Thrombin Induces Nitric-oxide Synthase via Gα12/13-coupled Protein Kinase C-dependent I-κBα Phosphorylation and JNK-mediated I-κBα Degradation

An imbalance between thrombin and antithrombin III contributed to vascular hyporeactivity in sepsis, which can be attributed to excess NO production by inducible nitric-oxide synthase (iNOS). In view of the importance of the thrombin-activated coagulation pathway and excess NO as the culminating factors in vascular hyporeactivity, this study investigated the effects of thrombin on the induction of iNOS and NO production in macrophages. Thrombin induced iNOS protein in the Raw264.7 cells, which was inhibited by a thrombin inhibitor, LB30057. Thrombin increased NF-κB DNA binding, whose band was supershifted with anti-p65 and anti-p50 antibodies. Thrombin elicited the phosphorylation and degradation of I-κBα prior to the nuclear translocation of p65. The NF-κB-mediated iNOS induction was stimulated by the overexpression of activated mutants of Gα12/13(Gα12/13QL). Protein kinase C depletion inhibited I-κBα degradation, NF-κB activation, and iNOS induction by thrombin or the iNOS induction by Gα12/13QL. JNK, p38 kinase, and ERK were all activated by thrombin. JNK inhibition by the stable transfection with a dominant negative mutant of JNK1 (JNK1(−)) completely suppressed the NF-κB-mediated iNOS induction by thrombin. Conversely, the inhibition of p38 kinase enhanced the expression of iNOS. In addition, JNK and p38 kinase oppositely controlled the NF-κB-mediated iNOS induction by Gα12/13QL. Hence, iNOS induction by thrombin was regulated by the opposed functions of JNK and p38 kinase downstream of Gα12/13. In the JNK1(−) cells, thrombin did not increase either the NF-κB binding activity or I-κBα degradation despite I-κBα phosphorylation. These results demonstrated that thrombin induces iNOS in macrophages via Gα12 and Gα13, which leads to NF-κB activation involving the protein kinase C-dependent phosphorylation of I-κBα and the JNK-dependent degradation of phosphorylated I-κBα.

Septic shock is the detrimental consequence of the host response to a bacterial infection. Septic shock involves hypotension, disseminated intravascular coagulation, and acute organ dysfunction accompanying the inflammatory and procoagulant responses. Hypotension and the signs of inadequate organ perfusion are the major manifestations of sepsis. The changes in the vascular reactivity do not depend on infectious pathogens but on disturbances in the coagulation and fibrinolytic cascade (1)(2)(3)(4)(5). The cascade of inflammatory and clotting reactions induces the development of disseminated intravascular coagulation, and microparticles cause the acute generation of thrombin (6). It has been shown that the level of the thrombin and antithrombin complexes is higher in patients with sepsis than in healthy control subjects (7). In sepsis, the level of antithrombin III (ATIII), 1 an endogenous coagulation inhibitor, is reduced as a result of complex formation with multiple activated clotting factors (8,9). The plasma ATIII level is low in septic patients as a consequence of ATIII consumption in severe sepsis (10). It is highly likely that the activation of prothrombin to thrombin by the coagulation pathway increases the level of unbound free thrombin in these patients.
Vascular hyporeactivity is attributable to excess nitric oxide (NO) production, a key gaseous molecule inducing the collapse of the cardiovascular system, by an inducible form of NOS (11). The inducible nitric-oxide synthase (iNOS) expression and NO production greatly affect the inflammatory processes (12,13). Proinflammatory cytokines such as the tumor necrosis factor-␣ induce NO production (14,15). Activated protein C, as a natural anticoagulant, regulates the coagulation system by inhibiting thrombin generation and attenuating the inflammatory responses induced by lipopolysaccharide (LPS). Protein C prevents LPS-induced hypotension by inhibiting excess NO production (16). ATIII inhibits nuclear factor-B (NF-B) activation in monocytes and endothelial cells (17). In addition, it has been shown that ATIII prevents LPS-induced hypotension by inhibiting NOS induction in animals (18).
