Molecular chaperoning by glucose-regulated protein 170 in the extracellular milieu promotes macrophage-mediated pathogen sensing and innate immunity

Abstract

Recognition of pathogen-associated molecular patterns by innate immune receptors is essential for host defense responses. Although extracellular stress proteins are considered as indicators of the stressful conditions (e.g., infection or cell injury), the exact roles of these molecules in the extracellular milieu remain less defined. We found that glucose-regulated protein 170 (Grp170), the largest stress protein and molecular chaperone, is highly efficient in binding CpG oligodeoxynucleotides (CpG-ODN), the microbial DNA mimetic sensed by toll-like receptor 9 (TLR9). Extracellular Grp170 markedly potentiates the endocytosis and internalization of CpG-ODN by mouse bone marrow-derived macrophages and directly interacts with endosomal TLR9 on cell entry. These molecular collaborations result in the synergistic activation of the MyD88-dependent signaling and enhanced production of proinflammatory cytokines and nitric oxide in mouse primary macrophages as well as human THP-1 monocyte-derived macrophages, suggesting that Grp170 released from injured cells facilitates the sensing of pathogen-associated “danger” signals by intracellular receptors. This CpG-ODN chaperone complex-promoted innate immunity confers increased resistance in mice to infection of Listeria monocytogenes compared with CpG-ODN treatment alone. Our studies reveal a previously unrecognized attribute of Grp170 as a superior DNA-binding chaperone capable of amplifying TLR9 activation on pathogen recognition, which provides a conceptual advance in understanding the dynamics of ancient chaperoning functions inside and outside the cell.—Zuo, D., Yu, X., Guo, C., Yi, H., Chen, X., Conrad, D. H., Guo, T. L., Chen, Z., Fisher, P. B., Subjeck, J. R., Wang, X.-Y. Molecular chaperoning by glucose-regulated protein 170 in the extracellular milieu promotes macrophage-mediated pathogen sensing and innate immunity.

Pattern recognition receptors (PRRs) in innate immune cells, e.g., macrophages and dendritic cells, are specialized for recognizing conserved motifs of microbial origins or pathogen-associated molecular patterns (PAMPs) and play crucial roles in innate immunity and host response to invading microorganisms (1, 2). Engagement of PRRs by their cognate ligands triggers activation of protein kinases and transcription factors involved in inflammation and immunity (3). Macrophages represent a major component of the innate immune system and play important roles in immune responses against microbial pathogens by producing tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-12, and nitric oxide (NO). Among biochemically diverse pathogen molecules, microbial CpG-DNA or its synthetic analog CpG-oligonucleotides (CpG-ODN) is a ligand for toll-like receptor 9 (TLR9; Refs. 4, 5). It is hypothesized that endocytosed CpG-DNAs or CpG-ODNs interaction with TLR9 results in MyD88-dependent signaling activation, production of inflammatory cytokines, induction of the Th1 response, and elimination of microbial pathogens (4, 6).

All living organisms respond to stressful conditions, such as heat shock, oxidative stress, or infection, by increasing the expression of specific sets of protective proteins that have been commonly referred to as stress proteins or heat shock proteins (HSPs). Many of these highly conserved proteins function as molecular chaperones within the cell to assist synthesis, folding, translocation, and degradation of intracellular proteins (7). Although HSPs are classically viewed as intracellular molecules, emerging evidence suggests that extracellular release of HSPs occurs during cell death or through physiological secretion mechanisms (8). Release of HSPs to the extracellular milieu may be triggered by physical trauma-caused cellular injury (9), pathogen infection (10), or exposure to certain stimuli (11, 12). Indeed, chaperoning of antigens by extracellular HSPs has been shown to facilitate antigen cross-presentation and T-cell activation (13). HSPs along with several other intracellular molecules have been proposed to serve as “alarmins” or damage-associated molecular patterns (DAMPs) to alert the host immune system at sites of stress or tissue injury (14–16).

Large stress proteins (LSPs), e.g., Hsp110 and Grp170, are highly evolutionarily diverged relatives of the Hsp70 family and highly abundant, conserved components of stress responses (17). Although the LSPs were identified almost 30 yr ago, they have received a fraction of the attention paid to their molecular cousins (18). As opposed to classical HSPs, LSPs are much more effective in stabilizing heat-denatured protein substrates (19, 20), representing the most powerful “holders” of client proteins in the cell. Indeed, our studies (20–22) demonstrated that in vitro reconstituted chaperone complexes of LSPs and protein antigens, which are believed to resemble the natural intracellular chaperone complexes, efficiently generated antigen-specific adaptive immunity. Although molecular chaperoning of protein or polypeptide substrates has been well established, the interaction between stress protein and deoxyribonucleic acid (DNA) remains undefined.

In this study, we show that Grp170 is a novel, highly efficient DNA-binding chaperone that forms a complex with CpG-ODN, the microbial DNA mimetic. Extracellular Grp170-facilitated delivery of CpG-ODN results in profoundly enhanced activation of MyD88-dependent inflammatory responses in macrophages. Furthermore, innate immunity potentiated by the CpG-ODN chaperone complex leads to increased resistance of mice to Listeria monocytogenes infection. The novel features of this ancient molecular chaperone elucidated in the present study provide a link between extracellular chaperone molecules and intracellular innate immune sensors, reinforcing the idea that molecular chaperoning underlies the essential functions of stress proteins.

