPONSES OF LUNG AND SPLEEN DCs TO CpG ODN Downloaded from http://www.jimmunol.org/

FIGURE 6. Cytokine profile of the culture supernatants of lung and spleen DCs from wt and CD40Ϫ/Ϫ mice cocultured with CD4ϩ T cells. DCs were isolated from the lungs and spleens of wt and CD40Ϫ/Ϫ mice and were cocultured with CD4ϩ T cells for 5 days, as described in the legend to Fig. 5. p Ͻ 0.01 ,ءء and ء .Cytokine levels were determined by ELISA. The data are shown as mean Ϯ SD and are representative of two independent experiments vs medium or control ODN, respectively; #, p Ͻ 0.05 vs LPS stimulation. by guest on September 24, 2021 in the presence of CpG ODN. CpG ODN did not inhibit OVA/ expression of additional Th2-dominant factors such as c-maf and alum-stimulated IL-6 production either. JunB (data not shown), none of which was differentially expressed under the three different immunization regimens. Collectively, Th2 cytokine repression despite high levels of GATA-3 these results show for the first time the ability of CpG ODN to expression cause a drastic inhibition of Th2 cytokine production in the ab- Having observed preferential inhibition of cytokine expression in sence of TLR9 expression and in the presence of high levels of the presence of CpG ODN in both the lung LN and the spleen, GATA-3 expression. mice were immunized via the i.n. route using the same regimen followed for cytokine assays. After ex vivo stimulation of lung LN Discussion cells in the presence of OVA, nuclear extracts were prepared of the In this study, we show that DCs in the lung and the spleen respond total LN cell population, and the extracts were subjected to West- differentially to the TLR agonists LPS and CpG ODNs. In the lung, ern blot analysis to determine activation/induction of specific tran- the best-studied effect of CpG ODN has been inhibition of Th2 scription factors. Although the results of the cytokine assay cell-mediated allergic disease (22–24, 52). In the spleens of mice, showed remarkable inhibition of Th2 cytokine expression in the CpG ODN has been shown to promote sensitivity to bacterial com- presence of CpG ODN, interestingly, GATA-3 expression was ponents. Administration of CpG ODN to mice induces transient similar to what was observed in the presence of OVA ϩ CT (Fig. splenomegaly (26). The immunomodulatory effects of Propi- 8). T-bet expression in contrast was higher in the presence of CpG onibacterium acnes on the spleen, such as the induction of spleno- ODN, which may explain the higher IFN-␥ levels when mice re- megaly and hypersensitivity (27, 28) to different TLR agonists, ceived OVA plus both adjuvants. We also investigated expression were recently shown to be dependent on TLR9 (53). The enhanced of IFN regulatory factor 8 and NFAT-2, factors known to be im- sensitivity upon P. acnes infection appears to be due to a rapid, portant for IL-12p40 gene expression. Although expression of IFN enhanced inflammatory response. Taken together, these studies regulatory factor 8 was similar under all three conditions, that of show that CpG ODN can cause either inhibition or stimulation of NFAT-2 was appreciably higher in the presence of CpG ODN, CD4ϩ T cell responses. Our investigations on the effects of CpG whether used alone or in conjunction with OVA and adjuvants. ODNs on DCs have revealed that the lung and spleen DCs show This high level of NFAT-2 expression may underlie the high level marked differences in their response to CpG ODN at the level of of IL-12p40 production in the context of CpG ODN (Fig. 8). Over- costimulatory molecule expression, cytokine secretion, and the all, while we did observe the appropriate transcription factor sig- ability to stimulate T cells. However, two common effects are in- nature to support IL-12p40 gene expression, we did not observe a duction of IL-12p40 expression in the DC and inhibition of Th2 profile that would explain inhibition of Th2 cytokine production, cytokine production, despite high levels of expression of GATA-3, such as inhibition of expression of GATA-3. We also analyzed the master regulator of Th2 differentiation (35, 36). The Journal of Immunology 2381

FIGURE 8. Effect of CpG ODN on OVA/CT-induced transcription factor expression. Mice were immunized i.n. with CpG ODN, OVA ϩ CT, or CpG ODN ϩ OVA ϩ CT, as described in the legend to Fig. 7. Lung LN cells were isolated on day 5 and cultured with OVA protein for 5 days. Nuclear extracts were prepared for analysis of transcription factor expres- Downloaded from sion by Western blot techniques. CREB-1 expression was assessed as a control for protein loading. The results are representative of two independent experiments.

