TLR4-mediated activation of dendritic cells by the heat ... - CiteSeerX

2 downloads 8 Views 505KB Size Report
TLR4-mediated activation of dendritic cells by the heat shock protein DnaK from Francisella tularensis. Amit R. Ashtekar,* Ping Zhang,† Jannet Katz,† Champion ...
TLR4-mediated activation of dendritic cells by the heat shock protein DnaK from Francisella tularensis Amit R. Ashtekar,* Ping Zhang,† Jannet Katz,† Champion C. S. Deivanayagam,‡ Prasad Rallabhandi,§ Stefanie N. Vogel,§ and Suzanne M. Michalek*,1 Departments of *Microbiology, †Pediatric Dentistry, and ‡Vision Sciences, University of Alabama at Birmingham, Birmingham, Alabama, USA; and §Department of Microbiology and Immunology, University of Maryland School of Medicine, Baltimore, Maryland, USA

Abstract: Francisella tularensis is the causative agent of tularemia, a severe, debilitating disease of humans and other mammals. As this microorganism is also classified as a “category-A pathogen” and a potential biowarfare agent, there is a need for an effective vaccine. Several antigens of F. tularensis, including the heat shock protein DnaK, have been proposed for use in a potential subunit vaccine. In this study, we characterized the innate immune response of murine bone marrow-derived dendritic cells (DC) to F. tularensis DnaK. Recombinant DnaK was produced using a bacterial expression system and purified using affinity, ionexchange, and size-exclusion chromatography. DnaK induced the activation of MAPKs and NF-␬B in DC and the production of the proinflammatory cytokines IL-6, TNF-␣, and IL-12 p40, as well as low levels of IL-10. DnaK induced phenotypic maturation of DC, as demonstrated by an up-regulation of costimulatory molecules CD40, CD80, and CD86. DnaK stimulated DC through TLR4 and the adapters MyD88 and Toll/IL-1R domain-containing adaptor-inducing IFN-␤ (TRIF) that mediated differential responses. DnaK induced activation of MAPKs and NF-␬B in a MyD88- or TRIF-dependent manner. However, the presence of MyD88and TRIF-dependent signaling pathways was essential for an optimal, DnaK-induced cytokine response in DC. In contrast, DnaK induced DC maturation in a TRIF-dependent, MyD88-independent manner. These results provide insight about the molecular interactions between an immunodominant antigen of F. tularensis and host immune cells, which is crucial for the rational design and development of a safe and efficacious vaccine against tularemia. J. Leukoc. Biol. 84: 1434 –1446; 2008. Key Words: Toll-like receptors 䡠 NF-␬B



MyD88



TRIF



MAP kinases

INTRODUCTION Francisella tularensis is a facultative, intracellular, gram-negative coccobacillus and the causative agent of tularemia, a 1434

Journal of Leukocyte Biology Volume 84, December 2008

severe, debilitating disease of humans and other mammals [1]. Tularemia can be initiated by bites from ticks or mosquitoes, handling carcasses of infected wildlife, drinking contaminated water, or inhaling infectious aerosols [1]. Depending on the route of infection, tularemia is characterized as ulceroglandular, occuloglandular, or respiratory, the latter being the most severe form of the disease [1]. Although the incidence of tularemia has declined, probably because of increased awareness, there has been a renewed interest in F. tularensis as a human pathogen as a result of its potential as a biowarfare weapon. Because of its virulence and ease of transmission, F. tularensis has been classified as a category A bioterrorism agent by the Centers for Disease Control and Prevention [1–3]. There are currently four subspecies of F. tularensis. The subspecies of F. tularensis of major clinical importance are subspecies tularensis (type A) and subspecies holarctica (type B), and the type A strains are most virulent in humans [4]. A live, attenuated vaccine strain (LVS), derived from a type B strain, has been available for over 50 years [5] but is not licensed as a result of shortcomings associated with its use that include a lack of information about the genetic basis for the attenuation [6]. Thus, a need exists for the development of an effective vaccine. A number of virulence factors of F. tularensis have been proposed [7–11]; however, the mechanisms involved in the interactions between virulence antigens and the host, which could help in the development of vaccines protective against tularemia or therapies for the treatment or amelioration of disease, are still poorly defined. Among the virulence antigens derived from F. tularensis, studies have shown the induction of long-term CD4 and/or CD8 T cell memory to heat shock or stress proteins (HSPs), including GroEL (Cpn-60, HSP60), GroES (Cpn-10, HSP10), and DnaK, following tularemia infection or the administration of F. tularensis LVS [7, 8]. A number of membrane proteins of F. tularensis also appear to be recognized by T cells [9 –11]. The HSPs represent a highly conserved family of proteins that are expressed constitutively and under stressful conditions in all prokaryotes and eu-

1

Correspondence: Department of Microbiology, University of Alabama at Birmingham, 845 19th Street South, BBRB 258/5, Birmingham, AL 352942170, USA. E-mail: [email protected] Received March 28, 2008; revised July 9, 2008; accepted July 25, 2008. doi: 10.1189/jlb.0308215

0741-5400/08/0084-1434 © Society for Leukocyte Biology

karyotes [12–14]. HSPs play essential roles as intracellular molecular chaperons for correct folding, stabilization, and translocation of proteins under all conditions. During stress (e.g., heat, anoxia, glucose starvation), the synthesis of HSPs is increased, and therefore, the range of such proteins may be up-regulated when a pathogen tries to overcome the host defense mechanisms and evade the immune system [15–17]. Apart from providing housekeeping and cytoprotective functions to the pathogen itself, bacterial HSPs can also have a range of stimulatory effects on the host immune system, which alerts the host of a pathogenic invasion, and thus, provides evidence for their use as appropriate antigens for the induction of protective immune responses. When a particular microbial component is proposed as a potential vaccine candidate, it is important to understand how the antigen interacts with the immune system and the underlying mechanisms responsible for such effects. In this regard, the TLRs, type I transmembrane pattern recognition receptors that are expressed differentially among host cells, are capable of specifically detecting various conserved microbial products that lead to the activation of innate immunity. For example, TLR4 is critical for the recognition of enterobacterial LPS [18]; TLR2 can heterodimerize with TLR1 [19] or TLR6 [20] and recognizes a broad range of ligands including lipoteichoic acids [21], peptidoglycans [22], and lipoproteins [23]; TLR5 recognizes bacterial flagellin [24]; and TLR3 and TLR9 are specific for double-stranded RNA and CpG-containing DNA, respectively [25, 26]. Recognition of various conserved, microbial products by TLRs leads to the activation of a variety of signal transduction pathways [22, 27–31]. Thus, TLRs have a crucial function in the innate arm of the immune system and also influence the development of an adaptive immune response to pathogens. Recognition of conserved microbial products by TLRs expressed on APC, e.g., dendritic cells (DC), leads to their activation and maturation, characterized by the production of inflammatory cytokines and enhanced expression of costimulatory and MHC molecules [32–35]. Mature DC are capable of processing and presenting peptides to naive T cells in the context of the MHC molecules, leading to the initiation of specific immunity, thus coupling ligand recognition by TLRs to the development of antigen-specific T cell responses. Therefore, DC activation by TLRs is a critical link between innate and adaptive immunity and crucial for the generation of protective immune responses. Central to the TLR signal transduction pathways are two intracytoplasmic adaptor molecules, MyD88 and Toll/IL-1R domain-containing adaptor-inducing IFN-␤ (TRIF), which differentially mediate downstream TLR signaling [36 –39]. MyD88 is a crucial adaptor molecule common to all TLRs except TLR3, which signals exclusively through TRIF [40 – 42]. TLR9 and TLR2 signal through MyD88 [36], whereas signaling through TLR4 involves MyD88 and TRIF. LPS stimulation via TLR4 can lead to MyD88-dependent activation of MAPKs, NF-␬B, and expression of various inflammatory mediators, including IL-6, TNF-␣, and IL-12 p40 [43, 44]. On the other hand, LPS stimulation via the TRIF-dependent pathway leads to activation of IFN regulatory factor-3 (IRF-3), induction of IFN-␤, and many IFN-inducible genes [40, 45– 47]. The TRIF-dependent pathway also activates MAPKs and NF-␬B,

although these events are delayed [40, 44, 45]. In addition to mediating cellular responses to TLR ligands, MyD88 and TRIF have been implicated in mediating the balance governing Th1 and Th2 immune responses. In the absence of MyD88, Th1 responses are mitigated, and Th2 responses remain intact [48, 49]; however, another report has shown that Th2 sensitization to intranasal protein antigen in the presence of low-dose LPS is MyD88-dependent [50]. Thus, recent in vitro and in vivo studies with MyD88- and TRIF-deficient mice have highlighted the consequences of specific adaptor use and suggested that MyD88- and TRIF-dependent pathways may work in concert to regulate particular cell responses, and some responses can be mediated entirely independent of each other. In the present study, we characterized the innate immune response induced by F. tularensis DnaK in murine bone marrow-derived DC cultures. Our results show that DnaK is a TLR4 ligand and induces the activation of MAPKs and NF-␬B in DC through a MyD88- or TRIF-dependent signaling pathway. DnaK induced the production of the proinflammatory cytokines IL-6, TNF-␣, and IL-12 p40 and the anti-inflammatory cytokine IL-10 by DC. However, production of IL-10 was low compared with IL-12 p40. MyD88 and TRIF were indispensable for an optimal, DnaK-induced cytokine response. However, DnaK induced DC maturation in a TRIF-dependent and MyD88-independent manner.