MATERIALS AND METHODS

Mice

Wild-type C57BL/6 mice were purchased from National Cancer Institute (Bethesda, MD, USA), and B6C3F1 (C57BL/6N×C3H/HeN) mice were purchased from Taconic (Germantown, NY, USA). All animal procedures used in these studies complied with guidelines of Institutional Animal Care and Use Committee of Virginia Commonwealth University.

Cell culture

Bone marrow-derived macrophages (BMMs) were collected after 7 days of bone marrow cells cultured in DMEM supplemented with 10% FBS, 100 U/ml of penicillin, 100 μg/ml of streptomycin, and 30% conditioned medium derived from the culture of L929 cells producing macrophage colony-stimulating factor. Immortalized wild-type and MyD88^−/− macrophage lines, kindly provided by Dr. Douglas Golenbock (University of Massachusetts, Waltham, MA, USA), were also cultured in DMEM supplemented with 30% L929 cell-conditioned medium. RAW264.7 cells were grown in DMEM supplemented with 10% FBS, 100 U/ml of penicillin, and 100 μg/ml of streptomycin in a humidified 5% CO₂ atmosphere. Human THP-1 cells were cultured in RPMI 1640 medium supplemented with 2 mM l-glutamine, 100 U/ml of penicillin, 100 μg/ml of streptomycin,, and 10% FBS.

Protein and oligodeoxynucleotides

Mouse recombinant Hsp70, Hsp110, Grp170, and Grp170 deletion mutants, including Grp170 lacking ATP-binding domain (BLH; 430–999 amino acids), Grp170 lacking ATP-binding domain and β-sheet domain (LH; 600–999 amino acids), H domain only (H; 715–999 amino acids), Grp170 lacking only the H domain (AB; 1–600 amino acids), and ATP-binding domain and β-sheet domain (ABL; 1–715 amino acids), were expressed in Sf21 insect cells using the BacPAK baculovirous expression system (Clontech, Palo Alto, CA, USA) as described previously (20). Proteins were purified using nickel-chelating resins (Qiagen, Valencia, CA, USA) and dialyzed against endotoxin-free PBS. Endotoxin levels in the protein preparations were undetected by a Limulus Amebocyte lysate kit (Biowhittaker, Walkersville, MD, USA). Endotoxin-free type B CpG-ODN 1826 (5′-TCCATGACGTTCCGACGTT-3′) and GpC-ODN 1982 (5′-TCCATGAGCTTCGCAGCTT-3′) were synthesized with a phosphorothioate backbone (Invitrogen, Carlsbad, CA, USA). FITC or biotin-labeled CpG-ODN1826 and CpG-ODN2006 (5′-TCGTCGTTTTGTCGTTTTGTCGTT-3′) were purchased from Invivogen (San Diego, CA, USA).

Direct binding of Grp170 to CpG-ODN

Chaperone proteins (2 μg) were incubated at room temperature with different concentrations of biotin-labeled CpG-ODN for 30 min. For competition studies, Grp170 was incubated with biotin-CpG-ODN in the presence of increasing concentrations of unlabeled CpG-ODN1826, GpC-ODN1982, peptides, or proteins. CpG-ODN chaperone complexes were separated on 10% SDS-PAGE under nonreducing conditions and probed with horseradish peroxidase-avidin after transferring to nitrocellulose membranes (Bio-Rad, Hercules, CA, USA). Reactions were visualized using enhanced chemiluminescence reagents (Thermo Fisher Scientific, Fremont, CA, USA) for immunoblotting. For ELISA assays, microtiter wells (Nunc, Kamstrup, Denmark) were coated overnight at 4°C with 10 μg/ml of Grp170 or BSA. Plates were incubated at room temperature for 1 h with different concentrations of biotin-CpG-ODN. The levels of bound CpG-ODN were determined using colorimetric assays after incubation with horseradish peroxidase-conjugated streptavidin.

Surface plasmon resonance (SPR) analysis

SPR analysis of binding affinity of chaperones for CpG-ODN was determined using the BIAcore T100 system (GE Healthcare, Piscataway, NJ, USA). Biotin-labeled ODN1826 were immobilized on streptavidin-coated SA biosensor chip. HBSP (10 mM HEPES, 3 mM EDTA, 150 mM NaCl, and 0.005% P20, pH 7.4) was used as the running buffer in all binding studies with a flow rate of 30 μl/min. Different concentrations of chaperone molecules were passed over the control (no CpG-ODN immobilized) at room temperature and test flow cells by the use of multichannel-flow option. The surface of sensor chip was fully regenerated using 50 mM NaOH containing 0.5% SDS. The interaction kinetics was analyzed using the BIAEVALUTION 3.2 software (GE Healthcare) by fitting to 1:1 Langmuir binding model.

Cell stimulation

BMMs (2×10⁵/well) in 6-well plates were incubated with CpG-ODN1826 (1 μg/ml) or GpC-ODN1982 in the presence or absence of Grp170 (50 μg/ml). Grp170 protein was preincubated with ODNs for 30 min at room temperature for complex formation before incubation with BMMs. For stimulation of human macrophages, THP-1 cells were differentiated with 10 nM of PMA for 24 h and then incubated with CpG-ODN2006 (5 μg/ml) with or without Grp170. In some cases, cells were pretreated with dynasore (80 μM; Sigma-Aldrich, St. Louis, MO, USA) or chloroquine (50 μM) at 37°C or fucoidan (100 μg/ml) at 4°C for 30 min before stimulation with Grp170-CpG-ODN.

