Published online before print August 7, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Departments of Critical Care Medicine and Pediatrics, Children’s Hospital of Pittsburgh,
Surgery,
Critical Care Medicine and Pathology, and
Critical Care Medicine, Surgery and Pharmacology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
1 Correspondence: Department of Critical Care Medicine, University of Pittsburgh School of Medicine and Children’s Hospital of Pittsburgh, 3705 5th Ave., Pittsburgh, PA 15213, USA. E-mail: anejar{at}upmc.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-differentiated human monocyte-like THP-1 cells. Treatment with 1 µg/ml HMGB1 18 h before exposure to LPS (1 µg/ml) decreased TNF release, NF-
B nuclear DNA-binding activity, phosphorylation, and degradation of IB
. Preconditioning with HMGB1 alone and HMGB1 in the presence of polymyxin B decreased LPS-mediated, NF-B-dependent luciferase reporter gene expression. The specificity of HMGB1 in tolerance induction was supported further by showing that boiled HMGB1 failed to induce tolerance, and antibodies against HMGB1 blocked the induction of LPS tolerance. Bone marrow-derived macrophages obtained from C57Bl/6 wild-type mice became LPS-tolerant following HMGB1 exposure ex vivo, but macrophages derived from RAGE-deficient mice failed to develop tolerance and responded normally to LPS. Mice preconditioned with HMGB1 (20 µg) 1 h before LPS injection (10 mg/kg) had lower circulating TNF compared with control mice preconditioned with saline vehicle. Similarly, decreased nuclear DNA binding of hepatic NF-B was observed in mice preconditioned with HMGB1. Taken together, these results suggest that extracellular HMGB1 induces LPS tolerance, and the RAGE receptor is required for this induction.
Key Words: signal transduction • RAGE • TLR
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
High mobility group box protein 1 (HMGB1) is a small DNA-binding protein, which facilitates the binding of several regulatory protein complexes to DNA, particularly members of the nuclear hormone-receptor family [5 6 ], variable(diversity)joining recombinases [7 ], and the tumor suppressor proteins, p53 and p73 [8 ]. In 1999, Wang et al. [9 10 11 ] identified HMGB1 as a late mediator of LPS-induced mortality in mice. Plasma levels of HMGB1 are increased in patients with severe sepsis, particularly nonsurvivors [12 ]. It has been postulated that the presence of extracellular HMGB1 is a result of its release from necrotic tissue damaged by trauma or ischemia as well as active secretion by macrophages or other cell types [13 ] in response to proinflammatory stimuli.
Recently, HMGB1 has come to be regarded as a member of endogenous compounds, called "alarmins," which serve as danger signals to promote activation of the innate immune system in response to tissue injury as a result of trauma, ischemia/reperfusion, or infection [14
]. Tsung and co-workers [15
] reported that pretreatment of mice with exogenous HMGB1 protected against hepatic ischemia/reperfusion injury and blunted the inflammatory response to this insult. These observations suggested that prior exposure to a low concentration of HMGB1 could blunt the response to subsequent challenge with a different proinflammatory stimulus, namely LPS. In other words, we hypothesized that exposure to HMGB1 is capable of inducing LPS tolerance. Herein, we document that exposing IFN--differentiated THP-1 human monocyte-like cells to HMGB1 induced LPS tolerance. We also examine LPS tolerance in bone marrow-derived macrophages obtained from C57Bl/6 wild-type and receptor for advanced glycation end products (RAGE)- and TLR2-deficient mice. More importantly, we show that injecting mice with a low dose of HMGB1 induces tolerance to the proinflammatory effects of LPS challenge in vivo.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(100 U/ml; Pierce Biotechnology, Rockford, IL, USA) for 18 h. THP-1 cells were cultured in RPMI-1640 medium containing 10% FBS, 1% penicillin/streptomycin, 0.35% 2-ME, and 2% glutamine, 10 mM Hepes (pH 7.35). Studies were performed in 12-well plates (Becton Dickinson, Mountain View, CA, USA) at a density of 1 x 106 cells. Where indicated, stimulation was performed with 1 µg/ml LPS (Escherichia coli, serotype O55:B5, Sigma Chemical Co., St. Louis, MO, USA). Recombinant HMGB1 was purchased from Sigma Chemical Co.
