Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A common feature of acute lung injury (ALI) is widespread intra-alveolar fibrin deposition leading to impaired gas exchange (1), presumably due to an imbalance between procoagulant and fibrinolytic pathways. The persistent procoagulant state activates an inflammatory response, perpetuates injury, and prevents repair. Procoagulant activity (PCA) in the lung is regulated by extravascular initiation of coagulation by tissue factor-factor VIIa (TF-FVIIa) complex, and TF-dependent PCA is elevated in ALI after a variety of insults, including pneumonia and interstitial lung diseases (1).

TF-FVIIa complex has the potential to contribute both directly and indirectly to the regulation of inflammation. In vitro work indicates that downstream products of extrinsic coagulation, e.g., factor Xa and thrombin, can affect vascular permeability, cytokine release, and adhesion molecule expression (4). In vivo, blockade of the initiation of extrinsic coagulation in baboon models of lethal Escherichia coli sepsis has improved survival and parameters of disseminated intravascular coagulation (7); however, the mechanism of protection has not been determined. This is partly due to the complexity of the responses. For example, a modified FVIIa, DEGR-FVIIa, attenuated plasma interleukin (IL)-6 and IL-8 levels (10) in live E. coli sepsis in baboons, whereas monoclonal antibody to TF given at the time of infusion had no effect on IL-6 levels in chimpanzees given sublethal endotoxin (11). In addition, the extent to which TF-FVIIa complex regulates inflammatory responses in specific tissues has not been studied during sepsis in vivo.

Although preliminary evidence suggests an important regulatory role for TF-FVIIa complex in systemic inflammation during sepsis, its role in local inflammatory responses in the lung, which is highly susceptible to acute injury, has not been studied. Because TF is a class II cytokine receptor (12) for which gene expression is regulated by lipopolysaccharide (LPS) (13), we hypothesized that blockade of the TF-FVIIa complex would prevent lung injury due to intratracheal LPS by attenuating local production of TF, IL-1, IL-6, and tumor necrosis factor (TNF)-. These early response cytokines are all regulated by NF-B and are associated with both sepsis and acute respiratory distress syndrome (ARDS) (8, 14). This mode of injury causes an intense local inflammatory response characterized by diffuse neutrophilic alveolitis and cell damage that evolves over several days (14). We blocked initiation of coagulation at the TF-FVIIa complex with FFR-FVIIa, a high-affinity competitive inhibitor of TF (18) that effectively inhibits TF-induced coagulation and thrombogenesis (19, 20). Treatment with FFR-FVIIa greatly attenuated LPS-mediated cytokine production and lung injury regardless of whether the agent was administered at the time of injury or 6 h later.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Unless otherwise specified, all chemicals were obtained from Sigma (St. Louis, MO).

Experimental and Animal Protocols

Male Sprague-Dawley rats (Charles River Laboratories, Research Triangle Park, NC) weighing ~ 250 g were used for the experiments. Animals were handled under protocols approved by the Duke University Institutional Animal Care and Use Committee (Durham, NC) in accordance with National Institutes of Health guidelines for the use of laboratory animals. The rats were anesthetized with halothane (Wyeth-Ayerst, Philadelphia, PA), the trachea was cannulated transorally with an 18-gauge catheter, and LPS (E. coli 0111:B4, 500 µl of 1 mg/ml in 0.9% NaCl) was rapidly instilled into the lungs. The animals were given either FFR-FVIIa (10 mg/kg) (Novo Nordisk, Copenhagen, Denmark) in 1 ml of 0.9% NaCl at the time of LPS instillation or an equal volume of NaCl solution by intravenous injection. The animals were killed after 6, 12, 24, or 48 h and lungs, blood, and bronchoalveolar lavage (BAL) were obtained from each group (Saline+ LPS or FFR-FVIIa + LPS). Blood and BAL were used for cytokine measurements and biochemistry (n = 5 at each time point). In addition, lungs tissue was flash frozen for reverse transcription-polymerase chain reaction (RT-PCR) or inflation-fixed at 20 cm H₂O pressure in 4% formaldehyde for histology.

A separate group of animals served as controls for drug effects. These rats were given FFR-FVIIa alone intravenously at 10 mg/kg. At appropriate times after the drug was given, blood and BAL were obtained from the animals for cytokines and biochemical measurements (6, 12, 24, and 48 h, n = 3 at each time point).

A second experiment was performed to evaluate the effects of blockade of coagulation on ALI at the level of thrombin. These rats received intravenous treatment with either the synthetic thrombin inhibitor peptide Hirulog (10 mg/kg; The Medicine Co., Cambridge, MA) in 1 ml of 0.9% NaCl (n = 6) or an equal volume of saline (n = 5) at the time of intratracheal instillation of LPS. BAL was obtained 24 h after LPS administration, the time of maximum lung injury in the control experiments. BAL fluid was used to measure protein, lactate dehydrogenase (LDH) activity, and cell counts.