MATERIALS AND METHODS Mice TLR2, TLR4, and MyD88 knockout (KO), TRIF-deficient, and C57BL/6 wild type (WT) mice were bred and maintained within an environmentally controlled, pathogen-free animal facility at the University of Alabama at Birmingham (UAB; Birmingham, AL, USA). The original TLR2, TLR4, and MyD88 KO breeding pairs were obtained under a Materials Transfer Agreement from Dr. Shizuo Akira (Osaka University, Osaka, Japan). The original TRIF-deficient breeding pairs were obtained from the Jackson Laboratories (Bar Harbor, ME, USA). Mice were 8 –12 weeks of age when used in the studies. All studies were performed according to the National Institutes of Health guidelines, and the UAB Institutional Animal Care and Use Committee approved the protocols.

Reagents Escherichia coli K12 LPS and N-palmitoyl-S-[2,3-bis(palmitoyloxy)-(2R,S)propyl]-(R)-cysteinyl-seryl-(lysly)(3)-lysine (Pam3CSK4) were obtained from InvivoGen (San Diego, CA, USA). E. coli K235 LPS was prepared by a modification of the method of McIntire et al. [51] using two rounds of hot phenol-water extraction. Polymyxin B sulfate (Sigma-Aldrich, St. Louis, MO, USA) and proteinase K (Fermentas Inc., Glen Burnie, MD, USA) were used in experiments to test for endotoxin contamination in the DnaK preparation.

Generation of DC Bone marrow-derived DC were generated as described initially [52] with minor modifications [53]. Briefly, the femurs and tibiae from mice were dissected, cleaned, and flushed with 15 ml ice-cold HBSS supplemented with 5% FBS into a sterile, bacteriological grade polystyrene petri dish. Erythrocytes were lysed by treatment with ammonium chloride. DC were generated by culturing the bone marrow-derived cells (2⫻105/ml) in the presence of 20 ng/ml recombinant GM-CSF (R&D Systems, Minneapolis, MN, USA) for 10 days in RPMI-1640 culture medium (Cellgro Mediatech, Washington, DC, USA) supplemented with 10% FBS, 2 mM L-glutamine, 50 ␮M 2-ME, 20 mM HEPES, 1 mM sodium pyruvate, 50 U/ml penicillin, and 50 ␮g/ml streptomycin (RPMI-1640 complete media) in a humidified 5% CO2 incubator at 37°C. The resulting, nonadherent DC were harvested after 10 days, and the purity was

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1435

determined by flow cytometry (FACSCalibur, BD Biosciences, San Jose, CA, USA). This procedure routinely yielded ⬎80% CD11c-positive cells.

and the culture supernatants were harvested after 20 h and assessed for IL-6 levels by ELISA.

Plasmid constructs and reporter assay

Cytokine ELISA and flow cytometry

Human embryo kidney (HEK)293T cells were cultured in DMEM (BioWhittaker, Walkersville, MD, USA) supplemented with 10% FBS, 2 mM Lglutamine, 100 units/ml penicillin, and 100 ␮g/ml streptomycin. All cells were maintained in a 37°C humidified atmosphere with 5% CO2. HEK293T cells were transfected with plasmids encoding TLR4, coreceptors, and reporter constructs as described previously [54]. Briefly, HEK293T cells (2⫻105) were seeded into 12-well plates (Corning Inc., Corning, NY, USA) and cotransfected for 4 h with pcDNA3-TLR4 (300 ng/well)and pcDNA3-human (hu)CD14 (30 ng/well) and with or without pEFBOS-hemagglutinin (HA)-human myeloid differentiation protein 2 (huMD-2; 3 ng/well), together with the NF-␬B reporter [endothelial-leukocyte adhesion molecule 1-luciferase reporter plasmid (pELAM-luc), 500 ng/well] and pCMV1-␤-galactosidase (␤-Gal; 100 ng/well) using the SuperFect transfection reagent (Qiagen Inc., Valencia, CA, USA). The plasmids pcDNA3-TLR4, pcDNA3-huCD14, pEFBOS-HA-huMD-2, and ELAM-1 luciferase pELAM-luc (NF-␬B reporter) and pCMV1-␤-Gal reporter constructs have been described elsewhere [54, 55]. After a 20-h recovery, the transfectants were stimulated with E. coli K235 LPS (5 ng/ml) or DnaK at different concentrations for 16 h. Cells were harvested and lysed, and the luciferase and ␤-Gal activities were determined as described elsewhere [54]. The supernatants were analyzed for IL-8 levels by ELISA by the Cytokine Core Laboratory (University of Maryland, Baltimore, MD, USA).

Expression and purification of F. tularensis DnaK Recombinant F. tularensis DnaK was produced using a bacterial expression system. The gene encoding DnaK (originally obtained from Anders Sjostedt, Umea University, Sweden) was cloned into the pET-23d vector (Novagen, Madison, WI, USA) and was transformed into E. coli strain BL21 (DE3) pLysS for protein expression (Novagen). Bacterial transformants were grown at 30°C in Luria-Bertani broth containing 50 ␮g/ml ampicillin and 50 ␮g/ml chloramphenicol to an OD of 0.6 at 600 nm, and protein expression was induced for 4 h following the addition of 0.5 mM isopropyl ␤-D-thiogalactoside. Cells were harvested by centrifugation and resuspended in a buffer containing 50 mM Tris and 300 mM NaCl at pH 8. Cell lysis was carried out using a sonicator (3–5 min) with a temperature probe that maintained the sample temperature below 16°C. The soluble fraction was collected after ultracentrifugation, and the DnaK was purified from this fraction using a three-step purification procedure. First, the soluble fraction was passed through a nickel sepharose column (HisPrep FF 16/10, GE Healthcare, Piscataway, NJ, USA) that selectively bound the C-terminal histidine tag of the DnaK, which was then eluted from the column with a linear imidazole gradient (0 –300 mM). The fractions containing DnaK (as determined by SDS-PAGE) were pooled, dialyzed in a buffer containing 25 mM Tris and 150 mM NaCl at pH 8, and applied on to an anion exchange column (Mono Q HR 10/10, GE Healthcare). DnaK was eluted using a linear NaCl gradient (150 mM–1.0 M), where DnaK eluted between 180 and 250 mM NaCl. The fractions containing DnaK were pooled and concentrated to a final volume of 5 ml and applied on to a size exclusion column (HiLoad 26/60 Superdex 200 pg, GE Healthcare). The purity of DnaK was confirmed by SDS-PAGE and Western blot analysis using a mouse anti-DnaK mAb (Stressgen/Assay Designs, Ann Arbor, MI, USA). The concentration of DnaK was estimated by a bicinchoninic acid (BCA) protein determination assay (Pierce, Rockford, IL, USA).

Testing for endotoxin To remove any possible contaminating endotoxin, purified DnaK preparations were passed through an EndoTrap Blue column (Profos AG, Germany). The level of endotoxin in the DnaK preparation was determined by a quantitative, chromogenic QCL-1000 Limulus amoebocyte lysate (LAL) assay (Cambrex Bio Science Walkersville, Inc., Walkersville, MD, USA) and was found to be ⬍2.1 ng/mg protein. To ensure that any trace amount of endotoxin did not contribute to the observed responses, DnaK (20 ␮g/ml) was left untreated or was preincubated with polymyxin B (10 ␮g/ml) for 15 min at room temperature, heated at 100°C for 30 min, or preincubated with proteinase K (5 ␮g/ml) for 1 h at 37°C and subsequently at 95°C for 5 min to inactivate proteinase K [56 –58]. These preparations and appropriate controls were then used to stimulate DC,

1436

Journal of Leukocyte Biology Volume 84, December 2008

DC derived from WT, TLR2, TLR4, MyD88 KO, and TRIF-deficient mice were cultured in 96-well tissue-culture plates (3⫻105/well) containing RPMI-1640 complete media in a humidified 5% CO2 incubator at 37°C. Following incubation with DnaK (20 ␮g/ml), LPS (100 ng/ml), or Pam3CSK4 (300 ng/ml) for 20 h, DC and culture supernatants were harvested. Culture supernatants were assessed for the levels of IL-1␤, IL-6, TNF-␣, IL-23 p19, and IFN-␥ (eBioscience, San Diego, CA, USA), IL-10 and IL-12 p40 (R&D Systems), and IFN-␤ (PBL InterferonSource, Piscataway, NJ, USA) by ELISA, according to the manufacturers’ instructions. The DC were suspended in FACS buffer and then stained with fluorescent-labeled antibodies against CD11c and CD80, CD86, CD40, or MHC-II (eBioscience) for 30 min on ice. The cells were washed twice and resuspended in 300 ␮l FACS buffer and analyzed immediately by flow cytometry (FACSCalibur with CellQuest software, BD Biosciences).

Preparation of whole cell lysates DC derived from WT, TLR2, TLR4, MyD88 KO, and TRIF-deficient mice were cultured in 24-well tissue-culture plates (3⫻106 cells/well) and incubated with DnaK (20 ␮g/ml), LPS (100 ng/ml), or Pam3CSK4 (300 ng/ml) for 0, 10, 30, 60, or 120 min in a humidified 5% CO2 incubator at 37°C. The cells were then harvested, washed twice with ice-cold PBS, and then lysed on ice for 10 min in radioimmunoprecipitation assay lysis buffer (Upstate Biotechnology, Lake Placid, NY, USA) freshly supplemented with 1 mM PMSF, 1 mM Na3VO4, 1 mM NaF, and protease inhibitor cocktail tablets (Complete, Mini, EDTA-free, Roche Applied Science, Indianapolis, IN, USA). The whole cell lysates were transferred to tubes and incubated on ice for an additional 20 min, and then supernatants were collected following centrifugation at 15,000 rpm for 15 min at 4°C.