Bone marrow-derived macrophage cultures
Isolation of murine bone marrow-derived macrophages has been described previously [16
]. Briefly, bone marrow cells were obtained from C57Bl/6, TLR 2–/– (Charles River Laboratories, Wilmington, MA, USA), and RAGE–/– mice and were centrifuged in lymphocyte separation medium (Mediatech Inc., Herndon, VA, USA). Cells harvested from the interface were cultured in 75 cm2 flasks in Eagle’s MEM (EMEM; Mediatech Inc.) supplemented with 10% FBS and 10 ng/ml GM-CSF (Thermo Scientific, Rockford, IL, USA). After a 24-h adherence step (Day 2), which allowed for the removal of mature monocytes and fibroblasts from the bone marrow, nonadherent cells from each flask were transferred to a second flask and were supplemented with medium containing GM-CSF (10 ng/ml). The cells were again supplemented with GM-CSF (10 ng/ml) on Day 4. After a total of 7 days in culture, macrophages were removed enzymatically with the neutral protease, dispase (Worthington Biochemical Corp., Lakewood, NJ, USA), and collected by gentle scraping. The cells were resuspended in complete EMEM and exposed to experimental conditions.
Model of LPS tolerance with HMGB1
THP-1 cells cultured in RPMI media containing 10% heat-inactivated FBS were incubated with IFN- for 18 h to allow for differentiation. The following day, cells were treated with 1 µg/ml HMGB1 or 10 ng/ml LPS in complete RPMI media and then incubated at 37°C for 18 h. The cells were then stimulated with 1 µg/ml LPS for the indicated time intervals. The chosen 18-h time-point between the preconditioning stimulus with HMGB1 and the LPS stimulus was based on previous tolerance literature in which the preconditioning stimulus was a low dose of LPS [2
].
Transient transfections and luciferase assays
A NF-B-luciferase reporter plasmid was used to measure LPS-dependent activation of NF-B. The plasmid (3x-NF-B-Luc) contains the luciferase reporter gene under the control of three tandem NF-B-binding motifs and a minimal IFN-β promoter. THP-1 cells were transfected using DEAE-dextran. THP-1 cells/ml (1x106) were seeded into the day before transfection. The next day, the cell suspension was washed twice with 25 mM Tris·Cl, pH 7.4, 137 mM NaCl, 5 mM KCl, 0.6 mM Na2HPO4, 0.7 mM CaCl2, and 0.5 mM MgCl2 (STBS) and pelleted. NF-B reporter plasmid (1 µg/ml) was mixed with DEAE-dextran (400 µg/ml) in 140 µl STBS buffer and added immediately to the pelleted THP-1 cells, which were incubated at 37°C for 15 min, washed twice with STBS, resuspended, and cultured in complete RPMI medium. The transfected cell lines were cultivated for 24 h and harvested. Where indicated, low-dose HMGB1 or low-dose LPS was added 24 h after transfection, followed by addition of LPS (1 µg/ml) for 4 h. After treatment, cellular proteins were extracted and analyzed for luciferase activity according to the manufacturer’s instructions (enhanced luciferase assay kit, BD Biosciences PharMingen, San Diego, CA, USA) by using a BioOrbit 1250 luminometer. Luciferase activity is reported as fold induction over control cells (transfected and treated with basal growth media) and corrected for total cellular protein.
Additional tolerance experiments to rule out endotoxin contamination
As described above, differentiated THP-1 cells were preconditioned with 1 µg/ml recombinant HMGB1 for 18 h and then treated with LPS (1 µg/ml) for 4 h. Further experiments were designed to demonstrate that induction of tolerance to LPS is secondary to HMGB1 and not a result of LPS contamination of HMGB1. For these studies, THP-1 cells were pretreated with 1 µg/ml HMGB1, which was denatured by immersion in boiling water for 60 min or boiled LPS (10 ng/ml). For comparison, other THP-1 cells were pretreated with the same concentration of HMGB1 or LPS, which was not subjected to prolonged heating. The cells were incubated with boiled or nonboiled HMGB1 or LPS for 18 h and then stimulated with 1 µg/ml LPS for 4 h. Cells were then harvested for protein extraction and assayed for luciferase activity as described above. To further test the contribution of HMGB1 in inducing tolerance, we pretreated THP-1 cells with HMGB1 and simultaneously added anti-HMGB1 neutralizing antibody (Sigma Chemical Co.) for 18 h. Additional experiments were conducted by simultaneous addition of HMGB1 and polymyxin B (50 µg/ml; Sigma Chemical Co.) for 18 h. Cell supernatant was analyzed for TNF levels. Endotoxin levels were determined using Limulus amoebocyte lysate (LAL) assay (BioWhitaker, Walkersville, MD, USA).
ELISA
TNF concentrations were measured in culture supernatants from treated cells using a commercially available sandwich ELISA (Biosource, Camarillo, CA, USA). All procedures were performed as recommended by the manufacturer.
