Portal de Periódicos da CAPES

Characterization of Placental Bikunin, a Novel Human Serine Protease Inhibitor

1997; Elsevier BV; Volume: 272; Issue: 18 Linguagem: Inglês

10.1074/jbc.272.18.12209

1083-351X

Katherine Delaria, Daniel K. Muller, Christopher W. Marlor, James E. Brown, Rathindra C. Das, Steven Roczniak, Paul P. Tamburini,

Hemophilia Treatment and Research

We reported previously the cloning of a novel human serine protease inhibitor containing two Kunitz-like domains, designated as placental bikunin, and the subsequent purification of a natural counterpart from human placental tissue (Marlor, C. W., Delaria, K. A., Davis, G., Muller, D. K., Greve, J. M., and Tamburini, P. P. (1997) J. Biol. Chem. 272, 12202–12208). In this report, the 170 residue extracellular domain of placental bikunin (placental bikunin(1–170)) was expressed in baculovirus-infected Sf9 cells using its putative signal peptide. The resulting 21.3-kDa protein accumulated in the medium with the signal peptide removed and could be highly purified by sequential kallikrein-Sepharose and C18 reverse-phase chromatography. To provide insights as to the potential in vivo functions of this protein, we performed an extensive investigation of the inhibitory properties of recombinant placental bikunin(1–170) and both of its synthetically prepared Kunitz domains. All three proteins inhibited a number of serine proteases involved in the intrinsic pathway of blood coagulation and fibrinolysis. Placental bikunin(1–170) formed inhibitor-protease complexes with a 1:2 stoichiometry and strongly inhibited human plasmin (K i = 0.1 nm), human tissue kallikrein (K i = 0.1 nm), human plasma kallikrein (K i = 0.3 nm) and human factor XIa (K i = 6 nm). Conversely, this protein was a weaker inhibitor of factor VIIa-tissue factor (K i = 1.6 μm), factor IXa (K i = 206 nm), factor Xa (K i = 364 nm), and factor XIIa (K i = 430 nm). This specificity profile was to a large extent mimicked, albeit with reduced potency, by the individual Kunitz domains. As predicted from this in vitrospecificity profile, recombinant placental bikunin(1–170)prolonged the clotting time in an activated partial thromboplastin time assay. We reported previously the cloning of a novel human serine protease inhibitor containing two Kunitz-like domains, designated as placental bikunin, and the subsequent purification of a natural counterpart from human placental tissue (Marlor, C. W., Delaria, K. A., Davis, G., Muller, D. K., Greve, J. M., and Tamburini, P. P. (1997) J. Biol. Chem. 272, 12202–12208). In this report, the 170 residue extracellular domain of placental bikunin (placental bikunin(1–170)) was expressed in baculovirus-infected Sf9 cells using its putative signal peptide. The resulting 21.3-kDa protein accumulated in the medium with the signal peptide removed and could be highly purified by sequential kallikrein-Sepharose and C18 reverse-phase chromatography. To provide insights as to the potential in vivo functions of this protein, we performed an extensive investigation of the inhibitory properties of recombinant placental bikunin(1–170) and both of its synthetically prepared Kunitz domains. All three proteins inhibited a number of serine proteases involved in the intrinsic pathway of blood coagulation and fibrinolysis. Placental bikunin(1–170) formed inhibitor-protease complexes with a 1:2 stoichiometry and strongly inhibited human plasmin (K i = 0.1 nm), human tissue kallikrein (K i = 0.1 nm), human plasma kallikrein (K i = 0.3 nm) and human factor XIa (K i = 6 nm). Conversely, this protein was a weaker inhibitor of factor VIIa-tissue factor (K i = 1.6 μm), factor IXa (K i = 206 nm), factor Xa (K i = 364 nm), and factor XIIa (K i = 430 nm). This specificity profile was to a large extent mimicked, albeit with reduced potency, by the individual Kunitz domains. As predicted from this in vitrospecificity profile, recombinant placental bikunin(1–170)prolonged the clotting time in an activated partial thromboplastin time assay. Blood clotting, resulting either from the extrinsic pathway following tissue injury or the intrinsic pathway following contact activation, involves tightly regulated proteolytic cascades (1Mann K.G. Lorand L. Methods Enzymol. 