To determine whether or not blockade of coagulation could protect the lung after injury by LPS, a rescue experiment was performed in which intravenous FFR-FVIIa (n = 5) or saline (n = 5) was administered 6 h after induction of ALI with intratracheal LPS. These animals were killed at 24 h after LPS, and blood, tissue, and BAL samples were collected by the protocol outlined in the primary experiment.

Sample Preparation

After the experiments, the animals were anesthetized with halothane. To isolate lung RNA for RT-PCR, the abdominal and chest cavities were opened, the aorta was transected, and the lungs were flushed gently through the right ventricle with 0.9% NaCl. The lungs were removed, snap frozen in liquid nitrogen, and stored at 80°C until RNA was extracted.

Blood was drawn from the vena cava at the time of death into tubes containing sodium citrate (Becton Dickinson, Franklin Lanes, NJ). These blood samples were centrifuged at 1,200 rpm for 10 min at 4°C. The plasma layer was carefully removed, placed into 1-ml aliquots, and stored at -80°C for subsequent analysis. All samples were kept on ice during processing.

BAL was performed in open chest animals using 10 ml of 0.9% normal saline. The volume recovered was recorded and 100 µl was set aside for total cell and differential count. All samples were kept on ice during processing. The remaining BAL fluid was centrifuged at 1,200 rpm for 10 min at 4°C. Supernatant was carefully withdrawn without disturbing the cell pellet, divided into 1-ml aliquots, and stored at 80°C for subsequent analysis.

Cell Count and Differential

BAL fluid was mixed in a 1:1 ratio with Trypan blue, counted in duplicate on a hemocytometer (Fisher Scientific, Pittsburgh, PA), and cell concentration calculated per milliliter of BAL. An unspun sample was processed for differential using a cytospin. Slides were stained with Wright stain and cell differential was tabulated using light microscopy at ×40 magnification. Differential was recorded as number of neutrophils and mononuclear cells per 100 total cells.

Biochemical Assays

Protein concentration was measured on all BAL samples with the bicinchoninic acid assay using bovine serum albumin (BSA) as a standard (21). LDH on BAL samples was determined by the methods of Bergmeyer (22).

Histology and Immunohistochemistry

To obtain lungs for routine histology and immunohistochemistry (IHC), the trachea was cannulated and the lungs were inflation-fixed en bloc with 4% formaldehyde at 20 cmH₂O fixative pressure for 10 min. Inflation-fixed lungs were paraffin-embedded and cut into 10-µm sections. Before the tissue sections were labeled, they were deparaffinized in xylene and then rehydrated in graded alcohol solutions. The sections were stained using either hematoxylin and eosin or antibodies to fibrin with the following method. The sections were blocked in a solution of 5% nonfat dry milk, 1% BSA, 5% goat serum in 0.01M phosphate-buffered saline, and 0.1% Triton X-100 before incubation overnight at 4°C with antibodies to fibrin (Novo-Nordisk) at a dilution of 1:3,000. The sections were washed three times with phosphate-buffered saline with 0.1% Triton X-100 (5 min each) and incubated with the secondary antibody, biotinylated goat anti-rabbit IgG (Jackson Laboratories, West Grove, PA), at a dilution of 1:1,000 at room temperature for 1 h. The signal was detected with peroxidase-conjugated avidin and diaminobenzidine. The slides were counterstained with hematoxylin. For negative controls, sections were processed as above except that the primary incubation was performed with nonimmune rabbit serum (Jackson Laboratories) instead of primary antibody.

Drug Levels

Drug levels were measured in timed plasma samples by enzyme-linked immunosorbent assay (ELISA) (Novo Nordisk).