Western blot analysis Equivalent amounts of protein samples from whole cell lysates were analyzed by SDS-PAGE on a 12.5% Tris-HCl gel. The proteins were then electrotransferred onto Immobilon-P polyvinylidene fluoride transfer membranes (Millipore, Billerica, MA, USA) and probed with specific antibodies against total p38 MAPK and the phosphorylated forms of p38 (Thr180/Tyr182), ERKs (p44/42, Thr202/Tyr204), or JNKs (stress-activated protein kinase/JNK, Thr183/ Tyr185; Cell Signaling Technology, Beverly, MA, USA). Detection of bands was carried out using HRP-linked anti-rabbit IgG, followed by ECL Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ, USA).

NF-␬B activation assay DC derived from WT, TLR2, TLR4, MyD88 KO, and TRIF-deficient mice were cultured in 24-well tissue-culture plates (3⫻106 cells/well) and incubated with DnaK (20 ␮g/ml) for 0, 10, 30, 60, or 120 min in a humidified 5% CO2 incubator at 37°C. Nuclear extracts were prepared using the TransAM nuclear extract kit (Active Motif, Carlsbad, CA, USA). The protein concentration in the nuclear extracts was estimated by a BCA protein determination assay (Pierce) and then adjusted to equal levels accordingly. Equivalent concentrations of nuclear extracts were used to measure the level of activated NF-␬B by using the ELISA-based TransAM NF-␬B p65 assay kit (Active Motif) and a specific anti-p65 mAb, according to the manufacturer’s instructions.

Statistical analysis To determine statistical significance, data were subjected to an unpaired ANOVA, followed by post-hoc analysis with the Tukey-Kramer multiple comparison test using the GraphPad InStat Version 3.0a (GraphpPad Software, San Diego, CA, USA). Differences between groups were considered significant at the level of P ⬍ 0.05.

http://www.jleukbio.org

Fig. 1. Dose-dependent production of cytokines by DC in response to DnaK. DC (3⫻105 cells), derived from WT mice, were stimulated with various amounts of DnaK. Culture supernatants were harvested after 20 h and assessed for IL-6, TNF-␣, IL-12 p40, and IL-10 levels by ELISA. Results are expressed as the mean ⫾ SD of triplicate cultures from one of five representative experiments.

RESULTS F. tularensis DnaK induces pro- and antiinflammatory cytokine production by DC Central to the development of protective immune responses is the ability of a vaccine antigen to interact and activate professional APC, such as DC, which are the only cells capable of activating naive T cells and thus, are involved in the initiation and development of antigen-specific immune responses [32, 35]. Therefore, we initially determined the ability of F. tularensis DnaK to stimulate DC and induce a response. DC from WT mice were incubated with 0 – 40 ␮g/ml DnaK for 20 h, and then culture supernatants were assessed for the production of IL-1␤, IL-6, TNF-␣, IL-12 p40, IL-23 p19, IL-10, and IFN-␥ by cytokine-specific ELISA. Stimulation of DC with DnaK resulted in a dose-dependent increase in the production of the proinflammatory cytokines IL-6, TNF-␣, and IL-12 p40 and the anti-inflammatory cytokine IL-10 (Fig. 1). Although DnaK induced the production of some IL-10 by DC, the concentration of secreted cytokine was significantly less (P⬍0.001; at DnaK, 20 ␮g/ml) compared with IL-12 p40. DnaK did not induce the secretion of IL-1␤, IFN-␥, and IL-23 p19 (data not shown). As an optimal induction of IL-6 and TNF-␣ by DnaK was observed at a dose of 20 ␮g/ml, this concentration was used in subsequent experiments.

F. tularensis DnaK induces phenotypic maturation of DC In addition to the induction of the appropriate inflammatory cytokines, the efficient activation of naive T cells requires the

expression of costimulatory molecules by DC [32, 34, 35]. Therefore, we next determined whether DnaK induces the phenotypic maturation of DC, as characterized by an upregulation of the costimulatory molecules CD80 and CD86 and of CD40 and MHC-II molecules. DC from WT mice were incubated with DnaK (20 ␮g/ml) or LPS (100 ng/ml) for 20 h and then analyzed for the expression of costimulatory and MHC-II molecules by flow cytometry. DnaK induced a significant up-regulation in the expression of CD80, CD86, and CD40, and this increase was almost comparable with that induced by LPS (Fig. 2). However, DnaK induced only a subtle increase in the expression of MHC-II molecules, unlike LPS, which up-regulated their expression significantly.

Stimulation of DC with F. tularensis DnaK leads to the activation of MAPKs and NF-␬B Cell recognition of extracellular stimuli can lead to the activation of various intracellular signaling pathways including MAPKs and NF-␬B [32, 35–39]. Therefore, we next examined whether DnaK induced the activation of MAPKs and NF-␬B in DC, which from WT mice, were stimulated with DnaK (20 ␮g/ml) for 0, 10, 30, 60, and 120 min, and whole cell lysates and nuclear extracts were prepared separately. Whole cell lysates were probed for the phosphorylation of p38, ERK, and JNK MAPKs by Western blotting, whereas nuclear extracts were analyzed for the DNA-binding activity of NF-␬B by an ELISA-based assay. Stimulation of DC with DnaK led to the phosphorylation of p38, ERK, and JNK MAPKs (Fig. 3A) and also led to the activation and nuclear translocation of NF-␬B p65 (Fig. 3B) in a time-dependent manner. The phosphorylaFig. 2. DnaK induces phenotypic maturation of DC. DC (3⫻105 cells) derived from WT mice were unstimulated (Un) or stimulated with DnaK (20 ␮g/ml) or LPS (100 ng/ml). Cells were harvested after 20 h and analyzed for the surface expression of CD80, CD86, CD40, and MHC-II molecules by flow cytometry. * and **, Significant differences at P ⬍ 0.001 and P ⬍ 0.01, respectively, compared with unstimulated cultures. Results are expressed as the mean fluorescence intensity (MFI) ⫾ SD of three representative experiments.

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1437

DnaK preparation used in our experiments to stimulate DC (20 ␮g/ml) contained ⬍0.42 EU/ml endotoxin, which is equivalent to ⬍0.042 ng/ml (⬍0.0002%), a level that does not stimulate DC. Under the conditions used in this study, at least 0.16 ng/ml LPS was required to stimulate the production, although very low, of IL-6 by DC (data not shown). To demonstrate that the trace amount of endotoxin did not contribute to the observed responses, we performed a series of experiments, where the endotoxin was neutralized by using polymyxin B, or DnaK was denatured by heat or proteinase K digestion before incubating with DC. Heat or proteinase K treatment of DnaK completely abolished DnaK-induced IL-6 production by DC (Fig. 4). This indicated that denaturation or digestion of DnaK renders it incapable of activating DC. However, DnaK was resistant to polymyxin B treatment, indicating that any endotoxin present had no effect on the stimulatory activity of DnaK. In contrast, the ability of LPS to induce IL-6 production by DC was resistant to heat and proteinase K digestion but was abolished completely with polymyxin B treatment (Fig. 4).

F. tularensis DnaK is a TLR4 ligand

Fig. 3. DnaK induced time-dependent activation of MAPKs and NF-␬B in DC. DC (3⫻106 cells) derived from WT mice were stimulated for the indicated times with DnaK (20 ␮g/ml), and then whole cell lysates and nuclear extracts were prepared separately. (A) Whole cell lysates were probed for the phosphorylated form of p38 (p-p38), ERK, and JNK MAPKs by Western blotting. Total p38 served as a protein-loading control. Results are representative of five independent experiments. (B) Equal amounts of nuclear extracts were probed for NF-␬B-binding activity using the TransAM NF-␬B p65 assay kit. **, Significant difference at P ⬍ 0.01 compared with unstimulated cultures at 0 min. Results are expressed as the mean OD ⫾ SD of three representative experiments.

Next, we assessed if a specific TLR was involved in the observed immune responses induced by F. tularensis DnaK. DC from WT, TLR2, and TLR4 KO mice were stimulated with Pam3CSK4 (300 ng/ml), LPS (100 ng/ml), or DnaK (20 ␮g/ml) for 20 h. The culture supernatants were then assessed for cytokine production by ELISA, and the cells were assessed for the expression of costimulatory molecules by flow cytometry. DnaK induced comparable levels of IL-6,

tion of p38 and ERK was observed by 10 min following stimulation, whereas the phosphorylation of JNK was not detected until 60 min. Nuclear translocation of NF-␬B p65 was detected at 30 min and peaked and plateaued at 60 and 120 min.