Quantitative RT-PCR

RNAs were extracted from cells using Trizol reagent (Invitrogen) 4 h after Grp170-CpG-ODN stimulation. RT was performed using RevertAid First Strand cDNA Synthesis Kit (Fermentas, Glen Burnie, MD, USA), followed by PCR assays using the PCR primers as follows, mouse inducible nitric oxide synthase (iNOS), sense primer: 5′-AGCATCACCCCTGTGTTCCACC-3′, antisense primer: 5′-TGGGACAGTCTCCATTCCCA-3′; β-actin, sense primer: 5′-GTCCCTCACCCTCCCAAAAG-3′, antisense primer: 5′-GCTGCCTCAACACCTCAACC-3′. For real-time PCR, all primers and reagents were purchased from Applied Biosystems (Foster City, CA, USA), and the reaction was run on the 7900HT Fast Real-time PCR System (Applied Biosystems). Gene expression was quantified relative to the expression of β-actin and normalized to that measured in control by standard 2^(−ΔΔCT) calculation.

Cytokine and NO assays

Culture supernatants were collected at 24 h after stimulation and cleared of debris by centrifugation. The production of NO was determined by measuring the quantity of nitrite in the supernatants using the Griess method (Invitrogen). TNF-α, IL-6, and IL-12p40 expression levels in the serum or culture supernatants were assayed with ELISA kits (Biolegend, San Diego, CA, USA) according to the manufacturer's instructions. IFN-γ and IL-10 levels in the serum were measured with ELISA kits from eBiosciences (San Diego, CA, USA).

Immunoprecipitation and immunoblotting

Protein lysates prepared using RIPA buffer (50 mM Tris, 150 mM NaCl, and 1% Nonidet P-40, pH 7.4.) were separated by SDS-PAGE and transferred onto nitrocellulose membranes. Membranes were blocked in TBS containing 0.05% Tween-20 and 5% nonfat milk or 5% BSA (for detecting phosphoproteins) for 1 h at room temperature and incubated overnight at 4°C with the appropriate primary antibodies against iNOS (BD Biosciences, San Jose, CA, USA), phospho-ERK1/2, phospho-JNK, phospho-p38, phospho-IκBα, ERK, JNK, p38, IκBα (Cell Signaling Technology, Danvers, MA, USA), TLR9 (clone 26C593.2; Imgenex, San Diego, CA), Grp94/gp96 (9G10; Stressgen, Victoria, BC, Canada), Grp78/BiP (SPA-826; Stressgen), anti-penta His antibody (Qiagen), or β-actin (AC-74; Sigma-Aldrich). For immunopreciptation, whole cell lysates were incubated with antibodies (1 μg) and protein A/G agarose (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at 4°C overnight. Eluted immunoprecipitates were resolved on SDS-PAGE and examined for association of proteins of interest using specific antibodies.

Immunofluorescence and confocal microscopy

For analysis of colocalization of Grp170 and TLR9, BMMs plated on glass cover slips were treated with FITC-labeled grp170 (100 μg/ml) for 1 h. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100 in PBS for 10 min, and stained with antibodies against TLR9 (1:1000) or Rab5 (1:200). All the images were acquired with a ×63 oil-immersion objective on a Leica TCS SP2 spectral confocal microscope (Leica Microsystems, Allendale, NJ, USA). Representative images were processed using Photoshop (Adobe Systems, San Jose, CA, USA), and the adjustments of brightness and contrast were applied to the whole images.

Endocytosis of CpG-ODN

Cells were incubated with FITC-CpG-ODN in the presence or absence of Grp170 for 1 h. Cells were extensively washed, trypsinized, and fixed in 2% paraformaldehyde. Data were immediately acquired using a FACSCalibur (Becton-Dickinson, Mountain View, CA, USA) and analyzed using Flowjo software 7.5.5 (Tree Star, Ashland, OR, USA). Alternatively, cells were fixed and analyzed using fluorescence microscopy.

CpG-ODN treatment in vivo

C57BL/6 mice were injected intraperitoneally with PBS, CpG-ODN (20 μg), Grp170 (200 μg), or a combination of both. Serum was collected 2 h later for cytokine assays. Cells were collected by peritoneal lavage 72 h after injection and were cultured in the presence of heat-killed L. monocytogenes for 48 h. The supernatants were harvested for ELISA assays.

Listeria infection in vivo

Sex- and age-matched B6C3F1 mice were injected intraperitoneally with 20 μg of CpG-ODN with or without 200 μg of Grp170 protein, followed by inoculation of 1 × 10⁵ colony-forming units (CFU) of L. monocytogenes intravenously 3 days later. On day 3 following infection, spleens and livers were removed from mice and homogenized in normal saline. Bacteria CFUs were determined by plating the homogenates with serial 10-fold dilutions on brain-heart infusion agar plates (BD Biosciences). The bacterial colonies were counted after incubating the plates at 37°C overnight.

Statistical analysis

Data are presented as means ± sd. Normal distribution of data was indicated by “goodness of fit” test using JMP9 software. Student's t test was used for comparisons between 2 groups. One-way ANOVA followed by Bonferroni adjustment was used for comparisons among multiple groups. A probability of P < 0.05 was used to define this significance.

