fgl2 prothrombinase, by its ability to generate thrombin, has been shown to be pivotal to the pathogenesis of viral-induced hepatitis, cytokine-induced fetal loss syndrome, and xeno- and allograft rejection. In this study, the molecular basis of fgl2 prothrombinase activity was examined in detail. Purified fgl2 protein generated in a baculovirus expression system had no measurable prothrombinase activity, whereas the activity was restored when the purified protein was reconstituted into phosphatidyl-l-serine-containing vesicles. Reconstituted fgl2 catalyzed the cleavage of human prothrombin to thrombin with kinetics consistent with a first order reaction, with an apparent Vmax value of 6 mol/min/mol fgl2 and an apparent Km value for prothrombin of 8.3 μM. The catalytic activity was totally dependent on calcium, and factor Va (500 nM) enhanced the catalytic efficiency of fgl2 by increasing the apparent Vmax value to 3670 mol/min/mol fgl2 and decreasing the apparent Km value for prothrombin to 7.2 μM. By a combination of site-directed mutagenesis and production of truncated proteins, it was clearly shown that residue Ser89 was critical for the prothrombinase activity of fgl2. Furthermore, fgl2 prothrombinase activity was not inhibited by antithrombin III, soybean trypsin inhibitor, 4-aminobenzamidine, aprotinin, or phenylmethylsulfonyl fluoride, whereas diisopropylfluorophosphate completely abrogated the activity. In this work we provide direct evidence that fgl2 cleaves prothrombin to thrombin consistent with serine protease activity and requires calcium, phospholipids, and factor Va for its full activity.

Coagulation proteins play important roles in the pathogenesis of human diseases including virus-induced hepatitis (1, 2), atherosclerosis (3, 4), systemic lupus encephalitis (5), septic shock (6), and cancer (7). The generation of thrombin is a crucial reaction in blood coagulation and hemostasis. Thrombin is formed from its inactive zymogen, prothrombin, under physiological conditions, by the action of a prothrombinase complex composed of the serine protease factor Xa, with calcium, a negatively charged phospholipid surface, and the nonenzymatic cofactor, factor Va (8). The serine protease, factor Xa, is the classically known physiological activator of prothrombin. Factor Xa is activated from its zymogen precursor factor X in the presence of calcium by complexes in the intrinsic (factors IXa/VIIIa) (9) and extrinsic (factor VIIa/tissue factor) (10) pathways.

Our laboratory has previously demonstrated the pivotal role of a unique direct prothrombinase, fgl2, in the pathogenesis of fulminant viral hepatitis secondary to murine hepatitis virus strain-3 (MHV-3)3 infection (11, 12, 13). The gene fgl2 was originally cloned from CTLs, and the encoded protein was named as a fibrinogen-like protein due to homology to the carboxyl terminus of the β- and γ-chains of fibrinogen (14). Several lines of evidence indicate fgl2 exhibits coagulation activity. First, following MHV-3 infection, mRNA transcripts of fgl2 and fgl2 protein can be detected predominantly in liver reticuloendothelial cells, followed by fibrin deposition and widespread hepatic necrosis (13). We have recently reported a similar relationship of fgl2 to fulminant hepatitis B virus infection (1). Second, neutralizing mAb to fgl2 prevents hepatic fibrin deposition, liver cell necrosis, and the lethality of MHV-3 infection (15). Third, fgl2 is also expressed in a model of Th1 cytokine-induced fetal loss syndrome in which fibrin deposition is a prominent feature (16, 17), and neutralizing Ab to fgl2 prevents cytokine-induced fetal loss (16). Recently, Ruegg and colleagues (18, 19) have reported the expression of FGL2 in human peripheral blood T lymphocytes and termed the encoded protein fibroleukin. Interestingly, FGL2 expressed by T cells as a secreted form lacks a coagulation activity, and the function for T cell generated fgl2/fibroleukin remains undefined.

Based on our previous findings, we believe induction of fgl2 represents a novel mechanism whereby macrophages and endothelial cells activate the coagulation cascade. However, the nature of this fgl2 coagulation activity has not been fully defined. In this paper, we generated murine fgl2 protein by a baculovirus expression system and examined the molecular basis of its prothrombinase activity. Using purified fgl2, we assessed the kinetic parameters of thrombin generation. We further demonstrated that phospholipids, calcium, and factor Va were required for the optimal catalytic activity of fgl2. A combination of site-directed mutagenesis and production of truncated proteins identified a domain and a single amino acid residue essential for the coagulation activity of fgl2.

Log-phase Spodoptera frugiperda ovarian cells 9 (Sf9) and High Five (H5) cells (Invitrogen, Carlsbad, CA) were maintained in Insect Pathology Laboratory 41 medium supplemented with 10% FBS, 10 μg/ml gentamicin, 3.3 μg/L yeastolate, 2.5 μg/ml fungizone, and 1× lipid concentrate at 27°C as monolayers. Recombinant baculovirus carrying murine fgl2 cDNA was generated in Sf9 cells and fgl2 protein was expressed in H5 cells. Chinese hamster ovary (CHO)-K1 cells were cultured in DMEM supplemented with 2 mM l-glutamine and 10% FBS. All tissue culture reagents were purchased from Life Technologies (Mississauga, Ontario, Canada).

Fgl2 protein with a tandem repeat of six histidine residues followed by an enterokinase cleavage site fused to the N terminus was prepared according to the Invitrogen Insect Expression System protocol. Briefly, a 1.4-kb cDNA encoding murine fgl2 was amplified using the forward primer 5′-TGCCGCACTGGATCCATGAGGCTTCCTGGT-3′ (with the methionine start codon underlined) and the reverse primer 5′-TTATGGCTTGAAATTCTTGGGC-3′ (nt 1283–1302 relative to the ATG start codon). Amplification was performed for 25 cycles for 2 min at 96°C, 2 min at 55°C, and 3 min at 72°C. The PCR product was cloned into the EcoRI and BamHI sites of the vector pBlueBacHis2A (Invitrogen).

Recombinant baculoviruses were generated by cotransfecting Sf9 cells with 2 μg pBlueBacHis2A carrying fgl2 insert plus 1 μg linearized Autographa californica multiple nuclear polyhedrosis virus DNA using the Bac-N-Blue transfection kit (Invitrogen). Putative recombinant viruses were screened for the presence of fgl2 by PCR followed by three rounds of viral plaque purification. The sequence of the recombinant baculovirus containing murine fgl2 cDNA was confirmed by an automated DNA sequencer (model no. 377; PE Applied Biosystems, Foster City, CA).

His-tagged fgl2 protein was purified using a nickel resin column according to Invitrogen’s protocol. Briefly, monolayers of H5 cells were infected with the recombinant baculovirus at a multiplicity of infection of 5. Seventy-two hours later, the infected cells were harvested by centrifugation and lysed in 6 M guanidinium hydrochloride, 20 mM sodium phosphate, and 500 mM NaCl. The soluble material was mixed with 50% slurry of ProBond nickel-nitrilotriacetic acid (Ni-NTA) resin (Invitrogen) for 1 h at 4°C. After washings, bound fgl2 protein was eluted with 8 M urea and 20 mM sodium phosphate (pH 5.3) with 150 mM NaCl. The pH of the eluted protein was adjusted to 7.2 immediately upon elution, and the protein was renatured by dialyzing against urea-saline buffers (150 mM NaCl (pH 7.2)) with successive decreases in urea concentrations (6, 4, 2, and 1 M) and finally against TBS (10 mM Tris, 150 mM NaCl (pH 7.2)). The dialyzed material was concentrated and soluble fgl2 protein was collected by spinning at 14,000 rpm for 10 min to remove insoluble particulates. Protein concentrations were determined by BCA assay (Pierce, Rockford, IL).

