Influence of lipoxin A4 and other lipoxygenase-derived eicosanoids on ...

Am J Physiol Cell Physiol 279: C945–C953, 2000.

Influence of lipoxin A4 and other lipoxygenase-derived eicosanoids on tissue factor expression PAOLA MADERNA, CATHERINE GODSON, GARY HANNIFY, MADELINE MURPHY, AND HUGH R. BRADY Centre for Molecular Inflammation and Vascular Research, Department of Medicine and Therapeutics, Mater Misericordiae Hospital, The Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Dublin 7, Ireland Received 18 June 1999; accepted in final form 6 April 2000

Maderna, Paola, Catherine Godson, Gary Hannify, Madeline Murphy, and Hugh R. Brady. Influence of lipoxin A4 and other lipoxygenase-derived eicosanoids on tissue factor expression. Am J Physiol Cell Physiol 279: C945–C953, 2000.—Lipoxins (LX) are eicosanoids generated via transcellular biosynthetic routes during inflammation, hypersensitivity reaction, and after angioplasty. LXs are modulators of leukocyte trafficking and vascular tone. Their influence on the coagulation cascade has not been determined. In this study, we evaluated the influence of LXs on the expression of tissue factor (TF), a key regulator of coagulation. TF activity was measured in lysates of monocytes, human umbilical vein endothelial cells, and ECV304 cells using a one-stage clotting assay. LXA4 stimulated TF activity in each cell type. The influence of LXA4 on TF activity by ECV304 cells was studied further to explore the mechanism of induction of TF expression. LXA4-induced TF activity was dose dependent, cycloheximide sensitive, and associated with increased TF mRNA levels. Induction of TF activity was specific for LXA4 and was not observed with LXB4, the other major lipoxin generated by mammalian cells. Furthermore, ECV304 cell TF expression was not influenced by 15(R/S)methyl-LXA4 or 16-phenoxy-LXA4, synthetic analogs of LXA4 that activate the myeloid LXA4 receptor, and was not modulated by SKF-104353, which blocks LXA4 bioactivities transduced through the putative shared LXA4/LTD4 receptor. LXA4-stimulated TF expression was blunted by pertussis toxin and by GF-109203X, an inhibitor of protein kinase C, and was not associated with degradation of I␬B␣. Our results establish that LXA4 induces TF activity via cell signaling pathways with different structural and receptor requirements from those described for inhibition of leukocyte-endothelial cell interactions. They suggest a role for LXA4 as a modulator of TF-related vascular events during inflammation and thrombosis. monocytes; endothelial cells; receptors; gene expression

(LX) are eicosanoids generated via biosynthetic pathways that involve the dual lipoxygenation of arachidonic acid by either 5- and 15-lipoxygenases or 5and 12-lipoxygenases (lipoxygenase interaction prod-

LIPOXINS

Address for reprint requests and other correspondence: H. R. Brady, Centre for Molecular Inflammation and Vascular Research, Dept. of Medicine and Therapeutics, Mater Misericordiae Hospital, Univ. College Dublin, 41 Eccles St., Dublin 7, Ireland (E-mail: [email protected]). http://www.ajpcell.org

ucts) (4). They are generated in vivo during host defense, inflammation, and hypersensitivity reactions and after percutaneous transluminal coronary angioplasty (PTCA) (4, 6). LXA4 and LXB4, the major LXs generated in mammalian systems, are inhibitors of neutrophil chemotaxis, adhesion, and transmigration induced by leukotrienes (LTs) and some other mediators and are putative endogenous “breaking signals” for leukocyte recruitment that promote resolution of inflammation (4). In addition to their actions with hematogenous cells, LXA4 and LXB4 have been reported to stimulate endothelial-dependent vasodilatation, inhibit endothelial cell P-selectin expression and hyperadhesiveness induced by LTs (28), and inhibit cytokine triggered interleukin (IL)-8 release by colonic epithelial cells (19). Thus LX formed during vascular events have the potential to influence the function of hematogenous cells and resident cells of the vessel wall. The influence of LX on the coagulation cascade has not been determined. As an initial assessment of this issue, we evaluated the influence of LX on tissue factor (TF) expression. TF is a transmembrane glycoprotein that is a key regulator of blood coagulation (17). When expressed on the surface of cells such as endothelium, TF binds with high affinity to factor VII, leading to rapid formation of factor VIIa. The resulting TF-factor VIIa complex activates factors IX and X, leading to thrombin generation and subsequent fibrin deposition (17). Hematogenous cells and endothelial cells do not express TF constitutively; however, TF is transiently induced in endothelial cells after physical or inflammatory perturbation of the vessel wall. During inflammation and atherosclerosis, TF is induced in monocyte and endothelial cells by a variety of agonists, including bacterial lipopolysaccharide (LPS), inflammatory cytokines, and oxidized low-density lipoproteins (2, 12, 13, 16). TF is also rapidly induced in the vessel wall as a consequence of acute arterial injury during PTCA (35). TF is constitutively expressed by many vascular wall cells external to the endothelial monolayer, and its The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

0363-6143/00 $5.00 Copyright © 2000 the American Physiological Society

C945

C946

LXA4 IS A MODULATOR OF TF ACTIVITY

expression is further induced in these cells by inflammatory stimuli. Here we explored the influence of LX on TF expression by monocytes, human umbilical vein endothelial cells (HUVEC), and ECV304 cells, cell-types chosen as models of hematogenous and vascular wall systems that 1) do not express TF under basal conditions but can be induced to express TF by inflammatory mechanisms (monocytes, HUVEC) and 2) nonendothelial parenchymal cells that constitutively express TF (ECV304 cells). We present evidence that LXA4 stimulates TF activity via a receptor-mediated signaling pathway that induces TF gene expression. METHODS