RT-PCR

mRNA expression for TF and glyceraldehyde phosphate dehydrogenase (GAPDH) was determined semiquantitatively by RT-PCR. Cytoplasmic RNA was extracted from rat lung using the Trizol Total RNA isolation kit (GIBCO BRL, Gaithersburg, MD). RNA concentration was determined spectrophotometrically at 260 nm. Sample quality was checked by running the RNA out on 1.5% agarose gels. Total RNA (1 µg) from each sample was reverse-transcribed into cDNA using oligo (dT) as a primer. PCR amplification was performed in a thermal cycler as follows: 35 cycles for GAPDH, and 25 cycles for rat tissue factor. The optimum number of cycles was determined by titration of visible product on GelStar (BioWhittaker Molecular Applications, Rockland, ME)-stained gels during the exponential phase of the PCR. The cycling parameters routinely used were: denaturation at 94°C for 40 s, annealing at 55°C for 30 s for GAPDH, and 40 s for TF, and extension at 72°C for 45 s for GAPDH and 120 s for TF. Primers specific for rat GAPDH sense (5'-CCATGGAGAAGGCTGGGG-3') and antisense (5'-CAAA GTTGTCATGGATGACC); and rat tissue factor sense (5'-GCT CAATGCCTTCTCTCAGG-3') and antisense (5'-CACCACTTG TAGCTCGGTGA) were used to amplify a 193-bp and 551-bp fragment of rat GAPDH and TF, respectively. The amplified products (10 µl) were electrophoresed in 2% agarose gels, stained with GelStar, and viewed under UV light. The GAPDH signals were used to control for variation in the efficiency of RNA extraction, reverse transcription, and PCR. TF mRNA expression was compared by densitometry in rats killed at times 0, 6, 12, 24, and 48 h after administration of intratracheal LPS.

Cytokine Measurements

Plasma and BAL IL-1, TNF-, IL-6, and IL-10 were measured by ELISA (R&D Systems, Minneapolis, MN).

Nuclear Protein Extraction

Nuclear protein was extracted from the lung tissue by homogenizing aliquots of 200 mg of fresh tissue in a Dounce tissue homogenizer with 2 ml of solution A (0.6% Nonidet P-40 [NP-40], 150 mM NaCl, 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid [Hepes] [pH 7.9], 1 mM EDTA, and 0.5 mM phenylmethylsulfonyl fluoride [PMSF]). The cells were lysed with five strokes of the pestle. After transfer to a 2-ml tube, debris was pelleted by briefly centrifuging at 2,000 rpm for 30 s. The supernatant was transferred to 2-ml tubes, incubated on ice for 5 min, and centrifuged for 10 min at 5,000 rpm. Nuclear pellets were then resuspended in 300 µl of solution B (25%glycerol, 20 mM Hepes [pH 7.9], 420 mM NaCl, 1.2 mM MgCl2, 0.2 mM EDTA, 0.5 mM dithiothreitol [DTT], 0.5 mM PMSF, 2 mM benzamidine, 5 µg/ml pepstatin, 5 µg/ml leupeptin, and 5 µg/ml aprotinin) and incubated on ice for 20 min. The mixture was transferred to microcentrifuge tubes, and nuclei were pelleted by centrifugation at 14,000 rpm for 1 min. Supernatants containing nuclear proteins were divided into aliquots, frozen, and stored at 80°C.

Protein quantitation was performed using the Bio-Rad protein assay dye reagent (Bio-Rad, Hercules, CA).

Electrophoretic Mobility Shift Assay

The oligonucleotide used for protein binding was to the TNF B3 binding site, 5'-CAAACAGGGGGCTTTCCCTCCTC-3' (Promega, Madison, WI). End labeling was performed by T4 kinase in the presence of [³²P]ATP. Labeled oligonucleotides were purified on a Sephadex G-50 M column (Pharmacia Biotech, Inc., Piscataway, NJ). An aliquot of 5 µg of nuclear protein was incubated with the labeled double-stranded probe (~ 50,000 cpm) in the presence of 5 µg of nonspecific blocker, poly(dI-dC) in binding buffer (10 mM Tris-HCl [pH 7.5], 100 mM NaCl, 1 mM EDTA, 0.2% NP-40, and 0.5 mM DTT) at 25°C for 20 min. Specific competition was performed by adding 100 ng of unlabeled double-stranded oligonucleotide; for nonspecific competition, 100 ng of unlabeled double-stranded OCT oligonucleotide (Promega) that does not bind NF-B was added. The mixture was separated by electrophoresis on a 5% polyacrylamide gel in 1× Tris glycine EDTA buffer. Gels were vacuum-dried and subjected to autoradiography and analysis on a PhosphorImager.

Statistics

Comparisons between experimental groups were made using a factorial analysis of variance (ANOVA) and a post hoc comparison (e.g., Fisher's exact test) or an unpaired Student's t test (two group comparisons) using a commercial software package (StatView, 4.01; SAS Institute, Cary, NC). A P value of 0.05 was accepted as significant and a value of 0.1 was considered a trend. P values are provided in the text and figures where comparisons were made.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Blockade of coagulation with FFR-FVIIa either at the time of intratracheal LPS or as a rescue, 6 h later, protected rats from acute pulmonary injury. This was reflected in histology, BAL measurements of LDH, protein and cell counts, local TF expression, and local proinflammatory cytokine production. Protection was independent of changes in NF-B translocation, and could not be wholly explained by reduction of fibrin or thrombin, downstream products that regulate inflammation. These findings are described in detail below.