Nuclear protein extraction
All nuclear protein extraction procedures were performed on ice with ice-cold reagents. Cells were treated with HMGB1 for 18 h before incubation with LPS for 45 min. Cells were washed twice with PBS and harvested by scraping. Cells were pelleted in 1 ml PBS at 14,000 rpm for 1 min. The pellet was washed twice with PBS and resuspended in lysis buffer [10 mM Tris-HCl (pH 7.8), 10 mM KCl, 1 mM EGTA, 5 mM MgCl2, 1 mM DTT, and 0.5 mM PMSF]. The suspension was incubated on ice for 15 min, and Nonidet P-40 was added, followed by centrifugation at 4°C at 2000 rpm for 5 min. The supernatant was discarded, and the cell pellet was dissolved in extraction buffer (20 mM Tris-HCl, pH 7.8, 32 mM KCl, 0.2 mM EGTA, 5 mM MgCl2, 1 mM DTT, 0.5 mM PMSF, and 25% v/v glycerol), added to the nuclear pellet, and incubated on ice for 15 min. Nuclear proteins were isolated by centrifugation for 10 min at 14,000 rpm. Protein concentrations of the resultant supernatants were determined using the Bradford assay. Nuclear proteins were stored at –70°C until used for EMSA.
EMSA
EMSA were performed as described previously [17
]. A double-stranded oligonucleotide probe corresponding to the NF-B oligonucleotide probe (5'-GTGGAATTTCCTCTGA-3') was labeled with -[32P] ATP using T4 polynucleotide kinase (Promega, Madison, WI, USA) and purified in Bio-Spin chromatography columns (GE Healthcare, Buckinghamshire, UK). For each sample, 4 µg nuclear proteins were incubated with Bandshift buffer (10 mM Tris-HCl, 40 mM KCl, 1 mM EDTA, 1 mM DTT, 50 ng/ml polydeoxyinosinic:polydeoxycytidylic acid, 10% glycerol) at room temperature with subsequent addition of the radiolabeled oligonucleotide probe for 30 min. Protein-nucleic acid complexes were resolved using a nondenaturing polyacrylamide gel consisting of 5% acrylamide (29:1 ratio of acrylamide:bisacrylamide) and run in 0.25 x 45 mM Tris-HCl, 45 mM boric acid, 1 mM EDTA for 1 h at constant current (30 mA). Gels were transferred to Whatman 3MM paper, dried under a vacuum at 80°C for 1 h, and used to expose to X-ray film at –70°C with an intensifying screen.
Western blot analysis
Western blot analyses were performed as described previously [17
]. Briefly, whole cell lysates containing 30 µg protein were boiled in equal volumes of loading buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, and 10% 2-β-ME). Proteins were separated electrophoretically on 8–16% and subsequently transferred to polyvinylidene difluoride (GE Healthcare). For immunoblotting, membranes were blocked with 5% nonfat dried milk in PBS for 1 h. Primary antibody against IB (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and phosphorylated IB (Cell Signaling Technology, Danvers, MA, USA) was applied at 1:250 and 1:500 dilution, respectively. After washing twice with PBS containing 0.5% Tween 20 (PBST), secondary antibody (HRP-conjugated goat anti-rabbit IgG, Stressgen, Victoria, British Columbia) was applied at 1:4000 dilution for 1 h. Blots were washed in PBST thrice for 10 min, incubated in ECL reagent (GE Healthcare), and used to expose X-ray film (GE Healthcare).
Cell viability
Cell viability after 18 h of exposure to HMGB1, followed by LPS treatment, was determined by trypan blue exclusion.
Animals
Male wild-type (C57BL/6) mice (8–12 weeks old) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). The Animal Care and Use Committee of the University of Pittsburgh (Pittsburgh, PA, USA) approved animal protocols, and the experiments were performed in adherence to the National Institutes of Health (NIH) Guidelines for the Use of Laboratory Animals. Mice were injected i.v. with HMGB1 or vehicle (PBS) 1 h before i.v. injection of LPS (10 mg/kg). The HMGB1 used for these studies contained undetectable amounts of LPS, as measured by the chromogenic LAL assay (Associates of Cape Cod, Cape Cod, MA, USA). The lower endotoxin detection limit for this assay is 0.06 ng/ml. Serum and hepatic tissue samples were obtained 8 h after LPS injection.
Isolation of nuclear and cytoplasmic proteins
Cytoplasmic and nuclear proteins were extracted from frozen liver tissues as described previously [18
]. Protein concentrations were determined with the bicinchoninic acid protein assay reagent (Pierce Biotechnology).