1993; 222: 1-10Google Scholar). The intrinsic pathway is initiated by activation of factor XII either through proteolysis or contact with negatively charged surfaces. Activated factor XIIa, in turn, converts plasma prekallikrein to kallikrein, which can then activate additional factor XII. Factor XIIa activates factor XI, which, in turn, activates factor IX. Activated factor IX forms a complex with factor VIIIa, phospholipid, and calcium, which converts factor X to factor Xa. Factor X is also activated by the factor VIIa-tissue factor complex operating within the extrinsic pathway. Thrombin generation by factor Xa in complex with factor Va leads ultimately to the formation of the fibrin clot. Thrombus formation is also regulated by the fibrinolytic system whereby plasmin, formed from plasminogen by the action of kallikrein, tissue plasminogen activator (tPA), 1The abbreviations used are: tPA, tissue plasminogen activator; TFPI, tissue factor pathway inhibitor; Suc, succinyl; AMC, 7-amido-4-methylcoumarin; Boc,t-butoxycarbonyl; Bz, benzoyl; pNa,p-nitroanilide; Tricine,N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; Fmoc,N-(9-fluorenyl)methoxycarbonyl; HPLC, high performance liquid chromatography; APTT, activated partial thromboplastin time. or urokinase, breaks down both fibrinogen and fibrin (2Marder V.J. Frances C.W. Doolittle R.F. Colman R.W. Hirsh J. Marder V.J. Salzman E.W. Hemostasis and Thrombosis: Basic Principles and Clinical Practice. Lippincott, Philadelphia1982: 145-163Google Scholar). Protease inhibitors play critical roles in the regulation of the coagulation and fibrinolytic systems. Tissue factor pathway inhibitor (TFPI), a multivalent Kunitz-type inhibitor, is a major regulator of the extrinsic pathway through factor Xa-dependent inhibition of factor VIIa-tissue factor (3Davie E.W. Fujikawa K. Kisiel W. Biochemistry. 1991; 30: 10363-10370Google Scholar). C1 inhibitor is thought to be the major physiological inhibitor of the intrinsic pathway enzymes plasma kallikrein and factor XII (4Travis J. Salvesen G.S. Annu. Rev. Biochem. 1983; 52: 655-709Google Scholar). More recently, TFPI-2, a serine protease inhibitor structurally related to TFPI, has been proposed to be an important physiological inhibitor of enzymes involved in both coagulation and fibrinolysis (5Petersen L.C. Sprecher C.A. Foster D.C. Blumberg H. Hamamoto T. Kisiel W. Biochemistry. 1996; 35: 266-272Google Scholar). In a previous study, we identified and cloned a human cDNA encoding a novel protein containing two Kunitz-like domains which we termed placental bikunin (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar). In this study a soluble recombinant fragment of this protein, placental bikunin(1–170), 2Residues of the placental bikunin sequence and fragments thereof are numbered as defined in Ref. 6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar according to their position within the open reading frame for full-length mature placental bikunin. In this scheme, placental bikunin(1–170) refers to a protein containing the NH2-terminal 170 amino acids of the mature protein produced by removal of the signal peptide. The