Lung Histology and Fibrin Deposition

Histologic examination was performed on the lungs of FFR-FVIIa- and saline-treated animals. Administration of FFR-FVIIa at the time of intratracheal LPS attenuated lung injury, but did not block inflammatory cell infiltration. Lungs from normal control rats had normal appearing alveolar septae with minimal parenchymal fibrin staining, no intra-alveolar inflammation, and no fibrin matrices (Figures 1A and 1D). Lung histopathology from LPS injury control animals showed a moderate inflammatory cell infiltrate in alveolar spaces and mild edema by 6 h that progressively increased in extent and distribution, until 24 h, when peak cellular inflammation occurred. The inflammatory cells were predominantly polymorphonuclear cells (PMNs) and were clumped within the alveolar spaces by 24 h (Figure 1B). Cell aggregates were adherent to alveolar walls and associated with dense deposits of fibrin along the alveolar septae and lumens of small vessels. Macrophages stained inconsistently for fibrin (Figure 1E). By 48 h, the inflammatory response had subsided leaving a few residual cells and minimal edema, but fibrin deposition remained similar to that seen at 24 h (micrographs not shown). In contrast, LPS-treated animals that received FFR-FVIIa had a peak inflammatory response at 24 h consisting of smaller, less vacuolated inflammatory cells within the alveoli that did not appear adherent to one another or to the alveolar epithelium. PMNs were again the predominant inflammatory cell, but these were fewer in number compared with LPS injury control animals (Figure 1C). Fibrin deposition was less prominent within alveoli and was not associated with aggregates of inflammatory cells. Fibrin staining along the alveolar septae was less confluent compared with the lungs of LPS-injured animals treated with saline (Figure 1F). By 48 h, resolution of inflammation was complete in animals treated with FFR-FVIIa (micrographs not shown).

View larger version (129K): [in this window] [in a new window]   View larger version (131K): [in this window] [in a new window]   View larger version (130K): [in this window] [in a new window]   View larger version (126K): [in this window] [in a new window]   View larger version (116K): [in this window] [in a new window]   View larger version (119K): [in this window] [in a new window]   Figure 1.   Lung histology after intratracheal LPS administration and intravenous FFR-FVIIa. Treatment with FFR-FVIIa protects against intra-alveolar consolidation of inflammatory cells and reduces fibrin deposition 24 h after intratracheal LPS. Sections on the left (A, B, and C  ) are stained with hematoxylin and eosin, and panels on the right (D, E, and F) are stained with antibodies to rat fibrin/ fibrinogen. Uninjured control rats had thin, nonedematous septae without intra-alveolar inflammation (A); fibrin staining was rare along alveolar septae and nonexistent within alveoli (D). Lungs from saline-treated, injured rats have diffuse septal edema as well as several areas of intra-alveolar consolidation mainly composed of neutrophils (B); fibrin was densely deposited over masses of consolidated inflammatory cells, particularly marked in small vessels and along alveolar septae (E ). Lungs from FFR-FVIIa-treated, injured animals also had an intra-alveolar inflammatory infiltrate, but with few areas of consolidation and less septal edema (C ); fibrin deposition was less confluent along alveolar septae and intra-alveolar inflammatory cells were not associated with dense deposits of fibrin (F). All images are at ×400 magnification except for A, which is at ×200.

BAL Protein and LDH

LPS-dependent increases in BAL protein and LDH were attenuated by FFR-FVIIa (Figures 2A and 2B). Maximal protein leakage occurred at 24 h after injury in both treated and untreated animals (Figure 2A). Animals treated with FFR-FVIIa had ~ 50% less protein in the BAL fluid at 24 h than untreated animals (P < 0.0001). As with the protein measurements, BAL LDH activity peaked at 24 h after injury in both groups, but was approximately one-third of the control value in animals that received FFR-FVIIa (Figure 2B) (P < .0001).

View larger version (26K): [in this window] [in a new window]   Figure 2.   BAL protein, LDH activity, and cell counts after intratracheal LPS administration and intravenous FFR-FVIIa. In BAL samples obtained at 0, 6, 12, 24, and 48 h after administration of intratracheal LPS, treatment with FFR-FVIIa (open bars) decreased protein (A) and LDH activity (B), and increased cell counts (C) compared with saline controls (filled bars). (D) Treatment with FFR-FVIIa (diamonds, dotted line) yielded proportionately more monocytes compared with saline controls (squares, solid line) at 12 h after injury. *P < 0.0001.

