The immunomodulator ginsan induces resistance to experimental sepsis by inhibiting Toll-like receptor-mediated inflammatory signals

Ginsan protects mice from the acute sepsis induced by S. aureus infection

We have previously demonstrated that ginsan increased the survival rate of septic C57BL/6 mice challenged with live S. aureus 13. In the present study, we first determined optimal efficacy dose of ginsan. Mice were i.v. injected with 0.012, 0.025, 0.5, 25, and 250 mg/kg ginsan 24 h before the S. aureus challenge (1.5 × 10⁸ CFU/mouse). Pretreatment with ginsan significantly increased survival in the mice infected with S. aureus, from 10% in control mice to 88% (p<0.0001; Fig. 1A). The mice given 25 μg/kg ginsan at 3, 24, or 48 h before the S. aureus challenge had survival rates of 58%, 83% (p=0.011), and 75% (p=0.034), respectively (data not shown). We also examined whether ginsan treatment could provide protection in other septic states, such as Gram-negative bacterial infection induced by live E. coli, and polymicrobial sepsis induced by cecal ligation and puncture (CLP), which closely mimics human acute peritonitis and is regarded as the most clinically relevant animal model of sepsis. The mortality induced by i.p. injection of 1.4 × 10⁷ live E. coli cells was almost completely inhibited, with the survival rate improving from 13.3% without ginsan to 100% with ginsan treatment (100 mg/kg, i.v., 24 h before E. coli injection; p<0.0001; Fig. 1B). Ginsan-conferred protection was also observed in mice challenged with CLP-induced polymicrobial sepsis in which the survival was significantly improved from 0% in control mice to 60% in ginsan pretreated mice on day 5 (100 mg/kg, i.v., 24 h before CLP; p=0.048; Figure 1C). The effective doses of ginsan in the E. coli infection and CLP models were very different from that in the S. aureus challenge experiment, and the biological reasons underlying the dose-response effects are not yet known.

Ginsan enhances in vivo bacterial clearance

To verify that the survival benefit of ginsan treatment is related to enhanced bacterial clearance, mice were culled 2 days after S. aureus infection. Samples of blood and homogenates of the spleen and kidney were cultured at 37°C for 24 h after which bacterial CFU were enumerated and compared (Fig. 2). Pretreatment of ginsan dramatically reduced the bacterial CFU counts in the bloodstream, spleen and kidney of the mice, compared with those of their control counterparts (p<0.001).

Ginsan increases bactericidal activity of macrophages

To identify possible mechanisms involved in the enhanced bacterial clearance observed in ginsan-treated mice, we assessed whether ginsan increased the bactericidal activity of macrophages isolated from intact or S. aureus-infected mice. Fig. 3A shows a typical FACS analysis of peritoneal macrophages (PM) incubated with ginsan for 3 h, following stimulation with heat-killed S. aureus for 30 min at 37°C. Phagocytosis of S. aureus by PM was significantly increased after ginsan treatment (by approximately 43% compared with the non-treated macrophages). Moreover, the PM isolated from ginsan-treated, S. aureus-infected mice also exhibited enhanced phagocytosis against S. aureus, although the magnitude of phagocytosis was not as great as that in the in vitro assays (Fig. 3B). Given our previous finding of increased nitric oxide production and intracellular killing of bacteria in PM isolated from ginsan-treated S. aureus-infected mice 13, ginsan appears to protect mice against S. aureus-induced sepsis via macrophages activation.