Macrophages, which are effector cells in eliminating microorganisms and other noxious elements, participate in many complex immunological and inflammatory processes. They produce the cytokines that recruit other inflammatory cells, which are responsible for the diverse effects of inflammation. Septic shock syndrome results from an excessive triggering of endogenous inflammatory mediators, which are released primarily by activated macrophages (19).
In view of the imbalance between thrombin and antithrombin in septic patients as a result of the exhaustion of antithrom-bin and excess NO as culminating factors in vascular hyporeactivity, this study investigated the effect of thrombin on the production of NO in macrophages. In addition, the signaling pathways responsible for the induction of iNOS by thrombin were examined. This study reports for the first time that thrombin induces iNOS through I-B␣ phosphorylation and subsequent NF-B activation via the protein kinase C (PKC) and c-Jun N-terminal kinase (JNK) pathways in macrophages. Thrombin exerts mitogenic proliferation through the receptors coupled with the G␣ 12 and G␣ 13 proteins belonging to the G␣ 12 subfamily (20 -22). We determined whether thrombin-induced NF-B activation occurred via a pathway involving G␣ 12/13 . In view of the uncertainty of the PKC linkage to the G␣ 12 subfamily, we assessed whether PKC was associated with the G␣ 12 family in the activation of NF-B by thrombin. The G␣ 12 protein is linked to the Rho-directed guanine nucleotide exchange factor, p115, and the GTPase-activating protein, RasGAP1 (23,24). The G␣ 12/13 proteins are implicated in the Rho-dependent cytoskeletal shape change and JNK activation (20,25,26). This study further determined that JNK was involved in the degradation of the phosphorylated I-B␣ downstream of G␣ 12/13 .

EXPERIMENTAL PROCEDURES
Reagents-LB30057 was the kind gift from LG Biotech Inc. (Daeduk, Korea). [␥-32 P]ATP (3000 mCi/mmol) was obtained from Amersham Biosciences. Horseradish peroxidase-conjugated goat anti-rabbit IgG and 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium were supplied from Invitrogen. Alkaline phosphatase-conjugated goat antimouse IgG was purchased from Kirkegaard & Perry Laboratories (Gaithersburg, MD). Anti-c-Rel (p65), anti-p50, and anti-I-B␣ antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-phospho-I-B␣ antibody was supplied from New England Biolabs (Beverly, MA). Anti-iNOS antibody was obtained from Transduction Laboratories (Lexington, KY) or Santa Cruz Biotechnology. Horseradish peroxidase-conjugated and fluorescein isothiocyanate-conjugated anti-rabbit IgG were obtained from Zymed Laboratories Inc. (San FIG. 1. Induction of iNOS by thrombin. A, the time course of iNOS induction and NO production. The macrophages were incubated with or without thrombin (10 units/ml) or LPS (1 g/ml), and the iNOS levels were immunochemically determined. The amount of nitrite in the medium was monitored, as described under "Experimental Procedures." B, the effect of varying concentrations of thrombin on iNOS induction. Both iNOS expression and NO production were measured in the cells incubated with 1, 10, or 100 unit(s) of thrombin per ml for 18 h. C, the effect of LB30057 on the induction of iNOS by thrombin. The cells were treated with 0.1-100 M of LB30057 for 30 min and further incubated with thrombin (10 units/ml) for 18 h. Subsequently, the iNOS expression level was assessed. Each lane was loaded with 30 g of the cytosolic proteins. The data represent the mean Ϯ S.E. with 6 separate experiments.