mine whether LPS can activate the TRIF/IFN-␤/IFN-RI signaling http://www.jimmunol.org/ axis in lung DCs and whether this pathway contributes to the higher level of CD40 expression induced by LPS on lung DCs. However, it is also important to note that LPS-stimulated lung DCs are still inferior to spleen DCs with regard to allostimulation. TLR4 is known to use both MyD88-Mal/Toll-IL-1R domain-con- FIGURE 7. CpG inhibits Th2 cytokine production in lung-draining LNs and spleens of mice. Mice were immunized i.n. with CpG ODN (5 ␮g/ taining adapter protein and TRIF/TRIF-related adapter molecule mouse) alone or OVA (100 ␮g) ϩ CT (1 ␮g/mouse), or CpG ODN (5 (MyD88-independent) pathways (2), and their relative use in lung ␮g/mouse) followed by OVA (100 ␮g) ϩ CT (1 ␮g/mouse). On day 5 after and spleen DCs remains to be determined. last immunization, lung LN cells were isolated and cultured with OVA Until recently, TLR9 was thought to be totally reliant on by guest on September 24, 2021 protein (100 ␮g/ml) for 5 days. Mice were also immunized i.p. with CpG MyD88-dependent signaling. Recent studies indicate that TLR9 ODN (25 ␮g/mouse) alone or OVA (10 ␮g) ϩ alum (2 mg/mouse), or CpG can signal via both TLR9-dependent and -independent mecha- ODN (25 ␮g/mouse) followed by OVA (10 ␮g) ϩ alum (2 mg/mouse). nisms (37, 38). However, the production of IL-6 has been shown After 5 days, mice received an i.p. boost with the same dose of Ag and to be TLR9 and MyD88 dependent (33, 38). Our results show that adjuvants. Splenocytes were isolated 24 h after the last booster dose and while CpG ODN can induce production of several cytokines in cultured in the presence of OVA protein (100 ␮g/ml) for 5 days. Cytokine lung LN cells, it is unable to induce IL-6 production, which can be profiles of the culture supernatants of lung LN and spleen cells were de- p Ͻ 0.05 spleen vs lung LN levels. The results readily detected from splenic cells. The absence of IL-6 production ,ء .tected by multiplex assay are representative of two independent experiments. from lung LN cells upon CpG ODN treatment is most likely due to absence of TLR9 protein expression in lung LN DCs. It is sur- prising that this response is selectively absent in the lung. Al- One distinguishing feature of CpG ODN- vs LPS-stimulated though IL-6 is induced in lungs upon infection with pathogens, it lung DCs is the level of CD40 expression on the stimulated cells. is interesting that CpG motifs are unable to induce IL-6 in lung Compared with CpG ODN, LPS induced more CD40 expression LNs that have many types of resident cells, including DCs, T cells, on the DCs. On spleen DCs, however, both LPS and CpG ODN B cells, and macrophages. It remains to be determined whether all induced high levels of CD40 expression. Interestingly, although lung cell types lack TLR9. The biological significance of absence both of these agonists augmented MHC class II, CD80, and CD86 of TLR9 in lung LN cells to restrict production of IL-6 may be a expression on the spleen DCs, they failed to have a significant protective measure that has evolved to protect the lung from the impact on the expression of these molecules on lung DCs. Clearly, development of diseases such as pulmonary fibrosis that are asso- CD40 expression by the