To further confirm the crucial role of macrophages in the ginsan-mediated protection of mice infected with S. aureus, we depleted mice of monocyte/macrophages by s.c. etoposide injection once a day for 3 consecutive days (71% mean monocyte reduction as determined by blood cell counts). A suboptimal dose of S. aureus (1 × 10⁷ CFU) was used in this experiment, because lethal bacterial dose would itself result in low survival, making it difficult to differentiate effect of monocyte depletion. The monocytopenic mice displayed higher mortality than the controls (50% vs. 20%), whereas pretreatment of ginsan demonstrated a still much higher survival benefit over the controls (Fig. 4A). In addition, pretreatment of the monocytopenic mice with ginsan led to a statistically significant increase in bacterial clearance from the blood, spleen and kidney at 48 h after the S. aureus challenges (Fig. 4B). Although ginsan augmented the phagocytosis of S. aureus by PM, ginsan still increased survival in macrophage-depleted animals and reduced the S. aureus burden in these animals, suggesting that macrophages may be partly involved in the phagocytosis and antimicrobial activity induced by ginsan in vivo.

S. aureus-associated release of inflammatory cytokines is attenuated by ginsan

It is generally accepted that poorly regulated inflammatory cytokines play an important role in the pathogenesis of sepsis. Therefore, the production of inflammatory cytokines in septic mice was examined at different times after ginsan treatment (Fig. 5). The highest levels of production of almost all the cytokines examined in this study were observed 8–12 h after S. aureus infection. The typical inflammatory cytokines, such as TNF-α, IL-1β, and IL-6, were robustly produced in S. aureus-infected mice, whereas ginsan pretreatment significantly decreased their production.

IFN-γ and TNF-α are known to aggravate sepsis synergistically, and IL-12 and IL-18 stimulate IFN-γ induction. Therefore, we determined the serum levels of these cytokines in septic mice and observed significant decreases in the levels of all of these cytokines with ginsan treatment. In contrast to this down-regulated synthesis of proinflammatory cytokines, ginsan treatment did not affect IL-2 and IL-4 (data not shown).

Ginsan suppresses TLR signaling pathway in macrophages

Given that TLR plays prominent roles in stimulating the protective immune and inflammatory response to infectious challenges, we next examined whether ginsan can modulate the expression of TLR2 and the adaptor molecule MyD88 in response to Gram-positive bacterial infections. PM treated with S. aureus had very high levels of TLR2 and MyD88, and moderate expression of TLR4 and TLR9 (Fig. 6). In striking contrast, PM pretreated with ginsan before S. aureus infection almost completely down-regulated the expression of TLR2 and MyD88. Moreover, the expression of TLR4 and TLR9 was blocked under the same experimental conditions.

Ginsan inhibits S. aureus-induced NF-κB and MAPK activation

To more critically evaluate whether the reduction in TLR and MyD88 was required for the suppression of inflammatory cytokines conferred by ginsan, we further investigated the downstream molecules connected to TLR signaling, including extracellular signal-regulated kinase (ERK), c-jun N-terminal kinase (JNK), p38 mitogen-activated protein kinases (MAPK), and NF-κB. The transduction of TLR signals in monocytes and macrophages activates MAPK, which subsequently leads to the activation of various transcription factors and the production of inflammatory cytokines 14. S. aureus stimulation induced phosphorylation of JNK1/2 and p38 kinases in PM, whereas a marked down-regulation of p38 and JNK1/2 phosphorylation were observed when PM were pretreated with 0.1 μg/mL ginsan (Fig. 7A). However, the expression of phospho-ERK1/2 in PM did not change significantly in response to S. aureus stimulation or ginsan pretreatment, indicating that ERK activation is not required during that period. The phosphorylation of MAPK is associated with the subsequent activation of transcription factors, including NF-κB, which plays a critical role in monocyte inflammatory cytokine production after exposure to bacterial stimuli 14–16. Therefore, we further examined the effect of ginsan on NF-κB activation. Fig. 7B shows that the nuclear extracts of PM that were pretreated with ginsan and subsequently co-cultured with S. aureus showed a remarkable reduction of free NF-κB. As expected, S. aureus infection per se increased NF-κB activation in PM, which was correlated with the findings for both cytokine production and MAPK activation.

Bacteria

S. aureus 25923 or E. coli 018ac:K1:H7, acquired from ATCC, was cultured at 37°C in Trypticase soy broth (Difco, Detroit, MI), harvested at the mid-logarithmic growth phase, washed twice, and resuspended in PBS. The concentration of resuspended bacteria was determined spectrophotometrically at 620 nm.

