Thrombin Induces NO Release from Cultured Rat Microglia via Protein Kinase C, Mitogen-activated Protein Kinase, and NF-κB*

Microglia, brain resident macrophages, become activated in brains injured due to trauma, ischemia, or neurodegenerative diseases. In this study, we found that thrombin treatment of microglia induced NO release/inducible nitric-oxide synthase expression, a prominent marker of activation. The effect of thrombin on NO release increased dose-dependently within the range of 5–20 units/ml. In immunoblot analyses, inducible nitric-oxide synthase expression was detected within 9 h after thrombin treatment. This effect of thrombin was significantly reduced by protein kinase C inhibitors, such as Go6976, bisindolylmaleimide, and Ro31-8220. Within 15 min, thrombin activated three subtypes of mitogen-activated protein kinases: extracellular signal-regulated kinase, p38, and c-Jun N-terminal kinase/stress-activated protein kinase. Inhibition of the extracellular signal-regulated kinase pathway and p38 reduced the NO release of thrombin-treated microglia. Thrombin also activated nuclear factor κB (NF-κB) within 5 min, andN-acetyl cysteine, an inhibitor of NF-κB, reduced NO release. However, thrombin receptor agonist peptide (an agonist of protease activated receptor-1 (PAR-1)), could not mimic the effect of thrombin, and cathepsin G, a PAR-1 inhibitor, did not reduce the effect of thrombin. These results suggest that thrombin can activate microglia via protein kinase C, mitogen-activated protein kinases, and NF-κB but that this occurs independently of PAR-1.

Microglia are major immune effector cells in the central nervous system, and activation of microglia is a common phenomenon that appears when the brain suffers injury. Many studies have reported that activated microglia produce inflammatory mediators such as NO, tumor necrosis factor ␣, and prostaglandins, factors that affect the onset and progression of brain diseases (1)(2)(3)(4). For example, microglial activation precedes neuronal cell death in ischemia, and profound microglial activation is commonly observed in seriously injured regions of the brain (5,6). Furthermore, neuronal cell death because of ischemia can be reduced by the inhibition of microglial activation (7). The risk and progression of Alzheimer's disease and Parkinson's disease can also be reduced by anti-inflammatory drugs that suppress microglial activation (8,9). However, it has not been clearly shown how the microglial activation occurs in the injured brain.
Thrombin is a well known protease involved in blood coagulation and wound healing. In addition, thrombin seems to have diverse functions in many different types of cells. Thrombin induces proliferation of fibroblasts (10) and smooth muscle cells (11), induces neurite retraction and synapse reduction (12,13), and changes the morphology of astrocytes (14). Many studies have reported inflammatory functions of thrombin. Thrombin acts as a chemotactic agent for inflammatory cells such as monocytes, macrophages, and neutrophils and induces secretion of cytokines from these kinds of cells (15)(16)(17). More importantly, thrombin injected into a rat brain induces the infiltration of inflammatory cells, brain edema, and reactive gliosis (18). However, the effect of thrombin on microglial cells has not yet been studied, although microglia are the major immune effector cells in the brain.
The intracellular signaling mechanisms activated by thrombin are diverse. In endothelial cells, thrombin activates Src family tyrosine kinases, phosphatidylinositol 3-kinase, and mitogen-activated protein kinases (MAPKs) (19). 1 In vascular smooth muscle cells and endothelial cells, thrombin produces reactive oxygen species (ROS) (20,21). In astrocytes, thrombin activates protein kinase C and small GTP-binding protein Rho (14,22,23). Protease activated receptor-1 (PAR-1) has been reported to serve as a thrombin receptor in many types of cells including platelets, endothelial cells, and astrocytes (24, 25). In activating PAR-1, thrombin cleaves the extracellular N terminus of this receptor. The newly formed N terminus then binds to the receptor and activates it (26). Non-PAR thrombin receptors have also been reported on fibroblasts (27). On these cells, the non-PAR has high affinity for thrombin, and activation of the non-PAR induces expression of genes that are not induced by the activation of PAR-1. However, the intracellular signaling mechanisms linked to non-PAR have not yet been clarified. The results of this study revealed that thrombin induces NO release/iNOS expression in microglia via PKC, MAPK, and NF-B. However, PAR-1 does not seem to be involved in this function of thrombin.
viously (1,28). Briefly, the cortices were triturated into single cells in minimal essential medium (Life Technologies, Inc.) containing 10% fetal bovine serum (Hyclone, Logan, UT) and plated in 75-cm 2 T-flasks (0.5 hemisphere/flask) for 2-3 weeks. Then the microglia were detached from the flasks by mild shaking and filtered through a nylon mesh to remove astrocytes and clumped cells. Cells were plated in 24-well plates (5 ϫ 10 4 cells/well) or 60-or 100-mm dishes (0.1-5 ϫ 10 6 cells/dish). One hour later, the cells were washed to remove unattached cells before being used in experiments.