Homogeneity of purified soluble fgl2 protein was evaluated by SDS-PAGE and confirmed by Western blot probed with anti-fgl2 Ab as previously described (13). Proteins were stained directly using Coomassie brilliant blue or were transferred to nitrocellulose, and then were probed using polyclonal rabbit anti-mouse fgl2 IgG as the primary Ab. The secondary Ab used for immunoblotting was affinity-purified donkey anti-rabbit IgG conjugated to HRP (Amersham, Little Chalfont, U.K.) and the blot was visualized with an immunochemiluminescent kit (Amersham).

Purified soluble fgl2 protein was incorporated into phospholipid vesicles based on the detergent dialysis technique previously described (20, 21, 22). Briefly, hen egg phosphatidyl-l-serine (PS) and bovine brain phosphatidylcholine (PC) (Sigma-Aldrich, St. Louis, MO) in chloroform were combined to give different ratios of PS:PC (ranging from 0:100 to 40:60) and dried to a thin layer on the walls of a borosilicate glass tube under a stream of dry nitrogen and evacuated for 2 h to remove the resident traces of solvent. The lipid film (16 mg of total lipid content) was resuspended in 3 ml of 20 mM HEPES, 150 mM NaCl, and 80 mM octyl glucoside (pH 7; Sigma-Aldrich), and 20 μM of the lipid mixture was mixed with various concentrations of purified fgl2 protein. The mixed micellar solution was placed in microdialysis chambers (Pierce) and dialyzed against successive changes of HEPES-saline buffer (20 mM HEPES, 0.15 M NaCl (pH 7.2)) for 96 h at room temperature. Following dialysis, samples were centrifuged at 14,000 rpm for 30 min to remove particulates. As a negative control, BSA was also reconstituted into the phospholipid membranes. All the preparations were stored at 4°C and 100 μl of each sample was used to assay for the coagulation activity.

The 1.4-kb cDNA encoding the murine fgl2 was amplified as previously described and the PCR product was cloned into the mammalian expression vector pCR3.1-Uni with the TA Cloning kit (Invitrogen) for the expression of wild-type fgl2 protein. Using primer combinations in Table I, various truncated fgl2 constructs were generated. All truncated PCR products were amplified with the same 5′ primer but various 3′ primers with a stop codon (underlined in Table I) incorporated in the primer sequence and cloned into the vector pCR3.1-Uni with the TA Cloning kit. Sequences of the truncated constructs were confirmed by an automated DNA sequencer.

Table I.

Sequences of primers used to generate fgl2 PCR products for preparing the truncated and mutated fgl2 proteinsa

ConstructsSequences
3′ Truncation constructs  
Forward primer 5′-TGC CGC ACT GCA AGG ATG AGG-3′ (−15 to 6) 
Reverse primers  
Fgl12-aa1/386 5′-CTG GTG GTA TTA TTT GCC ATT-3′ (1149–1169) 
Fgl2-aa1/336 5′-CT GAA ACG CTA GGC ATC CCC-3′ (1000–1019) 
Fgl2-aa1/284 5′-CTT ACT CTA GGT CAG AAG ATG-3′ (846–866) 
Fgl2-aa1/231 5′-CC CAT GGT CTA CAT GTC ACA-3′ (685–704) 
Fgl2-aa1/181 5′-GCT GGG ACA CTA GGA ACA CTT-3′ (534–554) 
Fgl2-aa1/135 5′-CTT CAG CTC TTA GGA CAG CTT G-3′ (396–407) 
Serine-mutated constructs  
Ser425Ala (sense primer) 5′-GGC TAC AAG GCC TCC TTC AAA-3′ (1266–1286) 
Ser135Ala (sense primer) 5′-AAG CTG TCC GCA GAG CTG AAG-3′ (396–416) 
Ser89Ala (sense primer) 5′-GCA GTG GAC GCT CTG AAG AAA-3′ (258–278) 
ConstructsSequences
3′ Truncation constructs  
Forward primer 5′-TGC CGC ACT GCA AGG ATG AGG-3′ (−15 to 6) 
Reverse primers  
Fgl12-aa1/386 5′-CTG GTG GTA TTA TTT GCC ATT-3′ (1149–1169) 
Fgl2-aa1/336 5′-CT GAA ACG CTA GGC ATC CCC-3′ (1000–1019) 
Fgl2-aa1/284 5′-CTT ACT CTA GGT CAG AAG ATG-3′ (846–866) 
Fgl2-aa1/231 5′-CC CAT GGT CTA CAT GTC ACA-3′ (685–704) 
Fgl2-aa1/181 5′-GCT GGG ACA CTA GGA ACA CTT-3′ (534–554) 
Fgl2-aa1/135 5′-CTT CAG CTC TTA GGA CAG CTT G-3′ (396–407) 
Serine-mutated constructs  
Ser425Ala (sense primer) 5′-GGC TAC AAG GCC TCC TTC AAA-3′ (1266–1286) 
Ser135Ala (sense primer) 5′-AAG CTG TCC GCA GAG CTG AAG-3′ (396–416) 
Ser89Ala (sense primer) 5′-GCA GTG GAC GCT CTG AAG AAA-3′ (258–278) 
a

Truncated fgl2 constructs were generated using the same forward primer with an ATG start site and their corresponding reverse primer with a stop codon (TTA or CTA) as shown (underlined). Mutated PCR products were prepared by mutating the serine codons into alanines (underlined) using primers containing appropriate base substitutions. The nucleotide position of the primers relative to the downstream of the ATG start codon of fgl2 cDNA are indicated in parentheses.

As we had postulated that fgl2 had serine protease activity, serine residues 89, 136, and 426 were individually mutated to alanines to evaluate the effect on fgl2 coagulation activity. Site-directed mutagenesis was performed using the QuickChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer’s protocol with primer containing the appropriate base substitutions as shown in Table I, such that serine residues were substituted individually by alanines. All sequences of the mutated constructs were confirmed by an automated DNA sequencer.

CHO-K1 cells were transfected with various plasmid DNA fgl2 constructs using the Lipofectamine Plus Reagent kit (Life Technologies). Three days after transfection, the transfected cells were trypsinized and split onto new plates. After incubation for 24 h to allow the cells attach onto the plate, a neomycin analog, G418 (800 μg/ml; Life Technologies), was added to the culture and incubated until the plate became ∼70% confluent for 10–14 days. Colonies were then selected and cultured with G418 (400 μg/ml) to generate stable cell lines by changing medium every 3–4 days. The cells were washed briefly with ice-cold PBS and spun at 2500 rpm for 5 min at 4°C. Cell pellets were then frozen immediately in liquid nitrogen and stored at −80°C until further analysis. Expression of fgl2 proteins was verified by Western blotting and their prothrombinase activity was analyzed in a one-stage clotting assay.

Samples of purified soluble fgl2 protein reconstituted into phospholipid vesicles were analyzed for their ability to accelerate the clotting time of recalcified normal human platelet-poor plasma in a one-stage clotting assay as previously described (23). Phospholipid vesicles alone or reconstituted with BSA were used as negative controls. In addition, stably transfected CHO-K1 cells (2 × 106 cells) subjected to three cycles of freeze/thawing and diluted in 100 μl of RPMI 1640 medium were assayed for their coagulation activity. In brief, samples (100 μl) were mixed with 100 μl of citrated normal human platelet-poor plasma and 100 μl of 15 mM CaCl2 (prewarmed to 37°C), and the time required for the appearance of an insoluble fibrin clot with manual rocking at 37°C was recorded.