F. tularensis DnaK-induced DC activation is not mediated by contaminating endotoxin Recombinant protein preparations and especially those obtained from bacterial expression systems can have endotoxin contamination, which if present in protein preparations, may lead to erroneous results and misinterpretation of data. LPS (endotoxin), a TLR4 agonist and a major component of the cell wall of gram-negative bacteria, can induce a variety of biological responses, including cytokine production by macrophages and DC, B cell proliferation, and endotoxin shock [43]. Therefore, to ensure a lack of endotoxin contamination in the DnaK preparation, the material derived from the final purification step was passed through an EndoTrap Blue column. The level of endotoxin detected in the DnaK preparation was ⬍21 EU/mg protein, as determined by the LAL assay. Therefore, the 1438

Journal of Leukocyte Biology Volume 84, December 2008

Fig. 4. DnaK-induced DC activation is not mediated by endotoxin contamination. DC (3⫻105 cells), derived from WT mice, were stimulated with DnaK (20 ␮g/ml) or LPS (100 ng/ml), which was untreated or treated with polymyxin B (10 ␮g/ml), proteinase K (5 ␮g/ml), or heat (100°C, 30 min). Culture supernatants were harvested after 20 h, and IL-6 levels were assessed by ELISA. *, Significant difference at P ⬍ 0.001 compared with cultures stimulated with untreated DnaK or LPS. Results are expressed as the mean ⫾ SD of triplicate cultures from one of five representative experiments.

http://www.jleukbio.org

Fig. 5. DnaK signals through TLR4. DC (3⫻105 cells), derived from WT, TLR2, and TLR4 KO mice, were unstimulated or stimulated with Pam3CSK4 (300 ng/ml), LPS (100 ng/ml), or DnaK (20 ␮g/ml) for 20 h. (A) Culture supernatants were harvested, and levels of IL-6, TNF-␣, IL-10, and IL-12 p40 were assessed by ELISA. * and ***, Significant differences at P ⬍ 0.001 and P ⬍ 0.05, respectively, compared with WT cultures. Results are expressed as the mean ⫾ SD of triplicate cultures from one of three representative experiments. (B) Flow cytometric analysis of the surface expression of CD40 and CD86. * and **, Significant differences at P ⬍ 0.001 and P ⬍ 0.01, respectively, compared with WT cultures. Results are expressed as the mean of the relative increase in MFI over unstimulated cultures ⫾ SD of three representative experiments. (C) DC (3⫻106 cells), derived from WT, TLR2, and TLR4 KO mice, were stimulated with DnaK (20 ␮g/ml) for the indicated times. Whole cell lysates were probed for the phosphorylated form of p38 MAPK by Western blotting. Total p38 served as a protein-loading control. Results are representative of five independent experiments. (D) DC (3⫻106 cells), derived from WT, TLR2, and TLR4 KO mice, were stimulated with DnaK (20 ␮g/ml) for the indicated times. Equal amounts of nuclear extracts were analyzed for NF-␬B-binding activity using the TransAM NF-␬B p65 assay kit. * and **, Significant differences at P ⬍ 0.001 and P ⬍ 0.01, respectively, compared with unstimulated cultures at 0 min. Results are expressed as the mean OD ⫾ SD of three representative experiments.

TNF-␣, IL-10, and IL-12 p40 in WT and TLR2 KO DC; however, this response was abolished completely in the absence of TLR4 (Fig. 5A). The LPS control also induced the production of inflammatory cytokines in WT and TLR2

KO DC, but not TLR4 KO DC. The levels of IL-6 and TNF-␣ induced by LPS were similar to that seen with DnaK; however, the levels of IL-12 p40 and IL-10 induced by LPS were higher than that seen with DnaK. The TLR2 agonist

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1439

Pam3CSK4 induced the production of IL-6, TNF-␣, IL-12 p40, and IL-10 in WT and TLR4 KO DC, but not in DC from TLR2 KO mice. The levels of cytokines induced by Pam3CSK4 were similar to that seen with DnaK (Fig. 5A). DnaK was also able to induce up-regulation of the maturation markers CD40 and CD86 on DC from WT and TLR2 KO, but not TLR4 KO mice (Fig. 5B). As expected, controls Pam3CSK4 and LPS were unable to induce up-regulation of the maturation markers CD40 and CD86 in TLR2 KO and TLR4 KO DC, respectively. In a separate series of experiments, DC from WT, TLR2 KO, and TLR4 KO mice were stimulated with DnaK (20 ␮g/ml) for 0, 10, 30, 60, and 120 min, and whole cell lysates and nuclear extracts were prepared separately and probed for MAPK and NF-␬B activation, respectively. DnaK stimulated the phosphorylation of p38 MAPK (Fig. 5C) and the activation of NF-␬B (Fig. 5D) in WT and TLR2 KO, but not in TLR4 KO DC. Precisely the same pattern of response was observed for ERK and JNK (data not shown). Taken together, these results demonstrate that DnaK is a TLR4 ligand.

DnaK activation of TLR4 is MD-2-dependent MD-2, a secreted glycoprotein, forms stable complexes with TLR4 on the cell surface and is indispensable for proper TLR4 activation [59, 60]. Although LPS and DnaK have different chemical and structural properties, both signal through TLR4. Therefore, we next assessed whether DnaK required MD-2 to activate TLR4, as shown previously for LPS [59, 60]. To address this possibility, we used the HEK293T cell system as a result of extreme sensitivity and clear “gain-of-function” readout. Also, HEK293T cells do not express any endogenous TLR4, MD-2, and CD14 proteins [61]. HEK293T cells were transiently transfected with vectors encoding huTLR4 and huCD14, in the presence or absence of MD-2 cotransfection, along with ELAM (NF-␬B) luciferase and ␤-Gal reporter constructs. TLR4/MD-2/ CD14-transfected cells were activated by DnaK in a dosedependent manner (Fig. 6, A and B), as evidenced by increased luciferase activity and IL-8 secretion. However, TLR4/CD14-transfected cells were unresponsive to DnaK, indicating that MD-2 is absolutely essential for DnaKmediated activation of TLR4. A similar requirement of MD-2 for signaling via TLR4 has been demonstrated for mycobacterial HSP65 [56] and chlamydial HSP60 [62]. Furthermore, pretreatment with polymyxin B did not affect DnaK-induced NF-␬B activation of HEK293T transfectants, which exhibited a similar requirement for MD-2 (Fig. 6, A and B). This is important, as it again demonstrates that endotoxin contamination is not a contributing factor for the observed responses. As expected, responses induced by E. coli K235 LPS in TLR4/MD-2/CD14-transfected cells were reduced significantly in the presence of polymyxin B and abolished completely in the absence of MD-2. Although much is known about the interactions of the TLR4-MD-2 complex with LPS [63], further studies are required to elucidate the interactions of a protein molecule such as DnaK at the TLR4-MD-2 interface. 1440

Journal of Leukocyte Biology Volume 84, December 2008

Fig. 6. DnaK activation of TLR4 is MD-2-dependent. HEK293T cells (2⫻105 cells/well) were cotransfected with the indicated amounts (in Materials and Methods) of untagged huTLR4 and huCD14 and with or without MD-2 expression constructs. After an overnight incubation, transfected cells were treated with E. coli K235 LPS (5 ng/ml) or with DnaK (10, 20, or 40 ␮g/ml) and with (right panels) or without (left panels) polymyxin B (10 ␮g/ml) for 16 h. (A) Luciferase and ␤-Gal activities were measured in cell lysates as described in Materials and Methods and represented as relative luciferase units (RLU). (B) Supernatants from the above cultures were analyzed for IL-8 levels by ELISA. *, Significant difference at P ⬍ 0.001 between LPS and LPS ⫹ polymyxin B treatments. Results are expressed as the mean ⫾ SEM and are representative of two independent experiments.

MyD88 and TRIF are indispensable for an optimal F. tularensis DnaK-induced cytokine response in DC TLR4 can signal through the MyD88- and TRIF-dependent pathways [40, 45, 47, 64, 65]. As certain TLR4 ligands such as LPS and monophosphoryl lipid A have been shown to modulate immune responses by differentially using the MyD88- and TRIF-dependent pathways [66], we sought to investigate the importance of the MyD88 and TRIF signaling pathways in DnaK-induced cytokine production by DC. In this set of experiments, we compared the cytokine responses in DC from WT, MyD88 KO, and TRIF-deficient mice. DnaK-induced production of IL-6, TNF-␣, IL-10, and IL-12 p40 was severely impaired in the absence of MyD88 or TRIF (Fig. 7). However, the requirement of MyD88 and TRIF in mediating the induction of these cytokines varied, in that differences were seen in IL-6 and IL-12 p40, but not IL-10 and TNF-␣ production. DnaK induced significantly higher (P⬍0.001) levels of IL-6 and IL-12 p40 in TRIFdeficient compared with MyD88 KO DC (Fig. 7), indicating that the adaptor molecule MyD88 is more important than TRIF for the induction of these cytokines. Furthermore, DnaK-induced IL-10 production was abolished completely in the absence of MyD88 or TRIF. The production of TNF-␣ was not abolished completely and was similarly affected in http://www.jleukbio.org

Fig. 7. Role of MyD88 and TRIF in DnaK-induced cytokine production in DC. DC (3⫻105 cells) derived from WT, MyD88 KO, and TRIF-deficient mice were stimulated with Pam3CSK4 (300 ng/ml), LPS (100 ng/ml), or DnaK (20 ␮g/ml) for 20 h. Culture supernatants were harvested and assessed for IL-6, TNF-␣, IL-10, and IL-12 p40 by ELISA. *, Significant difference at P ⬍ 0.001 compared with WT cultures or between groups, as indicated in the figures. Results are expressed as the mean ⫾ SD of triplicate cultures from one of three representative experiments.

the absence of MyD88 or TRIF. A similar requirement of MyD88 and TRIF for the induction of IL-6, TNF-␣, IL-10, and IL-12 p40 was observed with LPS-stimulated DC. Our results suggest that DnaK can induce limited amounts of proinflammatory cytokines in the absence of MyD88 or TRIF, whereas the presence of both of the adaptor molecules is crucial for an optimal, DnaK-induced cytokine response.

F. tularensis DnaK induces activation of MAPKs and NF-␬B through the MyD88- or TRIF-dependent pathway To explain the reduced and differential production of cytokines in the absence of MyD88 or TRIF, we compared the activation of p38 MAPK and NF-␬B in DnaK-stimulated DC derived from MyD88 KO and TRIF-deficient mice. DnaK induced p38 MAPK phosphorylation and NF-␬B activation in WT DC (Fig. 3) and also in MyD88 KO and TRIFdeficient DC in a time-dependent manner (Fig. 8, A and B). However, the early activation of p38 MAPK induced by DnaK in WT DC was impaired in MyD88 KO DC, whereas the late activation was impaired in TRIF-deficient DC. DnaK also induced activation of ERK and JNK in MyD88 KO and TRIF-deficient DC. Similarly, DnaK-induced activation of NF-␬B was delayed in the absence of MyD88 and did not peak until 60 min. However, in the absence of TRIF, the early activation of NF-␬B was intact, peaked at 30 min, but gradually decreased at 60 and 120 min. These results clearly indicate that the inability of MyD88 KO or TRIFdeficient cells to induce an optimal, DnaK-dependent cytokine response cannot be attributed to the lack of activation of MAPKs and NF-␬B.