Materials. Synthetic 5S,6R,15S-trihydroxy-7,9,13-trans11-cis-eicosatetraenoic acid (LXA4), 5S,14R,15S-trihydroxy6,10,12-trans-8-cis-eicosatetraenoic acid (LXB4), and 5Z,8Z, 11Z,13E,15(S)-15-hydroxyeicosatetraenoic acid (15-(S)-HETE) were purchased from Cascade Biochem (Berkshire, UK). Synthetic analogs 15(R/S)-methyl-LXA4 (5S,6R,15R/S-trihydroxy15-methyl-7,9,13-trans-11-cis-eicosatetraenoic acid methyl ester) and 16-phenoxy-LXA4 (15S-16-phenoxy-17,18,19,20tetranor-LXA4 methyl ester) were prepared by total synthesis by Dr. N. A. Petasis and Dr. V. V. Fokin as previously described (34). Human recombinant tumor necrosis factor (TNF)-␣ was obtained from R & D systems (Minneapolis, MN); arachidonic acid, bacterial endotoxin (LPS, Escherichia coli serotype 0111:B4), pertussis toxin, NG-monomethyl-L-arginine (L-NMMA), and cycloheximide from Sigma (St. Louis, MO); and GF-109203X and genistein from Calbiochem (San Diego, CA). SFK-104353 was a kind gift from SmithKline Beecham Pharmaceuticals. Rabbit anti-human TF antibody and recombinant human TF were purchased from American Diagnostica (Greenwich, CT). The anti-I␬B␣ rabbit anti-human polyclonal antibody and the purified protein I␬B␣ were from Santa Cruz Biotechnology (Santa Cruz, CA). A 641-bp fragment of the human TF cDNA was generated by EcoR I digestion of the pcTF314 containing TF cDNA plasmid (26). Cell culture and treatments. Human monocytes were obtained from whole blood collected from healthy donors and anticoagulated with 3.8% sodium citrate. Mononuclear cells were separated with Ficoll-Paque solution (Pharmacia, Biotec, Uppsala, Sweden) at 450 g for 20 min. To minimize platelet contamination, the cells were washed with PBS (Bio-Whittaker) containing 5 mM EDTA and suspended in RPMI 1640 supplemented with 10% autologous serum, 2 mM glutamine, 100 IU/ml penicillin, and 100 ␮g/ml streptomycin (GIBCO, Grand Island, NY). Cells were placed in 12-well plastic tissue culture plates (Costar, Cambridge, MA) at 2 ⫻ 106 cells per well and kept at 37°C in a 5% CO2 humidified atmosphere for 2 h. Supernatants containing nonadherent cells were removed, and adherent monocytes were washed twice with prewarmed medium. Primary cultures of HUVECs were obtained from collagenase-digested umbilical veins or from Clonetics (San Diego, CA). The cells used for this study were between the second and the fourth passage and propagated in medium 199 (Sigma) supplemented with 20% fetal calf serum (GIBCO), 2 mM L-glutamine, 100 IU/ml penicillin, and 100 ␮g/ml streptomycin in the presence of 50 ␮g/ml heparin and 50 ␮g/ml of endothelial cell growth factor (Sigma). ECV304 cells, a spontaneously immortalized cell line (36), were obtained from the European Collection of Animal Cell

Cultures (Salisbury, UK). ECV304 cells express TF under basal condition and display increased TF expression in response to cytokine stimulation (23), providing a convenient model for assessment of both constitutive and inducible TF expression. ECV304 cells were maintained in medium 199 supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 IU/ml penicillin, and 100 ␮g/ml streptomycin. For each experiment, cells were cultured in 12-well plastic tissue culture plates for assay of TF activity and in 100-mm dishes (Costar, Cambridge, MA) for RNA isolation. Cells were incubated in serum free medium with the appropriate stimuli as described. When inhibitors (pertussis toxin, LNMMA, GF-109203X, genistein, cycloheximide) were used, they were incubated with ECV304 cells before the addition of LXA4, as described. At the end of the indicated incubation times, the medium was discarded and the cells were washed twice with serum-free medium 199 and the plates stored at ⫺80°C until assayed. Cell viability was assessed by ability of cells to exclude trypan blue and by assay of lactate dehydrogenase (LDH) release (Sigma). Endotoxin contamination of all reagents used for cell isolation and culture was routinely excluded with chromogenic Limulus amebocyte lysate assay (Sigma). TF activity determination. For determination of TF activity, the cells were lysed with 16 mM octyl-␤-D-glycopyranoside at 37°C for 15 min. TF activity in cell lysates was determined by a one-stage clotting assay (39). The assay mixture contained 0.1 ml cell lysates; 0.1 ml citrated, pooled, normal plasma; and 0.1 ml of 25 mM CaCl2. Clotting times were determined at 37°C. The values were converted to arbitrary units of procoagulant activity by comparison with a standard curve of clotting times obtained by serial dilutions of rabbit brain thromboplastin (Sigma). This preparation was assigned a value of 2,000 U for a clotting time of 40 s. In some experiments, results were also expressed in nanograms of TF by comparison with a standard curve obtained by using a recombinant human TF lipidated lipoprotein standard. Protein concentration was determined on cell lysates using the method of Bradford (3). Analysis of I␬B␣ levels by Western blot. ECV304 cells in six-well plates were incubated with LXA4 or TNF-␣ for various times. Treatment was terminated by washing twice with ice-cold PBS. Cells were then scraped in 200 ␮l of ice-cold lysis buffer (40 mM HEPES, 150 mM NaCl, 5 mM MgCl2, 20 ␮M EDTA, 0.1% Triton-X 100) including a cocktail of protease inhibitors (2 ␮M phenylmethylsulfonyl fluoride, 1 mg/ml leupeptin, 20 mM ␤-glycerophosphate, 1 ␮M sodium orthophosphate, and 0.5 mM dithiothreitol). Samples were then centrifuged at 14,000 g for 20 min at 4°C, and the supernatant was removed as cell lysates. Equal amounts of protein of denatured cell extracts and a purified protein I␬B␣ standard were resolved by electrophoresis on a 10% SDS polyacrylamide gel and transferred to Immobilon-P polyvinyldiene difluoride membranes (Millipore). The membrane was probed with the anti-I␬B␣ rabbit anti-human polyclonal antibody at a dilution of 1:2,500. To detect antibody reaction, the blots were incubated with an alkaline phosphate conjugated anti-rabbit secondary antibody. Immunoreactive proteins were detected using an alkaline phosphatase liquid substrate system 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium (Sigma). RNA isolation and Northern blot analysis. Total cellular RNA was isolated using TRIZOL reagent (Life Technologies, Paisley, Scotland), a modification of the guanidinium isothiocyanate method (8). Total RNA (10 ␮g) was electrophoresed in denaturing 1% agarose-formaldehyde gels (31) and transferred overnight to nylon membranes (Hybond, Amersham,