DCs is important because its absence ciated with high IL-6 production and a high mortality rate (57–59). reduced/eliminated IL-12p40 or IL-12p70 production by the DCs CpG motifs of bacterial DNA have been shown to promote chronic (Fig. 6). Recently, TLR agonists and CD40 signaling were shown inflammation in colitis, in which the two key cytokines implicated to synergize in stimulation of CD8ϩ T cell proliferation (54). Sim- are IL-6 and IFN-␥ (29). It will be interesting to examine TLR9 ilarly, studies of Steinman and colleagues (55) have also shown a expression in cells present in mesenteric LNs to determine whether unique role for CD40-CD40L signaling in inducing functional within the mucosal tract there exists differential expression of maturation of DCs. The molecular nature of this distinctive CD40 TLR9 between DCs present in draining LNs of the lung vs those signaling pathway in DCs is currently unclear. LPS is known to draining the gastrointestinal tract. The production of appreciable up-regulate costimulatory molecule expression on APCs through levels of IL-12p40 by the CpG ODN-stimulated lung DCs shows activation of the Toll/IL-1 receptor domain-containing adaptor- that the lung actively responds to this agent. IL-12p40 is the com- inducing IFN-␤ (TRIF) pathway that through secretion of IFN-␤ mon subunit between IL-12p70 and IL-23. Because IL-12p70 is activates the IFN-RI receptor (56). It will be interesting to deter- not always detectable at high levels, high IFN-␥ levels in the lung 2382 DISTINCT RESPONSES OF LUNG AND SPLEEN DCs TO CpG ODN

LN cells suggest induction of sufficient IL-12p70 to induce IFN-␥ 2. Pulendran, B. 2005. Variegation of the immune response with dendritic cells and production. pathogen recognition receptors. J. Immunol. 174: 2457–2465. 3. Pollard, A. M., and M. F. Lipscomb. 1990. Characterization of murine lung Multiple studies have shown previously that CpG ODN inhibits dendritic cells: similarities to Langerhans cells and thymic dendritic cells. J. Exp. Th2-mediated disease in the lung (22–24, 52). Our data show that Med. 172: 159–167. the Th2 inhibition is also evident in the spleen. The ability of CpG 4. Lipscomb, M. F., and B. J. Masten. 2002. Dendritic cells: immune regulators in health and disease. Physiol. Rev. 82: 97–130. ODN to inhibit Th2 responses being similar in the lung LN and 5. Nicod, L. P., M. F. Lipscomb, J. C. Weissler, C. R. Lyons, J. Albertson, and spleen, the mechanism of Th2 suppression by CpG ODN is most G. B. Toews. 1987. Mononuclear cells in human lung parenchyma: characterization of a potent accessory cell not obtained by bronchoalveolar lavage. Am. likely distinct from that used for promotion of IL-6 gene expres- Rev. Respir. Dis. 136: 818–823. sion. Studies published recently have identified Src family kinases 6. Holt, P. G., M. A. Schon-Hegrad, and J. Oliver. 