F. tularensis DnaK leads to IFN-␤ production and DC maturation through a TRIF-dependent pathway Type I IFN signaling is absolutely required for the up-regulation of costimulatory molecules, especially when TRIF is the sole route for the production of type I IFN, as in the case of LPS signaling through TLR4 [40, 67, 68]. Therefore, to investigate the importance of TRIF in DnaK-induced DC maturation, we next determined if DnaK could induce IFN-␤ production and up-regulation of costimulatory molecules in DC through a TRIF-dependent signaling pathway. DnaK induced IFN-␤ production in WT and MyD88 KO, but not in TLR4 KO and TRIF-deficient DC (Fig. 9A), thus indicating that TRIF is indispensable for DnaK-induced IFN-␤ production. As expected, DnaK-induced production of IFN-␤ correlated with the up-regulation of CD86 and CD40, and their expression was abrogated in the absence of TLR4, as well as in the absence of TRIF (Fig. 9B). The expression of CD86 and CD40 with MyD88 KO DC was comparable with that observed with WT DC (Fig. 9B). These results suggest that DnaK can induce IFN-␤ production and DC maturation via a TRIF-dependent signaling pathway, and MyD88 is not required for this response.

DISCUSSION In the present study, we investigated the role of TLR signaling in mediating the innate immune response induced by F. tularensis DnaK in murine DC and characterized the cytokines produced and cell signaling pathways involved. Our studies showed that DnaK activates DC through TLR4 (in a MD-2-

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1441

Fig. 8. DnaK leads to MAPK and NF-␬B activation through the MyD88- or TRIF-dependent signaling pathway. (A) DC (3⫻106 cells), derived from MyD88 KO and TRIF-deficient mice, were stimulated with DnaK (20 ␮g/ml) for the indicated times. Whole cell lysates were probed for the phosphorylated form of p38 MAPK by Western blotting. Total p38 served as a protein loading control. Results are representative of three independent experiments. (B) DC (3⫻106 cells), derived from MyD88 KO and TRIF-deficient mice, were stimulated with DnaK (20 ␮g/ml) for the indicated times. Equal amounts of nuclear extracts were analyzed for NF-␬B-binding activity using the TransAM NF-␬B p65 assay kit. *, **, and ***, Significant differences at P ⬍ 0.001, P ⬍ 0.01, and P ⬍ 0.05, respectively, compared with unstimulated cultures at 0 min. Results are expressed as the mean OD ⫾ SD of three representative experiments.

dependent manner), leading to proinflammatory cytokine production and phenotypic maturation of DC. These characteristics are important for the potential use of DnaK as a vaccine antigen, as they constitute some of the prerequisites for the development of antigen-specific T cell and antibody responses [32, 35]. We further showed that the intracellular adapters MyD88 and TRIF are differentially involved in mediating the production of cytokines and phenotypic maturation of DC. MyD88- and TRIF-mediated signaling is essential for an optimal, DnaK-induced cytokine response by DC. Several studies have characterized the mechanisms involved in the recognition of and the innate immune responses induced by HSPs derived from various prokaryotic and eukaryotic cells. For instance, Helicobacter pylori HSP60 was shown to induce the activation of MAPKs and NF-␬B in human monocytes, leading to IL-8 production via TLR2 [69]. Similarly, HSP65 and HSP70 derived from Mycobacterium tuberculosis was

shown to induce NF-␬〉 activation in human endothelial cells and lead to IL-6 and TNF-␣ production from murine macrophages [56]. Furthermore, although Aosai et al. [70] showed that HSP70 derived from Toxoplasma gondii induced phenotypic maturation and IL-12 production in DC through a TLR4dependent pathway, Bulut et al. [56] showed that mycobacterial HSP70 signals through TLR2 and TLR4. Thus, during an infection, bacterial HSPs could contribute to the induction of host immune responses to the pathogen. In the case of F. tularensis, in vitro studies have shown that the innate immune response mediated by this bacterium is via TLR2 [53, 71, 72], which is likely a result of immunodominant surface lipoproteins [73]. However, following infection, it is possible that cytosol-derived proteins, such as DnaK, “hidden” epitopes, or nucleic acids, are expressed that can activate an innate immune response via different TLRs. Further studies are necessary to address this possibility.

Fig. 9. DnaK induces IFN-␤ production and phenotypic maturation through a TRIF-dependent pathway. (A) DC (3⫻105 cells), derived from WT, TLR4, MyD88 KO, and TRIF-deficient mice, were stimulated with LPS (100 ng/ml) or DnaK (20 ␮g/ml) for 20 h. Culture supernatants were harvested, and levels of IFN-␤ were assessed by ELISA. *, Significant difference at P ⬍ 0.001 compared with WT cultures. Results are expressed as the mean ⫾ SD of triplicate cultures from one of three representative experiments. (B) Flow cytometric analysis of the surface expression of CD86 and CD40. *, **, and ***, Significant differences at P ⬍ 0.001, P ⬍ 0.01, and P ⬍ 0.05, respectively, compared with WT cultures. Results are expressed as the mean of the relative increase in MFI over unstimulated cultures ⫾ SD of three representative experiments.

1442

Journal of Leukocyte Biology Volume 84, December 2008

http://www.jleukbio.org

In addition to TLRs, other cell surface receptors such as CD14, CD40, CCR5, and CD91 have been reported to mediate recognition of HSPs derived from other pathogens [74, 75]. Under the conditions used in this study, our results demonstrate that DnaK is rendered incapable of activating DC in the absence of TLR4, which underlines the prime role played by this receptor in mediating DnaK recognition. However, it remains to be determined whether other receptors might mediate recognition of DnaK, independently or in the form of a yetunknown receptor complex, to effectively activate DC or other host cells. Most of the knowledge about the role of MyD88 and TRIF in mediating cytokine responses via TLR4 comes from the studies using LPS as a TLR4 agonist. Earlier studies have also addressed the role of MyD88 in mediating the stimulatory properties of HSP70 family members [57, 62, 69, 76], but only a limited number of reports have suggested the involvement of a MyD88-independent pathway [56, 70]. Thus, a detailed, comparative analysis of pro- and anti-inflammatory cytokines induced by HSP70 family members, including F. tularensis DnaK, in the absence of MyD88- or TRIF-dependent signaling is lacking. Our findings in this regard indicated that MyD88 and TRIF are indispensable for an optimal, DnaK-induced cytokine response from DC. This observed cooperation between the MyD88- and TRIF-dependent pathways to induce an optimal cytokine response is in agreement with several other findings about LPS-mediated signaling [40, 43, 46, 47, 77]. However, it should be noted that not all TLR4 ligands exhibit a similar requirement for a subsequent MyD88- and TRIFmediated signaling, and indeed, monophosphoryl lipid A has been shown to preferentially activate the TRIF-dependent signaling pathway to mediate its stimulatory effects [66]. Earlier studies using murine macrophages demonstrated that cytokine production induced by LPS was abolished completely in the absence of MyD88 or TRIF [44, 47, 78]. Although we did not observe a complete lack of cytokine production by DC in response to DnaK or even LPS, we noted that MyD88 was more important than TRIF in mediating the production of certain proinflammatory cytokines, especially IL-6 and IL-12 p40, in response to DnaK or LPS. Conversely, the production of TNF-␣ and IL-10 seems to be similarly affected in MyD88 KO and TRIF-deficient DC. There are several possible explanations for why our results differ from some of the earlier findings with LPS. First, differences in the cell type that were used in earlier reports may contribute to the observed differences, as it has been shown that macrophages and DC respond differently to LPS [40, 77]. Second, the lipid A of LPS activates TLR4 through an interaction with MD-2, and the degree of lipid A acylation affects TLR4 responsiveness [55]. Therefore, variations in the type of LPS or lipid A preparations used in earlier reports could be responsible for the observed differences. In the case of DnaK signaling through TLR4, the MyD88dependent pathway leads to an early-phase activation of MAPKs and NF-␬B, whereas the MyD88-independent (TRIFdependent) pathway induces a late-phase activation of MAPKs and NF-␬B. These results are in line with the findings of others regarding LPS signaling via TLR4 [44 – 47]. Earlier studies suggested that the TRIF-dependent pathway has slower kinetics than the MyD88-dependent pathway for the activation of