iNOS Induction by Thrombin
Francisco, CA). PD98059 was obtained from Calbiochem. Thrombin and other reagents in the molecular studies were supplied from Sigma. The Limulus Amoebocyte Lysate test (i.e. an endotoxin test using the gel clot method) showed that the thrombin was endotoxin-free with the sensitivity limit of 0.06 enzyme units/ml. Activated mutants of G␣ 12/13 (G␣ 12/13 QL), wild types of G␣ 12/13 (G␣ 12/13 W), and JNK1 dominant negative mutant (KmJNK1) were kindly provided from Dr. N. Dhanasekaran (Fels Institute for Cancer Research and Molecular Biology, Department of Biochemistry, Temple University).
Cell Culture-The Raw264.7 cell line was obtained from American Type Culture Collection (ATCC, Manassas, VA) and maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 50 units/ml penicillin, and 50 g/ml streptomycin at 37°C in a humidified atmosphere with 5% CO 2 . Raw264.7 cells were plated at a density of 5 ϫ 10 6 /10-cm 2 dish and preincubated for 24 h at 37°C. For all experiments, cells were grown to 80 -90% confluency and were subjected to no more than 20 cell passages. To compare NO production, Raw264.7 cells were incubated with 1 g/ml of LPS (Escherichia coli 026:B6; Difco).
Assay of NO Production-NO production was monitored by measuring the nitrite level in the culture medium. This was performed by mixing with Griess reagent (1% sulfanilamide, 0.1% N-1-naphthylenediamine dihydrochloride, and 2.5% phosphoric acid). Absorbance was measured at 540 nm after incubation for 10 min.
Immunoblot Analysis-SDS-PAGE and immunoblot analyses were performed according to the procedures published previously (27). Cells were lysed in the buffer containing 20 mM Tris⅐Cl (pH 7.5), 1% Triton X-100, 137 mM sodium chloride, 10% glycerol, 2 mM EDTA, 1 mM sodium orthovanadate, 25 mM ␤-glycerophosphate, 2 mM sodium pyrophosphate, 1 mM phenylmethylsulfonyl fluoride, and 1 g/ml leupeptin. Cell lysates were centrifuged at 10,000 ϫ g for 10 min to remove debris. The proteins were fractionated using a 7.5% separating gel to assess the level of iNOS, whereas I-B␣ and its phosphorylated form were determined using a 12% separating gel. Briefly, the fractionated proteins were electrophoretically transferred to nitrocellulose paper. Cytosolic iNOS was immunoblotted with monoclonal anti-iNOS antibody, whereas polyclonal anti-I-B␣ and anti-phosphorylated I-B␣ antibodies were used to assess I-B␣ and its phosphorylated form, respectively. Cell lysates were centrifuged at 10,000 ϫ g for 10 min to remove debris. Activated JNK, p38 kinase, and ERK in cell lysates were immunochemically assessed using the specific antibodies, which recognized the active phosphorylated forms. The levels of unphosphorylated JNK, p38 kinase, and ERK1/2 were measured using the respective antibody directed against each MAP kinase. The secondary antibodies were horseradish peroxidase-or alkaline phosphatase-conjugated anti-IgG antibody. Nitrocellulose paper was developed using 5-bromo-4-chloro-3indolyl phosphate/4-nitro blue tetrazolium chloride or developed using ECL chemiluminescence system (Amersham Biosciences).