NH2-terminal(7–64) domain and COOH-terminal(102–159) domain refer to individual functional Kunitz domains normally present within residues 7–64 and 102–159, respectively, of mature placental bikunin. was expressed and found to be a potent serine protease inhibitor. Using the recombinant protein and both of its synthetic NH2- and COOH-terminal Kunitz domains, we dissect the specificity of the protein more comprehensively. In doing so, we demonstrate that placental bikunin is a potent inhibitor of plasma and tissue kallikreins, plasmin and factor XIa. Bovine chymotrypsin, bovine trypsin, and tPA (single chain form from human melanoma cell culture), urokinase, Suc-A-A-P-F-AMC, Boc-L-G-R-AMC, Suc-A-A-A-AMC, Boc-Q-G-R-AMC, Bz-P-F-R-pNa, and P-F-R-AMC were from Sigma. Neutrophil elastase was from Athens Research and Technology, Inc. (Athens, GA). Human plasmin, factor VIIa, tissue factor (lipidated), and (CH3SO2-D-cyclohexylpyrosyl)-G-R-pNa (Spectrozyme tPA substrate) were from American Diagnostica, Inc. (Greenwich, CT). Human plasma kallikrein and human factors IXa, X, Xa, XIa, and XIIa were from Enzyme Research Laboratories (South Lafayette, IN). Tos-G-P-K-AMC, Suc-A-A-P-R-pNa, and Boc-E(OBzl)-A-R-AMC, were from Bachem Bioscience (King of Prussia, PA). H-D-I-P-R-pNa was from Chromogenix AB. Bovine pancreatic kallikrein, human tissue kallikrein, and recombinant aprotinin were kindly provided by Bayer AG (Wuppertal, Germany). A cDNA fragment encoding an NH2-terminal 170-amino acid fragment of placental bikunin was expressed in Sf9 cells (7Luckow V.A. Summers M.D. Virology. 1989; 170: 31-39Google Scholar) as follows. Placental bikunin cDNA obtained by polymerase chain reaction and contained within the pCRII vector as described previously (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar) was liberated by digestion with HindIII and XbaI. This fragment was gel purified and then cloned into the M13mp19 vector (New England Biolabs, Beverly, MA).In vitro mutagenesis (8Kunkel T.A. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 488-492Google Scholar) was used to generate aPstI site between the upstream XbaI site and the sequence encoding the ATG start site at the 5′ end of the cDNA insert. The oligonucleotide used for the mutagenesis had the sequence CGC GTC TCG GCT GAC CTG GCC CTG CAG ATG GCG CAC GTG TGC GGG. A stop codon (TAG) and BglII/XmaI site was similarly engineered at the 3′ end of the cDNA using the oligonucleotide CTG CCC CTT GGC TCA AAG TAG GAA GAT CTT CCC CCC GGG GGG GTG GTT CTG GCG GGG CTG. The stop codon was in-frame with the sequence encoding placental bikunin and caused termination immediately following the lysine at amino acid residue 170 of the mature protein sequence. The mutated cDNA thus encoded a truncated placental bikunin protein containing both Kunitz domains but which was devoid of the putative transmembrane domain. The product from digestion with PstI andBglII was isolated and cloned into the BacPac8 vector (CLONTECH) for expression of placental bikunin(1–170). The expression of placental bikunin(1–170) by Sf9 insect cells was optimal at a multiplicity of infection of 1 when the medium was harvested at 72 h post-infection. Sf9 cell culture supernatant (2–3 liters) was harvested by centrifugation (1,500 × g for 30 min), adjusted to pH 8.0 by the addition of 1 m Tris-HCl (pH 8.0) to a final concentration of 50 mm, then applied at 2.0-ml min−1 to a 5-ml column comprised of bovine pancreatic kallikrein (70 mg) which had been immobilized onto CNBr-activated Sepharose 4B (Pharmacia Biotech Inc.) according to the manufacturer's instructions. After loading, the column was washed with 0.1m Tris-HCl (pH 8.0) containing 0.1 m NaCl