TNF measurements
Serum concentrations of immunoreactive TNF were determined using ELISA kits (Biosource International).
Statistical analyses
Differences among groups were evaluated by one-way ANOVA and Student-Newman-Keuls test. P 
0.05 was considered statistically significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B-dependent reporter gene expression. Differentiated THP-1 cells transfected with the 3x NF-B luciferase reporter plasmid were exposed to increasing half-log concentrations of recombinant HMGB1, ranging from 0.01 µg/ml to 10 µg/ml for 4 h. The minimum concentration of HMGB1 that induced detectable reporter gene expression was >3 µg/ml. Accordingly, the HMGB1 concentration chosen for further studies of HMGB1-mediated induction of LPS tolerance ranged from 0.01 to 3 µg/ml (i.e., we used a concentration of HMGB1, which was insufficient to induce NF-B-dependent luciferase expression in transfected THP-1 cells). HMGB1 concentrations less than 0.3 µg/ml failed to induce LPS tolerance. Pretreatment with HMGB1 concentrations of 0.3, 1, and 3 µg/ml decreased LPS-mediated NF-B activation by 5%, 20%, and 32%, respectively (data not shown). Hence, we used HMGB1 1 µg/ml as the pretreating concentration. A common criticism of using recombinant proteins for experiments is the presence of contaminating LPS. Recombinant HMGB1 was tested for the presence of LPS by using the Limulus lysate endotoxin assay kit. There was no detectable endotoxin present in the recombinant HMGB1 protein.
HMGB1 pretreatment induces tolerance in differentiated THP-1 cells
Differentiated THP-1 cells transiently transfected with the 3x NF-B luciferase reporter plasmid that were exposed to LPS expressed 53.2 ± 6.7-fold more luciferase activity compared with resting cells. Preconditioning with 1 µg/ml HMGB1 for 18 h before LPS stimulation significantly attenuated NF-B-dependent promoter activity (32.1±2.8-fold induction, P<0.05 compared with LPS alone; Fig. 1
).
 View larger version (10K): [in a new window] |
Figure 1. Effect of HMGB1 preconditioning on LPS-mediated NF-B promoter activity as assessed by luciferase assay. Differentiated THP-1 cells were transfected with a 3x NF-B promoter luciferase reporter plasmid and preconditioned with HMGB1 1 µg/ml or LPS (10 ng/ml) for 18 h, as indicated. Cells were then exposed to 1 µg/ml LPS for 4 h. Data represent mean ± SEM of three separate experiments with each condition performed in triplicate (*, P<0.05, vs. LPS alone).
|
B luciferase reporter plasmid-transfected THP-1 cells to varying concentrations of LPS (0.01–100 ng/ml), followed by LPS stimulation with 1 µg/ml (data not shown). The preconditioning LPS concentration needed to approximate the same degree of tolerance, as HMGB1 was 10 ng/ml. Cells preconditioned with LPS (10 ng/ml) demonstrated significant attenuation of NF-B-dependent promoter activity following LPS stimulation (38.0±3.2-fold induction, P<0.05 compared with LPS treatment of nontolerant cells; Fig. 1
). This observation is consistent with previous reports of LPS tolerance induction by pretreating with subthreshold concentrations of LPS. These data suggest that HMGB1 preconditioning induces tolerance to subsequent LPS-mediated NF-B activation in a manner similar to classical LPS tolerance.
TNF levels in culture supernatants
Decreased TNF secretion after exposure of cells to LPS is a reliable indicator of LPS tolerance [19
]. Hence, we determined the effect of HMGB1 preconditioning on LPS-induced production of this proinflammatory cytokine. TNF concentrations in culture supernatants were assessed by ELISA. As shown in Figure 2
, TNF levels in THP-1 culture supernatants were significantly lower in cells preconditioned with HMGB1 (1 µg/ml) as compared with LPS-stimulated cells without HMGB1 preconditioning (1420±420 vs. 2020±320 pg/ml, P<0.05; Fig. 2
).
 View larger version (9K): [in a new window] |
Figure 2. ELISA results demonstrating the effects of HMGB1 pretreatment on TNF levels in the supernatants of differentiated THP-1 cells. Control cells were maintained in a basal growth medium. Cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS 1 µg/ml for 4 h. Data represent the mean ± SEM of three separate experiments with each condition performed in triplicate (*, P<0.05, vs. LPS alone).
|
B promoter activity was comparable in THP-1 cells preconditioned with boiled HMGB1 and cells treated with LPS alone (47.5±3.5- vs. 49.4±5.0-fold induction; Fig. 3A
). As expected, the cells treated with nonboiled HMGB1 demonstrated LPS tolerance, as evidenced by attenuation of the NF-B-dependent reporter activity compared with cells treated with LPS alone (35.1±3.8 vs. 49.4±5.0 luciferase fold induction, P<0.05). Cells pretreated with 10 ng/ml boiled LPS were tolerant to a second exposure of LPS (1 µg/ml; 49.4±4.5- vs. 37.8±threefold induction, P<0.05; Fig. 3A
). These data demonstrate that boiled HMGB1 is unable to induce tolerance in contrast to boiled LPS, which retains the property to induce LPS tolerance.