BAL cell counts are shown in Figure 2C. The total number of recovered cells/mm³ reached a maximum at 24 h after injury in both treated and untreated animals. In contrast to LDH and protein measurements, total cell counts in BAL fluid were significantly higher in FFR-FVIIa-treated compared with untreated animals (P < 0.05) (Figure 2C). Animals that received FFR-FVIIa, but not LPS, did not have significant differences in BAL protein levels, LDH activity, or cell counts compared with normal control animals (data not shown).

FFR-FVIIa Levels

Plasma and BAL levels of FFR-FVIIa were measured to determine the amount of drug present at the time of maximum therapeutic effect, 24 h after LPS instillation. The large intravenous bolus of FFR-FVIIa used in this study produced detectable drug levels in plasma for 24 h after administration (Figure 3). Plasma levels of drug were no longer detectable by 48 h. BAL drug levels were obtained at 24 h and were similar to plasma levels at that time (data not shown).

View larger version (32K): [in this window] [in a new window]   Figure 3.   TF mRNA expression before and after intratracheal administration of LPS. Lungs were obtained from rats at 0, 6, 12, 24, and 48 h (n = 2 for each time point), flash frozen, homogenized, and RT-PCR was performed using primers for tissue factor (A, top panel) and GAPDH control (A, bottom panel). Densitometry of RT-PCR gel showed significantly less tissue factor mRNA transcription in FFR-FVIIa-treated animals over the course of the experiment compared with injured controls (B). Filled bars, LPS; open bars, LPS+FFR-FVIIa.

Tissue Factor Expression

One mechanism by which FFR-FVIIa may affect inflammation is through downregulation of TF production, the receptor for both endogenous FVIIa and FFR-FVIIa. To test this hypothesis, TF mRNA was measured in whole lung by PCR. In control rats, TF mRNA was constitutively limited in the lung, but within 6 h of injury it had increased significantly. LPS instillation into rat lung rapidly increased expression of TF. TF mRNA remained elevated throughout the study period, peaking at 24 h. By 48 h, TF transcription was decreasing. Treatment with FFR-FVIIa significantly decreased TF transcription at all time points after LPS instillation (Figure 4).

View larger version (14K): [in this window] [in a new window]   Figure 4.   FFR-FVIIa plasma levels. FFR-FVIIa was detectable in the plasma by ELISA up to 24 h after a single IV dose of 10 mg/kg.

Effect of Thrombin Inhibition on LPS Lung Injury

Another mechanism by which FFR-FVIIa may affect LPS-induced lung injury is by blocking generation of downstream products, such as thrombin. To test this hypothesis, the synthetic thrombin inhibitor Hirulog was given intravenously to rats at the time of intratracheal instillation of LPS. At 24 h, the time of maximal injury, animals treated with Hirulog had decreased protein concentration by 51% in the BAL fluid (P < 0.05). In contrast to animals treated with FFR-FVIIa, Hirulog treatment did not significantly reduce LDH activity or alter cell counts compared with LPS-injured rats that received saline.

Effect of FFR-FVIIa on Cytokine Responses

FFR-FVIIa may affect LPS-induced lung injury by reducing production of specific proinflammatory cytokines or augmenting the production of the anti-inflammatory cytokine IL-10. Cytokine profiles after lung injury with LPS were analyzed in plasma and BAL samples from FFR-FVIIa- and saline-treated rats. Plasma IL-1 and IL-10 levels reached a maximum 24 h after LPS and were not significantly altered by treatment with FFR-FVIIa (Figures 5A and 5C). LPS-dependent increases in plasma IL-6 were maximal at 6 h in both treated and untreated animals with no significant difference between the two groups (Figure 5B). In LPS-injured animals treated with saline, plasma TNF- peaked at 6 h, declined, and then rose again at 24 h. Treatment with FFR-FVIIa markedly attenuated the second, 24-h peak of TNF- (P < 0.0001) (Figure 5D).

View larger version (26K): [in this window] [in a new window]   Figure 5.   Plasma cytokines after LPS administration and intravenous FFR-FVIIa. In animals treated with intravenous saline (squares) or FFR-FVIIa (diamonds), plasma was obtained at 0, 6, 12, 24, and 48 h after intratracheal administration of LPS. TNF-, IL-1, IL-6, and IL-10 levels were measured by ELISA. Treatment with FFR-FVIIa had no significant effects on plasma levels of IL-1, IL-6, or IL-10 (A-C), but signficantly attenuated TNF- levels at 24 h after injury (D) (*P < 0.0001).

