LPS is recognized by a heterodimer consisting of TLR4 and its coreceptor MD-2. LPS signal causes excessive inflammation and tissue damage. In this study, we show that a mAb to TLR4/MD-2 protected mice from acute lethal hepatitis caused by LPS/d-galactosamine. The protective effect of the mAb was not due to inhibition of LPS response, because serum TNF-α, which was induced by LPS and caused lethal hepatitis, was 10 times up-regulated by the mAb pretreatment. Moreover, this mAb induced antiapoptotic genes in liver in a TLR4/MD-2-dependent manner. These results demonstrated that an agonistic mAb to TLR4/MD-2 protected mice from LPS/d-galactosamine-induced acute lethal hepatitis by delivering a protective signal activating NF-κB through TLR4/MD-2.

Innate immunity is the first line of defense against bacterial infection (1). The Toll family of receptors plays an essential role in innate recognition of microbial products (2). LPS is one of the most immunostimulatory glycolipids constituting the outer membrane of the Gram-negative bacteria (3, 4, 5, 6). LPS in the membrane poorly activates the immune system. LPS is transferred to CD14 by a lipid transferase LPS-binding protein (7, 8). LPS-CD14 stimulates TLR4/MD-2 on the cell surface (9). TLR4 is a type I transmembrane protein that contains large, leucine-rich repeats (LRR)3 in the extracellular region and a Toll/IL-1R homology domain in the cytoplasmic region (3). A coreceptor MD-2 is an extracellular molecule that is associated with the extracellular domain of TLR4, and is indispensable for cell surface expression and LPS recognition by TLR4 (9, 10, 11, 12).

Because LPS is one of the most immunostimulatory components of bacteria, LPS has been implicated in a variety of inflammatory diseases. LPS induces production of a variety of proinflammatory cytokines such as TNF-α, which leads to tissue damages and organ failure by mediating inflammation and apoptotic cell death. LPS challenge in mice can induce pathological reactions, including fever, hypotension, leukocyte infiltration, and inflammation in various organs, resulting in a syndrome resembling septic shock with a high mortality. d-Galactosamine (d-GalN) increases the susceptibility of mice to LPS-induced shock by impairing liver metabolism (13). Liver is a major target organ after challenge with low doses of LPS in conjunction with d-GalN. Massive apoptosis in the liver is caused by LPS/d-GalN and leads to lethal tissue damages. TNF-α plays a central role in low dose LPS-induced lethal acute hepatitis.

Given that LPS is recognized by TLR4/MD-2, TLR4/MD-2 is a target molecule for therapeutic intervention in LPS-induced tissue damages and organ failure. It is, therefore, interesting to study the effect of the mAbs to TLR4/MD-2 on LPS/d-GalN-induced acute lethal hepatic failure. In the present study, we demonstrated that a mAb to TLR4/MD-2 protected mice from acute lethal hepatitis induced by LPS/d-GalN not by inhibiting LPS signal, but by delivering a protective signal through TLR4/MD-2.

LPS from Escherichia coli 055:B5, lipid A purified from Salmonella minnesota, and d-GalN were purchased from Sigma-Aldrich. BALB/c, C3H/HeN, C3H/HeJ, and C57BL/6 mice were purchased from Japan SLC and used at 7–10 wk old. MD-2−/− mice were established in our laboratory (10). The experiments were conducted according to institutional ethical guidelines for animal experiments and safety guidelines for gene manipulation experiments at the University of Tokyo.

The cDNAs encoding the extracellular and transmembrane portions of mouse TLR4, human TLR4, and mouse human chimeric TLR4 (human TLR4 (1–286) and mouse TLR4 (286–661)) were followed by the cDNA encoding GFP. The constructed cDNAs were cloned into the pCAGGS vector (14). The cDNA encoding another chimeric TLR4 (mouse TLR4 (1–285) and human TLR4 (287–663)) was cloned into the pEGFP vector (BD Clontech). The cDNA encoding mouse MD-2 tagged with the flag epitope at the C terminus was cloned into the pEFBOS vector (15).

The human kidney cell line 293T was plated onto a 6-well plate at 5 × 105 cells/well the day before transfection. Plasmid DNA (total 2 μg/well) and LipofectAMINE 2000 dissolved in Opti-MEM (10 μl/well; Invitrogen Life Technologies) were added to 293T cells. The cells were incubated at 37°C in a CO2 incubator for 2 days with the medium replaced on the day after transfection. Cells were harvested with PBS containing 0.2 mM EDTA and used for FACS analysis.

Single-cell suspensions were incubated at 2 × 105 cells/100 μl on ice for 15 min with the biotinylated rat anti-mouse TLR4/MD-2 mAbs (MTS510, Sa15-21) diluted in staining buffer (PBS containing 2.5% FCS and 0.1% NaN3). Cells were washed in the staining buffer and incubated with R-PE-conjugated streptavidin (BD Pharmingen) for 15 min. After washing, cells were analyzed on a FACSCalibur system (BD Immunocytometry Systems).

BALB/c mice weighing 16–21 g were injected i.p. with 250 μl of PBS containing 100 μg of mAb, followed by another i.p. injection of 250 μl of PBS containing 500 ng of LPS and 25 mg of d-GalN.

Tissues were fixed in 10% Formalin and embedded in paraffin, and 3-μm sections were used for histopathological examinations. Sections were stained with H&E for histopathological and morphological analysis.

Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were measured using a standard clinical automatic analyzer.

The TUNEL assay was performed using Promega DeadEnd Fluorometric TUNEL System, according to the manufacturer’s protocol. Briefly, sections were treated with proteinase K (20 μg/ml) for 8–10 min, washed, and incubated with a mixture of equilibration buffer, TdT, and fluorescein-12-dUTP-labeled DNA mix. This sample was analyzed with a microscopy (Carl Zeiss).

The levels of serum TNF-α were determined by ELISA (BioSource International).