Arlington Heights, IL) by capillary transfer in 20⫻ sodium chloride-sodium citrate. RNA was fixed to the membrane by baking at 80°C for 2 h. The membranes were prehybridized for 1 h at 68°C using Quikhyb hybridization solution (Stratagene, La Jolla, CA). TF cDNA probe was labeled with [␣⫺32P]dCTP (3,000 Ci/mmol, DuPont-NEN) by random priming (Boehringer Mannheim). This was added to the Quikhyb solution, and the hybridization was carried out at 68°C for 2 h. Membranes were washed twice in 2⫻ SSC, 0.1% SDS for 30 min at room temperature, followed by one wash in 0.1⫻ SSC, 0.1% SDS at 65°C. The membranes were then exposed to X-Omat Kodak film (Eastman Kodak, Rochester, NY) with intensifying screens at ⫺70°C. Equal loading of RNA was assessed relative to glucose-6-phosphate dehydrogenase mRNA levels. Statistical analysis. Values are expressed as means ⫾ SE. Statistical significance was assessed by paired Student’s ttest. RESULTS

Influence of LXA4 on TF activity. The influence of LXA4 (10⫺9 M, 4 h) on TF activity in monocytes, HUVECs, and ECV304 cells was assessed and compared with TNF-␣ (10 ng/ml, 4 h) and LPS (10 ␮g/ml, 4 h), two well-defined inducers of TF activity. HUVECs and monocytes do not express appreciable amounts of TF under basal conditions. Exposure of adherent monocytes and HUVECs to LXA4 (10⫺9 M, 4 h) increased TF activity by almost two- and threefold over control, respectively. LXA4-induced TF activity in HUVECs was similar in magnitude to that observed with TNF-␣ treatment or with monocytes exposed to LPS (Fig. 1). ECV304 cells express TF under basal conditions and LXA4, similarly to TNF-␣, enhanced TF activity by approximately twofold by comparison with unstimulated cells (Fig. 1). Importantly, diluent controls for LXA4 containing 0.1% ethanol did not influence TF activity (ECV304 cells alone: 8.6 ⫾ 2.3 U TF/␮g protein; ECV304 cells ⫹ ethanol: 8.9 ⫾ 2.1 U TF/␮g protein). To confirm that the effect of LXA4 on clotting time was specific for TF activity, ECV304 cells were treated with LXA4 for 3 h and then incubated for a further 1 h with a rabbit anti-human TF antibody (50 ␮g/ml) at 37°C. In the presence of this antibody, LXA4triggered procoagulant activity was completely inhibited (LXA4: 19.2 ⫾ 2.6 U TF/␮g protein; LXA4 ⫹ antihuman TF antibody: 3.7 ⫾ 0.4 U TF/␮g protein; n ⫽ 3).

C947

LXA4-induced TF expression by ECV304 cells was studied further to assess structure-function relationships and mechanisms of TF induction. The effects of LXA4 and TNF-␣ on ECV304 cells were additive [TF activity expressed as multiples of increase over control: LXA4: 1.8 ⫾ 0.2; TNF-␣: 2.1 ⫾ 0.3; TNF-␣ ⫹ LXA4: 3.6 ⫾ 0.7 (P ⬍ 0.05 vs. LXA4 and TNF-␣); n ⫽ 4]. The dose dependence and kinetics of LXA4-triggered ECV304 cell TF expression is shown in Fig. 2. Maximal induction of TF activity was observed with 10⫺9 M LXA4 (Fig. 2A). Augmentation of TF activity by LXA4 was maximal after 4 h, with TF activity returning toward basal levels after 16 h incubation (Fig. 2B). Under certain experimental conditions, prolonged exposure of human endothelial cells to LXA4 has been reported to cause a slight increase in cell death via a nitric oxide-dependent mechanism (5). In the present study, incubation of ECV304 cells with LXA4 10⫺9 M for up to 4 h was not associated with loss of cell integrity as determined by trypan blue staining and by assay of LDH release. Furthermore, the competitive inhibitor of nitric oxide synthesis, L-NMMA (1 mM, 1 h), did not significantly modify LXA4-triggered TF activity (LXA4: 18.5 ⫾ 3.8 U TF/␮g protein; L-NMA ⫹ LXA4: 14.1 ⫾ 2.4 U TF/␮g protein, n ⫽ 10). LXA4-induced TF activity: structure-function relationship. Arachidonic acid did not induce ECV304 cell TF expression (Fig. 3), an important observation given that LXA4 causes arachidonic acid release by leukocytes and endothelial cells and that arachidonic acid can induce TF expression in monocytes (7). We next compared LXA4 with LXB4, the other major lipoxin generated in mammalian systems, and with 15(S)-HETE for their ability to activate TF. In contrast to LXA4, LXB4 and 15(S)-HETE did not influence ECV304 cell TF activity. 15(R/S)-methylLXA4 and 16-phenoxy-LXA4 are analogs of LXA4 that are relatively resistant to degradation in vitro. These compounds share the ability of LXA4 to inhibit neutrophil recruitment during inflammation in vitro and in vivo (34, 37). In the present study, neither 15(R/S)-methyl-LXA4 nor 16-phenoxy-LXA4 increased TF activity within the concentration range of 10⫺9 – 10⫺12 M (Table 1).