until the A 280 nm of the wash could no longer be detected. The column was washed further with 0.1 m Tris-HCl (pH 8.0) containing 0.5 m NaCl and then eluted with 0.2m acetic acid (pH 2.0). Fractions (2.0 ml) were collected in tubes containing 0.25 ml of 1 m Tris-HCl (pH 7.5) and assayed for the ability to inhibit bovine trypsin as described (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar). Active fractions were pooled, adjusted to pH 2.5 with trifluoroacetic acid and subjected to chromatography on a C18 reverse-phase column (1.0 × 25 cm) equilibrated with 0.1% trifluoroacetic acid containing 22.5% acetonitrile at a flow rate of 1 ml/min. The placental bikunin(1–170) was eluted with a linear gradient of 22.5–50% acetonitrile in 0.1% trifluoroacetic acid over 50 min. Fractions containing trypsin inhibitory activity were pooled, lyophilized, redissolved in 5 mm sodium acetate (pH 5.0), 0.1 m NaCl, and stored at −20 °C until needed. The highly purified protein exhibited an ε280 nm of 4.52 × 104 liters m−1cm−1 based upon composition analysis. Peptides corresponding to amino acids 7–64 and 102–159 of mature placental bikunin (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar) were synthesized on an Applied Biosystems model 433A peptide synthesizer using NMP-HBTU Fmoc chemistry. Synthesis was on a preloaded Gln resin (Nova Biochem, La Jolla, CA) with an 8-fold excess of Fmoc-amino acid/coupling. The side chain protecting groups used were as follows: for cysteine, histidine, tryptophan, and asparagine, trityl; for glutamic acid, serine, tyrosine, aspartic acid, and threonine, t-butyl; for lysine,t-Boc; and for arginine, 2,2,5,7,8-pentamethylchroman-6-sulfonyl. In addition, glycine was protected withN-α-Fmoc-N-α-(2-Fmoc-oxy-4-methoxybenzyl) when it occurred after aspartic acid. Cleavage and deprotection were performed in 84.5% trifluoroacetic acid, 4.4% thioanisole, 2.2% ethanedithiol, 4.4% liquefied phenol, and 4.4% H2O for 2 h at room temperature. The crude peptide was precipitated witht-butyl methyl ether, centrifuged at 3,000 rpm in a Sorval RT 6000D model table top centrifuge for 5 min, and washed twice int-butyl methyl ether. Reduced peptides were purified by HPLC (10 ml/min) on a Dynamax 60A C18 column (2.1 × 30 cm) using a 30-min linear gradient of either 23–34% acetonitrile in 0.1% trifluoroacetic acid (placental bikunin(7–64)) or a 30-min linear gradient of 20–30% acetonitrile in 0.1% trifluoroacetic acid (placental bikunin(102–159)). The purified synthetic peptides corresponding to placental bikunin(7–64) and placental bikunin(102–159) were refolded to yield the respective functional NH2-terminal(7–64) and COOH-terminal(102–159) Kunitz domains using an adaptation of the method of Tam et al. (9Tam J.P. Wu C.-R. Liu W. Zhang J.-W. J. Am. Chem. Soc. 1991; 113: 6657-6662Google Scholar) as follows. A solution containing 23% (v/v) dimethyl sulfoxide in 0.1 m Tris-HCl (pH 6.0) was added dropwise to a solution of reduced peptide (2.1 and 2.9 mg/ml for the NH2-terminal(7–64) and COOH-terminal(102–159) domain, respectively) in 0.1m Tris-HCl (pH 6.0), containing 8 m urea to obtain a final concentration of 0.3 mg/ml peptide in 20% dimethyl sulfoxide, 0.1 m Tris-HCl (pH 6.0), and 1 murea. The solutions containing either domain were each stirred at 25 °C for 24 h and then diluted 1:10 with 50 mmTris-HCl (pH 8.0) containing 0.1 m NaCl. Refolded material was isolated by affinity chromatography over immobilized bovine pancreatic kallikrein under conditions described above for the purification of recombinant placental bikunin(1–170). Fractions containing trypsin inhibitory activity were pooled, adjusted to pH 2.5 with trifluoroacetic acid, then