 View larger version (14K): [in a new window] |
Figure 3. (A) Effect of preconditioning with nonboiled HMGB1, boiled LPS, or boiled HMGB1 on LPS-mediated NF-B promoter reporter luciferase activity. Differentiated THP-1 cells were transfected with 3x NF-B promoter luciferase reporter plasmid and preconditioned with HMGB1 (1 µg/ml), boiled LPS 10 ng/ml, or boiled HMGB1 (1 µg/ml) for 18 h. Cells were then exposed to 1 µg/ml LPS for 4 h. Data represent mean ± SEM of three separate experiments with each condition performed in triplicate. (*, P<0.05, vs. LPS alone). (B) Effect of HMGB1 and polymyxin B preconditioning on LPS-mediated NF-B promoter activity, as assessed by luciferase assay. Differentiated THP-1 cells were transfected with a 3x NF-B promoter luciferase reporter plasmid and preconditioned with HMGB1 1 µg/ml and polymyxin (50 µg/ml) for 18 h, as indicated. Cells were then exposed to 1 µg/ml LPS for 4 h. Data represent mean ± SEM of three separate experiments with each condition performed in triplicate (*, P<0.05, vs. LPS alone).
|
HMGB1 antibody blocks induction of LPS tolerance by HMGB1
In a separate group of experiments, we determined if simultaneous addition of HMGB1 and anti-HMGB1 neutralizing antibody blocks induction of LPS tolerance. Consistent with our previous data, we observed that NF-B-dependent luciferase reporter activity was attenuated significantly in differentiated THP-1 cells preconditioned with HMGB1, as compared with cells treated with LPS alone (56.2±6.7- vs. 34.5±2.8-fold induction, P<0.05; Fig. 4
). Simultaneous addition of HMGB1 and anti-HMGB1 antibody during preconditioning blocked induction of LPS tolerance, as compared with cells preconditioned with HMGB1 alone (55.5±4 vs. 34.5±2.8 luciferase fold induction, P<0.05). As an additional control, we added an irrelevant antibody (anti-β-actin antibody), and this antibody failed to block the induction of tolerance by HMGB1 (37.2±3.2 vs. 56.16±6.7 luciferase fold induction, P<0.05). Collectively, these data provide further evidence that pretreatment with HMGB1 specifically induces tolerance to LPS in differentiated THP-1 cells.
 View larger version (12K): [in a new window] |
Figure 4. Luciferase assay demonstrating the comparison of preconditioning with HMGB1 or HMGB1 and simultaneous addition of a HMGB1 antibody on LPS-mediated NF-B promoter activity. Differentiated THP-1 cells were transfected with the 3x NF-B promoter luciferase reporter plasmid and preconditioned with HMGB1 alone, HMGB1 plus anti-HMGB1 antibody, or HMGB1 plus anti-β-actin antibody for 18 h. Cells were then exposed to 1 µg/ml LPS for 4 h. Data represent mean ± SEM of three separate experiments with each condition performed in triplicate (*, P<0.05, vs. LPS alone; #, P<0.05, vs. HMGB1 and LPS).
|
BB is a ubiquitous transcription factor that controls proinflammatory gene expression in response to LPS. We determined the effect of preconditioning with HMGB1 on LPS-mediated NF-B DNA binding in nuclear extracts prepared from THP-1 cells. Treatment with LPS alone increased DNA binding of NF-B as compared with control cells (Fig. 5A
). Nuclear NF-B DNA-binding activity was markedly inhibited compared with cells treated with LPS alone (Fig. 5A)
. Another mechanism proposed to explain decreased NF-B-dependent transcriptional activation following the induction of endotoxin tolerance with LPS is a switch in the NF-B heterodimer composition from p65/p50 to the transcriptionally inactive p50/p50 homodimer. We examined the composition of the NF-B dimers by supershift analyses of the protein complex bound to DNA. Nuclear proteins purified from THP-1 cells were incubated with antibody specific for p65 and p50 prior to assembling the DNA-binding reaction while performing EMSA. Preconditioning with HMGB1 did not affect the NF-B subunit composition in response to LPS (Fig. 5B)
.