Fig. 1. LXA4 stimulates tissue factor (TF) activity in monocytes, human umbilical vein endothelial cells (HUVECs), and ECV304 cells. Cells were exposed to LXA4 (10⫺9 M), lipopolysaccharide (LPS) (10 ␮g/ml), and tumor necrosis factor (TNF)-␣ (10 ng/ml) for 4 h at 37°C. TF activity was measured in cell lysates by a 1-stage clotting assay. The amount of TF protein was calculated by extrapolation of a standard curve of serial dilutions of recombinant human TF. Data are expressed as ng TF/106 cells and are the means ⫾ SE of 3 experiments for HUVECs, 6 experiments for monocytes, and 20 experiments for ECV304 cells. * P ⬍ 0.05; ** P ⬍ 0.01 vs. vehicle.

C948


Fig. 3. Comparison of effects of native lipoxins and other eicosanoids on TF activity. ECV304 cells were exposed to LXA4, LXB4, 5Z,8Z,11Z,13E,15(S)-15-hydroxyeicosatetraenoic acid [15(S)-HETE] (10⫺9 M), or arachidonic acid (AA; 15 ␮M) for 4 h. Data are expressed as fold induction of TF activity over control and represent the means ⫾ SE of 6 independent experiments. * P ⬍ 0.01 vs. vehicle.

Fig. 2. Dose-response curve and time course for induction of TF activity by LXA4. A: ECV304 cells were exposed to different concentrations of LXA4 for 4 h. TF activity is expressed as TF U/␮g protein and represents the means ⫾ SE of 4 independent experiments. * P ⬍ 0.05 vs. vehicle. B: ECV304 cells were exposed to LXA4 (10⫺9 M) for 2, 4, 6, and 16 h. TF activity is expressed as TF U/␮g protein and represents the means ⫾ SE of 4 independent experiments. * P ⬍ 0.05 vs. baseline (0 h).

Receptor mechanisms involved in LXA4-triggered TF activity. The results of our structure-function studies described above, namely stimulation of TF activity by nanomolar concentrations of LXA4, but not LXB4, and 15(S)-HETE suggest that ECV304 cells express specific high-affinity LXA4 receptors. To explore this possibility further, LXA4-stimulated TF activity was assessed after prior exposure of ECV304 cells to pertussis toxin (100 ng/ml, 37°C, 18 h). Pertussis toxin completely inhibited LXA4-induced TF activity (Fig. 4), suggesting involvement of an LXA4 receptor coupled to Gi proteins. Prior exposure of ECV304 cells to SKF-104353 (10⫺7 M, 15 min), an LTD4 receptor antagonist, did not influence the ability of LXA4 to induce TF activity (Fig. 4), excluding a role for the putative shared LXA4/LTD4 endothelial receptor.

The stable LXA4 analogs 15(R/S)-LXA4 and 16-phenoxy-LXA4 compete with LXA4 for binding to the neutrophil LXA4 receptor and share many biological activities of LXA4 with this cell type. In contrast, neither 15(R/S)-methyl-LXA4 nor 16-phenoxy-LXA4 shared the ability of LXA4 to trigger TF expression in ECV304 cells (see above). In a further set of experiments, prior exposure of ECV304 cells to 15(R/S)-methyl-LXA4 (10⫺9 M, 15 min) did not block LXA4-triggered (10⫺9 M, 4 h) TF expression. (Fig. 4). Cell signaling pathways involved in LXA4-triggered TF activity. To explore the role of the protein kinase C (PKC) in the regulation of TF activity by LXA4, we evaluated LXA4 bioactivity in the presence of GF-109203X, a pharmacological inhibitor of PKC. Whereas GF-109203X (10 ␮M, 15 min) did not affect basal TF activity, prior exposure of ECV304 cells to this PKC inhibitor markedly inhibited the LXA4-triggered response (Fig. 5). LXA4-stimulated TF activity was also assessed in the presence of genistein, an inhibitor of tyrosine kinase activity. In contrast to inhibiting PKC activity, genistein (100 ␮M, 15 min) Table 1. Dose-response curve for induction of TF activity by LXA4 stable analogs Vehicle 10⫺12 M 10⫺11 M 10⫺10 M 10⫺9 M

15 (R/S)-Methyl-LXA4

16-Phenoxy-LXA4

9.8 ⫾ 1.5 8.8 ⫾ 1.0 11.4 ⫾ 1.1 12.8 ⫾ 1.2 9.9 ⫾ 1.2

12.2 ⫾ 1.8 10.0 ⫾ 0.8 10.2 ⫾ 2.2 9.4 ⫾ 1.7 11.1 ⫾ 2.0

Data represent the means ⫾ SE of tissue factor (TF) activity in U/␮g protein of 5 independent experiments. ECV304 cells were exposed to different concentrations of 15(R/S)-methyl-LXA4 and 16phenoxy-LXA4 for 4 h.