Total RNA was isolated from the liver using an ISOGEN (Nippon Gene), according to the manufacturer’s instruction. For RT-PCR, total RNA (1 μg) was primed with a random primer and cDNA was synthesized with Revertra-Ash (Toyoba). The following PCR primers were used: A1α, forward, 5′-TGGCATCATTAACTGGGGAAG-3′, reverse, 5′-CTCTCTGGTCCGTAGTGTTACTTG-3′; A20, forward, 5′-TCGTGGCTCTGAAAACCAATG-3′, reverse, 5′-GATGGGTCTTCTGAGGATGTTGC-3′; Gadd45β, forward, 5′-CTTCTGGTCGCACGGGAAGG-3′, reverse, 5′-GCTCCACCGCGGCAGTCACC-3′; Bcl-2, forward, 5′-GCTACGAGTGGGATGCTGG-3′, reverse, 5′-GTGTGCAGATGCCGGTTCA-3′; Bcl-xL, forward, 5′-AGGATACAGCTGGAGTCAG-3′, reverse, 5′-TCTCCTTGTCTACGCTTTCC-3′; TNFR1, forward, GAACCTACTTGGTGAGTGAC-3′, reverse, 5′-CACAACTTCATACACTCCTC-3′; hypoxanthine phosphoribosyltransferase (HPRT), forward, 5′-TGATGAACCAGGTTATGACCTAG-3′, reverse, 5′-CCAGCAAGCTTGCAACCTTAACC-3′. Amplification conditions using a Biometra thermal cycler were 94°C for 1 min, 60°C for 2 min (Gadd45β: 65°C), and 72°C for 2 min for 1 cycle, and 94°C for 0.5 min, 60°C for 2 min (Gadd45β: 65°C), and 72°C for 2 min for 30 cycles. The PCR products were separated by electrophoresis on a 1.5–2% agarose gel and visualized by ethidium bromide staining.

Total RNA (30 μg/lane) was resolved by electrophoresis in a 1% agarose gel containing 2.2 M formaldehyde, and transferred to a Hybond-N+ membrane (Amersham Biosciences). The RNA was cross-linked to the membrane with UV cross-linker (Nippon Genetics) and then hybridized with a 32P-labeled cDNA probe encoding mouse A20 (from aa 92 to aa 207). The probe was labeled using a random primed labeling system (Rediprime II; Amersham Biosciences). Hybridization was conducted at 65°C for 16 h in a buffer containing 10% dextran sulfate, 1% SDS, 50 mM Tris-HCl (pH 7.5), 1 M NaCl, and denatured salmon sperm DNA (100 μg/ml). Blots were washed twice in 2× SSC/0.1% SDS at 65°C and three times in 0.2× SSC/0.1% SDS for 10 min at 65°C before exposure to imaging plate (Fuji film).

BALB/c mice were injected i.p. with MTS510 or Sa15-21 mAbs (100 μg/mice), and livers were collected 1 h after injection. The collected livers were instantly frozen and sonicated for total RNA extraction in ISOGEN (Nippon Gene) or RNeasy mini kit (Qiagen). The extracted liver RNAs were analyzed with the Affymetrix gene chip probe array of Mouse Genome 430 2.0.

Liver tissue was suspended in the lysis buffer containing 150 mM NaCl, 50 mM Tris-HCl (pH 7.4), 5 mM EDTA, 10 μg/ml aprotinin, 10 μg/ml leupeptin, 1 mM PMSF, 50 mM iodoacetamide, and 1% Triton X-100, and homogenized with a hand homogenizer (DREMEL). After 30 min of incubation on ice, the lysate was centrifuged and the supernatant was collected. mAb-coupled beads were added to the supernatant and rotated for 2 h at 4°C. Beads were washed in diluted lysis buffer, and bound proteins were subjected to SDS-PAGE and immunoprobing with anti-TLR4 Ab. The second reagent was alkaline phosphatase-conjugated goat anti-rat IgG (American Qualex).

NF-κB activity in the liver was determined with the TransAM NF-κB (Active Motif). BALB/c mice were injected i.p. with MTS510 (100 μg/mouse), Sa15-21 (100 μg/mouse), or lipid A (500 ng/mouse). Livers were removed 0, 0.5, 1, 2, and 3 h after injection of mAb or lipid A, and nuclear extract was prepared with TransAM nuclear extract kit (Active Motif). The NF-κB activity in the nuclear extract was determined by detecting the binding of p65 to the immobilized oligonucleotide containing the NF-κB consensus.

We previously established two mAbs to mouse TLR4/MD-2, Sa15-21 and MTS510. Sa15-21 reacted with the ligand receptor complex consisting of LPS/TLR4/MD-2, but MTS510 did not (16). Cross blocking studies demonstrated that the epitopes recognized with the two mAbs were distinct from each other (data not shown). To locate the epitopes on the TLR4 molecule, we took the advantage that these mAbs did not react with human TLR4. We made two plasmids encoding chimeric TLR4 consisting of mouse and human TLR4 (Fig. 1,a). The extracellular domain of mouse TLR4 has the 81-aa region (aa 286–366) that lacks LRR and hypervariable across species (17). The 81-aa region divides the rest of the extracellular domain into N-terminal LRR (N-LRR: aa 1–285) and membrane-proximal LRR. We made two constructs encoding chimeric protein consisting of mouse N-LRR followed by human TLR4, or of human N-LRR followed by mouse TLR4 (Fig. 1 a). These constructs and mouse MD-2 cDNA were transfected into 293T cells, and cell surface TLR4/MD-2 was stained with MTS510 and Sa15-21. Sa15-21 reacted with the mouse N-LRR, whereas MTS510 reacted with the rest of the extracellular domain of mouse TLR4, respectively.

FIGURE 1.

Characterization of two distinct mAbs to TLR4/MD-2, Sa15-21 and MTS510. a, A human kidney cell line, 293T, was transfected with expression vectors encoding the indicated molecules in conjunction with mouse MD-2. Harvested cells were stained with biotinylated MTS510 or Sa15-21, as indicated, followed by streptavidin-PE. Open histograms depicted the samples stained with the second reagent alone. The right scheme stands for the extracellular domain of mouse (▪), human (□), and chimeric TLR4. b, Mouse macrophage cell line RAW264.7 was preincubated with indicated mAbs (20 μg/ml) for 30 min and stimulated with lipid A (1 ng/ml) for 24 h. Concentrations of TNF-α in the culture supernatants were determined by ELISA. Results are shown as the mean value with SEs from triplicate wells.