RESULTS

Grp170 is a CpG-ODN-binding chaperone

Considering the extracellular release of stress proteins on cell injury, we investigated the potential interaction of the highly efficient large protein chaperone Grp170 with CpG-ODN (i.e., ODN1826), the microbial DNA mimetic. Freezing- and thawing-induced cell injury was adopted to mimic the cell death that often occurs at the sites of infection. After incubation of cell lysates with biotinylated CpG-ODN (biotin-CpG-ODN), CpG-ODN was recovered using streptavidin agarose. CpG-ODN was associated with Grp170 and Hsp70 released from the necrotic cells (Fig. 1A). Additional studies demonstrated that recombinant Grp170 protein (Fig. 1B), but not BSA (Fig. 1C), bound to CpG-ODN in a dose-dependent manner.

Figure 1.

Grp170 interacts with CpG-ODN directly by forming chaperone complexes. A) Grp170 released from necrotic cells binds to CpG-ODN1826. B16 cell lysates prepared after a freeze–thaw procedure were incubated with biotin-CpG-ODN and immunoprecipitated with avidin-agarose. Immunocomplexes were separated by SDS/PAGE and blotted with antibodies against the indicated HSPs. B) Dose-dependent binding of CpG-ODN to Grp170. Grp170 was incubated with increasing concentrations of biotin-CpG-ODN at room temperature, followed by immunoblotting (IB) analysis using streptavidin. C) Analysis of CpG-ODN binding to BSA and Grp170 using ELISA assays. Different concentrations of biotin-CpG-ODN were incubated with microtiter plates precoated with Grp170 or BSA. Levels of bound CpG-ODN were determined using colorimetric assays. D) Effect of reducing or oxidative conditions and ATP on the complex formation. Grp170 was incubated with biotin-CpG-ODN in the presence of DTT, H₂O₂, or ATP at different concentrations as indicated. E) Competition assays of Grp170 and CpG-ODN interaction. Grp170 was incubated with biotin-CpG-ODN1826 in the presence of excess unlabeled CpG-ODN1826, GpC-ODN1982, HPV E7_49–57 peptides, or gp100 protein, followed by immunoblotting analysis. F) Schematic illustration of the predicted domains of Grp170 and construction of Grp170 deletion mutants (BLH, LH, H, ABL, and AB). Grp170 has an NH₂-terminal ATP binding domain (A), followed by a β-sheet domain (B). These 2 domains are homologous to the ATP- and peptide-binding domains of Hsp70. This is followed by a predicted acidic loop domain (L) that is not seen in Hsp70 and by α-helical domain (H) that is somewhat similar to the COOH-terminal regions of Hsp70. All Grp170 deletion mutants contain a 6-His tag at the COOH terminus for purification. G) Immunoblotting analysis of the binding of CpG-ODN to chaperoning competent Grp170 mutants. H) SPR analysis of binding of Hsp70 superfamily members to CpG-ODN. Biotin-CpG-ODN was immobilized on a streptavidin-coated sensor chip. Different concentrations of Grp170, Hsp110, and Hsp70 protein were injected with a delay time of 180 s through the sensor chip surface. In addition, the assays were also run using 20 nM chaperone molecules and BSA was used as a negative control. Data represent 3 independent experiments with similar results.

We characterized the interaction between Grp170 and CpG-ODN under redox conditions or in the presence of ATP. The Grp170-CpG-ODN complex remained intact exposed to low concentrations of DTT, H₂O₂, or ATP. However, the complex formation was diminished when treated with high concentrations of these agents (Fig. 1D). The similar observation was made in the study of Grp170 interaction with antigenic peptides (22).

Competition study showed that the Grp170-biotin-CpG-ODN complex formation was blocked by the presence of excessive unlabeled or “cold” CpG-ODN1826 (Fig. 1E, left). Non-CpG-containing ODN (i.e., GpC-ODN1982) also efficiently competed with CpG-ODN1826 (Fig. 1E, middle), suggesting a promiscuous interaction between Grp170 and nucleic acids. Excess peptides or proteins also interfered with binding of CpG-ODN1826, suggesting that both CpG-ODN and polypeptides may bind to the same sites of Grp170 (Fig. 1E, right).

To examine the chaperoning competent domains responsible for binding to CpG-ODN, we prepared serial structural deletion mutants based on the predicted secondary structure of Grp170 by homology modeling with Hsp110 and Hsp70 (17, 20) (Fig. 1F). The mutant proteins, including BLH, LH, H, ABL, and AB were assayed for their ability to interact with biotin-CpG-ODN. Full-length Grp170 and all of its chaperoning functional mutants (BLH, LH, H, and ABL) bound to CpG-ODN, whereas the chaperone null mutant AB did not (Fig. 1G). This result is consistent with our previous study of functional domains involved in substrate chaperoning by Grp170 (20).

The binding of CpG-ODN to Grp170 was further assessed using SPR analysis. Biotin-CpG-ODN was coupled to a SA sensor chip as the immobilized ligand, and recombinant Grp170 was used as the soluble analyte. Various concentrations of the protein were passed over the CpG-ODN immobilized sensor chip to examine time- and concentration-dependent interactions (Fig. 1H). The best-fit interaction data between CpG-ODN and Grp170 protein was provided by the Langmuir 1:1 model with a K_d value of 2.05 × 10⁻¹¹. Two other Hsp70 superfamily members, Hsp110 and Hsp70, also bound to CpG-ODN (Fig. 1H). However, their affinity for CpG-ODN was ∼100-fold lower compared with that between CpG-ODN and Grp170 (Table 1).