1988. MHC class II antigen- Hck, Lyn, and DNA-dependent protein kinase as alternate TLR9- bearing dendritic cells in pulmonary tissues of the rat: regulation of antigen presentation activity by endogenous macrophage populations. J. Exp. Med. 167: independent mechanisms of CpG ODN action (37, 38). Whether 262–274. any of these molecules or other pathways are used by CpG ODN 7. Kradin, R. L., K. M. McCarthy, W. J. Xia, D. Lazarus, and E. E. Schneeberger. ϩ in the lung to induce its effects needs to be investigated in the 1991. Accessory cells of the lung. I. Interferon-␥ increases Ia dendritic cells in the lung without augmenting their accessory activities. Am. J. Respir. Cell Mol. future. Biol. 4: 210–218. It is interesting that the Th2 inhibition caused by CpG ODN 8. Xia, W. J., E. E. Schneeberger, K. McCarthy, and R. L. Kradin. 1991. Accessory ϩ occurs without any apparent inhibition of GATA-3 expression. Be- cells of the lung. II. Ia pulmonary dendritic cells display cell surface antigen heterogeneity. Am. J. Respir. Cell Mol. Biol. 5: 276–283. cause GATA-3 levels are maintained in the presence of CpG ODN, 9. Gong, J. L., K. M. McCarthy, J. Telford, T. Tamatani, M. Miyasaka, and the underlying mechanism is not a simple Th1/Th2 imbalance in- E. E. Schneeberger. 1992. Intraepithelial airway dendritic cells: a distinct subset duced by high IL-12/IFN-␥ levels that typically down-regulate the of pulmonary dendritic cells obtained by microdissection. J. Exp. Med. 175: Downloaded from 797–807. GATA-3 response. In a recent study, CpG-induced indoleamine 10. Toews, G. B. 1991. Pulmonary dendritic cells: sentinels of lung-associated lym- 2,3-dioxygenese (IDO) expression in lung cells was implicated in phoid tissues. Am. J. Respir. Cell Mol. Biol. 4: 204–205. the inhibition of experimental allergic airways disease in mice 11. Armstrong, L. R., P. J. Christensen, R. Paine III, G. H. Chen, R. A. McDonald, T. K. Lim, and G. B. Toews. 1994. Regulation of the immunostimulatory activity (24). Although lung epithelial cells were considered to be one of rat pulmonary interstitial dendritic cells by cell-cell interactions and cytokines. source of IDO in this study, lung DCs were also shown to express Am. J. Respir. Cell Mol. Biol. 11: 682–691. 12. McWilliam, A. S., D. Nelson, J. A. Thomas, and P. G. Holt. 1994. Rapid den- IDO. Whether IDO induction in a DC imparts a dominant-negative http://www.jimmunol.org/ ϩ dritic cell recruitment is a hallmark of the acute inflammatory response at mu- signal to a CD4 T cell via secretion of kynurenine metabolites cosal surfaces. J. Exp. Med. 179: 1331–1336. that cause Th2 inhibition without effects on GATA-3 remains to be 13. Lambrecht, B. N., H. C. Hoogsteden, and R. A. Pauwels. 2001. Dendritic cells as regulators of the immune response to inhaled allergen: recent findings in animal determined. Another potential mechanism for Th2 inhibition de- models of asthma. Int. Arch. Allergy Immunol. 124: 432–446. spite high GATA-3 levels is inhibition of GATA-3 DNA-binding 14. Vermaelen, K. Y., I. Carro-Muino, B. N. Lambrecht, and R. A. Pauwels. 2001. activity by tyrosine-phosphorylated T-bet (60). This could be im- Specific migratory dendritic cells rapidly transport antigen from the airways to portant because we detected an increase in T-bet expression when the thoracic lymph nodes. J. Exp. Med. 193: 51–60. 15. Penna, G., M. Vulcano, A. Roncari, F. Facchetti, S. Sozzani, and L. Adorini. CpG was administered in animals before treatment with OVA ϩ 2002. Cutting edge: differential chemokine production by myeloid and plasma- CT. Other mechanisms that would also functionally impair cytoid dendritic cells. J. Immunol. 169: 6673–6676.