Measurement of MAPK Activation-Microglia (5 ϫ 10 5 ) were incubated in serum-free medium overnight and treated with 10 units/ml thrombin for 5, 15, 30, 60, and 120 min. The cells were then washed with ice-cold phosphate-buffered saline three times and lysed with 2ϫ SDS-PAGE sample buffer, and the lysate was applied to an 8% SDSpolyacrylamide gel. After electrophoresis the proteins were blotted onto a polyvinylidene difluoride membrane. Activation of MAPKs was examined by immunoblot analysis using antibodies specific for the phosphorylated forms of ERK, p38, and JNK (New England Biolabs).

Measurement of NF-B Activation by Electrophoretic Mobility Shift
Assay-Electrophoresis mobility shift assays were carried out as described previously (30). Microglia (2 ϫ 10 6 cells) were harvested and suspended in 900 l of a hypotonic solution (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride) including 0.5% Nonidet P-40 on ice for 5 min. The suspension was then centrifuged at 500 ϫ g for 10 min at 4°C, and the pellet (nuclear fraction) was saved. The nuclear fraction was resuspended in a buffer containing 20 mM HEPES, pH 7.9, 20% glycerol, 0.4M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride, incubated on ice for 60 min with occasional gentle shaking, and centrifuged at 12,000 ϫ g for 15 min. The crude nuclear protein in the supernatant was collected and stored at Ϫ70°C and later used for electrophoretic mobility shift assays. Two synthetic oligonucleotides (Genosys, Woodlands, TX) containing the NF-B binding sequence of the murine immunoglobulin light chain gene (5Ј-GGGAGTTGAGGGGACTTTCCGAGG-3Ј) and its complementary sequence were end-labeled using Klenow fragment and [␣-32 P]dCTP. The labeled DNA probe (approximately 0.2 ng) was incubated with 0.5 g of nuclear proteins in a reaction buffer containing 8.5 mM EDTA, 8.5 mM EGTA, 8% glycerol, 0.1 mM ZnSO 4 , 50 g/ml poly(dI-dC), 1 mM dithiothreitol, 0.3 mg/ml bovine serum albumin, and 6 mM MgCl 2 for 30 min. The reaction mixture was applied to an 8% polyacrylamide gel. After electrophoresis, the gel was dried, and an autoradiogram was obtained. For supershift assays, the nuclear extract was preincubated with 1 g of anti-p50 or anti-p65 antibody (Santa Cruz Biotechnology) for 30 min. The reaction mixture was subjected to electrophoresis through a 6% polyacrylamide gel.
Reverse Transcription-Polymerase Chain Reaction-Total RNA was isolated using RNAzol TM B (TEL-TEST, Inc., Friendwood, TX), and cDNA was prepared using reverse transcriptase that originated from Avian Myeloblastosis Virus (TaKaRa, Japan) according to the manufacturer's instructions. The PCR primers for the iNOS gene were 5Ј-GCAGAATGTGACCATCATGG-3Ј (sense primer) and 5Ј-ACAACCTT-GGTGTTGAAGGC-3Ј (antisense primer). PCR products were separated by electrophoresis in a 1.5% agarose gel and detected under UV light.

Thrombin Induces Microglial NO Release and iNOS Expression-
To determine whether thrombin induced microglial activation, we looked at nitrite formed from NO in the medium and the expression of iNOS, a prominent marker of microglial activation. Microglia were treated with thrombin at 5-20 units/ml for 48 h. Nitrite production increased in a dose-dependent manner: 6.9 Ϯ 0.8 (mean Ϯ S.E. of three samples unless indicated otherwise), 10.2 Ϯ 0.8, and 16.1 Ϯ 0.3 M nitrite was detected from 5 ϫ 10 4 cells treated with 5, 10, and 20 units/ml of thrombin, respectively, whereas 3.7 Ϯ 1.3 M nitrite was detected from untreated cells (Fig. 1A). Basal level of nitrite was not likely produced by nitric-oxide synthase because neither iNOS inhibitors (2-amino-5,6-dihydro-6-methyl-4H-1,3-thiazine and 1400W) nor nonselective nitric-oxide synthase inhibitors (N G -nitro-L-arginine methyl ester and N Gmonomethyl-L-arginine) had little effect on it (data not shown). iNOS expression was also detected by immunoblot analysis within 9 h with further increase for up to 21 h and maintained to 25 h (Fig. 1B). This suggested that the effect of thrombin on nitrite production was due to the expression of iNOS.