Additional one-stage clotting assays were performed with human plasmas deficient in coagulation factor II, VII, or X (Helena Laboratories, Beaumont, TX) to determine the factor dependence of relipidated fgl2 prothrombinase activity. Data were expressed as the mean ± SE of triplicate assays.

Assessment of the susceptibility of fgl2 prothrombinase to protease inhibitors was performed as previously described (23). Purified fgl2 protein (150 nM) reconstituted into PCPS (75% PC:25% PS; 20 μM) was incubated with each of the following protease inhibitors: antithrombin III (10 μg/ml) in the presence of heparin (1 U/ml), soybean trypsin inhibitor (10 μg/ml), 4-aminobenzamidine (1 mM), aprotinin (10 U/ml), PMSF (1 mM), and diisopropylfluorophosphate (DFP; 2.5 and 10 mM; Sigma-Aldrich). Following a 15-min incubation with protease inhibitor at 22°C, 10 μl of 125I-labeled prothrombin and 10 μl of 25 mM CaCl2 were added, and the reactions were incubated for an additional 30 min at 37°C. To stop the reaction, 1% SDS and 1% EDTA were added and 125I-labeled prothrombin cleavage was assessed by SDS-PAGE and soft laser scanning densitometry (Zeineh soft laser; Biomed Instruments, Chicago, IL). Human factor Xa, derived from factor X by cleavage with RVV, in the presence of phospholipids was treated identically as positive control. The data were normalized to the maximum amplitude of 125I-labeled prothrombin cleavage when no inhibitor was used (defined as 0% inhibition) to facilitate comparison.

Kinetics of thrombin generation by relipidated fgl2 (150 nM plus 20 μM PL with 25% PS) was measured using a chromogenic substrate of thrombin, Spectrozyme TH (H-d-hexahydrotyrosyl-l-alanyl-l-arginine-p-nitroanalide-diacetate; American Diagnostica, Montreal, Canada). The initial rates of thrombin generation by the reconstituted fgl2 protein with 7 μM prothrombin over a range of CaCl2 concentrations (0–20 mM) were also quantitated. The effect of purified human factor Va (0.5–500 nM; Hematologic Technologies, Essex Junction, VT) on the murine fgl2 prothrombinase activity at 0.7 μM prothrombin was also determined. Kinetics parameters of thrombin generation were obtained by measuring the initial rates of thrombin generation at various concentrations of prothrombin (0.01–250 μM) with a final concentration of 5 mM CaCl2. The effects of factor Va (500 nM) on the Km and Vmax of relipidated fgl2 was determined after incubating with factor Va for 1 min at 37°C and addition of prewarmed prothrombin in the assay buffer. The assay buffer used in all experiments, unless otherwise noted, was 5 mM CaCl2, 10 mM HEPES, and 150 mM NaCl (pH 7.4). The initial rates of activation (<10% of substrate used) were determined. Reactions were monitored over time by quenching aliquots (10 μl) of the reaction in EDTA buffer (120 μl; 10 mM HEPES (pH 7), 100 mM NaCl, 5 mM EDTA, 1 mg/ml BSA, 1 mg/ml polyethylene glycol 8000) and measuring the rate of hydrolysis of Spectrozyme TH (20 μM), and the appearance of the chromophore, p-nitroanilide acetate, was monitored over time at 405 nm. The concentration of thrombin formed was calculated by comparison to a standard curve generated with purified human thrombin (Sigma-Aldrich). Pure prethrombin 1 and prethrombin 2 are known to have negligible amidase activity on Spectrozyme TH. Kinetic values were derived from the least-squares fit of the data based on the Michaelis-Menten equation (SigmaPlot). Rates of thrombin formation were calculated and expressed as moles of thrombin formed per minute per mole of fgl2. Km and Vmax values were determined by statistical analysis of the Lineweaver-Burk plots as described by Eisenthal and Cornish-Bowden (24) using the software SigmaPlot 2001 Enzymatic Kinetics (Sigma-Aldrich).

Female BALB/cJ mice, 6–8 wk of age, were purchased from The Jackson Laboratory (Bar Harbor, ME) and were fed a standard chow diet and allowed to acclimatize for 1 wk before use. Peritoneal exudative macrophages from BALB/cJ mice were harvested in ice-cold PBS 4 days after i.p. injection of 1.5 ml of 5% thioglycolate (Difco, Detroit, MI) as previously described (25, 26). The cells were washed twice in cold PBS and resuspended in RPMI 1640, 2% FBS at 2 × 106 cells/ml. This procedure consistently yielded a >85% macrophage cell population confirmed by staining with anti-Mac1 (CD11b/CD18) Ab (Cedarlane Laboratories, Hornby, Ontario, Canada), with >98% viability by trypan blue exclusion. Cells were incubated for 1 h at 37°C before experimentation. Infection of peritoneal macrophages by MHV-3 was performed as previously described (25, 26). In brief, MHV-3 with a titer of 5 × 106 PFU/ml was used to infect the purified peritoneal macrophages at a multiplicity of infection of 5. After 8 h of infection, infection was stopped by placing the cells on ice.

Expression of fgl2 on peritoneal macrophages was performed by flow cytometry. The MHV-3-infected macrophages were stained with biotinylated polyclonal rabbit anti-fgl2 Ab at 4°C for 30 min in the dark. After two washes, cells were stained with streptavidin-PE (Cedarlane Laboratories) for 30 min at 4°C and finally analyzed using a COULTER Epics-XL flow cytometer (Beckman Coulter, Fullerton, CA).

The results were calculated as means ± SEM in all figures and tables; the means were compared using the analysis of variance by Student’s t test using the software Statistix 7 (Analytical Software, Tallahassee, FL). A p value ≤ 0.05 was considered statistically significant.

Murine fgl2 protein was generated using the baculovirus expression system as described above. Fig. 1,a shows a SDS-PAGE of fgl2 baculovirus-infected H5 cells under reducing conditions followed by Western blot analysis. It can be seen that Coomassie blue staining (Fig. 1 b) showed a dominant band at the same molecular size of ∼65 kDa, comparable to the size of the fgl2 protein previously estimated (13, 25).

FIGURE 1.

Expression of fgl2 protein in a baculovirus system. a, Western blot analysis of fgl2 expression in H5 cells infected with the recombinant baculovirus containing the fgl2 insert. Homogenates of H5 cells infected with wild-type and recombinant baculoviruses were analyzed. Each lane was loaded with 20 μg of cell lysate and the blot was probed with polyclonal rabbit anti-mouse fgl2 IgG as described in Materials and Methods. Lane 1, lysate of the uninfected H5 cells; lane 2, lysate of the wild-type baculovirus-infected H5 cells; lane 3, lysate of the recombinant baculovirus-infected H5 cells. b, SDS-PAGE followed by Coomassie blue staining of fgl2 protein purified from recombinant baculovirus-infected H5 cells showed a dominant band consistent with the expression of fgl2. Cell lysates prepared under denaturing condition were incubated with Ni-NTA resin and proteins bound to Ni-NTA resin were eluted. All eluted proteins (10 μg) were analyzed on 10% SDS-PAGE and the gel was stained with Coomassie blue. Lane 1, eluted proteins from lysate of the uninfected H5 cells; lane 2, eluted proteins from lysate of the wild-type baculovirus-infected H5 cells; lane 3, eluted proteins from lysate of the recombinant baculovirus-infected H5 cells.