C949

well-defined inhibitor of protein synthesis, LXA4-induced TF activity (10⫺9 M, 4 h) in ECV304 cells was inhibited by 82 ⫾ 3% (n ⫽ 3). LXA4 enhanced ECV304 cells expression of TF mRNA as assessed by Northern blotting (Fig. 7). This LXA4-triggered increase in TF mRNA was also assessed after prior exposure of cells to pertussis toxin (100 ng/ml, 18 h) As also shown in Fig. 7, parallel to the TF activity data, LXA4 failed to induce TF mRNA expression in the presence of pertussis toxin. DISCUSSION

Fig. 4. Mechanisms of induction of TF activity by LXA4. ECV304 cells were exposed to LXA4 (10⫺9 M) for 4 h after pretreatment with either pertussis toxin (PTX, 100 ng/ml, 18 h), 15(R/S)-methyl-LXA4 (10⫺9 M, 15 min), SKF-104353 (10⫺7 M, 15 min), or their diluent. Data, expressed as TF U/␮g protein, represent LXA4-induced TF activity and were obtained by subtraction of TF basal levels. Results are the means ⫾ SE of 5 independent experiments. * P ⬍ 0.001 vs. LXA4 alone.

inhibited neither basal TF activity nor LXA4-stimulated TF activity (Fig. 5). Effect of LXA4 on I␬B␣ levels. In the present study, exposure of ECV304 cells to LXA4 (10⫺7 –10⫺9 M, 15 min–2 h) was not associated with degradation of I␬B␣, as assessed by Western blotting, in contrast to TNF-␣ (10 ng/ml), a well-defined stimulus for activation of the NF-␬B pathway (Fig. 6). Interestingly, LXA4-induced TF activity was inhibited by N␣-tosylphenylalanyl chloromethyl ketone (TPCK) and N␣-tosyl-L-lysine chloromethyl ketone (TLCK), inhibitors of chymotrypsin-like serine proteases that prevent nuclear factor (NF)-␬B activation (24), suggesting other intracellular actions of these pharmacological inhibitors (Table 2). Effect of LXA4 on TF mRNA levels. Induction of endothelial TF expression by LPS and cytokines is mediated through activation of gene transcription (32). In the presence of cycloheximide (10 ␮g/ml, 1 h), a

LXs are eicosanoids generated via biosynthetic pathways that initially involve the sequential actions of two lipoxygenases, either 5- and 15-lipoxygenases or 5- and 12-lipoxygenases, with arachidonic acid (4). More recently, an additional route of LX formation was demonstrated during neutrophil-endothelial cell interactions involving 5-lipoxygenase and aspirin-acetylated cyclooxygenase-2 (10). These lipid mediators are putative inhibitors of neutrophil trafficking in host defense and inflammation (4). LXs inhibit neutrophil chemotaxis, adhesion to endothelial cells, and transmigration across endothelial cell and epithelial cell monolayers induced by LTs and some other chemoattractants in vitro (11, 22, 28). The potential importance of LXA4 as an eicosanoid with anti-inflammatory actions is further supported by in vivo studies in animal models in which LXA4 and stable analogs of LXA4 inhibit neutrophil recruitment to inflammatory foci (20). In addition to their actions with leukocytes, LXA4 and LXB4 are modulators of bronchial and vascular smooth muscle tone and induce vasodilatation in several vascular beds through an endothelial-dependent mechanism (15). The formation of LXs within the vascular lumen and vessel wall during inflammation and after angioplasty places these lipid mediators in a strategically advantageous site for modulation of the function of a range of hematogenous and parenchymal cell types. LXA4 and LXB4, the major bioactive LXs, are well-defined regu-

Fig. 5. Exploration of the signaling pathways involved in the induction of TF activity by LXA4. ECV304 cells were exposed to vehicle or LXA4 (10⫺9M) for 4 h without or with pretreatment with GF-109203X (10 ␮M, 15 min) or genistein (100 ␮M, 15 min). Data are expressed as TF U/␮g protein. Results are the means ⫾ SE of 5 independent experiments. * P ⬍ 0.001 vs. LXA4.

C950


Fig. 6. LXA4 does not affect the degradation of I␬B␣. Cytoplasmic extracts were prepared from ECV304 cells treated with TNF-␣ (10 ng/ml) and LXA4 (10⫺9 M) at the indicated times. I␬B␣ protein levels were detected by Western blotting. A purified protein I␬B␣ standard (20 ng/ml) was used as internal control. Results are representative of 3 experiments.

lators of endothelial and mesangial cell adhesiveness for leukocytes, endothelial-dependent vasorelaxation, and chemokine release by epithelial cell lines (reviewed in Refs. 4, 33). In the present study, we demonstrate that LXA4, at nanomolar concentrations, induces TF activity by monocytes and HUVEC, cell types that do not express appreciable TF under basal conditions, and augments TF by ECV304 cells. Indeed, LXA4 induced TF expression by ECV304 cells to levels equivalent to those induced by TNF-␣ and LPS. TF is an inducible molecule expressed after vessel injury or exposure to cytokines (2, 35). TF initiates blood coagulation through its interaction with factor VII (17). TF appears to be an important trigger for thrombosis after acute arterial injury, and TF-mediated thrombosis may, in turn, be a key stimulus for subsequent intimal hyperplasia and luminal narrowing (21, 27). Interestingly, TF has recently been reported to promote recruitment of monocytes in inflammatory renal disease via a fibrinogen-independent mechanism, thereby expanding the pathogenic roles of this mediator (14). The results of our assessment of the structural requirements for LXA4-triggered TF activity are noteworthy for several reasons. First, LXA4 stimulates release of arachidonic acid from several cell types, including leukocytes and endothelial cells, and arachidonic acid enhances TF expression in mononuclear cells (7). Arachidonic acid did not induce TF expression in our study, suggesting that LXA4 regulates ECV304 cell TF expression via a different mechanism. Second, unlike LXA4, LXB4 did not trigger TF expression. LXB4

is the other major LX generated by mammalian cells and shares the ability of LXA4 to influence most other neutrophil and endothelial responses (4, 29). With specific regard to endothelial cells, both LXA4 and LXB4 stimulate prostacyclin generation and inhibit mobilization of P-selectin and endothelial hyperadhesiveness induced by peptidoleukotrienes (28). Third, neither 15(R/S)-methyl-LXA4 nor 16-phenoxy-LXA4 shared the ability of LXA4 to trigger TF expression. The major route for LX metabolism in human leukocytes involves dehydrogenation of alcohols and reduction of double bonds through pathways similar to the dehydrogenation-type systems involved in prostanoid metabolism. LXA4 is metabolized by phorbol myristate acetate-differentiated HL-60 cells and human monocytes to 15oxo LXA4, 13,14-dihydro-15-oxo-LXA4, and 13,14-dihydro-LXA4 in vitro. 15(R/S)-methyl-LXA4 and 16phenoxy-LXA4 are analogs of LXA4 that are relatively resistant to degradation in the latter model systems. These compounds share the ability of native LXA4 to inhibit neutrophil-endothelial cell adhesion in vitro and neutrophil trafficking to sites of inflammation induced by LTs in vivo (34, 37). In aggregate, these data establish that the structural requirements for induction of TF expression by LXA4 differ substantially from