applied directly to a Vydac C18 reverse-phase column (5 μm, 0.46 × 25 cm). The NH2-terminal(7–64) Kunitz domain was purified using a 40-min linear gradient of 22.5–50% acetonitrile in 0.1% trifluoroacetic acid at a flow rate of 1.0 ml/min. The COOH-terminal(102–159) Kunitz domain was purified using a 40-min linear gradient of 20–40% acetonitrile in 0.1% trifluoroacetic acid. Active fractions were pooled, lyophilized, dissolved in 0.1% trifluoroacetic acid, and stored at −20 °C until needed. To determine the stoichiometry with trypsin, the protease (2.2 nm) was incubated with recombinant placental bikunin (0–2 nm) or aprotinin (0–4 nm) in 1.0 ml of 50 mm Hepes (pH 7.5), 0.1 m NaCl, 2.0 mm CaCl2, 0.01% Triton X-100 (buffer A) for 30 min at room temperature, followed by the addition of Suc-A-A-P-R-pNa (90 μm, final concentration), after which residual enzymatic formation of pNa was monitored at 410 nm. To determine the stoichiometry with plasma kallikrein, 40 nm protease was incubated with recombinant placental bikunin (0–40 nm) in 1.0 ml of 50 mmTris-HCl (pH 8.0) containing 0.1 m NaCl and 0.01% Triton X-100 (buffer B). After 30 min at room temperature, Bz-P-F-R-pNa was added (300 μm, final concentration) and residual enzymatic activity monitored at 410 nm. Fractional protease activity was calculated asV (+I)/V (−I) and stoichiometries obtained from plots ofV (+I)/V (−I) versus ratio of inhibitor to enzyme concentrations. Apparent equilibrium dissociation constants (K i*), were determined as described previously (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar) using methods for tight binding inhibitors (10Boudier C. Bieth J.G. Biochim. Biophys. Acta. 1989; 995: 36-41Google Scholar) assuming enzyme:inhibitor stoichiometries of 1:1 and 2:1 for binding to single Kunitz domains and placental bikunin(1–170), (1Mann K.G. Lorand L. Methods Enzymol. 1993; 222: 1-10Google Scholar) respectively. Active site concentrations of trypsin, plasma and tissue kallikreins, and plasmin were determined by titration withp-nitrophenyl-p′-guanidinobenzoate as described (11Chase T. Shaw E. Methods Enzymol. 1970; 19: 20-27Google Scholar). The concentration of bovine chymotrypsin was determined by titration with N-trans-cinnamoylimidazole as described (12Schonbaum G.R. Zerner B. Bender M.L. J. Biol. Chem. 1961; 236: 2930-2935Google Scholar). The amount of the NH2-terminal(7–64) and COOH-terminal (102–159) Kunitz domain, and aprotinin were determined by titration with active site-titrated trypsin and by amino acid analysis. The concentration of placental bikunin(1–170) was quantified by amino acid analysis. The concentrations of factors VIIa, IXa, X, Xa, XIa, and XIIa, tissue factor, elastase, urokinase, and tPA were based on the manufacturer's specifications. Following preincubation of protease with inhibitor for 5 min at 37 °C in 990 μl of the appropriate buffer (see below), reactions were initiated with substrate to achieve the following initial component concentrations: bovine pancreatic kallikrein (buffer B) with [E 0] = 92 pm and 100 μm P-F-R-AMC (K m = 82 ± 8.8 μm); human tissue kallikrein, 50 mm Tris-HCl (pH 9.0), 50 mm NaCl, and 0.01% Triton X-100 with [E 0] = 0.35 nm and 10 μm P-F-R-AMC (K m = 5.7 ± 0.4 μm); human neutrophil elastase, 0.1 mTris-HCl (pH 8.0), and 0.05% Triton X-100 with [E 0] = 19 nm and 0.6 mm Suc-A-A-A-AMC (K m = 1,300 μm); human factor Xa, 20 mm Tris-HCl (pH 8.0), 0.1 m NaCl, 0.1% bovine serum albumin with [E 0] = 0.87 μm and 0.6 mm Boc-L-G-R-AMC (K m = 750 ± 98 μm); and human factor XIa (buffer A with 1 mg/ml bovine serum albumin) with [E 0] = 0.1 nmand 400 μm Boc-E(OBzl)-A-R-AMC (K m = 450 ± 50 μm). The following assays were performed following