FIGURE 1.

Expression of fgl2 protein in a baculovirus system. a, Western blot analysis of fgl2 expression in H5 cells infected with the recombinant baculovirus containing the fgl2 insert. Homogenates of H5 cells infected with wild-type and recombinant baculoviruses were analyzed. Each lane was loaded with 20 μg of cell lysate and the blot was probed with polyclonal rabbit anti-mouse fgl2 IgG as described in Materials and Methods. Lane 1, lysate of the uninfected H5 cells; lane 2, lysate of the wild-type baculovirus-infected H5 cells; lane 3, lysate of the recombinant baculovirus-infected H5 cells. b, SDS-PAGE followed by Coomassie blue staining of fgl2 protein purified from recombinant baculovirus-infected H5 cells showed a dominant band consistent with the expression of fgl2. Cell lysates prepared under denaturing condition were incubated with Ni-NTA resin and proteins bound to Ni-NTA resin were eluted. All eluted proteins (10 μg) were analyzed on 10% SDS-PAGE and the gel was stained with Coomassie blue. Lane 1, eluted proteins from lysate of the uninfected H5 cells; lane 2, eluted proteins from lysate of the wild-type baculovirus-infected H5 cells; lane 3, eluted proteins from lysate of the recombinant baculovirus-infected H5 cells.

Close modal

Initial studies using one-stage clotting analysis showed that the fgl2 protein expressed by lysates of recombinant baculovirus-infected H5 cells exhibited coagulation activity with a dependence on CaCl2. In contrast, uninfected H5 cells and cells infected with the wild-type baculovirus were devoid of coagulation activity (Table II). To characterize the molecular basis of fgl2 prothrombinase activity in detail, we first purified fgl2 protein. Purified fgl2 protein did not exhibit any measurable coagulation activity consistent with reports indicating a phospholipid backbone is required for most clotting enzymes to manifest their activity (27, 28, 29) (Table II). Table II shows that purified fgl2 protein reconstituted into phospholipid vesicles containing 10 or 25% PS had coagulation activity. Similar to data on the factor Xa/Va complex, phospholipid vesicles consisting of 25% PS also maximized recovery of prothrombinase activity, whereas no activity was detected when fgl2 protein was incorporated into phospholipid vesicles devoid of PS. Unlike factor Xa (30), soluble purified fgl2 protein exhibited coagulation activity only after incorporation into PS-containing vesicles, whereas no activity was detected when fgl2 was mixed with PS containing vesicles without reconstitution (data not shown).

Table II.

Coagulation activity of fgl2 protein

SamplesClotting Timea (sec/2 × 106 cells)
Insect cell lysatesb  
Uninfected H5 cells 256 ± 6 
H5 cells infected with WT-AcMNPV 126 ± 2c 
H5 cells infected with recombinant baculovirus 73 ± 3c 
Purified fgl2 protein 232 ± 6 
Relipidated fgl2d  
Fgl2 protein alone >240 
Fgl2+ 100% PC >240 
Fgl2+ 90% PC:10% PS 150 ± 5e 
Fgl2+ 75% PC:25% PS 110 ± 4e 
Fgl2+ 60% PC:40% PS >240 
SamplesClotting Timea (sec/2 × 106 cells)
Insect cell lysatesb  
Uninfected H5 cells 256 ± 6 
H5 cells infected with WT-AcMNPV 126 ± 2c 
H5 cells infected with recombinant baculovirus 73 ± 3c 
Purified fgl2 protein 232 ± 6 
Relipidated fgl2d  
Fgl2 protein alone >240 
Fgl2+ 100% PC >240 
Fgl2+ 90% PC:10% PS 150 ± 5e 
Fgl2+ 75% PC:25% PS 110 ± 4e 
Fgl2+ 60% PC:40% PS >240 
a

Data (measured in seconds for every 2 × 106 cells) represent the mean ± SE of three experiments done in triplicate (n = 9).

b

Lysate of recombinant baculovirus-infected H5 cells expressing fgl2 protein shows significant coagulation activity compared to cells expressing no fgl2 and to the purified fgl2 protein. The uninfected, wild-type baculovirus (WT-AcMNPV)-infected, recombinant baculovirus-infected H5 cells (2 × 106 cells) and the purified fgl2 protein were assayed for their ability to accelerate the clotting time of recalcified normal platelet-poor plasma.

c

Significantly different from uninfected H5 cells, with p < 0.01.

d

Purified fgl2 protein did not exhibit prothrombinase activity, but the activity was restored when fgl2 was reconstituted into membrane vesicles having appropriate PC to PS. Purified fgl2 protein (150 nM) was reconstituted into 20 μM of phospholipid vesicles and the mixtures were assayed for their ability to accelerate the clotting time of recalcified normal platelet-poor plasma.

e

Significant recovery of activity compared to purified fgl2 protein alone, with p < 0.001.

Using factor-deficient plasmas, we demonstrated that the coagulation activity of fgl2 was dependent only on factor II (prothrombin) and thus behaved as a direct prothrombinase (Table III).

Table III.

One-stage clotting assay using plasmas deficient in various coagulation factors suggests that the coagulation activity of fgl2 was dependent on factor II but not factor VII or Xa

PlasmaClotting Time (s)
Normal 110 ± 4 
Deficient in:  
Factor VII 108 ± 6 
Factor X 120 ± 4 
Factor II 250 ± 3b 
PlasmaClotting Time (s)
Normal 110 ± 4 
Deficient in:  
Factor VII 108 ± 6 
Factor X 120 ± 4 
Factor II 250 ± 3b 
a

Purified fgl2 protein (150 nM) reconstituted into 20 μM of PCPS (75% PC:25% PS) was assayed for their ability to shorten the clotting time of human plasmas deficient in factors II, VII, or X as described in Materials and Methods. Results represent the mean ± SE of three experiments done in triplicate (n = 9).

b

Significant loss of activity compared to result using normal plasma, with p < 0.001.

Purified fgl2 protein (150 nM) reconstituted into 20 μM PCPS (75% PC:25% PS) exhibited maximal coagulation activity and was used for kinetic analyses. As shown in Fig. 2, optimal fgl2 prothrombinase activity of this preparation was reached with 5 mM CaCl2. No prothrombinase activity was detected in the absence of CaCl2.

FIGURE 2.

Effect of Ca2+ on the rate of thrombin generation by fgl2 prothrombinase. Purified fgl2 protein (150 nM) reconstituted into 20 μM 75% PC:25% PS membrane vesicles in 10 mM HEPES, 150 mM NaCl, and 7 μM prothrombin as substrate with various amounts of CaCl2; at 0 mM-CaCl2, 1 mM EDTA was present in the reaction mixture. Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. The rate of thrombin formation at each calcium concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles.

FIGURE 2.

Effect of Ca2+ on the rate of thrombin generation by fgl2 prothrombinase. Purified fgl2 protein (150 nM) reconstituted into 20 μM 75% PC:25% PS membrane vesicles in 10 mM HEPES, 150 mM NaCl, and 7 μM prothrombin as substrate with various amounts of CaCl2; at 0 mM-CaCl2, 1 mM EDTA was present in the reaction mixture. Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. The rate of thrombin formation at each calcium concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles.