Table 1.

Comparison of the kinetic parameters of interactions between CpG-ODN and the Hsp70 superfamily members

Ligand	Ka (ms−1)	Kd (s−1)	KD	χ2
Hsp70	2.243 × 105	2.561 × 10−4	1.142 × 10−9	2.01
Hsp110	9.055 × 104	3.004 × 10−4	3.317 × 10−9	2.44
Grp170	5.993 × 106	1.226 × 10−4	2.045 × 10−11	1.78

CpG-ODN chaperone complex induces synergistic activation of the inflammatory response in mouse macrophages

The observation of Grp170-mediated binding of CpG-ODN, the microbial DNA mimetic recognized by TLR9, prompted us to examine the potential effect of Grp170 on CpG-ODN-induced cytokine production. BMMs were treated with immunostimulatory CpG sequences (i.e., ODN1826) or nonstimulatory corresponding GpC control (i.e., ODN1982), either alone or in complex with Grp170 that was generated by preincubation for 30 min at room temperature. The nonstimulatory ODN1982 was ineffective in inducing cytokine expression regardless of whether it was in a complex with Grp170, as indicated by QRT-PCR (Fig. 2A) and ELISA assays (Fig. 2B). In contrast, ODN1826 complexed with Grp170 was significantly more effective in eliciting production of TNF-α, IL-6, and IL-12p40 compared with ODN1826 alone (Fig. 2A, B). Additionally, enhanced IL-6 production by BMMs was also seen when titrated amounts of ODN1826 were complexed with Grp170 (Fig. 2C). Heat inactivation of Grp170 completely abolished the synergistic effect, which rules out the possibility of endotoxin contamination in the Grp170 preparations (Fig. 2D). Hsp70 failed to potentiate CpG-ODN-induced IL-6 production when used at the same molar concentration as Grp170 (Fig. 2E). Compared with simultaneous addition of Grp170 and CpG-ODN, incubation of Grp170 and CpG-ODN before addition to the cell cultures generated more active complexes associated with increased TNF-α production (Fig. 2F).

Figure 2.

Grp170 promotes CpG-ODN-induced inflammatory response. A, B) BMMs were stimulated with ODN1826 or ODN1982 (as a control) in the presence or absence of Grp170. mRNA (A), and protein levels (B) of cytokines were analyzed using QRT-PCR and ELISA assays, respectively. C) Effect of Grp170 complexed with titrated amounts of CpG-ODN on cytokine production. D) Heat inactivation of Grp170 abolishes the synergistic effect. E) Superior activity of Grp170 in promoting cytokine production compared with Hsp70. F) Incubation of CpG-ODN and Grp170 before addition to macrophage cultures results in enhanced cytokine production. BMMs were treated with CpG-ODN1826 and Grp170 with or without preincubation. ELISA was used to assess the levels of TNF-α in the culture. G, H) Increased iNOS expression and NO production on Grp170-ODN stimulation. BMMs were stimulated with CpG-ODN or CpG-ODN in complex with Grp170, followed by analyses of mRNA transcription (G, top) and protein expression (G, bottom) using PCR and immunoblotting, respectively. Levels of nitrite in culture supernatant were determined using Griess assays (H). I) Synergistic activity of Grp170 and CpG-ODN in human macrophages. PMA-differentiated THP-1 cells were stimulated with CpG-ODN2006 or CpG-ODN2006 plus Grp170 for 4 h. mRNA levels of TNF-α were examined using qRT-PCR. *P < 0.05; **P < 0.01; ***P < 0.001; NS, not significant. Data shown represent 3 independent experiments with similar results.

iNOS generates microbicidal NO to combat invading microorganisms (23). Both gene transcription and protein expression of iNOS were strongly induced by Grp170-CpG-ODN complex compared with CpG-ODN alone (Fig. 2G). The complex-stimulated cells produced higher levels of NO than those treated with CpG-ODN alone (Fig. 2H).

We also assessed potential synergistic activity of Grp170 and CpG-ODN in human macrophages. PMA-differentiated THP-1 cells treated with CpG-ODN2006 plus Grp170 induced a greater gene transcription of TNF-α compared with those treated with CpG-ODN alone (Fig. 2I).

Grp170-potentiated inflammatory response induced by CpG-ODN is MyD88 dependent

Recognition of CpG-ODN by TLR9 activates downstream signaling cascades and up-regulates inflammatory genes (24). Grp170-ODN complex induced higher levels of phosphorylation of ERK1/2 and JNK than did CpG-ODN alone (Fig. 3A). A slight increase in the phosphorylation of p38 and IκBα was also observed (Fig. 3A). To assess the role of MyD88 in Grp170-promoted cytokine production, wild-type and MyD88^−/− macrophages were treated with CpG-ODN alone or in complex with Grp170. The absence of MyD88 completely disrupted the increased expression of TNF-α and IL-6 (Fig. 3B). Up-regulation of iNOS expression was also diminished in MyD88^−/− macrophage stimulated with CpG-ODN, regardless of the presence or absence of Grp170 (Fig. 3C), suggesting that the augmentation by Grp170 of inflammatory responses to CpG-ODN is dependent on the TLR9-MyD88 pathway.

