IN VIVO AND IN VITRO EFFECTS OF THE ANTICOAGULANT, THROMBOMODULIN, ON THE INFLAMMATORY RESPONSE IN RODENT MODELS

Abstract

Sepsis remains a major health threat in intensive care medicine. The physiological functions of the coagulation cascade extend beyond blood coagulation and play a pivotal role in inflammation. We investigated whether the use of recombinant thrombomodulin (rTM), which has activity comparable with antithrombin, tissue factor pathway inhibitor, and activated protein C, could inhibit secretion of cytokines and high-mobility group box 1 (HMGB1) protein, thus reducing lung damage in a rat model of LPS-induced systemic inflammation. Rats treated with an intravenous injection of either rTM or saline were injected concurrently with intravenous LPS. In addition, mouse macrophage RAW264.7 cells were stimulated with LPS, with or without simultaneous rTM treatment. Histological examination revealed marked reductions of interstitial congestion, edema, inflammation, and hemorrhage in lung tissue harvested 12 h after treatment with both agents compared with LPS administration alone. LPS-induced secretion of proinflammatory cytokines and HMGB1 protein was inhibited by treatment with rTM. The presence of HMGB1 protein in the lung was examined by immunohistochemistry; the number of HMGB1-positive cells was significantly lower in LPS-treated animals that also received rTM. In the in vitro studies, rTM administration inhibited the activation of nuclear factor-kappa B by inhibiting I kappa B phosphorylation. The anticoagulant rTM blocked the LPS-induced inflammatory response and protected against acute lung injury normally associated with endotoxemia in this rat sepsis model. Given these results, rTM is a strong candidate as a therapeutic agent for various systemic inflammatory diseases.

INTRODUCTION

Sepsis is a significant cause of morbidity and mortality throughout the world. Accompanying complications, such as disseminated intravascular coagulation and acute respiratory distress syndrome (ARDS), remain refractory to therapy (^1-3). The pathophysiology of ARDS involves excessive, protracted inflammation characterized by increased vascular permeability, extravasation of plasma, and leukocyte infiltration. Acute respiratory distress syndrome, a serious condition often caused by systemic inflammation, can result in multiorgan dysfunction and death. Current therapeutic approaches for ARDS include supportive care, ventilator support, and pharmacological treatments (²). However, mortality rates for ARDS range from 40% to 60% (²).

Many studies on ARDS have implicated inflammation and derangement of the coagulation and fibrinolytic pathways (⁴). Several biomarkers of inflammation, such as IL-6, are associated with poor clinical outcomes in patients with acute lung injury (ALI)/ARDS (⁵). In addition, in vitro and in vivo studies indicate that high levels of TNF-α exacerbate inflammatory and pro-oxidative responses that are important in the pathogenesis of ARDS (⁶). Therefore, these cytokines play a significant role in inflammatory diseases of the lung.

The high-mobility group box 1 (HMGB1) protein is an intranuclear nonhistone DNA-binding protein originally identified as an important regulator of genetic information (⁷). However, a recent study indicated that HMGB1 protein has both nuclear and extracellular functions. For example, HMGB1 protein plays a key role as a late-phase mediator in the pathogenesis of sepsis (⁸). The HMGB1 protein is released from necrotic cells or secreted by cells such as activated monocytes or macrophages following cytokine stimulation, thus forming a proinflammatory loop (⁹). Secreted HMGB1 then binds to the receptor for advanced glycation end products (¹⁰) and the innate toll-like receptors 2 and 4 (TLR-2 and TLR-4, respectively) (¹¹). The HMGB1 subsequently initiates a signal cascade that results in further downstream cytokine release and contributes to lethality associated with endotoxemia (¹²). Furthermore, HMGB1 is associated with the pathogenesis of clinical and experimental ALI (¹³). These findings suggest that HMGB1 plays a critical role in ALI/ARDS.