Close modal

Before performing kinetic studies, we confirmed that the rate of thrombin generation by the reconstituted fgl2 protein was constant in time and proportional to the amount of fgl2 present in the reaction mixtures. Using three concentrations of fgl2 (15, 50, and 150 nM), the amount of thrombin generation was constant for up to 15 min and was proportional to the amount of fgl2 reconstituted into the vesicles. No further increase in the rate of thrombin generation was observed when 300 nM fgl2 reconstituted into PCPS vesicles was tested (data not shown). We then measured the catalytic function of fgl2 by evaluating the initial rate of thrombin generation at different concentrations of substrate, prothrombin. The data presented in the form of Michaelis-Menten (Fig. 3,a) and Lineweaver-Burk (Fig. 3 b) plots show that relipidated fgl2 behaves as an enzyme with a Km value for prothrombin of 8.3 ± 1.1 μM and Vmax value of 6 ± 0.2 mol of thrombin generated per minute per mole of fgl2.

FIGURE 3.

The Michaelis-Menten (a) and Lineweaver-Burk (b) plots of fgl2 prothrombinase converting prothrombin to thrombin. Fgl2 protein (150 nM) reconstituted into 75% PC:25% PS (20 μM membrane vesicles was incubated with varying concentrations of prothrombin (0–250 μM)). Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. The rate of thrombin formation at each prothrombin concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles. The Lineweaver-Burk figure is plotted as reciprocals of rate vs prothrombin concentration.

FIGURE 3.

The Michaelis-Menten (a) and Lineweaver-Burk (b) plots of fgl2 prothrombinase converting prothrombin to thrombin. Fgl2 protein (150 nM) reconstituted into 75% PC:25% PS (20 μM membrane vesicles was incubated with varying concentrations of prothrombin (0–250 μM)). Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. The rate of thrombin formation at each prothrombin concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles. The Lineweaver-Burk figure is plotted as reciprocals of rate vs prothrombin concentration.

Close modal

Because factor Va is known to greatly enhance the catalytic activity of factor Xa, we next examined the potential role of factor Va on the catalytic efficiency of fgl2. The rate of formation of thrombin at different concentrations of factor Va ranging from 0.5 to 500 nM with 0.7 μM prothrombin substrate was monitored. Factor Va enhanced the catalytic efficiency of fgl2 in a dose-dependent fashion, and the rate of thrombin generation was maximally reached when 500 nM of factor Va was used (Fig. 4,a). The Lineweaver-Burk plot (Fig. 4 b) suggested that factor Va (500 nM) enhances the catalytic efficiency of fgl2 by increasing the Vmax value to 3670 ± 87 mol of thrombin generated per minute per mole of fgl2 and decreasing the Km value to 7.2 ± 0.8 μM.

FIGURE 4.

Effect of purified human factor Va on the rate of thrombin generation by fgl2 prothrombinase. a, Rates of thrombin generation with different amounts of human factor Va. Initial rates of thrombin generation by purified fgl2 protein (150 nM) reconstituted into 20 μM 75% PC:25% PS membrane vesicles with 0.7 μM prothrombin as substrate in 10 mM HEPES and 150 mM NaCl containing various amounts of human factor Va (0.5–500 nM) were analyzed. Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. b, The Lineweaver-Burk plot of fgl2 prothrombinase with human factor Va (500 nM) converting prothrombin to thrombin. Fgl2 protein (150 nM) reconstituted into 75% PC:25% PS (20 μM) membrane vesicles with factor Va was incubated with varying concentrations of prothrombin (0–250 μM). The rate of thrombin formation at each prothrombin concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles. Data are plotted as reciprocals of rate vs prothrombin concentration.

FIGURE 4.

Effect of purified human factor Va on the rate of thrombin generation by fgl2 prothrombinase. a, Rates of thrombin generation with different amounts of human factor Va. Initial rates of thrombin generation by purified fgl2 protein (150 nM) reconstituted into 20 μM 75% PC:25% PS membrane vesicles with 0.7 μM prothrombin as substrate in 10 mM HEPES and 150 mM NaCl containing various amounts of human factor Va (0.5–500 nM) were analyzed. Following initiation of the reaction at 37°C, aliquots were quenched and initial rates of thrombin generation were analyzed by measurement of the hydrolysis of Spectrozyme TH (20 μM) as described in Materials and Methods. b, The Lineweaver-Burk plot of fgl2 prothrombinase with human factor Va (500 nM) converting prothrombin to thrombin. Fgl2 protein (150 nM) reconstituted into 75% PC:25% PS (20 μM) membrane vesicles with factor Va was incubated with varying concentrations of prothrombin (0–250 μM). The rate of thrombin formation at each prothrombin concentration was then calculated and expressed as moles of thrombin (IIa) per minute per mole of fgl2 protein reconstituted into the phospholipid vesicles. Data are plotted as reciprocals of rate vs prothrombin concentration.

Close modal

To further demonstrate that relipidated murine fgl2 protein had prothrombinase activity, we performed a 125I-labeled prothrombin cleavage assay as previously described (23). Fgl2 protein in the absence of phospholipid failed to cleave prothrombin. However, when fgl2 was reconstituted into PCPS-containing vesicles, in the presence or absence of factor Va, characteristic prothrombin cleavage products were observed. In comparison to cleavage fragments generated by factor Xa, a unique band of ∼24 kDa was seen when prothrombin was cleaved by fgl2, as we have previously reported (Ref. 17 and data not shown).

To determine the regions of fgl2 protein required for fgl2 coagulation activity, six truncated proteins were stably expressed in CHO-K1 cells. The wild-type fgl2 protein (fgl2-aa1/432) and the various truncated proteins were expressed at comparable levels as determined by Western blot analysis (data not shown). One-stage clotting assays were performed to determine whether the expressed proteins retained clotting activity. As shown in Fig. 5 a, truncated proteins with deletion of amino acids 387–432 or 337–432 (i.e., fgl2-aa1/386 and fgl2-aa1/336, respectively) retained coagulation activity. Truncated proteins with amino acids 285–432, 232–432, 182–432, and 136–432 deleted (i.e., fgl2-aa1/284, Fgl2-aa1/231, Fgl2-aa1/181, and Fgl2-aa1/135, respectively) did not exhibit coagulation activity. Inactive proteins lacked residues 285–335, suggesting that one or more of these amino acids were critical for fgl2 prothrombinase activity.

FIGURE 5.

Region(s) and residue(s) critical for fgl2 coagulation activity. a, Truncated fgl2 proteins stably expressed in CHO-K1 cells exhibited different coagulation activities. b, Site-directed mutagenesis was performed to identify the critical residue(s) for the fgl2 coagulation activity. One-stage clotting assay was performed to determine the ability of different truncated and serine-mutated fgl2 proteins to accelerate the clotting time of recalcified normal platelet-poor human plasma as shown in Materials and Methods. Results were expressed as coagulation time (seconds per 2 × 106 cells). ∗, Significant loss of activity compared with full-length unmodified fgl2 protein (Fgl2-aa1/432) with p < 0.001.

FIGURE 5.

Region(s) and residue(s) critical for fgl2 coagulation activity. a, Truncated fgl2 proteins stably expressed in CHO-K1 cells exhibited different coagulation activities. b, Site-directed mutagenesis was performed to identify the critical residue(s) for the fgl2 coagulation activity. One-stage clotting assay was performed to determine the ability of different truncated and serine-mutated fgl2 proteins to accelerate the clotting time of recalcified normal platelet-poor human plasma as shown in Materials and Methods. Results were expressed as coagulation time (seconds per 2 × 106 cells). ∗, Significant loss of activity compared with full-length unmodified fgl2 protein (Fgl2-aa1/432) with p < 0.001.