16. Grouard, G., M. C. Rissoan, L. Filgueira, I. Durand, J. Banchereau, and Y. J. Liu. by guest on September 24, 2021 GATA-3 despite high levels of expression include expression of 1997. The enigmatic plasmacytoid T cells develop into dendritic cells with in- repressor of GATA-3, which inhibits GATA-3 function through terleukin (IL)-3 and CD40-ligand. J. Exp. Med. 185: 1101–1111. inhibition of DNA-binding activity, and other as yet to be identi- 17. Cella, M., D. Jarrossay, F. Facchetti, O. Alebardi, H. Nakajima, A. Lanzavecchia, and M. Colonna. 1999. Plasmacytoid monocytes migrate to inflamed lymph fied mechanisms (61). nodes and produce large amounts of type I interferon. Nat. Med. 5: 919–923. In conclusion, the increase in GATA-3, T-bet, and concomitant 18. Rissoan, M. C., V. Soumelis, N. Kadowaki, G. Grouard, F. Briere, cytokine gene expression in T cells induced by LPS-stimulated R. de Waal Malefyt, and Y. J. Liu. 1999. Reciprocal control of T helper cell and dendritic cell differentiation. Science 283: 1183–1186. lung DCs has important ramifications in DC-mediated immuno- 19. Siegal, F. P., N. Kadowaki, M. Shodell, P. A. Fitzgerald-Bocarsly, K. Shah, therapy of lung cancer because the lung is frequently a site of S. Ho, S. Antonenko, and Y. J. Liu. 1999. The nature of the principal type 1 cancer metastasis. It may be useful to stimulate the TLR4 pathway interferon-producing cells in human blood. Science 284: 1835–1837. 20. Standiford, T. J., S. L. Kunkel, N. W. Lukacs, M. J. Greenberger, J. M. Danforth, in DCs used in cancer immunotherapy. Recently, expression of a R. G. Kunkel, and R. M. Strieter. 1995. Macrophage inflammatory protein-1␣ constitutively active mutant of TLR4 was shown to enhance allo- mediates lung leukocyte recruitment, lung capillary leak, and early mortality in stimulatory potential of DCs (62). The DCs expressing melanoma murine endotoxemia. J. Immunol. 155: 1515–1524. 21. Speyer, C. L., H. Gao, N. J. Rancilio, T. A. Neff, G. B. Huffnagle, J. V. Sarma, Ag and constitutively active mutant of TLR4 elicited significant and P. A. Ward. 2004. Novel chemokine responsiveness and mobilization of cytolytic activity in stimulated T cells (62). This approach may neutrophils during sepsis. Am. J. Pathol. 165: 2187–2196. also be important to enhance adaptive immune responses in 22. Jain, V. V., K. Kitagaki, T. Businga, I. Hussain, C. George, P. O’Shaughnessy, and J. N. Kline. 2002. CpG-oligodeoxynucleotides inhibit airway remodeling in chronic infectious diseases such as tuberculosis to efficiently elim- a murine model of chronic asthma. J. Allergy Clin. Immunol. 110: 867–872. inate pathogens. The ability of CpG ODNs to inhibit Th2 re- 23. Choudhury, B. K., J. S. Wild, R. Alam, D. M. Klinman, I. Boldogh, N. Dharajiya, sponses in the lung shows that appropriate TLR agonists may have W. J. Mileski, and S. Sur. 2002. In vivo role of p38 mitogen-activated protein kinase in mediating the anti-inflammatory effects of CpG oligodeoxynucleotide in inhibitory or stimulatory effects on lung DCs that could be ex- murine asthma. J. Immunol. 169: 5955–5961. ploited in DC-mediated immunotherapy of chronic lung diseases. 