Thrombomodulin (TM), which is normally expressed in endothelial cells, is a transmembrane protein that forms a complex with thrombin. The interaction between TM and thrombin results in a conformational change in the active site of thrombin, rendering it a potent activator of factor V, factor XIII, and protein C (¹⁴). Thrombomodulin exerts its anticoagulation activity through the activation of protein C.

Over the past decade, a concept of interest regarding the host response has been the importance of the association between inflammation and coagulation. The initiation of the coagulation cascade and the subsequent production of proinflammatory cytokines (particularly in response to the protein C system) are central to the pathogenesis of sepsis (¹⁵). Indeed, coagulation, particularly the generation of thrombin and factor Xa, is related to the acute inflammatory response (¹⁶). Moreover, activated protein C has been shown to reduce the inflammatory response by limiting thrombin generation and by modulating inflammatory mediators (¹⁷). Anti-inflammatory and anticoagulant agents have thus become a focus of new treatments for sepsis (^{18, 19}). Thrombomodulin is a transmembrane protein that forms a 1:1 complex with thrombin, and this interaction induces a conformational change in thrombin (²⁰). Once this complex is formed, its ability to cleave fibrinogen and to activate factor V, factor XIII, and platelets results in a 1000-fold increase in its efficacy as an activator of protein C (¹⁴). As such, TM has anticoagulant activity via its activation of protein C.

We hypothesized that TM would act as an inhibitor of systemic inflammation and prevent ALI in a rat model. To test this hypothesis, we investigated the impact of TM administration on serum and lung HMGBI levels, serum cytokines levels, and lung histopathology in rats during LPS-induced systemic inflammation. To further elucidate the mechanism of TM action, we assessed the impact of TM on HMGB1 and cytokine secretion by mouse macrophage RAW264.7 cells.

MATERIALS AND METHODS

In vitro study

Animals

All protocols conformed to the National Institutes of Health guidelines, and animal care was performed in compliance with the Principles of Laboratory Animal Care. The study was approved by the Ethical Committee on Animal Research of the College of Medicine, Oita University, Oita, Japan. Male Wistar rats (Charles River Laboratories, Japan Inc, Yokohama, Japan) weighing 250 to 300 g were used in all experiments. Animals had access to food and water ad libitum.

Drugs

The LPS (O127:B8; Sigma, St Louis, Mo) used in this study was derived from E. coli (O127) endotoxin and was dissolved in sterile saline for injections. Recombinant human TM (donated by Asahi Kasei Medical Co., Ltd, Tokyo, Japan) was dissolved in sterile saline for injections. Dosage information is described later.

Experimental protocols

Animals were randomly assigned to one of the following four groups (n = 18 for each group): (1) control group, rats received a bolus of a 0.9% NaCl solution (1.0 mL/kg) into the tail vein and immediately received another injection of 0.9% NaCl solution into the tail vein; (2) TM + LPS group, rats received a bolus of recombinant TM (1 mg/kg) into the tail vein and immediately received another injection of LPS (7.5 mg/kg) into the tail vein; (3) LPS group, rats received a bolus of a 0.9% NaCl solution (1.0 mL/kg) into the tail vein and immediately received another injection of LPS (7.5 mg/kg) into the tail vein; and the (4) TM group, rats received a bolus of recombinant TM (1 mg/kg) into the tail vein and immediately received another injection of 0.9% NaCl solution into the tail vein. Serum samples of venous blood were obtained from the left external jugular vein at the following time points: before treatment and 3, 6, 9, 12, and 24 h after treatment. All animals were breathing spontaneously during the experiments.

Histological analysis

Animals under pentobarbitone anesthesia were killed 12 h after LPS administration; left lungs were quickly removed and processed as indicated below. Lung tissue was stained with hematoxylin and eosin. A pathologist blinded to treatment assignment used the technique of Murakami et al. (²¹) to evaluate the extent of lung injury. Briefly, 24 areas of lung parenchyma were graded on a scale of 0 to 4 (0, abnormalities absent and tissue seems normal; 1, light; 2, moderate; 3, strong; 4, intense) for the degree of congestion, edema, inflammation, and hemorrhage. Then, a mean score for each parameter was calculated.