Close modal

Although no classical Ser/His/Asp triad of most serine proteases was detected in the critical fgl2 sequence, the presence of the SXXK motifs suggested that fgl2 belonged to the clan SE of serine proteases (31). By analyzing the amino acid sequence, we identified three serine residues (Ser89, Ser135, and Ser425) that might be critical if fgl2 were a serine protease. To test the relevance of these residues, we individually mutated them from serine into alanine. Wild-type and point-mutated murine fgl2 constructs were stably expressed in CHO-K1 cells, and the cell lysates were analyzed for clotting activity; it was not feasible to prepare baculovirus clones and purify these mutants. The wild-type fgl2 protein (Fgl2-aa1/432) and the various truncation mutants (i.e., Fgl2-Ser89Ala, Fgl2-Ser135Ala and Fgl2-Ser425Ala) were expressed at comparable levels as detected by Western blot analysis (data not shown). Fig. 5 b shows that mutating Ser135 or Ser425 to alanine did not alter the fgl2 activity. However, the mutation Ser89Ala completely abolished the activity, indicating that Ser89 residue was critical.

To compare fgl2 prothrombinase activity to that of factor Xa, we examined the effect of protein inhibitors. As shown in Table IV, fgl2 prothrombinase activity was not inhibited by antithrombin III in the presence of heparin, soybean trypsin inhibitor, 4-aminobenzamidine, aprotinin, or PMSF, whereas >80% inhibition of factor Xa activity was observed in the presence of these protease inhibitors. Furthermore, as little as 2.5 mM DFP completely inhibited fgl2 prothrombinase activity, whereas 10 mM DFP was required for inhibition of factor Xa. The data indicated that fgl2 protein directly cleaves prothrombin to thrombin in a manner that has the characteristic of a serine protease but differs from the serine protease activity of factor Xa.

Table IV.

Blockage of fgl2 prothrombinase activity by different protease inhibitors compared to their effects on factor Xa activitya

InhibitorsInhibition (%)
Fgl2 prothrombinaseFactor Xa
None (buffer alone) 
Antithrombin III 5 ± 3 90 ± 5b 
Soybean trypsin inhibitor 5 ± 2 100 ± 0b 
4-Aminobenzamidine 2 ± 1 90 ± 5b 
Aprotinin 2 ± 2 95 ± 2b 
PMSF 5 ± 3 100 ± 0b 
DFP (2.5 mM) 95 ± 3b 10 ± 5 
DFP (10 mM) 95 ± 2b 90 ± 3b 
InhibitorsInhibition (%)
Fgl2 prothrombinaseFactor Xa
None (buffer alone) 
Antithrombin III 5 ± 3 90 ± 5b 
Soybean trypsin inhibitor 5 ± 2 100 ± 0b 
4-Aminobenzamidine 2 ± 1 90 ± 5b 
Aprotinin 2 ± 2 95 ± 2b 
PMSF 5 ± 3 100 ± 0b 
DFP (2.5 mM) 95 ± 3b 10 ± 5 
DFP (10 mM) 95 ± 2b 90 ± 3b 
a

Purified fgl2 protein (150 nM) reconstituted into 20 μM of PCPS (75% PC:25% PS) vesicles were mixed with various protease inhibitors, and their ability to cleave prothrombin using 125I-labeled prothrombin cleavage assay was examined. Data represent the mean ± SE of three experiments done in triplicate (n = 9) and are expressed as the percentage of inhibition for each inhibitor relative to control with no inhibitor.

b

Significantly different from the sample that contained no inhibitor, with p < 0.01.

MHV-3-infected macrophages of susceptible strain BALB/cJ mice express fgl2 prothrombinase. Proteins involved in coagulation are either transmembrane proteins, such as tissue factor and thrombomodulin, or membrane-associated proteins, such as factors Xa, IXa, and VIIIa, that can be washed from the cell surface. To determine whether fgl2 prothrombinase was transmembrane or membrane associated, we infected peritoneal macrophages of BALB/cJ mice with MHV-3 and tested for fgl2 on the cell surface using flow cytometry analysis. The labeling procedure involves extensive washing. Fig. 6 shows a representative result. Mean fluorescence intensity of fgl2 expression detected by specific anti-fgl2 Ab increased from 0.8 ± 0.05 to 43.8 ± 3.8 after MHV-3 infection. Isotopic control Ab gave mean fluorescence intensity of 0.5 ± 0.08. We concluded that fgl2 is a transmembrane protein, unlike factor Xa.

FIGURE 6.

Flow cytometry analysis of MHV-3-infected macrophages suggests that fgl2 is a transmembrane protein. Representative histograms of fgl2 expression are shown. Shown are the uninfected (control) macrophages (a) and the MHV-3-infected macrophages (b) stained with biotinylated polyclonal anti-fgl2 Ab then streptavidin-PE. The ordinate and the abscissa represent cell number and fluorescence intensity (FL2-H), respectively.

FIGURE 6.

Flow cytometry analysis of MHV-3-infected macrophages suggests that fgl2 is a transmembrane protein. Representative histograms of fgl2 expression are shown. Shown are the uninfected (control) macrophages (a) and the MHV-3-infected macrophages (b) stained with biotinylated polyclonal anti-fgl2 Ab then streptavidin-PE. The ordinate and the abscissa represent cell number and fluorescence intensity (FL2-H), respectively.

Close modal

The data presented in this paper demonstrate that murine fgl2 protein exhibited coagulation activity dependent solely upon coagulation factor II (prothrombin), calcium, and a negatively charged phospholipid backbone. The kinetics of thrombin generation by relipidated fgl2 prothrombinase was slow, similar to what has been previously reported for factor Xa alone (32). Addition of factor Va to Xa induces conformational changes of factor Xa and thereby increases its catalytic efficiency by a factor of 3000-fold (33). Interestingly, factor Va also enhanced the catalytic efficiency of fgl2 in a dose-dependent fashion. Although the Vmax value of thrombin generation for fgl2 at 500 nM of factor Va was similar to that reported for factor Xa, the apparent Km value was significantly higher, suggesting that murine fgl2 has a lower affinity toward human prothrombin than human factor Xa. As the present studies were undertaken with molecules from different species, the calculated Km value may underestimate the interaction of these molecules, reflecting species incompatibility. Previous data using coagulation proteins from different species support this contention (34). Further investigations are required to answer this question and also to determine how factor Va enhances fgl2 activity.

Phospholipid is required for assembly of proteins of the coagulation cascade into functional form. For example, PS-containing membranes enhance the affinity of factor Xa for its cofactor Va by 75- to 800-fold (35). Also, PS is responsible for a 10-fold increase for factor IXa binding to factor VIIIa (29). In addition, PS-containing membranes increase the catalytic efficiency of the factors Va/Xa30, factor VII/tissue factor (35), and the factors VIIIa/IXa complexes (29). Similarly, purified fgl2 protein only exhibited prothrombinase activity when reconstituted into PS-containing vesicles. This supports our previous observation that murine fgl2 coagulation activity was maximal after rearrangement of the negatively charged phospholipid (i.e., PS) from the inner leaflet of the cell membranes to the exterior leaflet during cycles of freeze-thawing of fgl2-expressing cells. In this paper, we report that a similar ratio of PS:PC is crucial for optimal murine fgl2 coagulation activity. Whether relipidation of fgl2 increases binding of fgl2 to prothrombin or enhances catalysis of prothrombin by fgl2 requires further study.