Preparation of Nuclear Extracts-Culture dishes were washed with ice-cold PBS. The dishes were then scraped and transferred to microtubes. Cells were allowed to swell by adding 100 l of lysis buffer (10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.5% Nonidet P-40, 1 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride). Tubes were vortexed to disrupt cell membranes. The samples were incubated for 10 min on ice and centrifuged for 5 min at 4°C. Pellets containing crude nuclei were resuspended in 50 l of the extraction buffer, containing 20 mM HEPES (pH 7.9), 400 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride, and then incubated for 30 min on ice. The samples were centrifuged at 15,800 ϫ g for 10 min to obtain the supernatant containing nuclear extracts. The nuclear extracts were stored at Ϫ70°C until use. Gel Retardation Assay-A double-stranded DNA probe for the consensus sequence of nuclear factor-B (NF-B, 5Ј-AGTTGAGGGGACTT-TCCCAGGC-3Ј) was used for gel shift analysis after end-labeling of the probe with [␥-32 P]ATP and T 4 polynucleotide kinase. The reaction mixture contained 2 l of 5ϫ binding buffer containing 20% glycerol, 5 mM MgCl 2 , 250 mM NaCl, 2.5 mM EDTA, 2.5 mM dithiothreitol, 0.25 mg/ml poly(dI-dC), and 50 mM Tris⅐Cl (pH 7.5), 8 g of nuclear extracts, and sterile water in a total volume of 10 l. Incubations were initiated by addition of 1-l probe (10 6 cpm) and continued for 20 min at room temperature. In some experiments, an aliquot of nuclear extracts (8 g each) was incubated with 2 g of highly specific anti-p65 and/or p50 antibody (NF-B) at room temperature for 1 h, according to the method described previously (28). Samples were loaded onto 4% polyacrylamide gels at 100 V. The gels were removed, fixed, and dried, followed by autoradiography.
Immunocytochemistry of p65-Cells were grown on Lab-TEK chamber slides (Nalge Nunc International Corp.) and incubated in serumdeprived medium for 24 h. Standard immunocytochemical method was used as described previously (29). For immunostaining, cells were fixed in 100% methanol for 30 min and washed three times with PBS. After blocking in 5% bovine serum albumin in PBS for 1 h at room temperature or overnight at 4°C, cells were incubated for 1 h with polyclonal rabbit anti-p65 antibody (1:100) in PBS containing 0.5% bovine serum albumin. Cells were incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (1:100) after serial washings with PBS. Counter-staining with propidium iodide verified the location and integrity of nuclei. Stained cells were washed and examined using a laser scanning confocal microscope (Leica TCS NT, Leica Microsystems, Wetzlar, Germany).
Transient Transfection-Cells were plated at a density of 0.5 ϫ 10 6 cells/well in a 6-well dish and transfected the following day. Briefly, cells were incubated with G␣ 12/13 W or G␣ 12/13 QL plasmid (1 g each of the plasmid DNA) and 3 l of LipofectAMINE reagent (Invitrogen) in 1 ml of antibiotic-free MEM for 3 h. Culture medium was changed with serum-free MEM with antibiotics, and cells were further incubated for 12 h to assess iNOS expression and for 1-3 h to monitor NF-B activity. The transfection efficiency was ϳ50%, as determined by transfection containing each plasmid and Transfectam and then incubated at 37°C in a humidified atmosphere of 5% CO 2 for 6 h. After addition of 6.25 ml of MEM with 10% fetal bovine serum, cells were incubated for an additional 48 h, and geneticin was added to select the resistant colonies. In the present study, a mixture of stably transfected JNK1(Ϫ) clones were used. JNK1(Ϫ) cells had no inducible JNK activity, as monitored by phosphorylation of glutathione S-transferase-c-Jun (stimulation by 10 g/ml bovine collagen(I) for 1 h). The expression of JNK1, but not JNK2, was decreased 43% by JNK1(Ϫ) transfection.
Statistical Methods-One-way analysis of variance procedures were used to assess significant differences among treatment groups. For each significant effect of treatment, the Newman-Keuls test was used for comparisons of multiple group means. The criterion for statistical significance was set at p Ͻ 0.05 or p Ͻ 0.01.

Induction of iNOS by Thrombin-
The effect of thrombin on NO production in macrophages was first assessed. Thrombin (10 units/ml, 12 h) increased the NO production by 4 -5-fold. NO production was maintained for at least 48 h (Fig. 1A). Western blot analysis confirmed iNOS induction by thrombin at the time points examined (Fig. 1A). LPS (1 g/ml), which was used as a positive control, induced iNOS with a concomitant increase in NO production (Fig. 1A). The extent of iNOS induction by thrombin was comparable with that by LPS. iNOS induction by thrombin was potentiated by the presence of serum (data not shown). In order to characterize the effects of thrombin on the iNOS expression per se, the subsequent experiments were conducted with cells starved of serum for 24 h.