24. Hayashi, T., L. Beck, C. Rossetto, X. Gong, O. Takikawa, K. Takabayashi, D. H. Broide, D. A. Carson, and E. Raz. 2004. Inhibition of experimental asthma by indoleamine 2,3-dioxygenase. J. Clin. Invest. 114: 270–279. Acknowledgments 25. Juffermans, N. P., J. C. Leemans, S. Florquin, A. Verbon, A. H. Kolk, We thank Drs. Donna B. Stolz and Simon C. Watkins for assistance with P. Speelman, S. J. van Deventer, and T. van der Poll. 2002. CpG oligodeoxynucleotides enhance host defense during murine tuberculosis. Infect. Immun. scanning electron microscopy. 70: 147–152. 26. Sparwasser, T., L. Hultner, E. S. Koch, A. Luz, G. B. Lipford, and H. Wagner. 1999. Immunostimulatory CpG-oligodeoxynucleotides cause extramedullary mu- Disclosures rine hemopoiesis. J. Immunol. 162: 2368–2374. The authors have no financial conflict of interest. 27. Galanos, C., M. Gumenscheimer, P. Muhlradt, E. Jirillo, and M. Freudenberg. 2000. MALP-2, a Mycoplasma lipopeptide with classical endotoxic properties: end of an era of LPS monopoly? J. Endotoxin Res. 6: 471–476. References 28. Tsuji, H., N. Mukaida, A. Harada, S. Kaneko, E. Matsushita, Y. Nakanuma, 1. Maldonado-Lopez, R., and M. Moser. 2001. Dendritic cell subsets and the reg- H. Tsutsui, H. Okamura, K. Nakanishi, Y. Tagawa, et al. 1999. Alleviation of ulation of Th1/Th2 responses. Semin. Immunol. 13: 275–282. lipopolysaccharide-induced acute liver injury in Propionibacterium acnes-primed The Journal of Immunology 2383

IFN-␥-deficient mice by a concomitant reduction of TNF-␣, IL-12, and IL-18 46. Vermaelen, K., and R. Pauwels. 2004. Accurate and simple discrimination of production. J. Immunol. 162: 1049–1055. mouse pulmonary dendritic cell and macrophage populations by flow cytometry: 29. Obermeier, F., N. Dunger, U. G. Strauch, C. Hofmann, A. Bleich, N. Grunwald, methodology and new insights. Cytometry A 61: 170–177. H. J. Hedrich, E. Aschenbrenner, B. Schlegelberger, G. Rogler, et al. 2005. CpG 47. Szabo, S. J., S. T. Kim, G. L. Costa, X. Zhang, C. G. Fathman, and motifs of bacterial DNA essentially contribute to the perpetuation of chronic L. H. Glimcher. 2000. A novel transcription factor, T-bet, directs Th1 lineage intestinal inflammation. Gastroenterology 129: 913–927. commitment. Cell 100: 655–669. 30. Meng, Y., A. F. Carpentier, L. Chen, G. Boisserie, J. M. Simon, J. J. Mazeron, 48. Pulendran, B., P. Kumar, C. W. Cutler, M. Mohamadzadeh, T. Van Dyke, and and J. Y. Delattre. 2005. Successful combination of local CpG-ODN and radio- J. Banchereau. 2001. Lipopolysaccharides from distinct pathogens induce differ- therapy in malignant glioma. Int. J. Cancer 116: 992–997. ent classes of immune responses in vivo. J. Immunol. 167: 5067–5076. 31. Klinman, D. M., J. Conover, and C. Coban. 1999. Repeated administration of 49. Tsuyuki, S., J. Tsuyuki, K. Einsle, M. Kopf, and A. Coyle. 1997. Costimulation synthetic oligodeoxynucleotides expressing CpG motifs provides long-term pro- through B7-2 (CD86) is required for the induction of a lung mucosal T helper 2 tection against bacterial infection. Infect. Immun. 67: 5658–5663. (TH2) immune response and altered airway responsiveness. J. Exp. Med. 185: 32. Duramad, O., K. L. Fearon, B. Chang, J. H. Chan, J. Gregorio, R. L. Coffman, and 1671–1679. F. J. Barrat. 