In contrast to plasma values, BAL levels of IL-1 and IL-6 levels were significantly altered by treatment with FFR-FVIIa (Figure 6), suggesting a possible mechanism by which FFR-FVIIa attenuates LPS induced lung injury. BAL from LPS-injured animals yielded a bimodal pattern of IL-1 release with a smaller, initial peak at 6 h and maximal level at 24 h. In the animals treated with FFR-FVIIa, a single peak of IL-1 was measured at 12 h that was ~ 50% lower than the maximum level found at 24 h in the untreated animals (P < 0.01) (Figure 6A). The peak response in IL-6 to LPS occurred at 6 h and was also attenuated by FFR-FVIIa, but the time course of the IL-6 response was unaffected by the drug (P < 0.0001; Figure 6B).

View larger version (28K): [in this window] [in a new window]   Figure 6.   BAL cytokines after LPS administration and intravenous FFR-FVIIa. In animals treated with intravenous saline (squares) or FFR-FVIIa (diamonds)BAL was obtained at 0, 6, 12, 24, and 48 h after intratracheal LPS. TNF-, IL-1, IL-6, and IL-10 levels were measured by ELISA. Compared with saline-treated animals, treatment with FFR-FVIIa significantly altered the time course and decreased peak levels of IL-1 (*P = 0.002) (A) and IL-10 (*P < .0001) (C), and decreased peak levels of IL-6 (*P < 0.0001) (B). Treated animals also had small, but significant, increases in TNF- at 6 and 12 h after injury that were significantly lower by 24 h after injury compared with saline-treated animals (P = 0.0003) (D).

FFR-FVIIa treatment did not augment IL-10 production in BAL; therefore, increased production of this anti-inflammatory cytokine was not a major mechanism by which treatment attenuates lung injury. Maximum IL-10 level shifted from 24 h in LPS controls to 12 h after FFR-FVIIa therapy and was ~ 60% lower than the maximum level in the untreated LPS group (P < 0.0001) (Figure 6C).

The effect of FFR-FVIIa on the TNF- response to LPS in the BAL fluid was less pronounced than it was in plasma, suggesting that the mechanism of protection was not through a decrease in TNF- levels. In contrast, TNF- was maximal at 6 h in both treated and untreated animals, with a small but statistically significant increase in TNF- levels in the treated group compared with those of untreated animals (P < 0.01; Figure 6D). FFR-FVIIa infusion into uninjured control rats did not increase plasma or BAL cytokine levels over baseline (data not shown).

Nuclear Translocation of NF-B

NF-B is involved in transcription of TF, TNF-, IL-1, and IL-6 (16, 17) and its nuclear translocation is associated with TF-FVIIa complex signaling through pathways involving MAP kinase activation (23). It was therefore possible that decreases in NF-B translocation may underlie the observed reduction in local IL-1, IL-6, and TF production in FFR-FVIIa-treated animals. Compared with uninjured controls, intratracheal LPS increased NF-B nuclear translocation by 6 h after injury. Treatment with FFR-FVIIa did not significantly reduce nuclear translocation of NF-B at 6 h (Figure 7) or 24 h (data not shown) after intratracheal instillation of LPS compared with saline-treated animals.

View larger version (79K): [in this window] [in a new window]   Figure 7.   Effects of intratracheal LPS and intravenous FFR-FVIIa or saline on nuclear translocation of NF-B. Lungs were obtained from rats at 0, 6, and 24 h (n = 2 for each time point), flash frozen, homogenized, and nuclei were extracted per the protocol outlined in MATERIALS AND METHODS. EMSA was performed to assay nuclear translocation of NF-B. Translocation of NF-B was increased by 6 h after intratracheal LPS (lanes 3 and 4) compared with uninjured control animals (lanes 1 and 2). Treatment with intravenous FFR-FVIIa at the time of LPS injury did not significantly decrease nuclear translocation of NF-B (lanes 5 and 6). There was no effect on NF-B translocation at 24 h after injury (data not shown).

Rescue with FFR-FVIIa Attenuates Lung Injury

Based on the findings that FFR-FVIIa reduced both lung injury and the release of proinflammatory cytokines after instillation of LPS, a rescue experiment was performed in which animals were given FFR-FVIIa 6 h after injury because TF was already increased at that time point. Measurements were done 24 h after LPS instillation, the time of maximum injury. As with the initial experiment, LDH and protein levels were significantly reduced (P < 0.0001) and cell counts were significantly increased (P < 0.02) in FFR-FVIIa-treated compared with saline-treated animals (Figure 8). Additionally, LPS-dependent cytokine responses were also similar in rescue animals as compared with those treated with FFR-FVIIa at the time of injury (Figures 9A and 9B). Plasma TNF- was significantly reduced (P < 0.01) by delayed treatment, but plasma IL-1, IL-6, and IL-10 levels at 24 h were not significantly different from untreated animals. Treatment significantly reduced BAL levels of IL-6 (P < 0.01), IL-10 (P < 0.01), and IL-1 (P = 0.02). The BAL TNF- levels were not significantly affected by FFR-FVIIa treatment.