Table 2. Effect of protease inhibitors on LXA4 induction of TF activity TF Activity

Vehicle LXA4 LXA4 ⫹ TLCK LXA4 ⫹ TPCK

8.4 ⫾ 1.5 20.0 ⫾ 3.5 9.5 ⫾ 2.0* 10.3 ⫾ 2.2*

Data represent the means ⫾ SE in U TF/␮g protein of 5 independent experiments. ECV304 cells were exposed for 4 h to LXA4 (10⫺9 M) in the presence or in the absence of N␣-tosylphenylalanyl chloromethyl ketone (TLCK) or N␣-tosyl-L-lysine chloromethyl ketone (TPCK) (100 ␮M, 15 min). * P ⬍ 0.05 vs. LXA.

Fig. 7. Induction of TF mRNA expression by LXA4. ECV304 cells were exposed for 2 h to LXA4 (10⫺9 M) in the presence or absence of PTX (100 ng/ml, 18 h). Total RNA was isolated, and 10 ␮g of total RNA were analyzed for TF expression by Northern blotting using a radiolabeled human TF cDNA probe. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels were assessed to control for RNA loading. Results are representative of 5 experiments.


those described for neutrophil-directed and other parenchymal cell-directed action of LXA4 and suggest involvement of different cell signaling pathways (see below). In addition, the results of these structure-function studies have important implications for drug design, as they suggest that it is possible to discriminate between the well-established anti-inflammatory activities of LXA4 and its potential proinflammatory and prothrombotic effect on TF expression by chemical modification of the native compounds. Neutrophils, monocytes and epithelial cells express high-affinity (Kd ⬃1 nM) receptors for LXA4 that have been cloned and characterized and appear to be a member of the superfamily of G protein-coupled chemokine receptors (18, 19, 25). 15(R/S)-methyl-LXA4 and 16-phenoxy-LXA4 compete with LXA4 for binding to these neutrophil receptors and share the ability of LXA4 to inhibit neutrophil chemotaxis and adhesion. In contrast, LXB4 does not compete with LXA4 for binding to this receptor. Human endothelial cells also express high-affinity binding sites for LXA4 (33). There is increasing evidence that nonhematogenous cells, such as endothelial cells and glomerular mesangial cells, express at least two classes of membrane surface receptors for LXA4, one subtype that is specific for LXA4 and another shared by cysteinyl-LTs such as LTD4 (1, 9). In the present study, prior exposure of ECV304 cells to pertussis toxin blocked LXA4-induced TF expression, suggesting that these cells express Gi protein-coupled receptors for LXA4. The failure of 15(R/S)-methyl-LXA4 and 16-phenoxy-LXA4 to mimic the action of LXA4 with regard to TF expression suggests that ECV304 cells express a receptor for LXA4 that is distinct from the myeloid receptor. In further support of this contention, prior exposure of ECV304 cells to 15(R/S)-methyl-LXA4 did not influence the ability of the cells to generate TF on subsequent stimulation by LXA4. Some binding of [3H]LXA4 to endothelial cells is inhibited by the LTD4 receptor antagonist SKF104353, and both LXA4 and LXB4 antagonize the ability of LTD4 to mobilize endothelial cell P-selectin from Weibel-Palade bodies to the cell surface (28). In our study, SKF-104353 did not antagonize the activity of LXA4 on TF expression, excluding the involvement of this receptor. When viewed in the context of the failure of LXB4 to induce TF expression, the aggregate of these results suggests that LXA4 triggers ECV304 cell TF expression through engagement of novel LXA4 receptors that are distinct from the previously characterized myeloid LXA4 receptor and from the pharmacologically defined SKF-104353-sensitive shared LXA4/LTD4 receptor. The ability of LXA4 to induce ECV304 cell TF activity was blunted by the PKC inhibitor GF-109203X, but not by genistein, the tyrosine kinase inhibitor. These observations are interesting given that protein kinases C have been implicated in cytokine-triggered induction of TF expression in endothelial cells (38). Furthermore, the failure of genistein to influence LXA4 bioactivity excludes the involvement of tyrosine kinases and again distinguishes this endothelial cell-directed activity of