Ratio of wet weight to dry weight

Animals (n = 6 for each group) under sevoflurane anesthesia were killed 12 h after LPS administration. Lungs were removed, weighed, and dried in an oven at 80°C for 48 h to obtain pulmonary ratio of wet weight to dry weight (W/D).

Measurement of secreted cytokines and HMGB1

Secretion of IL-6, TNF-α, and HMGB1 was assayed by the enzyme-linked immunosorbent assay (ELISA) sandwich method. Ninety-six-well plates were precoated with monoclonal antibody specific to rat IL-6 (R&D Systems Inc, Minneapolis, Minn), TNF-α (R&D Systems Inc), or HMGB1 (Shino-Test Corporation, Tokyo, Japan). The secreted factors were detected according to the procedures described in the manufacturers' protocols. The A₄₅₀ values were measured using an ELISA reader (Bio-Rad Laboratories, Hercules, Calif).

Immunohistochemistry

Lungs were obtained from animals before and 12 h after LPS administration. Tissue samples were fixed immediately in 4% paraformaldehyde, embedded in OCT compound (Sakura Finetechnical Co, Tokyo Japan), and sectioned. Blocked sections were incubated with anti-HMGB1 (Shino-Test Corporation) polyclonal antibody (1:1000). Sections were then incubated with peroxidase-conjugated antimouse IgG and stained using an LSAB2 kit (Dako, Carpinteria, Calif) as the biotin-avidin-peroxidase complex system. After development, slides were counterstained with Mayer hematoxylin and mounted.

Western blot

Animals under pentobarbitone anesthesia were killed 12 h after LPS administration. The lung tissue specimens were homogenized with tissue protein extraction reagent (T-PER; Pierce Biotechnology, Rockford, Ill). Proteins were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in a 12% gel and transferred onto polyvinylidene difluoride sheets (Millipore, Bedford, Mass). Membranes were incubated with anti-HMGB1 (Shino-Test Corporation) and anti-β-actin (Abcam, Cambridge, UK) antibodies (1:1000) in PBS-T. After washes, membranes were incubated with peroxidase-labeled secondary antibodies (1:1000; ZYMED, South San Francisco, Calif) in PBS-T. Membranes were developed with ECL reagent (Amersham, Buckinghamshire, UK) and exposed to Hyperfilm ECL (Amersham).

In vitro study

Cells

The murine macrophage cell line, RAW264.7, was maintained in RPMI 1640 medium containing 5% heat-inactivated fetal bovine serum and antibiotics and cultured at 37°C under 5% CO₂. Opti-MEM (Sigma) was used in experiments designed to measure HMGB1 levels in conditioned medium. The following treatments were applied: control group, no drug treatment; LPS group, stimulated with LPS (100 ng/mL) only; TM + LPS group, simultaneously treated with recombinant TM (1000 nM) and LPS (100 ng/mL); and TM group, treated with recombinant TM (1000 nM) only. Samples of culture supernatant were obtained at the following time points: 0, 1.5, 3, 4.5, 6, 9, 12, and 24 h after treatment; "0" refers to the time point before LPS administration.

Nuclear factor-kappa B-binding assay

Murine RAW264.7 cells were harvested by scraping the adherent cell population from tissue culture flasks. Extracts were prepared using the NE-PER reagent (Pierce Biotechnology). The DNA-binding activity of nuclear factor-kappa B (NF-κB; p50/p65) was determined using an ELISA-based, nonradioactive NF-κB p50/p65 transcription factor assay kit (Chemicon, Temecula, Calif).

Preparation of protein from tissue and cultured cells

Tissue homogenates were boiled for 5 min, followed by the addition of dithiothreitol. In the cell culture experiments, murine RAW264.7 cells were harvested by scraping the adherent cell population from tissue culture flasks. Extracts were obtained using M-PER (Pierce Biotechnology). Cell homogenates were boiled for 5 min, followed by the addition of dithiothreitol.