FIGURE 1.

Characterization of two distinct mAbs to TLR4/MD-2, Sa15-21 and MTS510. a, A human kidney cell line, 293T, was transfected with expression vectors encoding the indicated molecules in conjunction with mouse MD-2. Harvested cells were stained with biotinylated MTS510 or Sa15-21, as indicated, followed by streptavidin-PE. Open histograms depicted the samples stained with the second reagent alone. The right scheme stands for the extracellular domain of mouse (▪), human (□), and chimeric TLR4. b, Mouse macrophage cell line RAW264.7 was preincubated with indicated mAbs (20 μg/ml) for 30 min and stimulated with lipid A (1 ng/ml) for 24 h. Concentrations of TNF-α in the culture supernatants were determined by ELISA. Results are shown as the mean value with SEs from triplicate wells.

Close modal

We further compared these anti-TLR4 mAbs with regard to their effect on LPS response. A mouse macrophage cell line RAW264.7 was pretreated with the mAbs and stimulated with lipid A (Fig. 1 b). Concentrations of TNF-α in supernatants were determined by ELISA. The mAb alone did not stimulate TNF-α production. Lipid A-dependent TNF-α production was appreciably inhibited by MTS510, but not by Sa15-21. Similar results were obtained in lipid A-induced B cell proliferation (data not shown).

Because TLR4/MD-2 was a target molecule for therapeutic intervention in endotoxin shock, we examined the effect of the anti-TLR4/MD-2 mAbs on endotoxin shock induced by LPS and d-GalN. Two hours before injection of LPS/d-GalN, mice were pretreated with PBS, MTS510, Sa15-21, or anti-CD14 mAb. Sa15-21-pretreated mice survived LPS/d-GalN challenge, whereas other pretreatments had no effect (Fig. 2,a). The protective effect of Sa15-21 was dose dependent, and as little as 10 μg of Sa15-21 mAb was still able to rescue more than half of mice (Fig. 2,b). Whereas the protective effect of Sa15-21 was not observed when Sa15-21 was injected simultaneously or subsequently with LPS/d-GalN, preceding injection 12 h before LPS/d-GalN still rescued half of mice (Fig. 2 c). We also examined the effect of Sa15-21 in bacterial infection. Sa15-21 was not able to protect mice from death caused by infection with Salmonella typhymurium infection (data not shown). We decided to focus on the experiment using LPS/d-GalN.

FIGURE 2.

An anti-TLR4 mAb Sa15-21 protected mice from death caused by LPS/d-GalN injection. a, Mice (n = 10 for each group) were injected i.p. with PBS or mAbs indicated in the figure (100 μg/mouse) 2 h before the injection of LPS (500 ng/mouse)/d-GalN (25 mg/mouse). Mice survival in the indicated times after LPS/d-GalN injection was shown. b, Mice (n = 5 for each group) were injected i.p. with the graded doses of Sa15-21 mAb 2 h before the injection of LPS (500 ng)/d-GalN (25 mg). Mice survival in the indicated times after LPS/d-GalN injection was shown. c, Mice (n = 10 for each group) were injected i.p. with Sa15-21 mAb (100 μg/mouse) at the indicated times and LPS (500 ng/mouse)/d-GalN (25 mg/mouse) at 0 h. Mice survival in the indicated times after LPS/d-GalN injection was shown.

FIGURE 2.

An anti-TLR4 mAb Sa15-21 protected mice from death caused by LPS/d-GalN injection. a, Mice (n = 10 for each group) were injected i.p. with PBS or mAbs indicated in the figure (100 μg/mouse) 2 h before the injection of LPS (500 ng/mouse)/d-GalN (25 mg/mouse). Mice survival in the indicated times after LPS/d-GalN injection was shown. b, Mice (n = 5 for each group) were injected i.p. with the graded doses of Sa15-21 mAb 2 h before the injection of LPS (500 ng)/d-GalN (25 mg). Mice survival in the indicated times after LPS/d-GalN injection was shown. c, Mice (n = 10 for each group) were injected i.p. with Sa15-21 mAb (100 μg/mouse) at the indicated times and LPS (500 ng/mouse)/d-GalN (25 mg/mouse) at 0 h. Mice survival in the indicated times after LPS/d-GalN injection was shown.

Close modal

In contrast to high dose LPS-induced shock, which induces a systemic disorder including multiple organ failure, liver is a major target organ after challenge with low doses of LPS in conjunction with d-GalN (13). We therefore examined livers from mice injected with LPS/d-GalN. Massive hemorrhage and tissue damage were apparent macroscopically and microscopically (Fig. 3, a and b). In sharp contrast, a liver from Sa15-21-treated mice was indistinguishable from normal liver. Liver damage can be estimated by elevated serum concentrations of liver-specific enzymes, AST and ALT. Serum AST and ALT levels were increased 8 h after LPS/d-GalN injection in mice pretreated with PBS, MTS510, and anti-CD14 mAb, but not in mice pretreated with Sa15-21 (Fig. 3,c). Massive hepatocyte apoptosis has been shown to cause liver damage in this model system using LPS/d-GalN. We examined hepatocyte apoptosis by TUNEL staining (Fig. 3 d). Six hours after LPS/d-GalN injection, apoptotic death of hepatocytes was apparent in mice pretreated with PBS or MTS510 mAb, but much less in mice pretreated with Sa15-21.

FIGURE 3.