 View larger version (44K): [in a new window] |
Figure 5. (A) DNA binding of NF-B in THP-1 cells stimulated with LPS. Representative autoradiograph of EMSA for NF-B is representative of three similar, separate experiments. Control cells were maintained in basal growth medium. LPS-treated cells were treated with LPS (1 µg/ml) for 1 h. HMGB1-treated cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS. (B) To analyze the composition of the DNA-binding complex, supershift of the DNA-binding complex was performed with p65 and p50 antibody. LPS-treated cells were treated with LPS (1 µg/ml) for 1 h. LPS-treated cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS. Nuclear protein purified from THP-1 cells was incubated with antibodies for p65 and p50 for 30 min before addition of the radiolabeled oligonucleotide.
|
B DNA binding but does not alter NF-B subunit composition.
HMGB1 pretreatment inhibits LPS-mediated degradation of IB
Inhibitor of B kinase phosphorylates IB, which is subsequently ubiquitinated and degraded by the 26 S proteasome. As a next step, we wanted to determine if HMGB1 pretreatment inhibits phosphorylation and subsequent degradation of IB. HMGB1 pretreatment significantly decreased LPS-mediated IB phosphorylation compared with cells treated with LPS alone (Fig. 6 A and B
).
 View larger version (23K): [in a new window] |
Figure 6. (A) Representative autoradiograph of Western blots analysis for IB, demonstrating the effect of HMGB1 preconditioning on IB phosphorylation following LPS stimulation. Control cells were maintained in basal growth medium. THP-1 cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS (1 µg/ml) for 3 min. The gel is representative of three experiments with similar results. (B) Image analysis of IB phosphorylation determined by densitometry. Increase calculated relative to IB phosphorylation in THP-1 cells treated with media alone (*, P<0.05, vs. LPS). (C) Representative autoradiograph of Western blot analysis for IB, demonstrating the effect of HMGB1 preconditioning on IB degradation following LPS stimulation. Control cells were maintained in basal growth medium. THP-1 cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS (1 µg/ml) for 2 and 7 min. The gel is representative of three experiments with similar results. (D) Image analysis of IB content determined by densitometry. Decrease calculated as percentage of THP-1 cells treated with media alone (*, P<0.05, vs. control; #, P<0.05, vs. LPS).
|
B, we examined if this protection extends to the subsequent LPS-induced IB degradation. In a time-course analysis, we noted that upon exposure to LPS, IB degradation is present as early as 2 min, and significant degradation is present by 7 min after LPS addition (Fig. 6 C
, Lane 4, and D
). Cells pretreated with HMGB1 demonstrated minimal-to-no degradation at 2 min after LPS addition (Fig. 6 C
, Lane 3, and D
) as compared with THP-1 cells exposed to LPS alone (Fig. 6 C
, Lane 2, and D
). Similarly, HMGB1 preconditioning resulted in significant attenuation of LPS-mediated IB degradation (Fig. 6 C
, Lane 5, and D
) as compared with cells treated with LPS alone at 7 min after LPS addition (Fig. 6 C
, Lane 4, and D
). Furthermore, we noted that exposure to LPS resulted in complete IB degradation at 30 and 45 min after LPS addition in HMGB1-pretreated and untreated cells (data not shown). Together, these data suggest that HMGB1 preconditioning decreases LPS-mediated IB phosphorylation and degradation early in the course after LPS addition.
HMGB1 pretreatment decreases serum TNF levels and hepatic NF-B activation in mice
To further verify our in vitro data, we treated mice with i.v. HMGB1 (20 µg) and subsequently treated with LPS (10 mg/kg). Serum TNF levels were measured 3 h after LPS treatment. Animals treated with 20 µg HMGB1 had significantly decreased levels of serum TNF following exposure to LPS when compared with animals treated with LPS alone (2990±300 vs. 4800±420 pg/ml, P<0.05; Fig. 7
). To further confirm the role of NF-B activation in LPS tolerance, EMSA was performed to examine NF-B DNA binding in hepatic tissue nuclear extracts. HMGB1-preconditioned animals demonstrated significant attenuation of NF-B DNA binding after LPS administration when compared with animals that were treated with LPS alone (Fig. 8
).