View larger version (13K): [in this window] [in a new window]   Figure 8.   Rescue with FFR-FVIIa: lung injury. Rescue with FFR-FVIIa 6 h after injury with intratracheal LPS significantly decreased BAL levels of protein (A) and LDH (B), and increased BAL cell counts (C) compared with treatment with saline at 24 h after injury, the time of maximum lung inflammation (*P < 0.0001).

View larger version (20K): [in this window] [in a new window]   Figure 9.   Rescue with FFR-FVIIa: cytokine levels. Rescue with FFR-FVIIa 6 h after injury with intratracheal LPS significantly decreased plasma TNF- (shaded bars) at 24 h (*P = 0.005), but did not affect IL-1 (open bars), IL-6 (filled bars), or IL-10 (hatched bars) plasma levels (A) compared with injury controls. BAL levels of IL-1, IL-6, and IL-10 were significantly reduced in FFR-FVIIa rescue animals 24 h after injury compared with injury controls (B). *P = 0.02, 0.003, and 0.0003, for IL-1, IL-6, and IL-10, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Blockade of initiation of coagulation at TF-FVIIa complex with FFR-FVIIa attenuated local cytokines, fibrin deposition, and lung injury in rats exposed to tracheal LPS regardless of whether treatment was administered at the time of injury or as a rescue, 6 h later. This is the first demonstration that initiation of coagulation has a major influence on the evolution of pulmonary inflammation after injury of the epithelial surface of the lung with LPS. Furthermore, protection was associated with a decrease in TF mRNA, but was independent of significant reductions in NF-B nuclear translocation at 6 and 24 h; consistent with the preserved initial responses of the early proinflammatory cytokines, TNF- and IL-1.

In several species, including humans, TF-FVIIa-dependent PCA is an important mediator of intra-alveolar fibrin deposition during ALI (1). Immunologically, fibrin is proinflammatory, influencing both vascular permeability and chemotaxis (26, 27). Physiologically, fibrin may contribute to impaired gas exchange, a hallmark of ALI (1). Yet TF-FVIIa blockade dramatically protected animals from LPS, despite only partial attenuation of fibrin deposition. In a baboon model of sepsis-induced ARDS, we recently showed that gas exchange and lung edema were improved after blockade of coagulation despite residual fibrin (28), a finding similar to those seen after treatment with tissue factor antisense plasmid in mouse kidneys in a model of endotoxemia (29). Absence of alveolar fibrin deposition may not be necessary to provide lung protection with TF blockade. Therefore, blockade of initiation of coagulation may reduce inflammation independent of its effects on fibrin deposition.

One likely mechanism by which blockade of TF-FVIIa complex protects the lung after LPS exposure is by reducing local production of proinflammatory cytokines IL-1 and IL-6. IL-1 increases vascular permeability and primes both endothelial and inflammatory cells for activation (30, 31), and when given intratracheally, it produces an inflammatory response similar to that of LPS (14). In humans with ARDS, both IL-1 and IL-6 are elevated in BAL fluid (32) and IL-1 is associated with increased proinflammatory activity and production of mediators of local tissue damage such as reactive oxygen species and myeloperoxidase (33). Additionally, IL-6 is associated with increased severity of septic shock and multiple organ failure in animal models of ARDS (9, 10, 34). TNF- is also found in the BAL of laboratory animals and humans with ALI, but the protective effect of FFR-FVIIa seems to be independent of TNF-. This finding supports previous experimental studies of ALI and sepsis (10, 33).

FFR-FVIIa-associated reductions in local IL-1 and IL-6, as well as these and other influences on cellular signal transduction, may have affected the recovery of inflammatory cells in the BAL. Although the increase in cell recovery at 24 h after injury in FFR-FVIIa-treated animals may appear surprising in view of the protection observed with TF-FVIIa blockade, this may simply reflect decreases in cell adhesion or activation, rather than an increase in migration, a theory consistent with the FFR-FVIIa- induced decreases in local IL-1 production (35). We are currently in the process of investigating these possibilities.