Sa15–21 inhibited hepatocyte apoptosis induced by LPS/D-GalN injection. a and b, Mice were injected i.p. with PBS (b, c, i, and j), MTS510 (d, e, k, and l), or Sa15–21 (f, g, m, and n) 2 h before the i.p. injection of LPS (500 ng)/d-GalN (25 mg). Livers were removed 6 or 9 h after LPS/d-GalN injection, and macroscopic and microscopic appearances (×150) were shown in conjunction with untreated normal liver (a and h). c, Mice (n = 5) were injected i.p. with PBS, or indicated mAbs (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. Sera were collected at 8 h after LPS/d-GalN injection, and the activities of AST and ALT were determined and expressed as mean values with SEs. d, Mice were injected with PBS (p), MTS510 (q), or Sa15–21 (r) (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. DNA cleavage during apoptosis was detected by TUNEL assay in normal liver (o) and livers that were treated as above and removed at 6 h after LPS/d-GalN injection. White bars: 50 μm.

FIGURE 3.

Sa15–21 inhibited hepatocyte apoptosis induced by LPS/D-GalN injection. a and b, Mice were injected i.p. with PBS (b, c, i, and j), MTS510 (d, e, k, and l), or Sa15–21 (f, g, m, and n) 2 h before the i.p. injection of LPS (500 ng)/d-GalN (25 mg). Livers were removed 6 or 9 h after LPS/d-GalN injection, and macroscopic and microscopic appearances (×150) were shown in conjunction with untreated normal liver (a and h). c, Mice (n = 5) were injected i.p. with PBS, or indicated mAbs (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. Sera were collected at 8 h after LPS/d-GalN injection, and the activities of AST and ALT were determined and expressed as mean values with SEs. d, Mice were injected with PBS (p), MTS510 (q), or Sa15–21 (r) (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. DNA cleavage during apoptosis was detected by TUNEL assay in normal liver (o) and livers that were treated as above and removed at 6 h after LPS/d-GalN injection. White bars: 50 μm.

Close modal

TNF-α induced by LPS stimulation was shown to cause hepatocyte apoptosis in mice injected with LPS/d-GalN (13). Concentrations of TNF-α were determined by ELISA (Fig. 4,a). TNF-α production was detected in response to LPS. No inhibitory effect was observed in mice pretreated with MTS510 or anti-CD14 mAb. Surprisingly, TNF-α in the sera was increased by 10-fold in mice pretreated with Sa15-21 when compared with mice pretreated with PBS. We could not see such up-regulation with a macrophage cell line RAW264 (Fig. 1 b).

FIGURE 4.

Sa15-21 pretreatment up-regulated LPS-induced TNF-α production and protected mice from death due to TNF-α/d-GalN injection. a, BALB/c mice (n = 5 for each group) were injected i.p. with PBS or indicated mAbs (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. Blood was collected 1 h after LPS/d-GalN. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SE. b, Mice were injected i.p. with PBS or indicated mAbs (100 μg/mouse) 3.5 h before TNF-α (500 ng)/d-GalN (25 mg) injection. The numbers of surviving mice are shown as a function of elapsed time.

FIGURE 4.

Sa15-21 pretreatment up-regulated LPS-induced TNF-α production and protected mice from death due to TNF-α/d-GalN injection. a, BALB/c mice (n = 5 for each group) were injected i.p. with PBS or indicated mAbs (100 μg/mouse) 2 h before LPS (500 ng)/d-GalN (25 mg) injection. Blood was collected 1 h after LPS/d-GalN. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SE. b, Mice were injected i.p. with PBS or indicated mAbs (100 μg/mouse) 3.5 h before TNF-α (500 ng)/d-GalN (25 mg) injection. The numbers of surviving mice are shown as a function of elapsed time.

Close modal

Sa15-21-pretreated mice survived LPS/d-GalN challenge despite a high concentration of TNF-α. To ask whether Sa15-21 made mice resistant to such a high concentration of serum TNF-α, acute lethal hepatitis was induced with TNF-α in conjunction with d-GalN. Sa15-21 pretreatment was still able to rescue mice from death caused by TNF-α/d-GalN (Fig. 4 b).

It is possible that d-GalN has an influence on TNF-α up-regulation by Sa15-21. To exclude this possibility, we used LPS alone. Mice were injected with LPS or Sa15-21, and the level of serum TNF-α was determined by ELISA (Fig. 5,a). TNF-α induction by Sa15-21 was minimal as compared with that by 500 ng of LPS injection. Next, mice were pretreated with Sa15-21 or LPS, and received LPS injection 2 h later. TNF-α production was drastically augmented by Sa15-21 pretreatment, but not by LPS pretreatment (Fig. 5 b). These results showed that Sa15-21 was able to augment LPS-mediated TNF-α production.

FIGURE 5.

Sa15-21 induced NF-κB production, but did not induce tolerance in the TNF-α production with LPS. a, BALB/c mice (n = 4 for each group) were injected i.p. with PBS or Sa15-21 (100 μg/mouse). Blood was collected from mouse heart 0, 1, 2, 3, and 4 h after injection. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SE. b, BALB/c mice (n = 4 for each group) were injected i.p. with Sa15–21 mAb (100 μg/mouse) or LPS (500 ng/mouse). Two hours later, LPS (500 ng/mouse) was injected. Blood was collected −2, 0, 1, 2, 3, and 4 h after LPS injection. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SD.

FIGURE 5.

Sa15-21 induced NF-κB production, but did not induce tolerance in the TNF-α production with LPS. a, BALB/c mice (n = 4 for each group) were injected i.p. with PBS or Sa15-21 (100 μg/mouse). Blood was collected from mouse heart 0, 1, 2, 3, and 4 h after injection. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SE. b, BALB/c mice (n = 4 for each group) were injected i.p. with Sa15–21 mAb (100 μg/mouse) or LPS (500 ng/mouse). Two hours later, LPS (500 ng/mouse) was injected. Blood was collected −2, 0, 1, 2, 3, and 4 h after LPS injection. Serum TNF-α was determined by ELISA. The results were represented as the mean value with SD.