Preparation of ginsan

Ginsan, a polysaccharide, was purified from the ethanol-insoluble fraction of the aqueous Panax ginseng extract, as described previously 12. Further purification was carried out successively using size exclusion and ion exchange column chromatography, and NMR analysis revealed that ginsan was composed of α(1→6)glucopyranoside and β(2→6) fructofuranoside in a 5:2 molar ratio. The endotoxin level in the purified ginsan preparation was less than 0.03 EU/mg as measured using the Limulus amebocytes lysates assay (Endosafe®, Charles River Laboratories, USA) according to the manufacturer's instructions. Moreover, the treatment of polymyxin B did not inhibit lymphocyte proliferation, excluding possible endotoxin contamination in ginsan. Ginsan was dissolved in PBS (pH 7.4) and filtered through a 0.25-μm Millipore membrane (Millipore, USA).

Sepsis model

Pyrogen-free BALB/c male mice, aged 5–7 weeks (18–22 g), were purchased from the Charles River Breeding Laboratory (Charles River Japan, Atsugi Breeding Center, Yokohama, Japan) and kept for 6 days in our animal quarters for acclimatization. Sterile standard mouse chow (NIH-7 open formula) and water were given ad libitum, and the mice were housed randomly at 60% humidity and 22 ± 2°C on a 12-h light-dark cycle. All animal experiments were performed according to the National Institutes of Health guidelines, based on the protocols approved by the Institutional Animal Care and Use Committee of the Korea Institute of Radiological and Medical Sciences (KIRAMS).

For the sepsis model, mice were injected i.p. with bacterial suspension containing either 1.5 × 10⁸ live S. aureus or 1.4 × 10⁷ live E. coli cells. For CLP, the cecum was exposed through a 1.0- to 1.5-cm abdominal midline incision in anesthetized mice, ligated at its base with 3–0 silk suture, punctured once with a 21-gauge needle, and gently squeezed to express fecal materials into the peritoneal cavity. The cecum was then returned to the peritoneal cavity, and the abdominal incision was closed. Survival was monitored for at least 7 days.

Reagents

Etoposide (Sigma, St. Louis, MO) was used to selectively depopulate monocytes/macrophages in the mice, as described elsewhere 43. Mice were injected s.c. for 3 consecutive days with 12.5 mg/kg etoposide in 1% DMSO-PBS before the S. aureus challenge, based on practices established in earlier studies 43, 44. Antibodies specific for ERK1/2, phospho-ERK1/2, JNK1, and phospho-JNK1/2 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA) and antibodies to p38 MAPK and phospho-p38 MAPK were from Cell Signaling Technology (Beverly, MA).

Analysis of cytokines

To measure the S. aureus-induced plasma cytokines, mice were given ginsan 24 h before the S. aureus challenge, and plasma was collected at various times after S. aureus infection. The concentrations of cytokines were measured with ELISA kits for detecting IFN-γ, IL-1β, IL-2, IL-4, IL-6, IL-10, IL-12, and TNF-α (R&D Systems, USA).

Enumeration of bacteria in blood and organs

Control and ginsan-treated mice were culled 48 h after i.p. injection of live S. aureus. Blood samples were taken, and the dissected spleen and kidney were homogenized. Serial tenfold dilutions of whole blood and organ homogenates in PBS were plated on blood agar and cultured for 24 h at 37°C to determine bacterial CFU.

Determination of phagocytic activity of macrophages

FITC-labeled S. aureus was prepared as described previously 45, 46. Briefly, S. aureus was heat-killed at 70°C for 90 min and labeled with 0.25% FITC at 4°C for 12 h. After washing out unlabeled S. aureus, the ten heat-killed FITC-labeled S. aureus per cell were incubated with PM (5 × 10⁶cells) for 30 min at 37°C. PM was treated with 0.1 μg/mL ginsan 3 h before the S. aureus administration. Bacterial uptake by the cells was assessed using FACScan analysis (Becton Dickinson, USA) with CellQuest software.