The effect of varying thrombin concentrations on the iNOS expression was next examined. Although 1 unit/ml thrombin (i.e. equivalent to the plasma concentration observed in healthy control animals (30)) affected iNOS expression minimally, 10 units/ml thrombin or greater notably induced the protein (Fig.  1B). This was in agreement with the Western blot results.
Therefore, thrombin exhibited a threshold effect. The level of thrombin was increased severalfold by sepsis with a reciprocal decrease in the ATIII levels (7,10). Hence, the concentration used in this study was considered appropriate for assessing the role of thrombin in sepsis.
LB30057 is a direct thrombin inhibitor with a K i of 0.38 nM (31). Treatment of the Raw264.7 cells with 1 M or greater LB30057 completely blocked the induction of iNOS by thrombin (18 h), which demonstrated that thrombin induces iNOS per se presumably through the thrombin receptor (Fig. 1C). Reverse transcriptase-PCR analysis confirmed that the iNOS induction by thrombin accompanied an increase in the mRNA level (data not shown).
Activation of NF-B Transcription Factor-iNOS expression is controlled primarily by the transcription factor, NF-B (32). In order to determine whether or not the iNOS induction by thrombin was mediated by NF-B activation, the nuclear extracts prepared from the cells treated with thrombin for 0.5-12 h were probed with the radiolabeled NF-B consensus oligonucleotide ( Fig. 2A). NF-B was activated by thrombin with a band intensity of a slow migrating p65/p50 complex being increased from 30 min to 12 h. The p50/p50 homodimer complex migrated slightly faster in the cells treated with thrombin. Supershift analysis was carried out using anti-p65 and anti-p50 antibodies to confirm whether or not the retarded band consisted of the p65 and p50 proteins. A 20-fold excess of the NF-B probe abolished the band retardation (Fig. 2B). Either anti-p65 or the anti-p50 antibodies supershifted the retarded band. The addition of both anti-p65 and anti-p50 antibodies also caused a supershift with the reduction in the band intensity of the p65/p50 complex (Fig. 2B). These results suggest that the p65 and p50 proteins were the components actively iNOS Induction by Thrombin binding to the NF-B-binding site. The specificity of the thrombin effect on NF-B activation was verified by LB30057 (10 M) (Fig. 2C).
Because p65 was the major component of the NF-B activated by thrombin, this study determined the translocation of p65 into the nucleus. The Raw264.7 cells were treated with thrombin for 30 min to 3 h, fixed, and permeabilized. Immunocytochemistry showed that the p65 protein was located mainly in the cytoplasm of the control cells (Fig. 3A). In contrast, the p65 protein moved into the nucleus after the thrombin treatment. The nuclear integrity was confirmed by propidium iodide staining of the identical cells (Fig. 3A). The proteolytic degradation of the I-B␣ subunit preceded the translocation of NF-B to the nucleus. These studies were extended to determine whether or not NF-B activation by thrombin resulted from the degradation of I-B␣ (Fig. 3B). I-B␣ phosphorylation preceded I-B␣ degradation. Thrombin (10 unit/ml) increased I-B␣ phosphorylation at 15 min (Fig. 3B). The I-B␣ level was subsequently decreased between 30 min and 1 h. Therefore, thrombin activates NF-B through I-B␣ degradation following its phosphorylation.