 View larger version (8K): [in a new window] |
Figure 7. ELISA results demonstrating the effects of HMGB1 pretreatment on serum TNF levels in mice, which were preconditioned with HMGB1 (20 µg) for 1 h before injecting LPS 10 mg/kg. Serum TNF levels were measured 3 h after injecting LPS in mice pretreated with 20 µg HMGB1 or vehicle. Data represent mean ± SE; n = 8 mice per group (*, P<0.05, vs. mice subjected to vehicle PBS).
|
 View larger version (51K): [in a new window] |
Figure 8. Representative autoradiograph of EMSA illustrating DNA binding of hepatic NF-B in two mice preconditioned with HMGB1. Mice were pretreated with HMGB1 (20 µg) or vehicle PBS 1 h, followed by injection of LPS 10 mg/kg. Three hours after LPS injections, murine hepatic tissue was harvested, and nuclear extracts were subjected to EMSA. Assay shown is representative of three experiments with similar results.
|
Bone marrow-derived macrophages were harvested from C57Bl/6 wild-type mice, TLR2–/–, and RAGE–/– mice, pretreated with HMGB1 for 18 h and then stimulated with LPS. Supernatant was collected for TNF detection. Consistent with our earlier results, following LPS stimulation, wild-type macrophages demonstrated a significant decrease in supernatant TNF levels as compared with untreated macrophages (946±336 vs. 1203±411 pg/ml, P<0.05; Fig. 9 ). In contrast, HMGB1 pretreatment failed to induce LPS tolerance in RAGE–/– macrophages (1297±226 vs. 1388±270 pg/ml; Fig. 9 ). This suggests that RAGE may potentially be a key receptor for induction of LPS tolerance. HMGB1-pretreated TLR2–/– bone marrow macrophages demonstrated a slight decrease in TNF secretion as compared with LPS-stimulated macrophages without preconditioning (851±366 vs. 897±322 pg/ml). Further studies are needed to confirm our results and verify the role of TLR2 in LPS tolerance induction by HMGB1.
 View larger version (11K): [in a new window] |
Figure 9. TNF ELISA demonstrating the effect of HMGB1 pretreatment ex vivo in bone marrow-derived macrophages from C57Bl/6, RAGE–/–, and TLR2–/– mice. Cells were preconditioned with HMGB1 (1 µg/ml) for 18 h before the addition of LPS 1 µg/ml for 4 h. Data represent the mean ± SEM of four separate experiments with each condition performed in triplicate (*, P<0.05, vs. LPS alone).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-helical segments that are important for DNA binding [21
22
]. The proinflammatory activities of HMGB1 are primarily attributed to the B box [23
]. Increased serum HMGB1 levels have been reported in patients with severe sepsis, septic shock [12
, 24
], and hemorrhagic shock [25
]. All of these clinical studies indicate that extracellular HMGB1 levels are increased in diverse inflammatory conditions. HMGB1 is viewed as a potential therapeutic target, as it appears much later than classical "alarmin phase" cytokines in response to the LPS-mediated inflammatory process [9
26
27
]. HMGB1 is actively released by activated macrophages and monocytes, subsequently leading to the release of proinflammatory cytokines from immune cells [10 ]. Depending on the cell type, HMGB1 is also passively released by necrotic and apoptotic cells [28 29 30 ]. The release of HMGB1 in inflammation supports the notion of HMGB1 as an endogenous danger signal alerting the immune system to the presence of inflammation and necrosis. We extend this hypothesis one step further and propose that extracellular HMGB1 modulates the innate immune system by inducing LPS tolerance.
Previously, it has been shown that upon pretreatment with endotoxin, cells demonstrate a reduced response to subsequent LPS challenges, a phenomenon referred to as LPS tolerance, which has also been described in vivo with reduced LPS lethality in mice pretreated with low-dose LPS [31
32
]. Tolerance is not a global down-regulation of signaling pathways but a reprogramming of the cellular signaling pathways [33
]. LPS-tolerant monocytes/macrophages demonstrate impaired activation of intracellular signaling pathways (e.g., NF-B) and subsequent, decreased, proinflammatory gene transcription and protein production, including TNF-, IL-β, and IL-6. These data have been corroborated in vitro with a reduced capacity of monocytes isolated from septic patients to release proinflammatory cytokines in response to further LPS stimulation [34
35
36
]. In this study, we report the novel finding that preconditioning with HMGB1, at a concentration that does not appear to activate NF-B, induces tolerance to LPS in THP-1 cells. Furthermore, we extend our observations in HMGB1-pretreated mice that demonstrate decreased, hepatic NF-B activation and serum TNF levels as compared with mice that were treated with LPS alone.