A series of truncated fgl2 proteins were expressed in CHO-K1 cells to determine which region(s) were critical for its activity. Truncation of residues 336–432 did not affect fgl2 prothrombinase activity, whereas deletion of amino acid residues of 285–335 completely abolished the activity. Truncations of murine fgl2 upstream of amino acid residue 284 also resulted in an inactive protein. Deletion of such a large number of amino acid residues might have affected the proper conformation of fgl2. Alternatively, these deletions might have removed an exosite (prothrombin binding site) for the initial interaction of murine fgl2 with prothrombin, distinct from the catalytically active site. A similar phenomenon has been reported in the recognition of prothrombin by the factor Xa/Va complex (36). A third mechanism for the loss of coagulation activity would be the removal of a domain required for a protein-phospholipid interaction. Fourth, the active site of fgl2 could have been deleted. At the present time, we do not have evidence to support any of these possibilities, and additional experiments will have to be performed.

Studies by others have demonstrated that many coagulation proteins, including factors VII, IX, X, XII, and thrombin, have serine protease activity (8, 31, 37). It has been suggested that there are two main types of serine proteases: the trypsin-like and subtilisin-like, which have an identical arrangement of the “classical” catalytic Ser/His/Asp triad but in quite different protein scaffolds. Over 20 families (S1–S27) of serine proteases have now been identified and have been grouped into six clans (clans SA, SB, SC, SE, SF, and SG) on the basis of structural similarity and other functional features (31, 37). In this study, we demonstrated that murine fgl2 protein has the characteristic of a serine protease based on the observation that the prothrombinase activity was inhibited by the serine protease inhibitor DFP. The concentration of DFP required for enzyme inhibition was well within the range that inhibits a number of known serine proteases that participate in the coagulation cascade (28). Moreover, the fact that murine fgl2 prothrombinase activity was not inhibited by antithrombin III, soybean trypsin inhibitor, aprotinin, or 4-aminobenzamidine suggests that murine fgl2 is a unique serine protease that behaves differently than other known coagulants.

Examination of the amino acid sequence of murine fgl2 protein shows the presence of serine protease clan SE motifs SXXK with serine residues at positions 89, 135, and 425. Proteases such as Escherichia D-Ala-D-Ala peptidase A, Streptomyces R61 D-Ala-D-Ala peptidase, and Actinomadura R39 peptidase have been categorized into the clan SE family of serine proteases (31). No homology at the amino acids level besides the SXXK motif is found between fgl2 protein and other members of the clan SE family. By site-directed mutagenesis, we have shown here that serine 89 is essential for murine fgl2 prothrombinase activity. Results obtained using the truncated murine fgl2 proteins are consistent with the site-directed mutagenesis results. Removal of residues 336–432 does not seem to be important for the expression of the clotting activity, consistent with the finding that mutation of serine 425 did not abolish the expression of coagulation activity. The mutation of serine 89 suggests the importance of the residue for the prothrombinase activity; however, as mentioned previously, the mechanism of the effect has not been determined. A similar case is observed for factor Xa when arginine 347 is mutated to asparagine, resulting in a selective reduction in the factor Va affinity of factor Xa (38).

Expression of murine fgl2 protein with a coagulation activity on activated macrophages and endothelial cells has been reported. However, Marazzi et al. (19) have recently reported that murine fgl2 protein is spontaneously secreted in vitro by freshly isolated human CD4+ and CD8+ T lymphocytes, but the secreted protein was devoid of coagulation activity. Thus, bioactive murine fgl2 protein can express either as a transmembrane form or as a soluble form under different circumstances, but coagulation activity of murine fgl2 requires phospholipid. The role of the soluble form of fgl2 protein lacking coagulation activity remains undetermined.

In this paper we have defined a unique coagulant, fgl2. From the hemostasis perspective, murine fgl2 bypasses both the intrinsic and extrinsic coagulation pathways to trigger prothrombin activation directly without the need of additional coagulation factors. However, similar to factor Xa, factor Va enhances the coagulation activity of murine fgl2. From the immunologic perspective, murine fgl2 results in direct generation of thrombin, which plays a significant role in inflammation by amplifying the inflammatory response. For instance, thrombin causes endothelial cells to produce IL-8, which induces endothelial P-selectin expression and early neutrophil adhesion and activation (39, 40). Also, thrombin is a potent chemotaxin for monocytes and polymorphonuclear cells, resulting in influx of massive infiltrates to the site of inflammation, which is the hallmark of diseases in which murine fgl2 has been implicated in the pathogenesis (1, 2, 16). The characterization of fgl2 and identification of serine 89 residue accounting for its prothrombinase activity will allow the development of novel strategies to prevent fulminant liver failure, cytokine-induced fetal loss, and other diseases in which fgl2 plays a pathogenic role.

1

This work was supported by a grant from the Canadian Institutes of Health Research.

3

Abbreviations used in this paper: MHV-3, murine hepatitis virus strain-3; Sf9, Spodoptera frugiperda ovarian cell 9; H5, High Five; CHO, Chinese hamster ovary; PS, phosphatidyl-l-serine; PC, phosphatidylcholine; DFP, diisopropylfluorophosphate; Ni-NTA, nickel-nitrilotriacetic acid.