iNOS Induction and NF-B Activation by G␣ 12/13 -Previously, it was reported (20) that thrombin binding to its receptor stimulates the guanine nucleotide exchange of G␣ 12 subfamily proteins. This study was interested in whether or not the G␣ 12/13 subunits were responsible for NF-B-mediated iNOS induction. G␣ 12 QL or G␣ 13 QL expression also notably increased NO production in the cells (Fig. 4A). In order to confirm that thrombin induced iNOS in the Raw264.7 cells through the G␣ 12/13 pathway, the cells were transfected with the either G␣ 12 W, G␣ 13 W, G␣ 12 QL, or G␣ 13 QL (G␣ 12 or G␣ 13 activated mutant) plasmid and then treated with thrombin. Thrombin induced iNOS in the cells transfected with the G␣ 12/13 W plasmid, which was comparable with that by either G␣ 12 QL or G␣ 13 QL. iNOS was not inducible in the cells transfected with the G␣ 15 QL (G␣ 15 activated mutant) plasmid, which was used as a negative control. This study next determined whether or not the activated mutants of G␣ 12/13 stimulated p65/p50 complex binding to the NF-B consensus oligonucleotide. Either G␣ 12 QL or G␣ 13 QL increased the band intensity of the p65/p50 DNA complex (Fig. 4B).
PKC-mediated NF-B Activation by Thrombin Downstream PMA completely inhibited both the iNOS induction and the increase in NO production by thrombin (Fig. 5A). PKC depletion also inhibited NF-B activation, I-B␣ phosphorylation, and I-B␣ degradation by thrombin (Fig. 5, B and C). These results show that PKC plays a role in the I-B␣ phosphorylation by thrombin.
In the case of the pathways of the G␣ 12 protein-coupled receptors, the PKC-dependent phosphorylation is linked to G␣ 12/13 activation (35). As an approach to determine whether or not PKC controlled iNOS induction downstream of G␣ 12/13 , iNOS expression was monitored in the G␣ 12/13 QL-transfected cells, which had been depleted of PKC. PKC depletion completely inhibited the iNOS induction by G␣ 12/13 QL (Fig. 5D).
The Role of MAP Kinases in the iNOS Induction by Thrombin-The MAP kinases were involved in iNOS expression (36). Subsequently, we determined whether the MAP kinases including JNK, p38 kinase, and ERK1/2 controlled iNOS induction by thrombin. Thrombin activated all three MAP kinases in the Raw264.7 cells (Fig. 6A). JNK and ERK1/2 phosphorylation was distinct at 30 min to 1 h, which gradually returned toward the control levels (i.e. 3-6 h). p38 kinase was only weakly phosphorylated by thrombin. Cells stably transfected with the dominant negative mutants or chemical inhibitors were used to assess the role of each MAP kinase in the induction of iNOS. JNK inhibition by the stable transfection with a dominant negative mutant of JNK1 (JNK1(Ϫ)) completely suppressed iNOS induction by thrombin (Fig. 6B). In contrast, p38 kinase inhibition by SB203580 (10 M) enhanced the enzyme expression (Fig. 6C). PD98059 (50 M) failed to affect the iNOS expression level. Hence, the induction of iNOS by thrombin was regulated by the distinct and opposed functions of JNK and p38 kinase.
Coupling of G␣ 12/13 to JNK-The activation of the G proteins stimulates the MAP kinases in a variety of cells (26,37,38). This study determined whether or not the MAP kinase pathways were connected with G␣ 12/13 by using either JNK1(Ϫ) cells or chemical inhibitors. First, the iNOS protein was monitored in the JNK1(Ϫ) cells that were transiently transfected with the plasmid encoding for either G␣ 12 QL or G␣ 13 QL. The expression of the active mutant of G␣ 12/13 failed to induce iNOS in the JNK1(Ϫ) cells (Fig. 7). SB203580 slightly enhanced iNOS induction in the cells expressing G␣ 12/13 QL (Fig. 7). PD98059 did not alter the induction of iNOS by G␣ 12 QL or G␣ 13 QL. Therefore, JNK and p38 kinase oppositely function downstream of G␣ 12/13 . JNK-dependent Degradation of Phosphorylated I-B␣-In order to assess whether or not the JNK pathway controls the NF-B activation by thrombin, the nuclear extracts, which were prepared from the cells treated with thrombin for 3 h, were probed with the radiolabeled NF-B consensus oligonucleotide. In the JNK1(Ϫ) cells, thrombin did not increase the NF-B binding activity (Fig. 8A, left). The basal NF-B binding to DNA was lower in the JNK1(Ϫ) cells than in the control cells (Fig. 8A, right). This is consistent with the observation that the nuclear translocation of p65 was blocked by the JNK1(Ϫ) stable transfection (Fig. 8B). However, I-B␣ was not degraded by the presence of thrombin in the JNK1(Ϫ) cells (Fig. 8C). The phosphorylated I-B␣ level in the JNK1(Ϫ) cells treated with thrombin was comparable with that of the control (Fig. 8C). These results support the notion that the JNK pathway is responsible for the degradation of phosphorylated I-B␣.