RT-PCR

Expression levels of RNA transcripts were quantified using RT-PCR. Total RNA from peritoneal macrophages was isolated using TRIzol reagent (Gibco BRL, USA). The RNA (1 μg/tube) was reverse transcribed into cDNA for 1 h at 37°C using 15 U MMLV RT (Promega, USA) in the presence of 1 μL pd(N)₆ random hexamer (Amersham Biotechnology, USA). The cDNA was used as a template for amplifying TLR-2, TLR-4, TLR-9, and MyD88 by PCR (Applied Biosystems, USA). The reaction mixture consisted of 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 2.5 mM dNTP (Takara, Japan), 1.25 U Taq DNA polymerase (Takara), and 0.05 mM of each primer. The primers were as follows, with the expected fragment size in parentheses: mTLR2 (375 bp), forward 5′-CAAACTGGAGACTCTGGAAG-3′ and reverse 5′-CTGTAGGAAACAAAGGCATC-3′; mTLR4 (266 bp), forward 5′-GACACCCTCCATAGACTTCA-3′ and reverse 5′-TGTTCA ACATTCACCAAGAA-3′; mTLR9 (297 bp), forward 5′-CCTGTCTCAAAATAACCTGC-3′ and reverse 5′-CTGAGGTTGACCTCTTTCAG-3′; mMyD88 (303 bp), forward 5′-CGATGCCTTTATCTGCTACT-3′ and reverse 5′-TTCTTCATCGCCTTGTATTT-3′; mGAPDH (285 bp), forward 5′-GAGTCTACTGGCGTCTTCAC-3′ and reverse 5′-CCATCCACAGTC TTCTGAGT-3′. The thermal cycle conditions were 94°C for 1 min, 52°–58°C for 30 s, and 72°C for 30 s for 32 cycles. The PCR products were electrophoresed on 1.5% agarose gels and quantified using an image analyzer (Bio-Rad, Hercules, CA).

Electrophoretic mobility shift assay

Nuclear extracts of PM that had been pretreated with ginsan (0.1 μg/mL) for 6 h and then co-cultured with heat-killed S. aureus for 40 min were prepared as described previously 29. The double-stranded oligonucleotide probes containing NF-κB binding sequences (5′-GGGGACTTTCCC-3′) were end-labeled with [γ-³²P]ATP using T4 polynucleotide kinase. The nuclear extract (15 μg) was incubated at room temperature for 20 min with ³²P-labeled probe in a binding buffer (10 mM Tris, pH 7.8, 420 mM KCl, 1.5 mM MgCl₂, 20% glycerol). DNA/nuclear protein complexes were electrophoresed on a 4% non-denaturing acrylamide gel. The gel was dried and exposed overnight to an X-ray film (Agfa, Belgium) in a cassette at –80°C. The film was developed for analysis.

Immunoblot analysis

To investigate the expression of ERK, JNK, and p38 MAPK, 2 × 10⁶ PM were pretreated with ginsan (0.1 μg/mL) for 6 h and further co-cultured with heat-killed S. aureus for 40 min. The cell lysates were prepared by extracting proteins with cell lysis buffer (Cell Signaling) supplemented with protease inhibitors. The protein samples (30 μg/well) were separated by 9% SDS-PAGE and transferred to nitrocellulose membranes (Schleicher & Schnell Bioscience, Germany). The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline and then incubated with primary antibodies for 1 h at room temperature. The blots were developed using peroxidase-conjugated secondary antibody, and the proteins were visualized using enhanced chemiluminescence (ECL) procedures (NEN) according to the manufacturer's recommendations.

Statistical analysis

The results are expressed as the mean ± SEM of data obtained from at least three replicate experiments. Survival data were analyzed using the log-rank test. Statistical significance and differences among groups were assessed by one-way analysis of variance and the Newman-Keuls test using GraphPad Prism v3.02 software (San Diego, CA).