preincubation of protease with inhibitor for 30 min at 37 °C: trypsin (buffer A) with [E 0] = 50 pm and 30 μm Tos-G-P-K-AMC (K m = 22 ± 2 μm); chymotrypsin (buffer A) with [E 0] = 50 pm and 80 μm Suc-A-A-P-F-AMC (K m = 70 ± 12 μm); human plasmin, 50 mm Tris-HCl (pH 7.5), 0.1 m NaCl, and 0.02% Triton X-100 with [E 0] = 50 pm and 500 μm Tos-G-P-K-AMC (K m = 726 ± 70 μm); human plasma kallikrein (buffer B) with [E 0] = 0.2 nm and 100 μm P-F-R-AMC (K m = 457 ± 28 μm); and factor XIIa (buffer B) with [E 0] = 4.0 nm, and 200 μm Boc-Q-G-R-AMC (K m = 200 ± 37 μm). Factor VIIa-tissue factor inhibition was measured by incubating enzyme (2.8 nm factor VIIa, 2.8 nmtissue factor) with inhibitor in 100 μl of buffer A for 30 min at 25 °C. Reactions were initiated with 1 mm final (CH3SO2-D-cyclohexylpyrosyl)-G-R-pNa, and the A 405 nm was monitored. Factor IXa inhibition was measured as described (5Petersen L.C. Sprecher C.A. Foster D.C. Blumberg H. Hamamoto T. Kisiel W. Biochemistry. 1996; 35: 266-272Google Scholar). Briefly, factor IXa (10 nm) was preincubated for 15 min at 37 °C with inhibitor in 50 mm Tris-HCl (pH 7.5) containing 0.1 mNaCl, 250 μm CaCl2, 60 nmpoly-d-lysine (M r = 209,200) and 0.1% polyethylene glycol 8000. Factor X (200 nm) was then added and the incubation continued for 30 min at 37 °C. Substrate (600 μm Boc-L-G-R-AMC) was then added and factor Xa activity measured. To determine the inhibition of tPA, tPA was preincubated with inhibitor for 2 h at room temperature in 20 mm Tris-HCl (pH 7.2) containing 150 mm NaCl and 0.02% sodium azide. Reactions were initiated with substrate to achieve initial component concentrations of: tPA (16.7 nm), inhibitor (0–6.6 μm), I-P-R-pNa (1 mm) in 28 mm Tris-HCl (pH 8.5) containing 0.004% (v/v) Triton X-100 and 0.005% (v/v) sodium azide. The amount of pNa formed after incubation for 2 h at 37 °C was determined from the increase in the A 405 nm. Hydrolysis of AMC-conjugated peptides was monitored on a Perkin-Elmer model LS50B fluorometer (excitation = 370 nm, emission = 432 nm) over the first 2 min of the reaction, while hydrolysis ofpNa conjugates was monitored at 405 nm on a Hewlett-Packard model HP8452 spectrophotometer. K i* values were determined from plots of fractional rate versus inhibitor concentration, which were fit by nonlinear regression analysis (Enzfitter by Biosoft, Cambridge, U. K.) using the following equation Vi/V0=1−([E]0+[I]0+Ki*)−[([E]0+[I]0Equation 1 +Ki*)2−4[E]0[I]0]1/2/2[E]0where V i and V 0 are the enzyme activities in the presence or absence of a total inhibitor concentration of [I]0, and [E]0is the total concentration of enzyme. K i values were obtained by correction for the effect of substrate using the following equation: K i = K i*/(1 + [S]0/K m). K i values for inhibition by placental bikunin(1–170) were determined assuming an inhibitor:protease stoichiometry of 1:2 and that the two Kunitz inhibitory domains are equivalent and act independently. The effect of recombinant placental bikunin(1–170) and aprotinin on the APTT was determined as follows. Inhibitor in 20 mm Tris-HCl (pH 7.2) containing 150 mm NaCl and 0.02% sodium azide was added (0.1 ml) to a cuvette within a MLA ElectraR 800 Automatic Coagulation Timer coagulometer (Medical Laboratory Automation, Inc., Pleasantville, NY). The instrument was set to APTT mode with a 300-s activation time, and samples were run in duplicate. Following the addition of 0.1 ml of plasma (Specialty Assayed Reference Plasma 1-6-5185, Helena Laboratories, Beaumont, TX), the APTT reagent (Automated APTT-lot 102345, Organon Teknika Corp., Durham, NC) and CaCl2 (final concentration of 25 mm) were automatically added to initiate clotting, which