PKC Inhibitors Reduce NO Release from Thrombin-treated Microglia-In several types of cells, PKC has been known to mediate the function of thrombin (19,22,31). Furthermore, PKC has been reported as an important mediator of iNOS expression in microglia and astrocytes (32)(33)(34). We therefore examined whether a PKC pathway could be involved in the thrombin-induced nitrite production. In the presence of PKC inhibitors, bisindolylmaleimide (2 M), Go6976 (0.1 M), and Ro31-8220 (0.5 M), nitrite production was reduced to 32.8 Ϯ 0.1, 27.6 Ϯ 7.9, and 10.9 Ϯ 4% of that induced by thrombin in the absence of any inhibitor ( Fig. 2A). PKC inhibitors had little effect on the level of nitrite from untreated cells (data not shown). Thrombin-induced iNOS expression was also inhibited in the presence of bisindolylmaleimide and Go6976 (Fig. 2B). The reduced nitrite production and iNOS expression were not caused by any toxicity of these reagents, as was confirmed by the exclusion of trypan blue observed through a light microscope. These data, therefore, strongly suggest that a PKC pathway is involved in the thrombin effect.
Thrombin Stimulates Mitogen-activated Protein Kinase-Because MAPKs have been known to mediate microglial activation (28, 35), we examined whether thrombin induced the ac- tivation of MAPKs. Within 15 min, thrombin activated ERK, p38, and JNK/SAPK, which was confirmed by increased phosphorylation of tyrosine residues of these kinases as determined by immunoblot analysis. The activity status of ERK and p38 remained sustained for 120 min, whereas that of JNK/SAPK lasted up to 60 min but fell back to the control level after 120 min (Fig. 3). To test whether activation of the MAPKs could be involved in the thrombin-induced nitrite production, we treated microglia with thrombin in the presence of PD98059 and SB203580, inhibitors of the ERK pathway and p38 pathway, respectively. PD98059 (5 M) and SB203580 (5 M) significantly reduced nitrite production to 20.9 Ϯ 14 and 19.3 Ϯ 4.9% of that induced by thrombin in the absence of any inhibitor (Fig. 4). These results suggest that MAPKs, particularly ERK and p38, mediate NO release from microglia treated with thrombin.
Thrombin Activates NF-B-Because NF-B binding sites are present in the promoter region of the iNOS gene, we examined whether thrombin activated NF-B. The electrophoretic mobility shift assay showed that thrombin activated NF-B within 5 min. The activity was sustained for 15 min but was significantly reduced after 30 min (Fig. 5A, arrows). To investigate which subtypes of NF-B was activated by thrombin, we carried out supershift assays using antibodies against p50 and p65 (Fig. 5B). The intensity of the shifted bands was decreased, and a supershifted band appeared in the presence of antibody against p50 (Fig. 5B, arrowhead) but not in the presence of antibody against p65, indicating that p50 could be activated by thrombin.
We further examined whether the activation of NF-B was directly related to the induction of NO release. For this, NAC, a known NF-B inhibitor, was applied together with thrombin. NAC dose-dependently reduced the thrombin-induced nitrite production; in the presence of 5, 10, and 20 mM NAC, nitrite production was reduced to 71.5 Ϯ 12.7, 7.7 Ϯ 6.4, and 12.6 Ϯ 5.4%, respectively (Fig. 6). Thus, the results suggest that NF-B could mediate thrombin effect in the induction of microglial NO release.
Thrombin-induced NO Release Is Independent of PAR-1-We also examined whether PAR-1 mediated the effect of thrombin to induce microglial NO release, because PAR-1 has been suggested to act as a thrombin receptor in many types of cells. For this, we looked at the effects of the agonist of PAR-1, TRAP. TRAP did not mimic the effect of thrombin even at a 500 M concentration: 50, 100, 200, and 500 M TRAP induced 1.5 Ϯ 0.5, 2.3 Ϯ 0.9, 1.5 Ϯ 0.7, and 1.1 Ϯ 0.6 M of nitrite, respectively, whereas 10 units/ml of thrombin induced 14.9 Ϯ 1.1 M nitrite (Fig. 7A). We then used RT-PCR to detect iNOS expression, because the RT-PCR method could be more sensitive than immunoblot analysis or measuring nitrite. Even with RT-PCR, the result was the same: iNOS was not detected in 500 M TRAP-treated cells, whereas it was detected within 3 h in thrombin-treated cells (Fig. 7B).