Close modal

The protective effect of Sa15-21 was not specific for LPS, because Sa15-21 worked not only in LPS/d-GalN-, but also in TNF-α/d-GalN-induced hepatic failure. We asked a possibility that Sa15-21 injection induced antiapoptotic genes in the liver. We conducted the DNA microarray analysis to see the antiapoptotic genes induced by Sa15-21 injection. TNF-α-induced apoptosis has been reported to be inhibited by a number of genes, including cellular inhibitors of apoptosis (c-IAP1 and c-IAP2), the Bcl-2 family members (A1, Bcl-xL, and Bcl-2), a transcriptional repressor A20, and Gadd45β (18). mRNA expression of these antiapoptotic genes in the liver was compared between mice injected with MTS510 or Sa15-21 (Fig. 6,a). c-IAP1, Gadd45β, A1, and A20 were appreciably induced by Sa15-21 injection, whereas Bcl-xL, Bcl-2, and cIAP2 were not. These results were confirmed by RT-PCR (Fig. 6,b). Constitutive expression of Bcl-2 and Bcl-xL in the liver was not influenced by Sa15-21 injection. In contrast, A1, A20, and Gadd45β were not constitutively expressed, but drastically induced by Sa15-21 mAb. Sa15-21 did not influence mRNA expression of the type 1 receptor for TNF-α. Among these antiapoptotic genes, A20 was of note, because overexpression of A20 in hepatocytes was reported to protect mice from acute hepatitis caused by LPS/d-GalN (19). A20 induction by Sa15-21 was, therefore, further confirmed by Northern hybridization (Fig. 6 c). These results suggested that Sa15-21 induced antiapoptotic genes in the liver, and thereby inhibited TNF-α-induced hepatocyte apoptosis.

FIGURE 6.

Sa15–21-induced antiapoptotic genes. a, Microarray analyses were conducted using liver RNA from mouse that had been treated with Sa15–21 or MTS510. Expression of the indicated genes with Sa15-21 treatment was compared with that with MTS510. b and c, RT-PCR and Northern hybridization were conducted to reveal expression of the indicated mRNAs in livers from mice that had been injected with PBS, Sa15-21 (100 μg/mouse), or MTS510 (100 μg/mouse). For RT-PCR, total RNA was reverse transcribed with a random primer, followed by PCR with the primers specific for the indicated mRNA. HPRT was used as a positive control. Northern hybridization was conducted with a probe specific for A20. The amount of loaded total RNA was shown by ethidium bromide staining of 28S RNA (lower panel). d, Mice (n = 4 at 0 and 1 h; n = 2 at 0.5, 1, 2, and 3 h) were injected with mAb (100 μg/mouse) or lipid A (500 ng/mouse), and livers were removed after the indicated periods of time. The NF-κB activity in the liver was determined by ELISA. The results were represented as the mean value.

FIGURE 6.

Sa15–21-induced antiapoptotic genes. a, Microarray analyses were conducted using liver RNA from mouse that had been treated with Sa15–21 or MTS510. Expression of the indicated genes with Sa15-21 treatment was compared with that with MTS510. b and c, RT-PCR and Northern hybridization were conducted to reveal expression of the indicated mRNAs in livers from mice that had been injected with PBS, Sa15-21 (100 μg/mouse), or MTS510 (100 μg/mouse). For RT-PCR, total RNA was reverse transcribed with a random primer, followed by PCR with the primers specific for the indicated mRNA. HPRT was used as a positive control. Northern hybridization was conducted with a probe specific for A20. The amount of loaded total RNA was shown by ethidium bromide staining of 28S RNA (lower panel). d, Mice (n = 4 at 0 and 1 h; n = 2 at 0.5, 1, 2, and 3 h) were injected with mAb (100 μg/mouse) or lipid A (500 ng/mouse), and livers were removed after the indicated periods of time. The NF-κB activity in the liver was determined by ELISA. The results were represented as the mean value.

Close modal

The antiapoptotic genes induced by Sa15-21 were targets of NF-κB (18), suggesting that Sa15-21 activated NF-κB in the liver. To address this possibility, we examined the NF-κB activity in the liver after Sa15-21 injection. Mice were injected with Sa15-21, MTS510, or lipid A, and the activity of NF-κB in the liver 2 h after infection was determined by ELISA (Fig. 6,d). Sa15-21 injection activated NF-κB in the liver as much as lipid A, whereas MTS510 had no effect. Sa15-21 was similar to lipid A in the time course of NF-κB activation. Despite comparable NF-κB activation, Sa15-21 was distinct from lipid A in that Sa15-21 failed to induce TNF-α production and that Sa15-21 drastically augmented LPS-induced TNF-α production (Fig. 5).

Finally, we asked whether the effect of Sa15-21 was dependent on TLR4/MD-2. Expression of the TLR4/MD-2 complex in the liver and a mouse hepatocyte line was first confirmed by immunoprecipitation with the TLR4/MD-2-specific MTS510 mAb and immunoprobing of the TLR4 protein (Fig. 7,a). We next used a TLR4-mutant mouse, C3H/HeJ. Sa15-21 pretreatment was not able to induce an antiapoptotic gene A20 in the liver and did not protect from lethal hepatitis induced by TNF-α/d-GalN injection in C3H/HeJ mice, demonstrating that the protective effect of Sa15-21 is TLR4 dependent (Fig. 7, b and c). We also examined A20 induction by Sa15-21 in MD-2-deficient mice, and we were unable to detect A20 induction (Fig. 7 d). These results clearly demonstrated that Sa15-21 induced antiapoptotic genes through TLR4/MD-2.

FIGURE 7.

The protective effect of Sa15–21 is dependent on TLR4/MD-2. a, Immunoprecipitation of TLR4/MD-2 and immunoprobing of TLR4 were conducted with cell lysate from liver or a cell line derived from mouse liver. b, RT-PCR was conducted to detect A20 mRNA expressed in the liver from mice that had been treated with Sa15-21 mAb (100 μg/mouse). Primers for β-actin were used as a positive control. c, C3H/HeN and C3H/HeJ mice were i.p. injected with PBS or Sa15-21 (100 μg/mouse). TNF-α (500 ng/mouse)/d-GalN (25 mg/mouse) was injected 3.5 h later. The numbers of mice surviving the challenge are shown. d, Wild-type or MD-2-deficient mice (n = 2) were injected with Sa15–21 (100 μg/mouse), and livers were removed 2 h later. RT-PCR was conducted to detect A20 mRNA expressed in the liver. Primers for HPRT were used as a positive control.