was monitored automatically. CNBr digestion was performed as follows. Purified placental bikunin(1–170) in water (165 pmol) was taken to dryness on a Speed-Vac concentrator (Savant Instruments, Farmingdale, NY), resuspended in 0.05 ml of 250 mm Tris-HCl containing 6 m guanidine HCl, 0.5% (v/v) 2-mercaptoethanol, and 1 mm EDTA, and incubated for 2 h in the dark at room temperature before the addition of 4% (v/v) final 4-vinylpyridine. Following an additional 2-h incubation, the sample was applied to a biphasic reaction cartridge containing C18 (Hewlett-Packard), eluted in methanol, then taken to dryness by Speed-Vac concentration. The sample was incubated with 0.05 ml of 70% (v/v) formic acid saturated with CNBr for 12 h, taken to dryness, reconstituted in 0.1 ml of 0.1% (v/v) trifluoroacetic acid, then subjected to NH2-terminal sequencing. NH2-terminal sequence analysis was performed essentially as described (13Miller C.G. Methods: Companion Methods Enzymol. 1994; 6: 315-333Google Scholar), but with minor modification (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar). Amino acid analysis was performed as described previously (14Tamburini P.P. Dreyer R.N. Hansen J. Letsinger J. Elting J. Gore-Willse A. Dally R. Hanko R. Osterman D. Kamarck M.E. Yoo-Warren H. Anal. Biochem. 1990; 186: 363-368Google Scholar). Reducing SDS-polyacrylamide gel electrophoresis was performed with 10–20% Tricine-buffered polyacrylamide gels (Novex, San Diego, CA) according to the manufacturer's instructions, and protein was visualized with Coomassie Brilliant Blue R-250 using standard protocols. Electrospray ionization mass spectrometry was performed on a Finnigan TSQ7000 mass spectrophotometer capable of unit resolution, with the following operational parameters: scan rate, 200–2,000 over 3 s; spray voltage, 5 kV; capillary temperature, 220 °C; nitrogen pressure, 50 p.s.i.; auxillary nitrogen, 30 ml/min. Samples were loaded on a PLRP cartridge (Microm Bioresources, Auburn, CA), washed with water, and eluted at 0.2 ml/min with 90% (v/v) acetonitrile, 2% (v/v) acetic acid. To explore the protease inhibitory capacity of placental bikunin, a cDNA encoding the entire extracellular domain of 170 amino acids of placental bikunin plus its natural signal peptide (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar) was prepared by site-directed mutagenesis and expressed in the baculovirus/Sf9 system. The protein was purified from the cell culture supernatant by sequential kallikrein-Sepharose affinity chromatography and C18 reverse-phase HPLC (Table I). An overall enrichment of 240-fold was achieved based on the increase in specific activity; however, the actual fold enrichment of the protein was likely to be much higher as only a fraction of the inhibitory activity of the culture supernatant was due to placental bikunin(1–170). This latter conclusion is based on the fact that the majority (>99%) of the trypsin inhibitory activity present in the starting supernatant did not bind to the kallikrein-Sepharose column, whereas all the placental bikunin(1–170) bound the column based on immunoblot analysis of the column starting material and flow-through (not shown). Fractionation of 2 liters of medium typically yielded between 0.7 and 1 mg of placental bikunin(1–170) based on amino acid analysis.Table IPurification of recombinant placental bikunin(1–170) from Sf9 cellsPurification stepVolumeA 280totalUnitsaA unit of activity is the amount of inhibitor needed to inhibit trypsin activity by 50% with the conditions specified under "Experimental Procedures."Specific activitymlunits/ASupernatant2,300.020,7006,150,000297Kallikrein affinity23.02.7640,70014,746C180.41.5411,11172,150a