DISCUSSION
Overproduction of the proinflammatory cytokines in sepsis (e.g. tumor necrosis factor-␣) is critically involved in activating the coagulation system, leading to vascular hyporeactivity and disseminated intravascular coagulation (39,40). LPS and/or the cytokines induce iNOS expression in many cell types (41).
NO and the major inflammatory mediators are produced mainly by activated macrophages. Excessive NO production by iNOS plays a crucial role in activating the immune system and in the proinflammatory effects during LPS-induced septic shock (42)(43)(44). In particular, NO induces the collapse of vascular reactivity and the pathologic alterations (13,45). In clinical situations, there is a discrepancy between the serum endotoxin level and the mortality of the patients with Gram-negative sepsis (46). The persistent NO production in septic patients may result from other unknown mediator(s) that are generated by LPS. The endogenous coagulation inhibitors reverse the vascular hyporeactivity induced by LPS (16,18). Protein C and ATIII have been studied extensively as the protective modulators of septic shock (47,48).
Thrombin, a serine protease of the trypsin family, is a key enzyme of the blood clotting system. It is the key coagulant molecule that is commonly involved in two independent (e.g. contact and extrinsic systems) activation pathways. The coagulation pathway involves a series of reactions, which culminate in the production of sufficient thrombin (6). Thrombin converts fibrinogen to fibrin and participates in regulating numerous physiological and pathological processes. Thrombomodulin serves as a thrombin receptor. When thrombin is bound to thrombomodulin, it loses its procoagulant activity, and its inactivation by ATIII is accelerated with an enhancement of protein C activation (6). The generation of excessive thrombin leads to thrombosis, which is the major cause of morbidity and mortality. Thrombin in close proximity to the active mediators also plays a role in the diverse cellular responses in the vascular and avascular tissues (49). It has been reported that thrombin potentiates both interferon-␥ and tumor necrosis factor-␣induced NO production in C6 glioma cells (50). Thrombin stimulates the proliferation of smooth muscle cells and vascular disturbances through NF-B activation (51). This study found for the first time that thrombin induces iNOS in the macrophages via members of the G␣ 12 family. The inflammatory cytokines exert their biological effects through the cytokine receptor superfamily. Hence, the regulatory mechanisms for iNOS induction by thrombin appear to differ from those by inflammatory cytokines. This study found a threshold effect for thrombin in iNOS induction. The concentration of 10 units/ml markedly increased iNOS expression, whereas 1 unit/ml thrombin has a minimal effect. The threshold effect may reflect the septic pathological situation, wherein the level of unbound activated thrombin is increased as a result of the conversion of prothrombin to thrombin and the reciprocal consumption of ATIII in sepsis.
NF-B is a pleiotropic regulator of many genes (e.g. iNOS) involved in the immune and inflammatory responses (32). This study found that iNOS induction and excess NO production by thrombin are mediated primarily by NF-B activation. NF-B exists in the cytoplasm of unstimulated cells in a quiescent form bound to its inhibitor (52). Thrombin was found to activate the p65/p50 NF-B DNA binding complex and to induce the nuclear translocation of the p65 protein.