MAPKs and NF-␬B [40, 47, 64, 65]. However, Covert et al. [45] reported that MyD88- and TRIF-dependent pathways are likely to have similar activation kinetics. They further suggested that instead of directly activating MAPKs and NF-␬B, the TRIF-dependent pathway leads to the production and secretion of TNF-␣ in a NF-␬B-independent manner. The secreted TNF-␣ can signal in an autocrine or a paracrine manner, leading to the activation of MAPKs and NF-␬B [45, 79], which could explain the delayed activation of MAPKs and NF-␬B via the TRIF-dependent pathway. However, based on our studies, we cannot conclusively state if the activation of MAPKs and NF-␬B is direct or indirect following DnaKmediated signaling through TRIF. Moreover, studies are still required to explain if differences in the activation profiles of MAPKs and NF-␬B are capable of affecting cytokine production in MyD88- or TRIF-deficient DC. Combined null mutations for MyD88 and TRIF completely abrogated TLR4-mediated signal transduction [40, 47]. This indicates that both of these adapters are necessary for the induction of MAPKs and NF-␬B in response to TLR4 agonist activation. It can also be speculated that in the case of TLR4, only when a signal is propagated through the MyD88- and TRIF-dependent pathways may all of the necessary adapters be activated, which then might subsequently lead to an optimal induction of cytokines and other effector proteins. This is strongly supported by the work of Toshchakov et al. [80], in which they demonstrated that cell-permeable BB loop peptide inhibitors of the TRIF-related adaptor molecule blocked MyD88-dependent and MyD88-independent gene expression. Little is known about the importance of TRIF and MyD88 in mediating the up-regulation of costimulatory molecules induced by HSP70 family members. Our results showed that TRIF is indispensable for DnaK-induced up-regulation of costimulatory molecules, whereas MyD88 is not essential for this response. Up-regulation of costimulatory molecules via the TRIF-dependent signaling pathway might result from direct transcriptional activation of genes encoding costimulatory proteins or alternatively, from a secondary, cytokine-induced effect in an autocrine or paracrine manner. Lim et al. [81, 82] showed that LPS-induced CD80 and CD86 transcription in human monocytic cells is regulated by JNK. They also demonstrated a role of JNK-mediated activation of the IRF-7 transcription factor in the regulation of CD80 expression. Hoebe et al. [67] demonstrated that macrophages derived from mice deficient in type I IFNR do not show up-regulation of costimulatory molecules in response to LPS, indicating that type I IFN signaling is absolutely essential for the up-regulation of costimulatory molecules. IFN-␣, IFN-␤, and IFN-␥ can each induce the up-regulation of costimulatory molecules in WT and even in TRIF-deficient macrophages, indicating that up-regulation of costimulatory molecules is a TRIF-independent process [67]. However, as we observed in the present study that TRIF is indispensable for DnaK-induced up-regulation of CD40 and CD86, it would appear that TRIF is the sole route for type I IFN production, as IFN-␤ production was intact in the absence of MyD88 but was abolished completely in the absence of TLR4 or TRIF. Thus, DnaK-induced IFN-␤ can lead to DC maturation. In the present study, production of

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1443

IFN-␤ correlated with the up-regulation of costimulatory molecules, and these results are in line with previous findings [40, 67, 68]. DnaK induced a stronger up-regulation in the expression of CD86 compared with that of CD80. Whether CD80 and CD86 deliver unique, costimulatory signals that directly influence Th1 and Th2 development is an unresolved issue, as previous findings are controversial concerning the individual contributions of CD80 versus CD86 signals in different experimental models in vivo and in vitro [83, 84]. Therefore, immunization studies are required to investigate the type of immune response that is induced by DnaK and also the role of CD80 and CD86 in modulating these responses. DnaK also induced CD40 upregulation, and this is most important, as CD40 interacts with CD154 (CD40 ligand) expressed on activated T cells and NK cells. This interaction is key for the priming and expansion of antigen-specific T cells, up-regulation of costimulatory molecules and cytokine production, and also humoral immunity [85]. Lastly, MHC-II expression was only up-regulated slightly by DnaK, likely because a high basal expression level of MHC-II molecules was observed in unstimulated DC cultures, which was in agreement with previous studies of others [52]. As bacterial HSPs are predominantly available during microbial infections, they have been proposed as potential vaccine candidates [12–14, 16, 17]. However, HSPs represent a highly conserved group of proteins that are expressed in eukaryotic and prokaryotic cells. Immune responses to conserved regions shared between pathogen and self-HSPs could lead to autoimmune reaction in the host, which has been an issue of concern and debate. Some literature supports the use of HSPs as protective vaccine antigens against various pathogens, whereas others imply that HSPs derived from pathogens are involved in the initiation and progression of autoimmune diseases (reviewed in ref. [17]). As antibodies and T cells crossreactive for epitopes shared between pathogen HSPs and mammalian HSPs have been identified in healthy individuals, which is likely a result of the abundant presence of commensal organisms, it is argued that HSPs do not have a diseaseprovoking role in autoimmunity [17, 86, 87]. Moreover, a picture is emerging in that the expression of HSPs or immune reactivity to HSPs seems to be associated with the downregulation of various inflammatory responses and thus, protective effects, rather than with the induction of inflammation and the propagation of autoimmune diseases [88, 89]. To establish protective immunity against F. tularensis, it is of prime importance to identify bacterial components that can serve as T cell antigens during infection and also lead to effective memory responses. The fact that long-lived memory CD4 and CD8 T cells specific for HSPs, including DnaK, exist in individuals previously infected with F. tularensis LVS [7, 8] demonstrates that these bacterial components are being processed and presented to T cells in conjunction with MHC-II/I molecules, respectively. These studies indicate that HSPs can contribute to the overall host immune response and therefore, lend support to the concept that HSPs can be used as vaccine candidates that may help in mounting an early recall and thus, a protective response against a F. tularensis challenge. 1444

Journal of Leukocyte Biology Volume 84, December 2008

ACKNOWLEDGMENTS This work was supported by U.S. Public Health Service grant AI-56460 (to S. M. M.) and a subproject grant of the Middle Atlantic Regional Center of Excellence for Biodefense and Emerging Infectious Diseases (grants U54 AI-57168 and AI18797 to S. N. V) from the National Institute of Allergy and Infectious Diseases. We sincerely thank Matt Larson for his help in preparing the bacterial constructs and in DnaK purification.

REFERENCES 1. Dennis, D. T., Inglesby, T. V., Henderson, D. A., Bartlett, J. G., Ascher, M. S., Eitzen, E., Fine, A. D., Friedlander, A. M., Hauer, J., Layton, M., Lillibridge, S. R., McDade, J. E., Osterholm, M. T., O’Toole, T., Parker, G., Perl, T. M., Russell, P. K., Tonat, K. (2001) Tularemia as a biological weapon. Medical and public health management. JAMA 285, 2763–2773. 2. Ellis, J., Oyston, P. C. F., Green, M., Titball, R. W. (2002) Tularemia. Clin. Microbiol. Rev. 15, 631– 646. 3. Oyston, P. C., Sjostedt, A., Titball, R. W. (2004) Tularemia: bioterrorism defense renews interest in Francisella tularensis. Nat. Rev. Microbiol. 2, 967–978. 4. Tarnvik, A. (1989) Nature of protective immunity to Francisella tularensis. Rev. Infect. Dis. 11, 440 – 451. 5. Eigelsbach, H. T., Downs, C. M. (1961) Prophylactic effectiveness of live and killed tularemia vaccines. I. Production of vaccine and evaluation in the white mouse and guinea pig. J. Immunol. 87, 415– 425. 6. Elkins, K. L., Cowley, S. C., Bosio, C. M. (2003) Innate and adaptive immune responses to an intracellular bacterium, Francisella tularensis live vaccine strain. Microbes Infect. 5, 135–142. 7. Ericsson, M., Golovliov, I., Sandstrom, G., Tarnvik, A., Sjostedt, A. (1997) Characterization of the nucleotide sequence of the groE operon encoding heat shock proteins chaperone-60 and -10 of Francisella tularensis and determination of the T cell response to the proteins in individuals vaccinated with F. tularensis. Infect. Immun. 65, 1824 –1829. 8. Ericsson, M., Kroca, M., Johansson, T., Sjostedt, A., Tarnvik, A. (2001) Long-lasting recall response of CD4⫹ and CD8⫹ ␣␤ T cells, but not ␥␦ T cells, to heat shock proteins of Francisella tularensis. Scand. J. Infect. Dis. 33, 145–152. 9. Sandstrom, G., Tarnvik, A., Wolf-Watz, H. (1987) Immunospecific Tlymphocyte stimulation by membrane proteins from Francisella tularensis. J. Clin. Microbiol. 25, 641– 644. 10. Sjostedt, A., Sandstrom, G., Tarnvik, A., Jaurin, B. (1990) Nucleotide sequence and T-cell epitopes of a membrane protein of Francisella tularensis. J. Immunol. 145, 311–317. 11. Surcel, H. M. (1990) Diversity of Francisella tularensis antigens recognized by human T lymphocytes. Infect. Immun. 58, 2664 –2668. 12. Pockley, A. G. (2003) Heat shock proteins as regulators of the immune response. Lancet 362, 469 – 476. 13. Srivastava, P. (2002) Roles of heat-shock proteins in innate and adaptive immunity. Nat. Rev. Immunol. 2, 185–194. 14. Stewart, G. R., Young, D. B. (2004) Heat-shock proteins and the hostpathogen interaction during bacterial infection. Curr. Opin. Immunol. 16, 506 –510. 15. Ericsson, M., Tarnvik, A., Kuoppa, K., Sandstrom, G., Sjostedt, A. (1994) Increased synthesis of DnaK, GroEL, and GroES homologs by Francisella tularensis LVS in response to heat and hydrogen peroxide. Infect. Immun. 62, 178 –183. 16. Kaufmann, S. H. (1990) Heat shock proteins and the immune response. Immunol. Today 11, 129 –136. 17. Zu¨gel, U., Kaufmann, S. H. (1999) Role of heat shock proteins in protection from and pathogenesis of infectious diseases. Clin. Microbiol. Rev. 12, 19 –39. 18. Poltorak, A., He, X., Smirnova, I., Liu, M. Y., Van Huffel, C., Du, X., Birdwell, D., Alejos, E., Silva, M., Galanos, C., Freudenberg, M., Ricciardi-Castagnoli, P., Layton, B., Beutler, B. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085–2088. 19. Wyllie, D. H., Kiss-Toth, E., Visintin, A., Smith, S. C., Boussouf, S., Segal, D. M., Duff, G. W., Dower, S. K. (2000) Evidence for an accessory protein

http://www.jleukbio.org

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33. 34. 35. 36. 37. 38. 39. 40.

41.

42.

43.