C951

LXA4 from its inhibitory actions with neutrophils that are blunted by genistein (29). The TF gene is an immediate early gene that is rapidly induced in various cell types in response to different stimuli. In endothelial cells, TF gene transcription is activated by LPS, TNF-␣, and IL-1␤ (2, 12, 13). The induction of endothelial cell TF gene expression is controlled by several regulatory elements on the human TF promoter. The LPS response element is the major cytokine and endotoxin response element. It includes two activator protein-1 sites and an NF-␬B site that selectively binds cRel/p65 heterodimers (30). Several observations in the present study suggested that LXA4 augments ECV304 cell gene expression, at least in part, by stimulating gene transcription. First, most other mediators that trigger TF expression do so by stimulating de novo protein synthesis. Second, the time course of induction of TF activity by LXA4 in our study was compatible with de novo protein synthesis. Third, both cytokine-triggered and LXA4-triggered TF activity are modulated by inhibitors of PKC, suggesting overlap of signal transduction pathways. Fourth, LXA4 failed to stimulate ECV304 cell TF activity in the presence of cycloheximide, a well-defined inhibitor of protein synthesis. Analysis of TF mRNA levels by Northern analysis provided further evidence that LXA4 triggered TF expression through a receptor activated cell signaling pathway involving activation of gene transcription. Treatment of ECV304 cells with LXA4 was associated with an increase in TF mRNA levels. This increase in TF mRNA expression was inhibited by prior exposure of ECV304 cells to pertussis toxin, a profile of inhibition that paralleled that noted for inhibition of LXA4stimulated TF activity. Interestingly, induction of TF gene expression appeared to be dissociated from NF-␬B activation in this system, thereby distinguishing LXA4triggered TF expression from TF expression triggered by other inflammatory mediators. It should be noted, however, that there is evidence that LPS and cytokines also stabilize TF mRNA, thereby suggesting at least two levels of regulation of TF activity (13). Studies are in progress to investigate the possibility that LXA4 in addition to inducing TF gene transcription may also influence TF activity through the latter mechanism. In summary, the results of the present study demonstrate a novel role for LXA4 in regulating TF expression and activity. This effect appears to be dose dependent, stereospecific, and mediated through unique G protein-coupled receptors. Complementary studies using assays of TF bioactivity and mRNA levels suggest that engagement of LXA4 receptors triggered TF gene expression through a PKC-dependent and NF-␬B-independent mechanism. Importantly, studies of structurefunction indicate that it is possible to disassociate these potentially proinflammatory and prothrombotic effects of LXA4 from its previously reported capacity to inhibit neutrophil-endothelial cell interactions and neutrophil recruitment to sites of inflammation. Finally, these studies establish precedence for induction of parenchymal cell gene expression by lipoxygenase

C952


products. The identification of other genes regulated by these lipid mediators should advance our knowledge of the pathophysiology of inflammation, hypersensitivity, and thrombosis and may suggest novel strategies for therapeutic intervention. We acknowledge Dr. N. A. Petasis and Dr. V. V. Fokin (University of Southern California, Los Angeles, CA) for providing the LXA4 analogs, Dr. N. Mackman (Scripps Research Institute, La Jolla, CA) for providing the pcTF314 containing TF cDNA plasmid, Dr. L. A. J. O’ Neill (Trinity College, Dublin, Ireland) for supplying ECV304 cells, and D. Fitzgerald (Royal College of Surgeons, Dublin, Ireland) for supplying HUVECs. This study was funded by grants from the Irish Heart Foundation (to P. Maderna), the Health Research Board of Ireland (to H. Brady and C. Godson), and the Wellcome Trust (to H. Brady). REFERENCES 1. Badr KF, DeBoer DK, Schwartzberg M, and Serhan CN. Lipoxin A4 antagonizes cellular and in vivo actions of leukotriene D4 in rat glomerular mesangial cells: evidence for competition at a common receptor. Proc Natl Acad Sci USA 86: 3438–3442, 1989. 2. Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS, and Gimbrone MA Jr. Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: characterization and comparison with the actions of interleukin 1. Proc Natl Acad Sci USA 83: 4533–4537, 1986. 3. Bradford MMA. Rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254, 1976. 4. Brady HR and Serhan CN. Lipoxins: putative breaking signals in host defense, inflammation and hypersensitivity. Curr Opin Nephrol Hypertens 5: 20–27, 1996. 5. Bratt J and Gyllenhammar H. The role of nitric oxide in lipoxin A4-induced polymorphonuclear neutrophil-dependent cytotoxicity to human vascular endothelium in vitro. Arthritis Rheum 38: 768–776, 1995. 6. Brezinski DA, Nesto RW, and Serhan CN. Angioplasty triggers intracoronary leukotrienes and lipoxin4. Impact of aspirin therapy. Circulation 86: 56–63, 1992. 7. Cadroy Y, Dupouy D, and Boneu B. Arachidonic acid enhances the tissue factor expression of mononuclear cells by the cyclo-oxygenase-1 pathway: beneficial effect of n-3 fatty acids. J Immunol 160: 6145–6150, 1998. 8. Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156–159, 1987. 9. Christie PE, Spur BW, and Lee TH. The effects of lipoxin A4 on airway response in asthmatic subjects. Am Rev Respir Dis 145: 1281–1284, 1992. 10. Claria J and Serhan CN. Aspirin triggers novel bioactive eicosanoids by human endothelial cell-leukocyte interactions. Proc Natl Acad Sci USA 92: 9475–9479, 1995. 11. Colgan SP, Serhan CN, Parkos CA, Delp-Archer C, and Madara JL. Lipoxin A4 modulates transmigration of human neutrophils across intestinal epithelial cell monolayers. J Clin Invest 92: 75–82, 1993. 12. Colucci M, Balconi G, Lorenzet R, Pietra A, Locati D, Donati MB, and Semeraro N. Cultured human endothelial cells generate tissue factor in response to endotoxin. J Clin Invest 71: 1893–1896, 1983. 13. Crossman DC, Carr DP, Tuddenham EGD, Pearson JD, and McVey JH. The regulation of tissue factor mRNA in human endothelial cells in response to endotoxin or phorbol ester. J Biol Chem 265: 9782–9787, 1990. 14. Cunningham MA, Holdsworth SR, and Tipping PG. Coagulation independent, pro-inflammatory effects of tissue factor/ factor VIIa interactions in glomerulonephritis (Abstract). J Am Soc Nephrol 9: 2421, 1998. 15. Dahlen SE, Veale CA, Webber SE, Manon BE, Nicolaou KC, and Serhan CN. Pharmacodynamics of lipoxin A4 in airway smooth muscle. Agents Actions 26: 93–95, 1989.