2005. Inhibitors of TLR-9 act on multiple cell subsets in mouse and 50. Schulz, O., A. D. Edwards, M. Schito, J. Aliberti, S. Manickasingham, A. Sher, man in vitro and prevent death in vivo from systemic inflammation. J. Immunol. and C. Reis e Sousa. 2000. CD40 triggering of heterodimeric IL-12 p70 produc- 174: 5193–5200. tion by dendritic cells in vivo requires a microbial priming signal. Immunity 13: 33. O’Keeffe, M., R. J. Grumont, H. Hochrein, M. Fuchsberger, R. Gugasyan, 453–462. D. Vremec, K. Shortman, and S. Gerondakis. 2005. Distinct roles for the NF-␬B1 51. Pillarisetty, V. G., A. B. Shah, G. Miller, J. I. Bleier, and R. P. DeMatteo. 2004. and c-Rel transcription factors in the differentiation and survival of plasmacytoid Liver dendritic cells are less immunogenic than spleen dendritic cells because of and conventional dendritic cells activated by TLR-9 signals. Blood 106: differences in subtype composition. J. Immunol. 172: 1009–1017. 3457–3464. 52. Shirota, H., K. Sano, T. Kikuchi, G. Tamura, and K. Shirato. 2000. Regulation of 34. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Im- murine airway eosinophilia and Th2 cells by antigen-conjugated CpG oligode- munol. 21: 335–376. oxynucleotides as a novel antigen-specific immunomodulator. J. Immunol. 164: 35. Zhang, D. H., L. Cohn, P. Ray, K. Bottomly, and A. Ray. 1997. Transcription 5575–5582. factor GATA-3 is differentially expressed in murine Th1 and Th2 cells and con- 53. Kalis, C., M. Gumenscheimer, N. Freudenberg, S. Tchaptchet, G. Fejer, A. Heit, trols Th2-specific expression of the interleukin-5 gene. J. Biol. Chem. 272: S. Akira, C. Galanos, and M. A. Freudenberg. 2005. Requirement for TLR9 in the Downloaded from 21597–21603. immunomodulatory activity of Propionibacterium acnes. J. Immunol. 174: 36. Zheng, W., and R. A. Flavell. 1997. The transcription factor GATA-3 is neces- 4295–4300. sary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89: 54. Ahonen, C. L., C. L. Doxsee, S. M. McGurran, T. R. Riter, W. F. Wade, 587–596. R. J. Barth, J. P. Vasilakos, R. J. Noelle, and R. M. Kedl. 2004. Combined TLR 37. Dragoi, A. M., X. Fu, S. Ivanov, P. Zhang, L. Sheng, D. Wu, G. C. Li, and and CD40 triggering induces potent CD8ϩ T cell expansion with variable de- W. M. Chu. 2005. DNA-PKcs, but not TLR9, is required for activation of Akt by pendence on type I IFN. J. Exp. Med. 199: 775–784. CpG-DNA. EMBO J. 24: 779–789. 55. Fujii, S., K. Liu, C. Smith, A. J. Bonito, and R. M. Steinman. 2004. The linkage

38. Sanjuan, M. A., N. Rao, K. T. Lai, Y. Gu, S. Sun, A. Fuchs, W. P. Fung-Leung, of innate to adaptive immunity via maturing dendritic cells in vivo requires CD40 http://www.jimmunol.org/ M. Colonna, and L. Karlsson. 2006. CpG-induced tyrosine phosphorylation oc- ligation in addition to antigen presentation and CD80/86 costimulation. J. Exp. curs via a TLR9-independent mechanism and is required for cytokine secretion. Med. 199: 1607–1618. J. Cell Biol. 172: 1057–1068. 56. Hoebe, K., E. M. Janssen, S. O. Kim, L. Alexopoulou, R. A. Flavell, J. Han, and 39. Swanson, K. A., Y. Zheng, K. M. Heidler, Z. D. Zhang, T. J. Webb, and B. Beutler. 