FIGURE 7.

The protective effect of Sa15–21 is dependent on TLR4/MD-2. a, Immunoprecipitation of TLR4/MD-2 and immunoprobing of TLR4 were conducted with cell lysate from liver or a cell line derived from mouse liver. b, RT-PCR was conducted to detect A20 mRNA expressed in the liver from mice that had been treated with Sa15-21 mAb (100 μg/mouse). Primers for β-actin were used as a positive control. c, C3H/HeN and C3H/HeJ mice were i.p. injected with PBS or Sa15-21 (100 μg/mouse). TNF-α (500 ng/mouse)/d-GalN (25 mg/mouse) was injected 3.5 h later. The numbers of mice surviving the challenge are shown. d, Wild-type or MD-2-deficient mice (n = 2) were injected with Sa15–21 (100 μg/mouse), and livers were removed 2 h later. RT-PCR was conducted to detect A20 mRNA expressed in the liver. Primers for HPRT were used as a positive control.

Close modal

The present study demonstrated that an agonistic anti-TLR4/MD-2 mAb Sa15-21 protected mice from LPS/d-GalN-induced lethal hepatic failure. A previous report showed that a mAb to LPS-binding protein protected mice from LPS/d-GalN-induced endotoxin shock (20). The mAb to LPS-binding protein was able to protect mice at a lower dose of LPS (50 ng/mouse), but not at a higher dose of LPS (250 ng/mouse). In contrast, Sa15-21 was effective at an even higher dose of LPS (500 ng/mouse). Anti-CD14 mAb was not effective, as shown in Fig. 2. Sa15-21 is the most potent in protecting mice from LPS/d-GalN-induced lethal hepatitis among the mAbs to molecules mediating LPS processing/recognition. Recently, an antagonistic anti-TLR2 mAb was shown to be effective in preventing lethal shock-like syndrome induced by a TLR2 ligand (21). These results demonstrated that TLRs were novel targets for therapeutic intervention in lethal inflammatory damages caused by microbial stimuli.

The mAb Sa15-21 protected mice from LPS/d-GalN-induced hepatic failure by delivering an antiapoptotic signal through TLR4/MD-2 that inhibited TNF-α-induced apoptosis. The following results support this conclusion. First, serum TNF-α production induced by LPS/d-GalN was not inhibited, but augmented by the Sa15-21 pretreatment (Fig. 4,a). Second, the Sa15-21 pretreatment protected mice from lethal liver failure induced by TNF-α/d-GalN, demonstrating that the protective effect of Sa15-21 was not specific for LPS. Third, the protective effect of Sa15-21 was dependent on TLR4, because Sa15-21 failed to protect TLR4-mutant mice from fatal hepatitis induced by TNF-α/d-GalN (Fig. 7). Finally, Sa15-21 activated NF-κB in the liver and induced several antiapoptotic molecules in the liver (Fig. 6). c-IAP1, one of the best studied antiapoptotic proteins, was up-regulated by Sa15-21 treatment (22). Gadd45β, an inducible factor associated with cell cycle control and DNA repair, completely suppresses TNF-α-induced apoptosis in RelA−/− cells by inhibiting JNK signaling (23). A1 was cloned as a Bcl-2 homologue, and its overexpression prevents apoptosis mediated by TNF-α (24). A20, a zinc finger protein, was originally identified as a TNF-α-inducible gene product and later found to prevent endothelial cell apoptosis mediated by TNF-α, LPS, or the Fas ligand (25, 26). Taken together with the results that induction of an antiapoptotic gene A20 in the liver was dependent on TLR4 and MD-2 (Fig. 7, b and d), we conclude that Sa15-21 acted on TLR4/MD-2 and induced antiapoptotic genes in the liver.

Antiapoptotic genes induced by Sa15-21 are also inducible by TNF-α. Although Sa15-21 alone induced only a little amount of TNF-α in the serum (Fig. 5 a), it was still possible that a trace amount of TNF-α induced the antiapoptotic genes. To investigate whether TNF-α is required for induction of antiapoptotic genes, we used TNF-α−/− mice (27). Sa15-21 induced A20 expression in the liver of the TNF-α−/− mice (data not shown), demonstrating that the induction of antiapoptotic genes was not dependent on the TNF-α. It remains to be determined whether Sa15-21 acts directly on hepatocytes to induce antiapoptotic genes or indirectly through other cell lineages like Kupffer cells.

Among the antiapoptotic genes induced by Sa15-21, A20 is most intriguing, because adenoviral mediated hepatic expression of A20 protected mice from lethal hepatic failure induced by LPS/d-GalN (19). Moreover, A20 has been shown to have a role in termination of inflammatory responses triggered by TNF-α and LPS. The anti-inflammatory function of A20 is well documented in A20−/− mice, which are born cachectic and die within 3 wk of birth due to severe inflammation (28). Given that mice lacking A20 are highly susceptible to endotoxin shock (29), up-regulated expression of A20 by Sa15-21 is likely to be beneficial for mice recovering from endotoxin shock.

Whereas LPS pretreatment has been shown to induce endotoxin tolerance (30), an LPS-refractory state, in vivo, Sa15-21 did not induce endotoxin tolerance, but rather augmented LPS response. LPS-induced TNF-α production was not attenuated, but augmented by Sa15-21 pretreatment (Fig. 5,b). It is possible that a signaling pathway used by Sa15-21 is distinct from LPS signaling pathway. Whereas a macrophage line RAW264 produced TNF-α in response to lipid A (Fig. 1 b), Sa15-21 pretreatment was not able to enhance subsequent LPS-induced TNF-α production. In contrast, preincubation of Sa15-21 slightly augmented the LPS-induced TNF-α production in bone marrow-derived dendritic cells (DCs) (data not shown). TNF-α production was up-regulated with increasing concentrations of Sa15-21. MTS510 pretreatment did not augment LPS-induced TNF-α production. These results suggest that Sa15-21 directly acts on DCs and augments LPS-induced TNF- α production by DCs. DCs in the liver may be responsible for Sa15-21-mediated augmentation of LPS-induced TNF-α production.