Measurement of secreted cytokines and HMGB1

IL-6, TNF-α, and HMGB1 secretion was assayed by the ELISA sandwich method as described in the In vivo section.

Western blot

Proteins were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in a 12% gel and transferred onto polyvinylidene difluoride sheets (Millipore). Membranes were incubated with the following primary antibodies: antiphosphorylated I kappa B (IκB)-alpha (Cell Signaling Technology, Beverly, Mass), anti-IκB-α (Cell Signaling Technology), and anti-β-actin (Abcam) at 1:1000 dilutions in PBS-T. Membranes were then incubated with peroxidase-labeled secondary antibodies (1:1000; ZYMED) in PBS-T and processed as described in the In vivo section.

Statistical analysis

All data are presented as mean ± SD. Data were analyzed as repeated measurements followed by post hoc testing for group pairs for multiple comparisons and Mann-Whitney U test for comparisons between two independent groups. Survival data were analyzed using the Kaplan-Meier program in the Prism 4.0 software package (San Diego, Calif). P < 0.05 was considered statistically significant.

RESULTS

In vivo study

Mortality

Rat survival was analyzed 24 h after the various treatments. Only 6 of 12 rats in the LPS group survived, whereas 11 of 12 rats in the TM + LPS group survived, yielding a statistically significant increase in survival rate (P < 0.05). All rats in the control and the TM groups survived.

The effect of recombinant TM on histology and weight ratios of lung tissue after LPS administration

Lung tissue specimens were obtained 12 h after LPS administration, with or without treatment with recombinant TM. Although no histological alterations were observed in the control (saline-treated) group (Fig. 1A), marked interstitial edema and inflammatory cell infiltration were seen in the LPS group (Fig. 1B). Interstitial edema and inflammatory cell infiltration were markedly reduced in the TM + LPS group (Fig. 1C) compared with the LPS group. No histological changes in lung tissue were observed in the TM group (data not shown). Furthermore, the histology scores were all significantly higher in the LPS group than that in the control group, whereas the scores were intermediate in the TM + LPS group (Fig. 2). The W/D ratios of the lungs of animals of the LPS group were significantly higher than those of control animals (Fig. 3). The W/D ratios of the lungs in animals of the TM + LPS group were significantly lower than those in the LPS group; that is, the ratios in TM + LPS group were more similar to those in the control and the ENA groups.

Fig. 1: Effects of recombinant TM on lung injury in LPS-treated rats. Rats were treated with saline (control group), LPS (7.5 mg of LPS/kg i.v.; LPS group), and recombinant TM and LPS (TM, 1 mg/kg i.v.; LPS, 7.5 mg/kg i.v.; TM + LPS group). Shown are representative lung specimens obtained from the control (A, 100× magnification), LPS (B, 100× magnification), and TM + LPS (C, 100× magnification) groups, respectively (hematoxylin and eosin staining).

Fig. 2: Changes in lung histology score. The identified histological changes included congestion, edema, inflammatory cells, and hemorrhaging 12 h after administration of LPS: control (gray bars), LPS (black bars), and TM + LPS (white bars). The data are expressed as the mean score ± SD. ^#Significant difference compared with the control group (P < 0.05). *Significant difference relative to the administration of LPS (P < 0.05).

Fig. 3: Effects of recombinant TM on W/D ratio of LPS-treated rats. The W/D ratio in lungs was determined 12 h after LPS treatment of each group (n = 6 for each group): control (black bar), LPS (white bar), and TM + LPS (gray bar). The data are expressed as mean ± SD. ^#Significant difference compared with the control group (P < 0.05). *Significant difference compared with the LPS group (P < 0.05).

Effects of recombinant TM on serum levels of IL-6, TNF-α, and HMGB1

IL-6, TNF-α, and HMGB1 were undetectable in the serum of animals in the control group (Fig. 4, A-C). After LPS administration, with or without TM before treatment, serum was sampled from animals at various time points over the course of 24 h. IL-6 levels were elevated at all time points, peaking at 3 h after LPS administration in both the LPS and the TM + LPS groups. However, the induction of IL-6 was significantly diminished in the TM + LPS group (Fig. 4A).