FFR-FVIIa induced reductions in IL-6 and IL-1 most likely by direct effects on TF signaling, reductions in thrombin activity, or both. That the effects of treatment with FFR-FVIIa affect inflammation and proinflammatory cytokine production directly is supported by two major independent lines of evidence. First, our in vivo data show that treatment with FFR-FVIIa affected lung protein leakage, LDH activity, and cell counts after LPS instillation, whereas inhibition of thrombin with Hirulog only affected protein leakage. Second, FFR-FVIIa treatment may reduce activation of thrombin-independent signaling pathways due to the drug's intrinsic lack of catalytic activity (17, 18). Proteolytically active FVIIa induces phosphatidylinsoitol-specific phospholipase C (PI-PLC) signaling independent of thrombin in cell lines that constitutively express TF (23). Furthermore, catalytic FVIIa is required for p42/44 and p38 mitogen activated protein kinase (MAPK) activation by TF (16, 17) independent of thrombin, FXa, and agonists for the thrombin receptor, protease-activated receptor-1 (PAR-1) (23, 36). This is important because in addition to its role in signaling by TF-FVIIa complex (36), in vitro MAP kinase activation results in gene expression of multiple cytokines including TF, IL-1, TNF, and IL-6 (37,38). Therefore, blockade of TF-FVIIa complex may reduce proinflammatory cytokine release and protect against lung injury by several mechanisms, the understanding of which will require the ability to distinguish the direct effects of TF-FVIIa on signal transduction in specific lung cells from those resulting from the lack of production of downstream coagulation products.

Alternatively, FVIIa may bind to a variety of cell surface receptors or combinations of those receptors resulting in activation of inflammatory pathways that precede the production of cytokines. Given the proteolytic nature of the FVIIa molecule, PARs have been investigated as possible transducers for signaling by MAPK and calcium-dependent pathways, both activated by FVIIa binding to the cell surface (23, 39). In some in vitro systems, PAR-2 appears to be involved in FVIIa signaling (23), but in other systems, FVIIa activation of MAPK pathways has been shown to be independent of activation of any of the known PARs (40). This does not exclude the possibility that an undiscovered PAR may be activated by FVIIa, or that its involvement in FVIIa signaling may be required. Furthermore, the mechanism by which FVIIa stimluates increases in intracellular calcium is not well defined. One possibility involves the PI-PLC pathway (23), yet a cell surface receptor or receptor complex for transducing this signal is not known, and is not one of the known PARs (40).

Previous studies have indicated that activation of downstream products of coagulation also contribute to tissue injury (4, 5, 41). Factor Xa and thrombin are serine proteases downstream from TF with independent inflammatory signaling functions (5, 42). Yet only inhibition of thrombin, not FXa, has been shown to be protective against shock in models of sepsis or endotoxemia (45, 46). In contrast, thrombin inhibition with Hirulog in the present study offered only partial protection against ALI. Therefore, it will be important to distinguish the direct and indirect signaling pathways activated by formation of TF-FVIIa complex to determine the potential effects of therapies targeting different levels of the extrinsic coagulation cascade.

As FFR-FVIIa treatment decreases MAP kinase activation in vitro, we had hypothesized that NF-B translocation would be decreased by treatment. Yet treatment with FFR-FVIIa was independent of NF-B translocation. Therefore, this mechanism is unlikely to explain the profound reductions in local IL-1, IL-6, and TF. Although previous studies have shown that NF-B activation is central to transcription of TF and IL-6 (21, 22), reduction of NF-B does not appear to be an integral part of the mechanism by which FFR-FVIIa provided protection after LPS. It is possible that blockade of coagulation may instead affect one or more other factors that regulate TF and IL-1 gene expression (47). Alternatively, FFR-FVIIa may not affect gene expression through a single factor, but instead may affect the manner in which several factors synergistically activate transcription (48, 51). We have not excluded the possibility that changes in NF-B after FFR-FVIIa are cell-specific and the EMSA was not sensitive enough to detect site-specific changes.

Rescue with FFR-FVIIa 6 h after intratracheal LPS attenuated lung injury as effectively as administration at the time of injury. By 6 h, TF transcription, which correlates with its cell surface expression (52), was already upregulated. The success of rescue intervention highlights two major points. First, evolution of LPS-related inflammation in the rat lung is reversible at the level of the TF-FVIIa complex. Second, protection in the rescue setting suggests that FFR-FVIIa may have promise as a novel therapy for established ARDS.

In summary, blockade of TF with intravenous FFR-FVIIa given to rats after LPS instillation into the lung attenuates lung injury and fibrin deposition in association with local decreases in inflammatory cytokine elaboration. The positive effect of this intervention as long as 6 h after LPS administration may offer a promising new approach to therapy of ALI. Future study will be needed to further clarify both the specific sequence of cellular mechanisms underlying lung protection after coagulation blockade, as well as the role of specific inflammatory cell activation and migration pathways upon activation of TF by FVIIa.