Sa15-21 activated NF-κB in the liver as much as lipid A (Fig. 6,d), whereas Sa15-21 failed to induce TNF-α production and to induce tolerance to the following LPS stimulation (Fig. 5). These results may suggest a difference between LPS and Sa15-21 in triggering a signal through TLR4/MD-2. LPS binds to TLR4/MD-2 and induces potent homotypic interaction of the receptor-ligand complex, LPS/TLR4/MD-2 (31). Given that LPS is able to further stimulate TLR4/MD-2 that has been already ligated by Sa15-21, Sa15-21-mediated ligation would be distinct from LPS-mediated receptor clustering. Whereas the MTS510 epitope disappeared by LPS stimulation, the Sa15-21 epitope was not influenced (16). The N-terminal half of the extracellular domain of TLR4 might not be important for LPS binding and subsequent homotypic interaction. It is important to understand a difference of the TLR4/MD-2 ligation by Sa15-21 from LPS-mediated receptor oligomerization with regard to downstream signaling pathways. Although Sa15-21-mediated ligation of TLR4/MD-2 is not a physiological situation, there might be an Sa15-21-like ligand, because a variety of TLR4/MD-2 ligands other than LPS was reported, such as the fusion protein from respiratory syncytial virus (32). In this regard, it is of note that the preceding adenoviral infection enhanced TNF-α production and protected mice from lethal hepatic failure induced by LPS/d-GalN (33).

It should be stressed that up-regulated TNF-α is not necessarily deleterious, but can be beneficial for liver regeneration, because TNF-α is a comitogen required for hepatocyte proliferation (34). Importantly, Sa15-21 up-regulated A20 in the liver that inhibited TNF-α-mediated apoptosis, but allowed for TNF-α-mediated liver regeneration (19). Given that LPS induces A20 and A1 also in endothelial cells (26), Sa15-21 might be able to act on endothelial cells as well and inhibit endothelial cell apoptosis that has an important role in ischemic/reperfusion injury.

In conclusion, we demonstrated a unique agonistic mAb to TLR4/MD-2 that protected mice from acute lethal liver failure caused by LPS/d-GalN.

We thank Kaori Tomita, Kazumi Shinoda, and Yuji Motoi for technical assistance.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This study was supported by Special Coordination Funds of the Ministry of Education, Culture, Sports, Science, and Technology of the Japanese Government; Uehara Memorial Foundation; the Takeda Foundation; the Naito Foundation; Mochida Memorial Foundaion for Medical and Pharmaceutical Research; and Sankyo.

3

Abbreviations used in this paper: LRR, leucine-rich repeat; ALT, alanine aminotransferase; AST, aspartate aminotransferase; c-IAP, cellular inhibitor of apoptosis; d-GalN, d-galactosamine; DC, dendritic cell; HPRT, hypoxanthine phosphoribosyltransferase; N-LRR, N-terminal LRR.