Fig. 4: Temporal changes in IL-6, TNF-α, and HMGB1 protein concentration in serum after LPS administration in rats. Concentrations of the proinflammatory factors were determined in serum taken from animals in the LPS and the TM + LPS groups (n = 8 for each group) at the time points indicated. IL-6, TNF-α, and HMGB1 protein levels were analyzed by ELISA. Squares represent values for the LPS group; circles represent values for the TM + LPS group; "0 h" refers to a time point immediately before LPS administration. All data are expressed as mean ± SD. *Significant difference relative to administration of LPS alone (P < 0.05). A, IL-6; B, TNF-α; C, HMGB1.

Induction of TNF-α peaked 3 h after LPS administration in both the LPS and the TM + LPS groups, with TM again exerting a suppressive effect on the induction of TNF-α, especially at the 3-h time point (Fig. 4B).

Serum levels of HMGB1 increased steadily over time after LPS administration in both the LPS and the TM + LPS groups; peak levels were not reached until 12 h after treatment. Treatment with TM resulted in significantly reduced levels of HMGB1; the largest difference was observed at the 12-h time point (Fig. 4C). IL-6, TNF-α, and HMGB1 were undetectable in the serum of animals in the TM group (data not shown).

Effect of recombinant TM on HMGB1 levels in the lung

Immunohistochemical analysis revealed that the number of lung cells expressing HMGB1 increased after LPS administration alone (compare Fig. 5A with Fig. 5B). In contrast, the number of lung cells expressing HMGB1 was dramatically reduced in the TM-treated group (Fig. 5C).

Fig. 5: Changes in HMGB1 protein expression in lung tissue specimens after LPS administration in rats. Immunohistochemical analysis was used to detect HMGB1 protein in lung sections obtained 12 h after LPS administration. All photographs are at 400× magnification. Representative specimens from the control group (A), LPS group (B), and TM + LPS group (C) are presented. The arrows in panel B indicate cells staining positive for HMGB1. D, HMGB1 expression in the lung 12 h after LPS administration in control, LPS, and TM + LPS groups was detected by Western blot.

On the basis of Western blot analysis, HMGB1 in lung tissue increased following LPS administration (Fig. 5D). Consistent with the immunohistochemical analysis, this increase was less pronounced in lungs of animals in the TM + LPS group.

In vitro study

Effect of recombinant TM on HMGB1 secretion

High-mobility group box 1 was undetectable in control supernatants. Secreted HMGB1 levels were elevated at 24 h after LPS administration but were inhibited by cotreatment with TM (Fig. 6).

Fig. 6: Effect of recombinant TM on HMGB1 protein secretion by LPS-stimulated murine macrophages. LPS-stimulated (100 ng/mL) murine macrophages were cotreated with the indicated concentrations of recombinant TM for 24 h (n = 8 for each group). High-mobility group box 1 protein levels were measured by ELISA using supernatants of cell cultures treated with either LPS or each concentration of TM + LPS (n = 8 for each group). All data are expressed as mean ± SD. *Significant difference compared with the LPS group (P < 0.05).

Effect of recombinant TM on secreted cytokines

Cultured macrophages secreted elevated levels of IL-6 after LPS administration. This induction was significantly inhibited by cotreatment with TM (Fig. 7A). TNF-α levels in supernatants increased at 3 h after LPS administration. Cotreatment with recombinant TM significantly inhibited TNF-α secretion (Fig. 7B). Levels of these cytokines were undetectable in the supernatants of the control group (data not shown).

Fig. 7: Effect of recombinant TM on IL-6 and TNF-α production by LPS-stimulated murine macrophages. Murine macrophages were stimulated with LPS, with or without recombinant TM treatment (100 nM) for the indicated durations. Supernatants were collected, and levels of IL-6 (A) and TNF-α (B) were determined by ELISA (n = 8 for each group): LPS (squares) and TM + LPS (circles). All data are expressed as mean ± SD. *Significant difference compared with LPS-treated cells (P < 0.05).