In this study, the role of the G␣ 12 family members downstream of thrombin signaling (35) in NF-B-mediated iNOS induction was verified by experiments using an activated mutant of G␣ 12 or G␣ 13 . It was shown that thrombin induces the stress fiber assembly via G␣ 12 -or G␣ 13 -coupled receptor activation (20). The G q -coupled receptor activates the downstream signals in a PKC-dependent, fully PKC-independent, or partially PKC-dependent pathway (54,55). Thrombin differentiates the normal lung fibroblasts to a myofibroblast phenotype via its receptor via a protein kinase C-dependent pathway (34). Macrophage activation by external stimuli causes the phosphorylation and degradation of I-B␣. I-B kinase activation by LPS is dependent on PKC and ERK (56). This study found that PKC was involved in the NF-B-mediated iNOS induction by thrombin via the phosphorylation and degradation of I-B␣. PKC depletion prevented the induction of iNOS by the acti-vated mutants of G␣ 12/13 , which raised the notion that the PKC pathway functions downstream of G␣ 12/13 activation.
Thrombin activated ERK in the endothelial cells (57). While this report was being revised, it was reported that thrombin activated p38 kinase in the platelets, which led to NF-B-dependent leukocyte recruitment (58). In this study, it was found that all three MAP kinases JNK, p38 kinase, and ERK1/2, were activated by thrombin. Among the MAP kinases, the JNK pathway was responsible for the induction of iNOS by thrombin, which was strongly supported by the lack of iNOS induction in the thrombin-treated JNK1(Ϫ) cells. The p38 kinase pathway oppositely regulated iNOS induction by thrombin. ERK1/2 activation was not responsible for iNOS induction, as evidenced by the results from the chemical inhibitor. The lack of iNOS induction by G␣ 12/13 in the JNK1(Ϫ) cells supports the concept that JNK serves as an essential pathway downstream of the G␣ 12/13 proteins. Again, JNK and p38 kinase oppositely control the NF-B-mediated iNOS induction by the activated mutants of G␣ 12/13 . Thus, the induction of iNOS by thrombin was regulated by the opposed functions of JNK and p38 kinase downstream of G␣ 12/13 .
The gel shift and immunoblot analyses revealed that the pathway involving JNK controlled NF-B activation in response to thrombin. The increase in NF-B DNA binding activity and the nuclear translocation of p65 were both completely abolished by the JNK1(Ϫ) transfection. The diminished NF-B DNA binding activity in the JNK1(Ϫ) cells may be due in part to the decrease in the basal NF-B activity. This study found for the first time that thrombin failed to degrade I-B␣ in the JNK1(Ϫ) cells despite its I-B␣ phosphorylation. These results strongly support the belief that the JNK pathway was responsible for degrading the phosphorylated I-B␣. The time course in I-B␣ phosphorylation and degradation by thrombin paralleled that in JNK activation.
The ubiquitin-proteasome pathway controls the timed destruction of the phosphorylated I-B␣ in order to activate NF-B (33,59). Recently, it was shown that the activation of the stress-activated protein kinase, JNK, by the forced expression of the constitutively active mutants of JNKK2 and members of the Jun family leads to the accumulation of ␤-TrCP, which mediates the ubiquitination of the phosphorylated I-B␣ via the recruitment of a ubiquitin ligase complex (53). Therefore, it is highly likely that the accumulation of phosphorylated I-B␣ and the failure of I-B␣ degradation by thrombin in the JNK1(Ϫ) cells might result from the inhibition of the ubiquitinproteasome pathway.
In summary, this study demonstrated that thrombin plays an important role in the vascular responsibility by inducing iNOS and NO production. In addition, thrombin activates the pathway coupled with G␣ 12/13 for enzyme induction. G␣ 12/13 activation then leads to the PKC-dependent phosphorylation of I-B␣ and the JNK-mediated I-B␣ degradation. The cellular signaling pathways, by which thrombin induces iNOS, may serve as the pharmacological targets for both preventing and treating vascular hyporeactivity in septic patients.