Ziegler-Heitbrock and colleagues [33
] previously suggested enhanced expression of transcriptionally inert p50 NF-B homodimer and reduction in p50/p65 NF-B heterodimer expression as a possible mechanism for endotoxin tolerance induction in an in vitro LPS-preconditioning model. We tested this hypothesis in LPS-tolerance induction with HMGB1 preconditioning. Although HMGB1 pretreatment decreased LPS-mediated NF-B activation, there was no change in the p50/p65 heterodimer composition as evaluated by supershift analyses. HMGB1 pretreatment also preserved the NF-B inhibitory protein IB as compared with LPS-mediated, marked degradation of IB.
Decreased TNF production is a marker for endotoxin tolerance; hence, we examined TNF levels in our experimental paradigm. We demonstrate decreased serum TNF levels in mice pretreated with HMGB1. These data suggest that extracellular HMGB1 exercises a unique, functional role by modulating innate immunity.
A common criticism of recombinant proteins is the potential of endotoxin contamination and its role as a confounding factor in cell activation. It has been shown that highly purified HMGB1 has minimal cytokine-like activity in vitro in contrast to E. coli-derived recombinant HMGB1 that exhibits potent proinflammatory activity [34 35 ]. It has been suggested that HMGB1 binds to substances derived from microbes and/or injured tissues, creating complexes that modulate the innate immune response by mediating proinflammatory cytokine release. In our study, endotoxin could not be detected by the standard Limulus lysate endotoxin assay, suggesting that it was below the threshold of detection. Our tolerance experiments using boiled HMGB1 and simultaneous addition of HMGB1 antibody along with HMGB1 affirm that induction of the LPS tolerance is a property of HMGB1 and not endotoxin contamination.
Since the first description of endotoxin tolerance [1 ], there has been immense curiosity about the potential pathways involved in mediating endotoxin tolerance. Debate continues about the benefits of inducing tolerance in patients admitted with diverse inflammatory states. Although endotoxin tolerance decreases the severity of infections and ischemia/reperfusion damage, conversely, it can be responsible for diminished responsiveness to repeated bacterial challenge, leading to increased sensitivity to nosocomial infections [40 ]. Although tremendous progress has been made, the precise pathway for inducing LPS tolerance is also unclear.
It has been suggested that at least three distinct receptors, i.e., TLR2, TLR4, and RAGE, are involved in HMGB1-induced, proinflammatory activity [41 ]. The contribution of these receptors in cellular activation following HMGB1 stimulation continues to be a work in progress and unclear at the present time. Our data about bone marrow-derived macrophages suggest that RAGE may play a key role in induction of LPS tolerance, as we were unable to induce tolerance in RAGE-deficient macrophages. The RAGE receptor has been suggested as the major receptor for HMGB1 in rodent macrophages. Kokkola et al. demonstrated that HMGB1 stimulation leads to significantly decreased secretion of TNF, IL-1β, and IL-6 in RAGE-deficient macrophages compared with wild-type mice [42 ]. Hence, it is possible that absence of RAGE precludes induction of LPS tolerance with HMGB1 preconditioning. In contrast, coimmunoprecipitation studies done by Park et al. suggest that there is no interaction between HMGB1 and RAGE [43 ]. These authors suggest that similar to LPS, HMGB1 activates the proinflammatory process as a result of its early interaction with TLR4 and TLR2. Our data from TLR2-deficient macrophages are inconclusive, and further work needs to be done to delineate the role of TLR2 in induction of LPS tolerance.
In conclusion, we have demonstrated that HMGB1 pretreatment can induce LPS tolerance using in vitro and in vivo experiments. Potential mechanisms involved in HMGB1-mediated LPS tolerance include decreased phosphorylation of IB, cytosolic preservation of IB, and decreased NF-B activation. RAGE plays a key role in induction of LPS tolerance by HMGB1. Our data also suggest that LPS contamination of the HMGB1 preparation does not account for our observations.
Received January 4, 2008; revised June 3, 2008; accepted July 9, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B kinase activation and IL-8 gene expression in respiratory epithelium Inflammation 26,233-241[CrossRef][Medline] B with predominance of p50 homodimers J. Biol. Chem. 269,17001-17004This article has been cited by other articles:
 |
|
 |
|
  K. Tsoyi, T. Y. Lee, Y. S. Lee, H. J. Kim, H. G. Seo, J. H. Lee, and K. C. Chang Heme-Oxygenase-1 Induction and Carbon Monoxide-Releasing Molecule Inhibit Lipopolysaccharide (LPS)-Induced High-Mobility Group Box 1 Release in Vitro and Improve Survival of Mice in LPS- and Cecal Ligation and Puncture-Induced Sepsis Model in Vivo Mol. Pharmacol., July 1, 2009; 76(1): 173 - 182. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