Recombinant TM inhibits IκB kinase and modulates NF-κB DNA-binding activity

Levels of the transcription factor NF-κB (p50/p65) in the nucleus of RAW264.7 cells increased by 1 h after LPS stimulation as measured by DNA-binding activity. However, cotreatment with TM partially suppressed this increase (Fig. 8).

Fig. 8: Effect of recombinant TM on the LPS-induced increase in p50/p65 DNA binding. Cells were harvested 1 h after LPS treatment, nuclear fractions were prepared, and p50/DNA binding was measured: control (gray bar), LPS (black bar), and TM + LPS (white bar). All data are expressed as mean ± SD. ^#Significant difference compared with control cells (P < 0.05). *Significant difference compared with LPS treatment alone (P < 0.05).

LPS administration resulted in IκB degradation, and this was inhibited by TM treatment (Fig. 9). In addition, an increase in IκB phosphorylation was observed in RAW264.7 cells after LPS administration, and this was suppressed by cotreatment with TM (Fig. 9).

Fig. 9: Effect of recombinant TM on the LPS-induced phosphorylation of IκB-α. RAW264.7 murine macrophages were stimulated with LPS, with or without recombinant TM (100 nM) cotreatment, and harvested 1 h later. Cytoplasmic levels of phosphorylated IκB-α (p-IκB) were determined by Western blot using antibodies against IκB-α (IκB), p-IκB, and β-actin as a control.

DISCUSSION

In this study, we demonstrated that treatment with recombinant TM significantly improved LPS-induced ALI and was associated with a reduction in cytokine and HMGB1 levels in rats. Our results also suggest that the inhibition of cytokine and HMGB1 secretion was the result of an inhibition of NF-κB activity. Thus, the improvement in LPS-induced ALI by TM administration seems to be related to the altered expression of these mediators.

We found that TM administration significantly improved LPS-induced lung injury by reducing the secretion of cytokines and HMGB1. Activated protein C has anti-inflammatory effects in addition to its anticoagulant properties, which include suppression of proinflammatory cytokine production and inhibition of leukocyte attachment to the endothelium (¹⁶). However, in sepsis and ALI, coagulation is propagated due to suppression of the protein C system (^{22, 23}). This suppression is in part due to a reduction in TM on endothelial cell surfaces (^{22, 23}). Protein C is transformed to its active form on the cell surface by the TM-thrombin complex (²⁴). In addition, TNF-α down-regulates endothelial production of TM (²²). Thus, TM may be effective in treating sepsis and ALI.

In our rat model of LPS-induced systemic inflammation, we examined serum levels of TNF-α and IL-6. These cytokines are normally strongly induced by LPS. However, treatment with TM significantly suppressed this induction, suggesting that TM can reduce the inflammatory response in this rat model of sepsis. Cytokines such as TNF-α and IL-6 are secreted during the early phase of the inflammatory response and play an important role in the development of ARDS (²⁵). As a primary mediator of inflammation, TNF-α plays an important role in coordinating the inflammatory response and in activating other cytokines (²⁶). Generation of high levels of TNF-α leads to exacerbation of inflammatory and pro-oxidative responses that are important in the pathogenesis of pulmonary disorders (⁶). In ALI, high levels of cytokines, such as IL-6, are present both locally in the lung and systemically. Also, in experimental models of sepsis, IL-6 levels positively correlate with mortality. Furthermore, measurement of IL-6 levels in at-risk patients has been proposed for accurate prediction of individuals who are at significant risk of death from sepsis (²⁷). The clinical relevance of LPS-induced ALI led us to investigate the effect of TM specifically in lung tissue. Indeed, histological changes (including interstitial and intra-alveolar inflammation, edema, congestion, and hemorrhage) in the lungs that were observed after LPS administration were ameliorated in animals cotreated with TM. Therefore, the reduction of cytokines (such as TNF-α and IL-6) due to TM administration might be responsible for the improvement of LPS-induced lung injury.