Figure 3.

Grp170 potentiates CpG-ODN-induced activation of MyD88-dependent signaling. A) BMMs were treated with CpG-ODN or CpG-ODN in complex with Grp170 for the different times indicated. Cell lysates were analyzed using immunoblotting for phosphorylation of IκBα and MAPKs (ERK1/2, p38, and JNK1/2). β-actin serves as a loading control. B) Immortalized wild-type and MyD88-deficient macrophages were treated with CpG-ODN, CpG-ODN plus Grp170 or left untreated. Levels of TNF-α and IL-6 in the supernatants were determined by ELISA. **P < 0.01; ***P < 0.001; NS, not significant; ND, not detectable. C) Immunoblot analysis of iNOS expression following CpG-ODN and Grp170 stimulation. Data shown represent 3 independent experiments with similar results.

Grp170 facilitates endocytosis of CpG-ODN and interacts with TLR9 on cell entry

Our previous study (25) showed that Grp170 binds to macrophages in a receptor-dependent manner. We therefore assessed whether Grp170 could facilitate the internalization of CpG-ODN. BMMs treated with FITC-labeled CpG-ODN (FITC-CpG-ODN) in complex with Grp170 displayed increased endocytosis of CpG-ODN compared with those treated with FITC-CpG-ODN alone, as indicated by fluorescence microscopy (Fig. 4A) and FACS analysis (Fig. 4B).

Figure 4.

Grp170 facilitates the uptake and delivery of CpG-ODN by macrophages. A) BMMs were treated with FITC-labeled CpG-ODN in the presence or absence of Grp170 protein for the indicated times. Internalization of CpG-ODN was examined using fluorescence microscopy (×200; scale bar=200 μm). Phase contrast images are also shown. B) Following incubation, FITC-positive cells were determined by FACS analysis. C) Direct interaction between TLR9 and the internalized His-Grp170. Raw264.7 cells were incubated with His-Grp170 for 30 or 60 min. Cells were extensively washed and TLR9 was immunoprecipitated using anti-TLR9 antibodies (top panels). Immune complexes were analyzed using anti-His antibodies for the presence of His-Grp170. In addition, His-Grp170 was pulled down and analyzed for association with TLR9 using anti-TLR9 antibodies (bottom panels). Levels of TLR9 and His-Grp170 in the whole cell lysate were also examined. Cells incubated with His-Grp170 for 60 min was immunoprecipitated (IP) with normal mouse IgG as a control. D) Raw264.7 cells were stimulated with Grp170 in the presence or absence of CpG-ODN1826. TLR9 immune complexes were analyzed for association of His-Grp170 using immunoblotting. E) Colocalization of TLR9 and internalized Grp170 in the endosomal compartment. BMMs were incubated with FITC-labeled Grp170 protein and stained with anti-Rab5 and TLR9 antibodies, followed by confocal microscopy analyses (×630; scale bar=5 μm). F) Association of TLR9 with endogenous ER-resident chaperones. Macrophages were stimulated with or without CpG-ODN1826 (20 μg/ml) for 60 min. TLR9 was pulled down and subjected to immunoblotting for Grp170, Grp94, Grp78, and TLR9. G) BMMs pretreated with dynasore or chloroquine were incubated with CpG-ODN or CpG-ODN plus Grp170. Production of TNF-α and IL-6 was determined by ELISA. H) BMMs were pretreated with fucoidan before stimulation. ELISA was used to examine the TNF-α levels in the supernatants. **P < 0.01. Data shown represent 3 independent experiments with similar results.

In light of the superior chaperoning property of Grp170, we examined the possibility of TLR9 interaction with exogenous Grp170 after entry into the cell. Indeed, the internalized His-tagged Grp170 was physically associated with TLR9, as indicated by reciprocal immunoprecipitation assays (Fig. 4C). The presence of CpG-ODN in the Grp170 chaperone complex modestly increased the association of Grp170 with TLR9 (Fig. 4D). Confocal microscopy showed that FITC-labeled Grp170 localized with the early endosomal marker Rab5 and TLR9 after entering the cell (Fig. 4E). Surprisingly, TLR9 also associated with several endogenous stress proteins or chaperones, including Grp170, Grp94/gp96, and Grp78/Bip, and this interaction appeared to be strengthened in the presence of CpG-ODN signal (Fig. 4F).

Disruption of endocytosis using dynasore or endosomal functions using chloroquine severely impaired TNF-α and IL-6 production in BMMs in response to the stimulation with CpG-ODN alone or in complex with Grp170 (Fig. 4G). Pretreatment of macrophages with fucoidan to block scavenger receptors significantly reduced the TNF-α production augmented by the complex (Fig. 4H), indicating that scavenger receptors, which are suggested to be putative receptors for Hsp70 superfamily members (25–28), may be involved in the uptake of the complexes.