A unit of activity is the amount of inhibitor needed to inhibit trypsin activity by 50% with the conditions specified under "Experimental Procedures." Open table in a new tab The final preparation was highly pure as judged by SDS-polyacrylamide gel electrophoresis (Fig. 1) and exhibited a molecular mass of 21.3 kDa, consistent with the expected size of placental bikunin(1–170). NH2-terminal sequence analysis of the protein (26 cycles) yielded the same NH2 terminus for as obtained for the natural protein (6Marlor C.W. Delaria K.A. Davis G. Muller D.K. Greve J.M. Tamburini P.P. J. Biol. Chem. 1997; 272: 12202-12208Google Scholar). The sequence started at residue +1 and continued with the sequence ADRER-. Furthermore, the amino acid composition was > 95% accurate relative to the theoretical composition of placental bikunin(1–170). Purified placental bikunin(1–170) (100 pmol) was then pyridylethyl alkylated, CNBr digested, and then sequenced without purification of the resulting peptide fragments. Sequencing for 20 cycles yielded the sequence for each of the NH2 termini expected from the digestion of placental bikunin(1–170)(Table II).Table IINH2-terminal sequences of fragments resulting from the digestion of purified placental bikunin(1–170) with CNBrSequenceaLowercase letters denote tentative amino acid assignments.AmountPlacental bikunin residue no.pmolLRCFrQQENPP–PLG21154 –168ADRERSIHDFCLVSKVVGRC201 –20FNYeEYCTANAVTGPCRASF16100 –119Pr–Y–V–dGS–Q–F–Y–G625 –43Digestion was performed according to "Experimental Procedures."a Lowercase letters denote tentative amino acid assignments. Open table in a new tab Digestion was performed according to "Experimental Procedures." Titration of fixed concentrations of either bovine trypsin or human plasma kallikrein with purified placental bikunin(1–170)showed that each molecule of inhibitor can interact with two molecules of protease simultaneously (Fig. 2). On the other hand, titration of trypsin with aprotinin yielded a stoichiometry of 1:1 in the enzyme inhibitor complex (Fig. 2). To attempt to dissect the specificities of the individual domains within this bikunin, synthetic peptides were made corresponding to the NH2-terminal(7–64) and COOH-terminal(102–159 Kunitz domains of placental bikunin. Purified, reduced NH2-terminal(7–64)([M-H]+ = 6,567.5), and COOH-terminal(102–159) domain ([M-H]+ = 6,835.6) were prepared, each of which had the expected amino acid composition. Refolding of these peptides using dimethyl sulfoxide as the oxidizing agent, followed by purification on a C18column, yielded purified refolded NH2-terminal(7–64) (2% of the starting reduced peptide) and COOH-terminal(102–150) domains (1.4% of the starting reduced peptide) which exhibited [M-H]+values of 6,561.2 and 6,829.3, respectively. The mass reduction upon refolding of each peptide (6 ± 1 mass units) suggests that refolding had in each case resulted in the formation of three intrachain disulfide bonds from the six cysteines present within each of the reduced peptides. A panel of serine proteases was used to determine the relative specificity and potency of recombinant placental bikunin(1–170) and its individual Kunitz domains (Table III). Aprotinin, a Kunitz inhibitor of defined specificity, was included in these studies for comparison. Recombinant placental bikunin(1–170) was a potent inhibitor of several trypsin-like proteases and also of chymotrypsin. Of potential relevance to its putative physiological role, this protein was a potent inhibitor of human plasma and tissue kallikreins, human plasmin and factor XIa. The potency of recombinant placental bikunin(1–170)against plasmin, plasma kallikrein, and factor XIa were 2-, 45-, and 47-