44.

function for Toll-like receptor 1 in anti-bacterial responses. J. Immunol. 165, 7125–7132. Takeuchi, O., Kawai, T., Mu¨hlradt, P. F., Morr, M., Radolf, J. D., Zychlinsky, A., Takeda, K., Akira, S. (2001) Discrimination of bacterial lipoproteins by Toll-like receptor 6. Int. Immunol. 13, 933–940. Takeuchi, O., Hoshino, K., Kawai, T., Sanjo, H., Takada, H., Ogawa, T., Takeda, K., Akira, S. (1999) Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 11, 443– 451. Yang, R. B., Mark, M. R., Gray, A., Huang, A., Xie, M. H., Zhang, M., Goddard, A., Wood, W. I., Gurney, A. L., Godowski, P. J. (1998) Toll-like receptor-2 mediates lipopolysaccharide-induced cellular signaling. Nature 395, 284 –288. Aliprantis, A. O., Yang, R. B., Mark, M. R., Suggett, S., Devaux, B., Radolf, J. D., Klimpel, G. R., Godowski, P., Zychlinsky, A. (1999) Cell activation and apoptosis by bacterial lipoproteins through Toll-like receptor-2. Science 285, 736 –739. Hayashi, F., Smith, K. D., Ozinsky, A., Hawn, T. R., Yi, E. C., Goodlett, D. R., Eng, J. K., Akira, S., Underhill, D. M., Aderem, A. (2001) The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410, 1099 –1103. Alexopoulou, L., Holt, A. C., Medzhitov, R., Flavell, R. A. (2001) Recognition of double-stranded RNA and activation of NF-␬B by Toll-like receptor 3. Nature 413, 732–738. Hemmi, H., Takeuchi, O., Kawai, T., Kaisho, T., Sato, S., Sanjo, H., Matsumoto, M., Hoshino, K., Wagner, H., Takeda, K., Akira, S. (2000) A Toll-like receptor recognizes bacterial DNA. Nature 408, 740 –745. Chaudhary, P. M., Ferguson, C., Nguyen, V., Nguyen, O., Massa, H. F., Eby, M., Jasmin, A., Trask, B. J., Hood, L., Nelson, P. S. (1998) Cloning and characterization of two Toll/interleukin-1 receptor-like genes TIL3 and TIL4: evidence for a multi-gene receptor family in humans. Blood 91, 4020 – 4027. Chuang, T. H., Ulevitch, R. J. (2000) Cloning and characterization of a subfamily of human Toll-like receptors: hTLR7, hTLR8 and hTLR9. Eur. Cytokine Netw. 11, 372–378. Medzhitov, R., Preston-Hurlburt, P., Janeway, C. A. (1997) A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388, 394 –397. Sebastiani, G., Leveque, G., Lariviere, L., Laroche, L., Skamene, E., Gros, P., Malo, D. (2000) Cloning and characterization of the murine Toll-like receptor 5 (Tlr5) gene: sequence and mRNA expression studies in Salmonella-susceptible MOLF/Ei mice. Genomics 64, 230 –240. Takeuchi, O., Kawai, T., Sanjo, H., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Takeda, K., Akira, S. (1999) TLR6: a novel member of an expanding Toll-like receptor family. Gene 231, 59 – 65. Reis e Sousa, C., Sher, A., Kaye, P. (1999) The role of dendritic cells in the induction and regulation of immunity to microbial infection. Curr. Opin. Immunol. 11, 392–399. Pasare, C., Medzhitov, R. (2004) Toll-like receptors: linking innate and adaptive immunity. Microbes Infect. 6, 1382–1387. Medzhitov, R., Janeway Jr., C. (1997) Innate immunity: impact on the adaptive immune response. Curr. Opin. Immunol. 9, 4 –9. Banchereau, J., Steinman, R. M. (1998) Dendritic cells and the control of immunity. Nature 392, 245–252. Akira, S., Takeda, K. (2004) Toll-like receptor signaling. Nat. Rev. Immunol. 4, 499 –511. Janeway Jr., C., Medzhitov, R. (2002) Innate immune recognition. Annu. Rev. Immunol. 20, 197–216. Takeda, K., Akira, S. (2005) Toll-like receptors in innate immunity. Int. Immunol. 17, 1–14. Takeda, K., Kaisho, T., Akira, S. (2003) Toll-like receptors. Annu. Rev. Immunol. 21, 335–376. Hoebe, K., Du, X., Georgel, P., Janssen, E., Tabeta, K., Kim, S. O., Goode, J., Lin, P., Mann, N., Mudd, S., Crozat, K., Sovath, S., Han, J., Beutler, B. (2003) Identification of Lps2 as a key transducer of MyD88-independent TIR signaling. Nature 424, 743–748. Medzhitov, R., Preston-Hurlburt, P., Kopp, E., Stadlen, A., Chen, C., Ghosh, S., Janeway Jr., C. (1998) MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol. Cell 2, 253–258. Oshiumi, H., Matsumoto, M., Funami, K., Akazawa, T., Seya, T. (2003) TICAM-1, an adaptor molecule that participates in Toll-like receptor 3-mediated interferon-␤ induction. Nat. Immunol. 4, 161–167. Kaisho, T., Takeuchi, O., Kawai, T., Hoshino, K., Akira, S. (2001) Endotoxin-induced maturation of MyD88-deficient dendritic cells. J. Immunol. 166, 5688 –5694. Kawai, T., Adachi, O., Ogawa, T., Takeda, K., Akira, S. (1999) Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 11, 115–122.

45. Covert, M. W., Leung, T. H., Gaston, J. E., Baltimore, D. (2005) Achieving stability of lipopolysaccharide-induced NF-␬B activation. Science 309, 1854 –1857. 46. Kawai, T., Takeuchi, O., Fujita, T., Inoue, J., Muhlradt, P. F., Sato, S., Hoshino, K., Akira, S. (2001) Lipopolysaccharide stimulates the MyD88independent pathway and results in activation of IFN-regulatory factor 3 and the expression of a subset of lipopolysaccharide-inducible genes. J. Immunol. 167, 5887–5894. 47. Yamamoto, M., Sato, S., Hemmi, H., Hoshino, K., Kaisho, T., Sanjo, H., Takeuchi, O., Sugiyama, M., Okabe, M., Takeda, K., Akira, S. (2003) Role of adaptor TRIF in the MyD88-independent Toll-like receptor signaling pathway. Science 301, 640 – 643. 48. Kaisho, T., Hoshino, K., Iwabe, T., Takeuchi, O., Yasui, T., Akira, S. (2002) Endotoxin can induce MyD88-deficient dendritic cells to support Th2 cell differentiation. Int. Immunol. 14, 695–700. 49. Muraille, E., De Trez, C., Brait, M., De Baetselier, P., Leo, O., Carlier, Y. (2003) Genetically resistant mice lacking MyD88-adapter protein display a high susceptibility to Leishmania major infection associated with a polarized Th2 response. J. Immunol. 170, 4237– 4241. 50. Piggott, D. A., Eisenbarth, S. C., Xu, L., Constant, S. L., Huleatt, J. W., Herrick, C. A., Bottomly, K. (2005) MyD88-dependent induction of allergic Th2 responses to intranasal antigen. J. Clin. Invest. 115, 459 – 467. 51. McIntire, F. C., Sievert, H. W., Barlow, G. H., Finley, R. A., Lee, A. Y. (1967) Chemical, physical, biological properties of a lipopolysaccharide from Escherichia coli K-235. Biochemistry 6, 2363–2372. 52. Lutz, M. B., Kukutsch, N., Ogilvie, A. L., Rossner, S., Koch, F., Romani, N., Schuler, G. (1999) An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223, 77–92. 53. Katz, J., Zhang, P., Martin, M., Vogel, S. N., Michalek, S. M. (2006) Toll-like receptor 2 is required for inflammatory responses to Francisella tularensis LVS. Infect. Immun. 74, 2809 –2816. 54. Rallabhandi, P., Bell, J., Boukhvalova, M. S., Medvedev, A., Lorenz, E., Arditi, M., Hemming, V. G., Blanco, J. C., Segal, D. M., Vogel, S. N. (2006) Analysis of TLR4 polymorphic variants: new insights into TLR4/ MD-2/CD14 stoichiometry, structure, and signaling. J. Immunol. 177, 322–332. 55. Rallabhandi, P., Awomoyi, A., Thomas, K. E., Phalipon, A., Fujimoto, Y., Fukase, K., Kusumoto, S., Qureshi, N., Sztein, M. B., Vogel, S. N. (2008) Differential activation of human TLR4 by Escherichia coli and Shigella flexneri 2a lipopolysaccharide: combined effects of lipid A acylation state and TLR4 polymorphisms on signaling. J. Immunol. 180, 1139 –1147. 56. Bulut, Y., Michelsen, K. S., Hayrapetian, L., Naiki, Y., Spallek, R., Singh, M., Arditi, M. (2005) Mycobacterium tuberculosis heat shock proteins use diverse Toll-like receptor pathways to activate pro-inflammatory signals. J. Biol. Chem. 280, 20961–20967. 57. Cohen-Sfady, M., Nussbaum, G., Pevsner-Fischer, M., Mor, F., Carmi, P., Zanin-Zhorov, A., Lider, O., Cohen, I. R. (2005) Heat shock protein 60 activates B cells via the TLR4-MyD88 pathway. J. Immunol. 175, 3594 – 3602. 58. Jung, S. B., Yang, C. S., Lee, J. S., Shin, A. R., Jung, S. S., Son, J. W., Harding, C. V., Kim, H. J., Park, J. K., Paik, T. H., Song, C. H., Jo, E. K. (2006) The mycobacterial 38-kilodalton glycolipoprotein antigen activates the mitogen-activated protein kinase pathway and release of proinflammatory cytokines through Toll-like receptors 2 and 4 in human monocytes. Infect. Immun. 74, 2686 –2696. 59. Nagai, Y., Akashi, S., Nagafuku, M., Ogata, M., Iwakura, Y., Akira, S., Kitamura, T., Kosugi, A., Kimoto, M., Miyake, K. (2002) Essential role of MD-2 in LPS responsiveness and TLR4 distribution. Nat. Immunol. 3, 667– 672. 60. Shimazu, R., Akashi, S., Ogata, H., Nagai, Y., Fukudome, K., Miyake, K., Kimoto, M. (1999) MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J. Exp. Med. 189, 1777–1782. 61. Medvedev, A. E., Vogel, S. N. (2003) Overexpression of CD14, TLR4, and MD-2 in HEK 293T cells does not prevent induction of in vitro endotoxin tolerance. J. Endotoxin Res. 9, 60 – 64. 62. Bulut, Y., Faure, E., Thomas, L., Karahashi, H., Michelsen, K. S., Equils, O., Morrison, S. G., Morrison, R. P., Arditi, M. (2002) Chlamydial heat shock protein 60 activates macrophages and endothelial cells through Toll-like receptor 4 and MD2 in a MyD88-dependent pathway. J. Immunol. 168, 1435–1440. 63. Kim, H. M., Park, B. S., Kim, J. I., Kim, S. E., Lee, J., Oh, S. C., Enkhbayar, P., Matsushima, N., Lee, H., Yoo, O. J., Lee, J. O. (2007) Crystal structure of the TLR4-MD-2 complex with bound endotoxin antagonist Eritoran. Cell 130, 906 –917. 64. Jiang, Z., Mak, T. W., Sen, G., Li, X. (2004) Toll-like receptor 3-mediated activation of NF-␬B and IRF3 diverges at Toll-IL-1 receptor domain-