Synergistic activation of innate immunity by Grp170 and CpG-ODN protects mice from microbial infection

To assess the potential effect of Grp170 on CpG-ODN-induced immune response in vivo, sera were collected from C57BL/6 mice after injections with PBS, Grp170, CpG-ODN, or CpG-ODN in complex with Grp170. The presence of Grp170 in the CpG-ODN chaperone complex resulted in enhanced production of TNF-α, IL-6, and IL-12p40 (Fig. 5A) compared with treatment with CpG-ODN alone. Similar to in vitro assays, the enhancing effect of Grp170 was also observed when different concentrations of CpG-ODN were tested in vivo (Fig. 5B). The complex-treated mice also displayed a profound increase in the serum level of IFN-γ, a cytokine that drives Th1 immunity and contributes to microbicidal defense (Fig. 5C). The levels of IL-10, Th2 cytokine, were not significantly different in mice receiving CpG-ODN alone or Grp170-CpG-ODN complex (Fig. 5C). We examined the ability of immune cells from the treated mice to respond to heat-killed L. monocytogenes (HKLM). Peritoneal cells isolated from mice receiving the CpG-ODN chaperone complex produced significantly greater NO, a potent microbicidal agent, than those from mice treated with CpG-ODN alone (Fig. 5D).

Figure 5.

Synergistic effect of CpG-ODN and Grp170 enhances resistance to L. monocytogenes challenge in mice. A) Promotion of CpG-ODN-induced innate immune response by Grp170. ELISA assays of TNF-α, IL-6 and IL-12p40 in serum collected 2 h after mice were administrated intraperitoneal with Grp170, CpG-ODN or combination of Grp170 and CpG-ODN. B) Immune activation in vivo by different amounts of CpG-ODN in the presence of Grp170. C) ELISA assays of the serum IFN-γ and IL-10 levels 3 days after the injection. D) Peritoneal cells were isolated by lavage and cultured in vitro in the presence of HKLM for 48 h. Nitrite levels in culture supernatants were quantified using Griess assays. E) Mice (n=8) were administered intraperitoneally with CpG-ODN alone or CpG-ODN in combination with Grp170, and inoculated intravenously with Listeria 3 days later. Bacterial load in spleens and livers was measured 72 h after infection. Symbols represent individual mice. Data shown represent 2 independent experiments. F) Hypothetic model for extracellular Grp170-enhanced recognition of the pathogen-associated danger signals by macrophages. Stress proteins (e.g., Grp170) are released from injured cells during pathogen infection. Chaperoning of PAMPs, e.g., CpG-ODN, by the extracellular Grp170 amplifies TLR9 signaling in macrophages by facilitating the ligand uptake and possibly ligand interaction with its cognate receptor, which results in improved pathogen recognition associated with enhanced innate immunity. *P < 0.05; **P < 0.01; ***P < 0.001.

We investigated the potential impact of the Grp170-enhanced innate immune response on CpG-ODN-mediated protection against L. monocytogenes challenge. Three days after infection, the number of L. monocytogenes organisms in the liver and spleen of infected mice was quantitated by culturing the organ homogenates. CpG-ODN-treated mice exhibited decreased bacterial burden in the livers and spleens compared with those treated with PBS or Grp170 (Fig. 5E), which was consistent with previous reports (29, 30). However, treatment with CpG-ODN in combination with Grp170 resulted in significantly more reduction in the number of L. monocytogenes than did CpG-ODN alone (Fig. 5E).

DISCUSSION

In addition to the previously established role of Grp170 as a chaperone preferentially holding client proteins (20), we now reveal another novel aspect of chaperoning activity of Grp170 in interactions with DNA substrate. The efficient binding of CpG-ODN by Grp170 compared with other Hsp70 superfamily members is quite striking. However, the superior holding capability of Grp170 has been demonstrated in stabilizing heat-denatured protein substrates (22) and in transporting polypeptides (31). The significantly increased size of the Grp170 protein caused mainly by the expansion of the highly divergent central loop domain may play a role (32). Intriguingly, 2 regions capable of substrate binding have been identified in Grp170 (20). Therefore, the high binding affinity of Grp170 for CpG-ODN is reminiscent of its exceptional ability to hold client proteins. It should be noted that Grp170 readily forms complexes with CpG-ODNs, at least in vitro, while sensing and capture of protein substrate by chaperones often occur in response to stressors (e.g., heat shock). Although it is not clear whether a common molecular mechanism is used in the Grp170-mediated chaperoning of clients of different natures, HSPs (e.g., Hsp110 and Hsp70) have been implicated in interactions with RNAs (33).

We have also demonstrated that Grp170 in the extracellular milieu acts as an important modulator of CpG-ODN-TLR9 activation via its superior chaperoning activity, which accentuates the importance of molecular chaperoning in facilitating pathogen recognition and amplifying innate immune responses. Grp170 is more efficient than other chaperones, e.g., Hsp70, not only in interacting with CpG-ODN, but also in promoting CpG-ODN-induced cytokine production and innate immunity. That molecular chaperoning is involved in Grp170 interactions with microbial molecules and subsequent immune responses is also in line with our earlier observations implicating the critical importance of the ancient chaperoning activity in Grp170 interactions with protein antigens and consequent T-cell immunity (25). Although the phenomenon of HSP interaction with PAMPs (e.g., LPS) was previously reported (34, 35), the present study provides the first in vivo evidence showing that Grp170 potentiates CpG-ODN-elicited innate immunity, as indicated by increased serum levels of cytokines, such as TNF-α, IL-12, and IFN-γ, which are established contributors to Th1 immunity and microbicidal defense. Furthermore, treatment of mice with a single injection of CpG-ODN in complex with Grp170 significantly increased the resistance of mice to the challenge of L. monocytogenes compared with treatment with CpG-ODN alone. The burden of L. monocytogene was reduced by >1 log in the complex-pretreated mice compared with mice receiving CpG-ODN alone. Given the clinical relevance of the disease, future studies to evaluate the therapeutic potential of Grp170-CpG-ODN and to define the optimal therapeutic window for treating infections of L. monocytogene or other pathogens are warranted.