We also tested the effect of cathepsin G, an inhibitor of PAR-1. As we expected based on the results of the experiments Nuclear extracts were prepared from microglia treated with 10 units/ml thrombin for the indicated time periods. A, the NF-B-specific oligonucleotide-protein complexes are marked by arrows. B, in the supershift assay, antibodies against p50 and p65 were mixed with nuclear extracts obtained from microglia treated with thrombin for 30 min. A supershifted band was detected in the presence of p50 antibody (Ab) and is marked by an arrowhead. The electrophoresis mobility shift assay was carried out as described under "Experimental Procedures." using TRAP, cathepsin G (40 milliunits/ml) did not reduce nitrite production from thrombin-treated microglia. Microglia pretreated with cathepsin G for 30 min produced 12.6 Ϯ 0.5 M nitrite in response to 10 units/ml of thrombin compared with 12.6 Ϯ 0.9 M of nitrite produced by microglia treated with thrombin alone. Cathepsin G (40 milliunits/ml) alone had no discernible effect (Fig. 8). DISCUSSION The results in this study indicate that thrombin could be a mediator of brain inflammation, because thrombin induces NO release and iNOS expression in microglia, which are major immune effector cells in the brain. The intracellular signaling mechanisms that mediate the function of thrombin in microglia can be compared with those of other microglial activators. LPS and ␤-amyloid peptide have been known to be microglial activators. Recently we reported that gangliosides, components of the plasma membrane, could induce microglial activation (35). The microglial activation of thrombin could be achieved via similar intracellular signaling pathways. PKC may thus be the common mediator of microglial activation. PKC inhibitors reduce NO release by microglia treated with LPS (33, 34) and gangliosides (data not shown) as they inhibited thrombin-induced NO release. (Fig. 2). Recently, it has been reported that in microglia PKC isoforms, such as ␣, ␦, and ⑀, are activated by ␤-amyloid (36). MAPKs are other common mediators of microglial activation. ERK, p38, and JNK/SAPK are activated by thrombin, LPS, ␤-amyloid, and gangliosides (28,35). However, some discrepancies still exist with regard to the effect of inhibitors of the ERK pathway (PD98059) and p38 (SB203580) on the action of these activators. LPS-and thrombin-induced nitrite production was significantly reduced by PD98059 and SB203580, whereas ganglioside-induced nitrite production was inhibited by PD98059 but not by SB203580 (28,35). NF-B is also a common mediator of microglial activation. Microglial activators, such as ␤-amyloid, interferon-␥, and gangliosides activate NF-B (35,37), and inhibition of NF-B reduced NO release (35).
In addition to PKC, MAPK, and NF-B, Src family tyrosine kinases, PI3-K, and ROS have also been reported to mediate the function of thrombin in many kinds of cells. Thrombin has been known to activate a Src family tyrosine kinase in astrocytes (22) and endothelial cells (19). In microglia, however, selective inhibitors of Src family tyrosine kinases, PP2 and lavendustin A, did not inhibit the action of thrombin (data not shown). PI3-K has been known to mediate thrombin-induced proliferation of smooth muscle cells (38), and ROS have been known to induce the expression of chemokines and cell adhesion molecules in endothelial cells, epithelial cells, and lymphoma cells (39,40). However, PI3-K may not be involved in the thrombin effect on microglia, because PI3-K inhibitors, such as wortmannin and LY294002, had little effect on thrombin-induced NO release (data not shown). Although NAC that increases reduced form of glutathione level dose-dependently reduced thrombin-induced NO release (Fig. 6), ROS scavengers, such as catalase and superoxide dismutase did not inhibit thrombin-induced nitrite production (data not shown). Thus, ROS that could be removed by these scavengers do not seem to mediate thrombin-induced NO release.
PAR-1 has been suggested as a thrombin receptor. In astrocytes and C6 glioma cells, thrombin-induced Ca 2ϩ mobilization is mimicked by TRAP and blocked by cathepsin G (41,42). Activation of platelets could be achieved by both thrombin and TRAP (43). However, in microglia TRAP did not induce NO release and iNOS mRNA expression (Fig. 7). Furthermore, cathepsin G had no effect on the thrombin-induced NO release (Fig. 8). Although the effect of thrombin was reduced by hirudin, which inhibits protease function of thrombin (44 -46) (data not shown). We also tested the effect of prothrombin in the absence of factor VII and X that change prothrombin into thrombin. Prothrombin dose-dependently induced microglial NO release (data not shown). These results suggest that, in microglia, thrombin induces NO release via a PAR-1-independent pathway. Recent studies in other laboratories could support this conclusion. In myoblasts, thrombin protected myoblasts from apoptosis, whereas TRAP did not (47). In fibroblasts, TRAP and TP508 (a peptide fragment of thrombin different from TRAP) induced differential patterns of gene expression, suggesting additional nonproteolytic action of thrombin (27). However, until recently, information regarding thrombin receptors other than PAR-1 has been limited, although PAR-3 and 4 also have been suggested. Therefore, further studies are needed to characterize thrombin receptors in microglia.