2003. Up-regulation of costimulatory molecules induced by lipopoly- D. S. Wilkes. 2004. Flt3-ligand, IL-4, GM-CSF, and adherence-mediated isola- saccharide and double-stranded RNA occurs by Trif-dependent and Trif-inde- tion of murine lung dendritic cells: assessment of isolation technique on pheno- pendent pathways. Nat. Immunol. 4: 1223–1229. type and function. J. Immunol. 173: 4875–4881. 57. Baecher-Allan, C. M., and R. K. Barth. 1993. PCR analysis of cytokine induction 40. Agrawal, A., J. Lingappa, S. H. Leppla, S. Agrawal, A. Jabbar, C. Quinn, and proﬁles associated with mouse strain variation in susceptibility to pulmonary B. Pulendran. 2003. Impairment of dendritic cells and adaptive immunity by ﬁbrosis. Reg. Immunol. 5: 207–217. anthrax lethal toxin. Nature 424: 329–334. 58. Takizawa, H., M. Satoh, H. Okazaki, G. Matsuzaki, N. Suzuki, A. Ishii, M. Suko, 41. Ahmad-Nejad, P., H. Hacker, M. Rutz, S. Bauer, R. M. Vabulas, and H. Wagner. H. Okudaira, Y. Morita, and K. Ito. 1997. Increased IL-6 and IL-8 in bronchoal-

2002. Bacterial CpG-DNA and lipopolysaccharides activate Toll-like receptors at veolar lavage fluids (BALF) from patients with sarcoidosis: correlation with the by guest on September 24, 2021 distinct cellular compartments. Eur. J. Immunol. 32: 1958–1968. clinical parameters. Clin. Exp. Immunol. 107: 175–181. 42. Das, J., C.-H. Chen, L. Yang, L. Cohn, P. Ray, and A. Ray. 2001. A critical role 59. Pantelidis, P., G. C. Fanning, A. U. Wells, K. I. Welsh, and R. M. Du Bois. 2001. for NF-␬B in Gata3 expression and Th2 differentiation in allergic airway inflam- Analysis of tumor necrosis factor-␣, lymphotoxin-␣, tumor necrosis factor re- mation. Nat. Immunol. 2: 45–50. ceptor II, and interleukin-6 polymorphisms in patients with idiopathic pulmonary 43. Oriss, T. B., M. Ostroukhova, C. Seguin-Devaux, B. Dixon-McCarthy, fibrosis. Am. J. Respir. Crit. Care Med. 163: 1432–1436. D. B. Stolz, S. C. Watkins, B. Pillemer, P. Ray, and A. Ray. 2005. Dynamics of 60. Hwang, E. S., S. J. Szabo, P. L. Schwartzberg, and L. H. Glimcher. 2005. T dendritic cell phenotype and interactions with CD4ϩ T cells in airway inflam- helper cell fate specified by kinase-mediated interaction of T-bet with GATA-3. mation and tolerance. J. Immunol. 174: 854–863. Science 307: 430–433. 44. Yang, L., L. Cohn, D.-H. Zhang, R. Homer, A. Ray, and P. Ray. 1998. Essential 61. Miaw, S. C., A. Choi, E. Yu, H. Kishikawa, and I. C. Ho. 2000. ROG, repressor role of nuclear factor ␬B in the induction of eosinophilia in allergic airway in- of GATA, regulates the expression of cytokine genes. Immunity 12: 323–333. flammation. J. Exp. Med. 188: 1739–1750. 62. Cisco, R. M., Z. Abdel-Wahab, J. Dannull, S. Nair, D. S. Tyler, E. Gilboa, 45. Zhang, D. H., L. Yang, L. Cohn, L. Parkyn, R. Homer, P. Ray, and A. Ray. 1999. J. Vieweg, Y. Daaka, and S. K. Pruitt. 2004. Induction of human dendritic cell Inhibition of allergic inflammation in a murine model of asthma by expression of maturation using transfection with RNA encoding a dominant positive Toll-like a dominant-negative mutant of GATA-3. Immunity 11: 473–482. receptor 4. J. Immunol. 172: 7162–7168.