It should be noted that a type B CpG-ODN (i.e., ODN1826) is used in the current study, which is known to preferentially activate Th1-like immune responses, as indicated by IL-12 and TNF-α production, and has been tested as an adjuvant in numerous animal models (6). In contrast, type A CpG-ODNs are weak stimulators of TLR9-dependent NF-κB signaling but efficiently stimulate plasmacytoid dendritic cells to produce IFNα/β, which is important for antiviral immunity, indicating distinct pathways triggered by different types of CpG-ODNs (36, 37). It remains interesting whether Grp170 also synergizes with type A CpG-ODNs in driving a type I IFN response.

TLR9 activation requires CpG-ODN endocytosis and translocation of TLR9 to the endosomal compartment (38). Our studies demonstrate that at least 2 molecular events are involved in the exogenous Grp170-enhanced inflammatory response to CpG-ODN. Formation of CpG-ODN chaperone complexes with Grp170 facilitates the internalization and delivery of CpG-ODN to the intracellular compartment or enhances CpG-ODN access to its cognate receptor. The increased endocytosis of CpG-ODN in the presence of Grp170 may be attributed to HSP-binding structures on the cell surface of macrophages (39, 40). However, the exact contributions of these individual receptors to the Grp170-enhanced CpG-ODN uptake and TLR9 activation remain to be defined. Moreover, given the colocalization and association of the exogenous Grp170 with intracellular TLR9, the synergistic activity of the outside-in Grp170 and CpG-ODN may also involve direct interaction of the Grp170 with TLR9 in the endosomal compartment. It is possible that chaperoning of TLR9 and its ligand by Grp170 enhances the association of CpG-ODN with TLR9 because of their close proximity to each other and lowers the threshold of CpG-ODN detection. Alternatively, Grp170 may assist with maintaining TLR9 in a signaling-competent conformation or status, as described previously for the hormone receptor activity regulated by molecular chaperones (41). Interestingly, in addition to exogenous Grp170, TLR9 is associated with several endogenous stress proteins or chaperones, including Grp170, Grp94/gp96 and Grp78/Bip. Indeed, recent work of Yang et al. (42) reported that chaperoning of TLR9 by intracellular Grp94/gp96 inside the cell is important for TLR9 functions in macrophages. Thus the outside-in Grp170 may function in concert with intracellular chaperone networks in modifying TLR9 recognition of its ligand. Further study is necessary to define the molecular interactions within the multicomponent complex and how it may influence the magnitude of CpG-ODN-augmented inflammatory response.

Our results extend the concept of stress proteins as DAMPs or alarmins (15) by revealing an additional aspect of molecular chaperoning in modifying the sensing of PAMPs by PRRs and host defense (Fig. 5F). Grp170 and other stress proteins are passively released or actively secreted from the infected or stressed cells due to bacterial intrusion. Promiscuous interaction between these chaperone molecules and immunogenic nucleic acids, e.g., CpG-ODN, facilitates the recognition of microbial molecules by specific PRRs in danger-sensing cells, such as macrophages, and alerts the host immune system to the tissue injury or stress provoked during early bacterial infection. Together with the previously described Grp170 activity in chaperoning and presenting polypeptide antigens for T-cell activation (22, 25, 43), the actions of extracellular Grp170 may provide an important link between the innate and adaptive immune systems. However, the specifics of this mechanism are expected to be highly complex and potentially redundant given the presence of other stress proteins or DAMPs, such as HMGB-1 (44). Nonetheless, molecular chaperoning appears to be one of the essential common denominators underlying the observed immunological activities of Grp170.

Although TLR9 was originally believed to specifically recognize microbial DNA, it is now evident that mammalian self-DNA can also be an effective TLR9 ligand (45, 46), which is thought to contribute to the pathogenesis of autoimmune diseases (44). The interaction with exogenous stress proteins or chaperones could provide an alternative mechanism for self-DNA entry into the cell resulting in TLR9 activation. Indeed, extracellular stress proteins have been implicated in several autoimmune disorders (47). Chronic expression of Grp94/gp96 on cell surfaces produced spontaneous systemic lupus erythematosus-like phenotypes in mice (48). Therefore, the potential involvement of Grp170 interactions with self-protein antigens or DNAs in the pathogenesis of autoimmunity remains an interesting possibility worth exploring.

Our data collectively identify a previously unknown mechanism by which microbial DNA-chaperone complexes induce an increased inflammatory response and protective innate immunity. Physical association of immunostimulatory molecules (e.g., CpG-ODNs) with Grp170, a stress sensor and a molecular chaperone broadly found in all living organisms, and subsequent translocation of the complex from the outside to the inside of the cell facilitate the efficient recognition of microbial molecules by discriminative TLRs (Fig. 5F). Consequently, active interaction between an evolutionarily conserved chaperone molecule and PAMPs in the extracellular milieu may play a critical role in augmentation of innate immunity and maintenance of host homeostasis. Our findings provide an important new perspective on the dynamics of ancient chaperone molecules and the diverse activities of these molecules, Grp170 in particular, inside as well as outside the cell.