Ashtekar et al. TLR4-mediated responses to F. tularensis DnaK

1445

65.

66. 67.

68.

69.

70.

71.

72. 73.

74. 75.

76.

containing adapter inducing IFN-␤. Proc. Natl. Acad. Sci. USA 101, 3533–3538. Sato, S., Sugiyama, M., Yamamoto, M., Watanabe, Y., Kawai, T., Takeda, K., Akira, S. (2003) Toll/IL-1 receptor domain-containing adaptor inducing IFN-␤ (TRIF) associates with TNF receptor-associated factor 6 and TANK-binding kinase 1, and activates two distinct transcription factors, NF-␬B and IFN-regulatory factor-3, in the Toll-like receptor signaling. J. Immunol. 171, 4304 – 4310. Mata-Haro, V., Cekic, C., Martin, M., Chilton, P. M., Casella, C. R., Mitchell, T. C. (2007) The vaccine adjuvant monophosphoryl lipid A as a TRIF-biased agonist of TLR4. Science 316, 1628 –1632. Hoebe, K., Janssen, E. M., Kim, S. O., Alexopoulou, L., Flavell, R. A., Han, J., Beutler, B. (2003) Upregulation of costimulatory molecules induced by lipopolysaccharide and double-stranded RNA occurs by Trifdependent and Trif-independent pathways. Nat. Immunol. 4, 1223–1229. Yamamoto, M., Sato, S., Mori, K., Hoshino, K., Takeuchi, O., Takeda, K., Akira, S. (2002) Cutting edge: a novel Toll/IL-1 receptor domain-containing adapter that preferentially activates the IFN-␤ promoter in the Tolllike receptor signaling. J. Immunol. 169, 6668 – 6672. Zhao, Y., Yokota, K., Ayada, K., Yamamoto, Y., Okada, T., Shen, L., Oguma, K. (2007) Helicobacter pylori heat-shock protein 60 induces interleukin-8 via a Toll-like receptor (TLR) 2 and mitogen-activated protein (MAP) kinase pathway in human monocytes. J. Med. Microbiol. 56, 154 –164. Aosai, F., Rodriguez Pena, M. S., Mun, H. S., Fang, H., Mitsunaga, T., Norose, K., Kang, H. K., Bae, Y. S., Yano, A. (2006) Toxoplasma gondiiderived heat shock protein 70 stimulates maturation of murine bone marrow-derived dendritic cells via Toll-like receptor 4. Cell Stress Chaperones 11, 13–22. Cole, L. E., Shirey, K. A., Barry, E., Santiago, A., Rallabhandi, P., Elkins, K. L., Puche, A. C., Michalek, S. M., Vogel, S. N. (2007) Toll-like receptor 2-mediated signaling requirements for Francisella tularensis live vaccine strain infection of murine macrophages. Infect. Immun. 75, 4127– 4137. Li, H., Nookala, S., Bina, X. R., Bina, J. E., Re, F. (2006) Innate immune response to Francisella tularensis is mediated by TLR2 and caspase-1 activation. J. Leukoc. Biol. 80, 766 –773. Thakran, S., Li, H., Lavine, C. L., Miller, M. A., Bina, J. E., Bina, X. R., Re, F. (2008) Identification of Francisella tularensis lipoproteins that stimulate the Toll-like receptor (TLR) 2/TLR1 heterodimer. J. Biol. Chem. 283, 3751–3760. Binder, R. J., Vatner, R., Srivastava, P. (2004) The heat-shock protein receptors: some answers and more questions. Tissue Antigens 64, 442– 451. Floto, R. A., MacAry, P. A., Boname, J. M., Mien, T. S., Kampmann, B., Hair, J. R., Huey, O. S., Houben, E. N., Pieters, J., Day, C., Oehlmann, W., Singh, M., Smith, K. G., Lehner, P. J. (2006) Dendritic cell stimulation by mycobacterial Hsp70 is mediated through CCR5. Science 314, 454 – 458. Vabulas, R. M., Ahmad-Nejad, P., da Costa, C., Miethke, T., Kirschning, C. J., Hacker, H., Wagner, H. (2001) Endocytosed HSP60s use Toll-like

1446

Journal of Leukocyte Biology Volume 84, December 2008

77. 78.

79. 80.

81.

82.

83. 84.

85. 86.

87.

88. 89.

receptor 2 (TLR2) and TLR4 to activate the Toll/interleukin-1 receptor signaling pathway in innate immune cells. J. Biol. Chem. 276, 31332– 31339. Weighardt, H., Jusek, G., Mages, J., Lang, R., Hoebe, K., Beutler, B., Holzmann, B. (2004) Identification of a TLR4- and TRIF-dependent activation program of dendritic cells. Eur. J. Immunol. 34, 558 –564. Jeyaseelan, S., Young, S. K., Fessler, M. B., Liu, Y., Malcolm, K. C., Yamamoto, M., Akira, S., Worthen, G. S. (2007) Toll/IL-1 receptor domain-containing adaptor inducing IFN-␤ (TRIF)-mediated signaling contributes to innate immune responses in the lung during Escherichia coli pneumonia. J. Immunol. 178, 3153–3160. Baud, V., Karin, M. (2001) Signal transduction by tumor necrosis factor and its relatives. Trends Cell Biol. 11, 372–377. Toshchakov, V. U., Basu, S., Fenton, M. J., Vogel, S. N. (2005) Differential involvement of BB loops of Toll-IL-1 resistance (TIR) domain-containing adapter proteins in TLR4- versus TLR2-mediated signal transduction. J. Immunol. 175, 494 –500. Lim, W., Gee, K., Mishra, S., Kumar, A. (2005) Regulation of B7.1 costimulatory molecule is mediated by the IFN regulatory factor-7 through the activation of JNK in lipopolysaccharide-stimulated human monocytic cells. J. Immunol. 175, 5690 –5700. Lim, W., Ma, W., Gee, K., Aucoin, S., Nandan, D., Diaz-Mitoma, F., Kozlowski, M., Kumar, A. (2002) Distinct role of p38 and c-Jun Nterminal kinases in IL-10-dependent and IL-10-independent regulation of the costimulatory molecule B7.2 in lipopolysaccharide-stimulated human monocytic cells. J. Immunol. 168, 1759 –1769. Salomon, B., Bluestone, J. A. (2001) Complexities of CD28/B7: CTLA-4 costimulatory pathways in autoimmunity and transplantation. Annu. Rev. Immunol. 19, 225–252. Zhang, P., Martin, M., Yang, Q. B., Michalek, S. M., Katz, J. (2004) Role of B7 costimulatory molecules in immune responses and T-helper cell differentiation in response to recombinant HagB from Porphyromonas gingivalis. Infect. Immun. 72, 637– 644. Grewal, I. S., Flavell, R. A. (1998) CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16, 111–135. Lamb, J. R., Bal, V., Mendez-Samperio, P., Mehlert, A., So, A., Rothbard, J., Jindal, S., Young, R. A., Young, D. B. (1989) Stress proteins may provide a link between the immune response to infection and autoimmunity. Int. Immunol. 1, 191–196. Munk, M. E., Schoel, B., Modrow, S., Karr, R. W., Young, R. A., Kaufmann, S. H. (1989) T lymphocytes from healthy individuals with specificity to self-epitopes shared by the mycobacterial and human 65kilodalton heat shock protein. J. Immunol. 143, 2844 –2849. Van Eden, W., van der Zee, R., Prakken, B. (2005) Heat-shock proteins induce T-cell regulation of chronic inflammation. Nat. Rev. Immunol. 5, 318 –330. Van Eden, W., Wick, G., Albani, S., Cohen, I. (2007) Stress, heat shock proteins, and autoimmunity: how immune responses to heat shock proteins are to be used for the control of chronic inflammatory diseases. Ann. N. Y. Acad. Sci. 1113, 217–237.

http://www.jleukbio.org