16. Drake TA, Hannani K, Fei H, Lavi S, and Berliner JA. Minimally oxidized low-density lipoproteins induce tissue factor expression in cultured human endothelial cells. Am J Pathol 138: 601–607, 1991. 17. Edgington TS, Mackman N, Brand K, and Ruf W. The structural biology of expression and function of tissue factor. Thromb Haemost 66: 67–79, 1991. 18. Fiore S, Maddox JF, Perez HD, and Serhan CN. Identification of a human cDNA encoding a functional high affinity lipoxin A4 receptor. J Exp Med 180: 253–260, 1994. 19. Gronert K, Gewirtz A, Madara JL, and Serhan CN. Identification of a human enterocyte lipoxin A4 receptor that is regulated by interleukin (IL)-13 and interferon ␥ and inhibits tumor necrosis factor-␣-induced IL-8 release. J Exp Med 187: 1285– 1294, 1998. 20. Hedqvist P, Raud J, Palmertz J, Haeggstrom J, Nicolaou KC, and Dahlen SE. Lipoxin A4 inhibits leukotriene B4-induced inflammation in the hamster cheek pouch. Acta Physiol Scand 137: 571–581, 1989. 21. Jang Y, Guzman LA, Lincoff AM, Gottsauner-Wolf M, Forudi F, Hart CE, Courtman DW, Ezban M, Ellis SG, and Topol EJ. Influence of blockade at specific levels of the coagulation cascade on restenosis in a rabbit atherosclerotic femoral artery injury model. Circulation 92: 3041–3050, 1995. 22. Lee TH, Horton CE, Kyan-Aungm V, Haskard D, Crea AEG, and Spur W. Lipoxin A4 and lipoxin B4 inhibit chemotactic responses of human neutrophils stimulated by LTB4 and N-formyl-L-methionyl-L-leucyl-L-phenylalanine. Clin Sci (Lond) 77: 195–203, 1989. 23. Lopez-Pedrera C, Jardi M, Ingles-Esteve J, MunozCanoves P, Dorado G, Velasco F, and Felez J. Characterization of tissue factor expression on the human endothelial cell line ECV304. Am J Hematol 56: 71–78, 1997. 24. Mackman N. Protease inhibitors block lipopolysaccharide induction of tissue factor gene expression in human monocytic cells by preventing activation of c-Rel/p65 heterodimers. J Biol Chem 269: 26363–26367, 1994. 25. Maddox JF, Hachicha M, Takano T, Petasis NA, Fokin VV, and Serhan CN. Lipoxin A4 stable analogs are potent mimetics that stimulate human monocytes and THP-1 cells via a Gprotein-linked lipoxin A4 receptor. J Biol Chem 272: 6972–6978, 1997. 26. Morrissey JH, Fakhrai H, and Edgington TS. Molecular cloning of the cDNA for tissue factor, the cellular receptor for the initiation of the coagulation protease cascade. Cell 50: 129–135, 1987. 27. Oltrana L, Speidel CM, Recchia D, and Abeudschein DR. Inhibition of tissue factor-mediated thrombosis markedly attenuates stenosis after balloon-induced arterial injury in hyperlipidemic minipigs (Abstract). Circulation 90: I-344, 1994. 28. Papayianni A, Rennke H, Serhan CN, and Brady HR. Transcellular biosynthesis of lipoxin A4 during adhesion of platelets and neutrophils in experimental immune complex glomerulonephritis. Kidney Int 47: 1295–1302, 1995. 29. Papayianni A, Serhan CN, and Brady HR. Lipoxins inhibit leukotriene-stimulated interactions of human neutrophils and endothelial cells. J Immunol 156: 2264–2272, 1996. 30. Parry GCN and Mackman N. Transcriptional regulation of tissue factor expression in human endothelial cells. Arterioscler Thromb Vasc Biol 15: 612–621, 1995. 31. Sambrook J, Fritsch EF, and Maniatis T (Editors). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. 32. Scarpati EM and Sadler JE. Regulation of endothelial cell coagulant properties. J Biol Chem 264: 20705–20713, 1989. 33. Serhan CN. Lipoxins and novel aspirin-triggered 15-epi-lipoxins (ATL): a jungle of cell-cell interactions or a therapeutic opportunity? Prostaglandins 53: 107–137, 1997. 34. Serhan CN, Maddox JF, Petasis NA, Akritopoulou-Zanze I, Papayianni A, Brady HR, Colgan SP, and Madara JL. Design of lipoxin A4 analogs that block transmigration and

LXA4 IS A MODULATOR OF TF ACTIVITY adhesion of human neutrophils. Biochemistry 34: 14609–14615, 1995. 35. Speidel CM, Eisenberg PR, Ruf W, Edgington TS, and Abendschein DR. Tissue factor mediates prolonged procoagulant activity on the luminal surface of balloon-injured aortas in rabbits. Circulation 92: 3323–3330, 1995. 36. Takahashi K, Sawasaki Y, Hata J, Mukai K, and Goto T. Spontaneous transformation and immortalization of human endothelial cells. In Vitro Cell Dev Biol 25: 265–274, 1990. 37. Takano T, Clish CB, Gronert K, Petasis N, and Serhan CN. Neutrophil-mediated changes in vascular permeability are inhibited by topical application of aspirin-triggered 15-epi-lipoxin A4 and novel lipoxin B4 stable analogues. J Clin Invest 101: 819–826, 1998.

C953

38. Terry CM and Callahan KS. Protein kinase C regulates cytokine-induced tissue factor transcription and procoagulant activity in human endothelial cells. J Lab Clin Med 127: 81–93, 1996. 39. Tremoli E, Eligini S, Colli S, Maderna P, Rise P, Pazzucconi F, Marangoni F, Sirtori CR, and Galli C. n-3 fatty acid ethyl ester administration to healthy subjects and to hypertriglyceridemic patients reduces tissue factor activity in adherent monocytes. Arterioscler Thromb 14: 1600–1608, 1994. 40. Tsao BP, Fair DS, Curtiss LK, and Edgington TS. Monocytes can be induced by lipopolysaccharide-triggered T lymphocytes to express functional factor VII/VIIa protease activity. J Exp Med 159: 1042–1057, 1984.