1
Levy, G. A., M. Liu, J. Ding, S. Yuwaraj, J. Leibowitz, P. A. Marsden, Q. Ning, A. Kovalinka, M. J. Phillips.
2000
. Molecular and functional analysis of the human prothrombinase gene (HFGL2) and its role in viral hepatitis.
Am. J. Pathol.
156
:
1217
2
Li, C., L. S. Fung, S. Chung, A. Crow, N. Myers-Mason, M. J. Phillips, J. L. Leibowitz, E. Cole, C. A. Ottaway, G. Levy.
1992
. Monoclonal antiprothrombinase (3D4.3) prevents mortality from murine hepatitis virus (MHV-3) infection.
J. Exp. Med.
176
:
689
3
Sueishi, K., M. Kumamoto, H. Sakuda, K. Tanaka.
1993
. Angiogenic processes in the pathogenesis of human coronary atherosclerosis.
Curr. Top. Pathol.
87
:
47
4
Tanaka, K., K. Sueishi.
1993
. The coagulation and fibrinolysis systems and atherosclerosis.
Lab. Invest.
69
:
5
5
Merrill, J. T., R. G. Lahita.
1996
. The antiphospholipid syndrome and SLE: is there a clue in the link between complement and coagulation?.
Lupus
5
:
6
6
Gando, S., S. Nanzaki, S. Sasaki, K. Aoi, O. Kemmotsu.
1998
. Activation of the extrinsic coagulation pathway in patients with severe sepsis and septic shock.
Crit. Care Med.
26
:
2005
7
Bromberg, M. E., M. Cappello.
1999
. Cancer and blood coagulation: molecular aspects.
Cancer J. Sci. Am.
5
:
132
8
Dahlback, B..
2000
. Blood coagulation.
Lancet
355
:
1627
9
Burri, B. J., T. S. Edgington, D. S. Fair.
1987
. Molecular interactions of the intrinsic activation complex of coagulation: binding of native and activated human factors IX and X to defined phospholipid vesicles.
Biochim. Biophys. Acta
923
:
176
10
Marlar, R. A., A. J. Kleiss, J. H. Griffin.
1982
. An alternative extrinsic pathway of human blood coagulation.
Blood
160
:
1353
11
Parr, R. L., L. Fung, J. Reneker, N. Myers-Mason, J. L. Leibowitz, G. Levy.
1995
. Association of mouse fibrinogen-like protein with murine hepatitis virus-induced prothrombinase activity.
J. Virol.
69
:
5033
12
Ding, J. W., Q. Ning, M. F. Liu, A. Lai, J. Leibowitz, K. M. Peltekian, E. H. Cole, L. S. Fung, C. Holloway, P. A. Marsden, et al
1997
. Fulminant hepatic failure in murine hepatitis virus strain 3 infection: tissue-specific expression of a novel fgl2 prothrombinase.
J. Virol.
71
:
9223
13
Ding, J. W., Q. Ning, M. F. Liu, A. Lai, K. Peltekian, L. Fung, C. Holloway, H. Yeger, M. J. Phillips, G. A. Levy.
1998
. Expression of the fgl2 and its protein product (prothrombinase) in tissues during murine hepatitis virus strain-3 (MHV-3) infection.
Adv. Exp. Med. Biol.
440
:
609
14
Koyama, T., L. R. Hall, W. G. Haser, S. Tonegawa, H. Saito.
1987
. Structure of a cytotoxic T-lymphocyte-specific gene shows a strong homology to fibrinogen β and γ chains.
Proc. Natl. Acad. Sci. USA
84
:
1609
15
Fung, L. S., G. Neil, J. Leibowitz, E. H. Cole, S. Chung, A. Crow, G. A. Levy.
1991
. Monoclonal Ab analysis of a unique macrophage procoagulant activity induced by murine hepatitis virus strain 3 infection.
J. Biol. Chem.
266
:
1789
16
Clark, D. A., G. Chaouat, P. C. Arck, H. W. Mittruecker, G. A. Levy.
1998
. Cytokine-dependent abortion in CBA × DBA/2 mice is mediated by the procoagulant fgl2 prothrombinase.
J. Immunol.
160
:
545
17
Clark, D. A., J. W. Ding, G. Yu, G. A. Levy, R. M. Gorczynski.
2001
. Fgl2 prothrombinase expression in mouse trophoblast and decidua triggers abortion but may be countered by OX-2.
Mol. Hum. Reprod.
7
:
185
18
Ruegg, C., R. Pytela.
1995
. Sequence of a human transcript expressed in T-lymphocytes and encoding a fibrinogen-like protein.
Gene
160
:
257
19
Marazzi, S., S. Blum, R. Hartmann, D. Gundersen, M. Schreyer, S. Argraves, V. von Fliedner, R. Pytela, C. Ruegg.
1998
. Characterization of human fibroleukin, a fibrinogen-like protein secreted by T lymphocytes.
J. Immunol.
161
:
138
20
Bach, R., R. Gentry, Y. Nemerson.
1986
. Factor VII binding to tissue factor in reconstituted phospholipid vesicles: induction of cooperativity by phosphatidylserine.
Biochemistry
25
:
4007
21
Galvin, J. B., S. Kurosawa, K. Moore, C. T. Esmon, N. L. Esmon.
1987
. Reconstitution of rabbit thrombomodulin into phospholipid vesicles.
J. Biol. Chem.
262
:
2199
22
Mimms, L. T., G. Zampighi, Y. Nozaki, C. Tanford, J. A. Reynolds.
1981
. Phospholipid vesicle formation and transmembrane protein incorporation using octyl glucoside.
Biochemistry
20
:
833
23
Schwartz, B. S., G. A. Levy, D. S. Fair, T. S. Edgington.
1982
. Murine lymphoid procoagulant activity induced by bacterial lipopolysaccharide and immune complexes is a monocyte prothrombinase.
J. Exp. Med.
155
:
1464
24
Cornish-Bowden, A., R. Eisenthal.
1978
. Estimation of Michaelis constant and maximum velocity from the direct linear plot.
Biochim. Biophys. Acta
523
:
268
25
Fingerote, R. J., M. Abecassis, M. J. Phillips, Y. S. Rao, E. H. Cole, J. Leibowitz, G. A. Levy.
1996
. Loss of resistance to murine hepatitis virus strain 3 infection after treatment with corticosteroids is associated with induction of macrophage procoagulant activity.
J. Virol.
70
:
4275
26
Chung, S., G. Downey, O. Rotstein, G. Levy.
1993
. Early cellular events in the induction of murine hepatitis virus (MHV-3) induced macrophage procoagulant activity (PCA).
Adv. Exp. Med. Biol.
342
:
385
27
Galli, M., E. M. Bevers.
1994
. Inhibition of phospholipid-dependent coagulation reactions by “antiphospholipid Abs”: possible modes of action.
Lupus
3
:
223
28
Gilbert, G. E., A. A. Arena.
1996
. Activation of the factor VIIIa-factor IXa enzyme complex of blood coagulation by membranes containing phosphatidyl-L-serine.
J. Biol. Chem.
271
:
11120
29
Falls, L. A., B. Furie, B. C. Furie.
2000
. Role of phosphatidylethanolamine in assembly and function of the factor IXa-factor VIIIa complex on membrane surfaces.
Biochemistry
39
:
13216
30
Krishnaswamy, S., K. A. Field, T. S. Edgington, J. H. Morrissey, K. G. Mann.
1992
. Role of the membrane surface in the activation of human coagulation factor X.
J. Biol. Chem.
267
:
26110
31
Barrett, A. J., N. D. Rawlings.
1995
. Families and clans of serine peptidases.
Arch. Biochem. Biophys.
318
:
247
32
Husten, E. J., C. T. Esmon, A. E. Johnson.
1987
. The active site of blood coagulation factor Xa: its distance from the phospholipid surface and its conformational sensitivity to components of the prothrombinase complex.
J. Biol. Chem.
262
:
12953
33
Krishnaswamy, S., K. C. Jones, K. G. Mann.
1988
. Prothrombinase complex assembly: kinetic mechanism of enzyme assembly on phospholipid vesicles.
J. Biol. Chem.
263
:
3823
34
Rosenthal, G. A., G. A. Levy, O. D. Rotstein.
1989
. Induction of macrophage procoagulant activity by Bacteroides fragilis.
Infect. Immun.
57
:
338
35
Ruf, W., A. Rehemtulla, J. H. Morrissey, T. S. Edgington.
1991
. Phospholipid-independent and -dependent interactions required for tissue factor receptor and cofactor function.
J. Biol. Chem.
266
:
16256
36
Krishnaswamy, S., A. Betz.
1997
. Exosites determine macromolecular substrate recognition by prothrombinase.
Biochemistry
36
:
12080
37
Rawlings, N. D., A. J. Barrett.
1994
. Families of serine peptidases.
Methods Enzymol.
244
:
19
38
Rudolph, A. E., R. Porche-Sorbet, J. P. Miletich.
2000
. Substitution of asparagine for arginine 347 of recombinant factor Xa markedly reduces factor Va binding.
Biochemistry
39
:
2861
39
Kaplanski, G., M. Fabrigoule, V. Boulay, C. A. Dinarello, P. Bongrand, S. Kaplanski, C. Farnarier.
1997
. Thrombin induces endothelial type II activation in vitro: IL-1 and TNF-α-independent IL-8 secretion and E-selectin expression.
J. Immunol.
158
:
5435
40
Anrather, D., M. T. Milan, A. Palmetshofter, S. C. Robson, C. Geczy, A. J. Ritchie, F. H. Bach, B. M. Ewenstein.
1997
. Thrombin activates nuclear factor-κB and potentiates endothelial cell activation by TNF.
J. Immunol.
159
:
5620