We demonstrated that HMGB1 induction was inhibited by TM treatment in our in vivo and in vitro models. High-mobility group box 1 mediates acute inflammation in animal models of lung injury and endotoxemia and plays an important role in the development of sepsis and LPS-induced lung injury (^{13, 28}). Supporting this, HMGB1 causes acute lung inflammation when administered intratracheally (²⁹). In the current study, we demonstrate that TM before treatment can dramatically reduce serum and lung HMGB1 levels in an LPS-induced septic shock model. Thrombomodulin has been reported to sequester HMGB1 (³⁰). In addition, increased HMGB1 secretion results in NF-κB activation (³¹). Therefore, the inhibition of NF-κB activation we observed might be a result of decreased HMGB1 levels. The observation that TM treatment also ameliorated LPS-induced ALI suggests a direct relationship between lung injury and HMGB1 levels.

We found that TM reduced serum and supernatant levels of cytokines in response to LPS treatment. The transcription factor NF-κB regulates the expression of inflammatory genes, including TNF-α and IL-6 (³²), and its aberrant activation is associated with septic shock (³³). Some drugs have been reported to suppress the NF-κB pathway and may be beneficial for the prevention or treatment of septic shock (³⁴). Our results suggest that TM blocks the activation of NF-κB-regulated genes, offering a potential mechanism by which TM exerts its anti-inflammatory effect and ameliorates LPS-induced lung injury. The anti-inflammatory properties of the N-terminal lectin-like domain of TM may be responsible for the inhibition of NF-κB activation (³⁵).

In this study, we showed that IκB phosphorylation and NF-κB activation were inhibited by TM in LPS-stimulated murine macrophage cells. LPS is recognized by monocytes and macrophages of the innate immune system. Upon binding to LPS-binding protein in plasma, LPS is delivered to the cell surface receptor, CD14, and transferred to the transmembrane signaling receptor, TLR-4 (³⁶). LPS stimulation of murine macrophages activates several intracellular signaling pathways, including the IκB kinase-NF-κB pathway (³⁷). Nuclear factor-kappa B is composed of dimers of different subunits that exist in the cytoplasm as inactive forms bound to the inhibitory protein, IκB. Phosphorylation of IκB leads to its proteolytic degradation and dissociation of NF-κB. Upon translocation into the nucleus, NF-κB induces the transcription of target genes. Therefore, IκB phosphorylation represents a key step in NF-κB activation (³⁸). Importantly, TM seems to suppress NF-κB activation by preventing IκB phosphorylation. However, we did not examine the mechanism of inhibition of IκB phosphorylation. Further studies will be required to address this issue.

The current study has several limitations. First, we used human recombinant TM but did not examine effects of recombinant TM administered after the onset of disease caused by exposure to live bacteria. Furthermore, given that TM was used concomitantly with LPS treatment in our study, effects of delayed TM administration in LPS-induced systemic inflammation models are unclear. Further experimentation under such conditions will be required. Second, the LPS-induced systemic inflammation model is certainly not a true reflection of the actual clinical setting. However, LPS-induced inflammation is generally considered to adequately reproduce the full spectrum of inflammatory changes observed in patients with sepsis (³⁹). Finally, potential differences in TM function and the pathophysiology of inflammation and ALI in rodents and humans may also influence the interpretation of our results.

In conclusion, recombinant human TM protected rats against ALI associated with LPS-induced systemic inflammation. In addition, recombinant TM prevented HMGB1 induction normally observed after LPS exposure. These effects were attributed to IκB kinase inhibition. Our findings provide significant contributions to the understanding of the pathophysiology of ALI in sepsis. Thus, pharmacological blockade of actions of the coagulation process may be a promising approach for improving other systemic inflammatory conditions. Our results point to the coagulation system as an important novel therapeutic target for sepsis-related ALI.

ACKNOWLEDGMENTS

The authors thank Hiroaki Kawazato and Aiko Yasuda for helpful advice on the preparation of lung specimens.