1
Janeway, C. A., Jr, R. Medzhitov.
2002
. Innate immune recognition.
Annu. Rev. Immunol.
20
:
197
-216.
2
Takeda, K., T. Kaisho, S. Akira.
2003
. Toll-like receptors.
Annu. Rev. Immunol.
21
:
335
-376.
3
Medzhitov, R., P. Preston-Hurlburt, C. A. Janeway, Jr.
1997
. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity.
Nature
388
:
394
-397.
4
Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. V. Huffel, X. Du, D. Birdwell, E. Alejos, M. Silva, C. Galanos, et al
1998
. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene.
Science
282
:
2085
-2088.
5
Qureshi, S. T., L. Lariviere, G. Leveque, S. Clermont, K. J. Moore, P. Gros, D. Malo.
1999
. Endotoxin-tolerant mice have mutations in Toll-like receptor 4 (Tlr4).
J. Exp. Med.
189
:
615
-625.
6
Hoshino, K., O. Takeuchi, T. Kawai, H. Sanjo, T. Ogawa, Y. Takeda, K. Takeda, S. Akira.
1999
. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product.
J. Immunol.
162
:
3749
-3752.
7
Schumann, R. R., S. R. Leong, G. W. Flaggs, P. W. Gray, S. D. Wright, J. C. Mathison, P. S. Tobias, R. J. Ulevitch.
1990
. Structure and function of lipopolysaccharide binding protein.
Science
249
:
1429
-1431.
8
Ulevitch, R. J., P. S. Tobias.
1995
. Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin.
Annu. Rev. Immunol.
13
:
437
-457.
9
Shimazu, R., S. Akashi, H. Ogata, Y. Nagai, K. Fukudome, K. Miyake, M. Kimoto.
1999
. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4.
J. Exp. Med.
189
:
1777
-1782.
10
Nagai, Y., S. Akashi, M. Nagafuku, M. Ogata, Y. Iwakura, S. Akira, T. Kitamura, A. Kosugi, M. Kimoto, K. Miyake.
2002
. Essential role of MD-2 in LPS responsiveness and TLR4 distribution.
Nat. Immunol.
3
:
667
-672.
11
Schromm, A. B., E. Lien, P. Henneke, J. C. Chow, A. Yoshimura, H. Heine, E. Latz, B. G. Monks, D. A. Schwartz, K. Miyake, D. T. Golenbock.
2001
. Molecular genetic analysis of an endotoxin nonresponder mutant cell line: a point mutation in a conserved region of MD-2 abolishes endotoxin-induced signaling.
J. Exp. Med.
194
:
79
-88.
12
Miyake, K..
2004
. Endotoxin recognition molecules, Toll-like receptor 4-MD-2.
Semin. Immunol.
16
:
11
-16.
13
Lehmann, V., M. A. Freudenberg, C. Galanos.
1987
. Lethal toxicity of lipopolysaccharide and tumor necrosis factor in normal and d-galactosamine-treated mice.
J. Exp. Med.
165
:
657
-663.
14
Niwa, H., K. Yamamura, J. Miyazaki.
1991
. Efficient selection for high-expression transfectants with a novel eukaryotic vector.
Gene
108
:
193
-199.
15
Mizushima, S., S. Nagata.
1990
. pEF-BOS, a powerful mammalian expression vector.
Nucleic Acids Res.
18
:
5322
16
Akashi, S., S. Saitoh, Y. Wakabayashi, T. Kikuchi, N. Takamura, Y. Nagai, Y. Kusumoto, K. Fukase, S. Kusumoto, Y. Adachi, et al
2003
. Lipopolysaccharide interaction with cell surface Toll-like receptor 4-MD-2: higher affinity than that with MD-2 or CD14.
J. Exp. Med.
198
:
1035
-1042.
17
Hajjar, A. M., R. K. Ernst, J. H. Tsai, C. B. Wilson, S. I. Miller.
2002
. Human Toll-like receptor 4 recognizes host-specific LPS modifications.
Nat. Immunol.
3
:
354
-359.
18
Karin, M., A. Lin.
2002
. NF-κB at the crossroads of life and death.
Nat. Immunol.
3
:
221
-227.
19
Arvelo, M. B., J. T. Cooper, C. Longo, S. Daniel, S. T. Grey, J. Mahiou, E. Czismadia, G. Abu-Jawdeh, C. Ferran.
2002
. A20 protects mice from d-galactosamine/lipopolysaccharide acute toxic lethal hepatitis.
Hepatology
35
:
535
-543.
20
Le Roy, D., F. Di Padova, R. Tees, S. Lengacher, R. Landmann, M. P. Glauser, T. Calandra, D. Heumann.
1999
. Monoclonal antibodies to murine lipopolysaccharide (LPS)-binding protein (LBP) protect mice from lethal endotoxemia by blocking either the binding of LPS to LBP or the presentation of LPS/LBP complexes to CD14.
J. Immunol.
162
:
7454
-7460.
21
Meng, G., M. Rutz, M. Schiemann, J. Metzger, A. Grabiec, R. Schwandner, P. B. Luppa, F. Ebel, D. H. Busch, S. Bauer, et al
2004
. Antagonistic antibody prevents Toll-like receptor 2-driven lethal shock-like syndromes.
J. Clin. Invest.
113
:
1473
-1481.
22
Wang, C. Y., M. W. Mayo, R. G. Korneluk, D. V. Goeddel, A. S. Baldwin, Jr.
1998
. NF-κB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and c-IAP2 to suppress caspase-8 activation.
Science
281
:
1680
-1683.
23
De Smaele, E., F. Zazzeroni, S. Papa, D. U. Nguyen, R. Jin, J. Jones, R. Cong, G. Franzoso.
2001
. Induction of gadd45β by NF-κB down-regulates pro-apoptotic JNK signalling.
Nature
414
:
308
-313.
24
Lin, E. Y., A. Orlofsky, M. S. Berger, M. B. Prystowsky.
1993
. Characterization of A1, a novel hemopoietic-specific early-response gene with sequence similarity to bcl-2.
J. Immunol.
151
:
1979
-1988.
25
Daniel, S., M. Arvelo, C. Ferran.
2001
. Overexpression of A20 in endothelial cells of vascularized grafts creates a protective barrier against TNF- and FAS-mediated apoptosis.
Transplant. Proc.
33
:
225
26
Hu, X., E. Yee, J. M. Harlan, F. Wong, A. Karsan.
1998
. Lipopolysaccharide induces the antiapoptotic molecules, A1 and A20, in microvascular endothelial cells.
Blood
92
:
2759
-2765.
27
Doi, T. S., M. W. Marino, T. Takahashi, T. Yoshida, T. Sakakura, L. J. Old, Y. Obata.
1999
. Absence of tumor necrosis factor rescues RelA-deficient mice from embryonic lethality.
Proc. Natl. Acad. Sci. USA
96
:
2994
-2999.
28
Lee, E. G., D. L. Boone, S. Chai, S. L. Libby, M. Chien, J. P. Lodolce, A. Ma.
2000
. Failure to regulate TNF-induced NF-κB and cell death responses in A20-deficient mice.
Science
289
:
2350
-2354.
29
Boone, D. L., E. E. Turer, E. G. Lee, R. C. Ahmad, M. T. Wheeler, C. Tsui, P. Hurley, M. Chien, S. Chai, O. Hitotsumatsu, et al
2004
. The ubiquitin-modifying enzyme A20 is required for termination of Toll-like receptor responses.
Nat. Immunol.
5
:
1010
-1012.
30
Cross, A. S..
2002
. Endotoxin tolerance: current concepts in historical perspective.
J. Endotoxin Res.
8
:
83
-98.
31
Saitoh, S., S. Akashi, T. Yamada, N. Tanimura, M. Kobayashi, K. Konno, F. Matsumoto, K. Fukase, S. Kusumoto, Y. Nagai, et al
2004
. Lipid A antagonist, lipid IVa, is distinct from lipid A in interaction with Toll-like receptor 4 (TLR4)-MD-2 and ligand-induced TLR4 oligomerization.
Int. Immunol.
16
:
961
-969.
32
Kurt-Jones, E. A., L. Popova, L. Kwinn, L. M. Haynes, L. P. Jones, R. A. Tripp, E. E. Walsh, M. W. Freeman, D. T. Golenbock, L. J. Anderson, R. W. Finberg.
2000
. Pattern recognition receptors TLR4 and CD14 mediate response to respiratory syncytial virus.
Nat. Immunol.
1
:
398
-401.
33
Yarovinsky, T. O., L. S. Powers, N. S. Butler, M. A. Bradford, M. M. Monick, G. W. Hunninghake.
2003
. Adenoviral infection decreases mortality from lipopolysaccharide-induced liver failure via induction of TNF-α tolerance.
J. Immunol.
171
:
2453
-2460.
34
Yamada, Y., I. Kirillova, J. J. Peschon, N. Fausto.
1997
. Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lacking type I tumor necrosis factor receptor.
Proc. Natl. Acad. Sci. USA
94
:
1441
-1446.