Three-dimensional structure of human cytomegalovirus protease.

Herpesviruses encode a serine protease that specifically cleaves assembly protein. This protease is critical for replication, and represents a new target for antiviral drug design. Here we report the three-dimensional structure of the protease from human cytomegalovirus (hCMV) at 2.27 angstroms resolution. The structure reveals a unique fold and new catalytic strategy for cleavage. The monomer fold of the enzyme, a seven-stranded beta-barrel encircled by a chain of helices that form the carboxy terminus of the molecule, is unrelated to those observed in classic serine proteases such as chymotrypsin and subtilisin. The serine nucleophile at position 132 is activated by two juxtaposed histidine residues at positions 63 and 157. Dimerization, which seems to be necessary for activity, is observed in the crystals. Correlations of the structure with the sequences of herpesvirus proteases suggest that dimerization may confer specificity and recognition in substrate binding.

Recombinant E. coli-derived tissue factor pathway inhibitor reduces coagulopathic and lethal effects in the baboon gram-negative model of septic shock.

Excessive coagulation is a typical response to the vascular injury occurring in gram negative sepsis. This study evaluated the pharmacological effects of the use of a recombinant Escherichia coli derived form of tissue factor pathway inhibitor (ala-TFPI) in a baboon model of septic shock. Several doses of ala-TFPI were administered either 30 or 120 min after the initiation of a lethal intravenous infusion of E. coli into baboons. Treatment at 30 min with either 2.7 or 7.4 mg/kg of ala-TFPI resulted in the same survival rates and attenuation of both the coagulation response and cellular injury, as measured by clinical chemistry. When administration of ala-TFPI was delayed for 120 min, a dose of ala-TFPI protein continued to provide a benefit to survival. Ala-TFPI reduced the drop in mean systemic arterial pressure compared to control baboons in addition to partially attenuating the coagulopathic response. Baboons given ala-TFPI also maintained lower levels of plasma interleukin-6 (IL-6) and thrombin-antithrombin. These results suggest that the site of action of the protein may involve the later stage components of the coagulation and inflammatory pathways.

Refold and characterization of recombinant tissue factor pathway inhibitor expressed in Escherichia coli.

Human tissue factor pathway inhibitor (TFPI) was expressed in E. coli as a non-glycosylated protein with an additional alanine attached to the aminoterminus of the wild type molecule. High-level expression was obtained with pMON6875, a plasmid containing a tac promoter, Gene 10 leader from bacteriophage T7, methionine-alanine-TFPI coding sequence, and the p22 transcriptional terminator. In this system, TFPI accounted for about 5-10% of the total cell protein. The inclusion bodies containing. TFPI were sulfitolyzed, purified by anion-exchange chromatography, refolded through a disulfide interchange reaction, and further fractionated by Mono S cation exchange chromatography. The Mono S resin resolved a peak of highly active TFPI from relatively inactive and possibly misfolded molecules. The E. coli TFPI was shown to be about two-fold more active, on a molar basis, than full-length human SK hepatoma TFPI in a tissue factor-induced clotting assay in human plasma.

Clearance of recombinant tissue factor pathway inhibitor (TFPI) in rabbits.

The pharmacokinetics of recombinant tissue factor pathway inhibitor (TFPI) after an intravenous bolus injection was studied in rabbits. Clearance of TFPI was followed by measurement of the radioactivity of the 125I-labelled compound in the whole plasma or the trichloroacetic acid precipitate and by quantitation of the functional TFPI activity of the unlabelled compound using a tissue factor-induced coagulation assay. When iodinated TFPI was used, the ratios of the trichloroacetic acid precipitable counts vs. that of the whole plasma was about 1 in the first 10 min after TFPI injection, but this ratio gradually decreased to less than 0.5 after 2 h. This result suggested that the iodinated TFPI in the plasma was partially degraded after prolonged circulation in the animal. When unlabelled TFPI was used, the clearance of TFPI activity from the plasma exhibited bi-exponential elimination kinetics with a rapid alpha phase half-life (t1/2 alpha) of 2.3 min, and a terminal beta phase half-life (t1/2 beta) of 79 min. The plasma clearance was 4.2 ml kg-1 min-1. The tissue distribution of intravenously administered 125I-TFPI in the rabbit was studied using whole-body autoradiography. At 3 min after dosing, significant levels of TFPI were apparent in the liver, kidney, and other highly blood perfused tissues. Significant levels of 125I-TFPI-derived radioactivity were also apparent in the liver and kidney at 30 min after intravenous administration. The localization within the liver demonstrated a mottled appearance, suggesting regions of higher uptake within the liver. In the kidney, the outer cortex consistently revealed the highest activity.

Comparison of recombinant tissue factor pathway inhibitors expressed in human SK hepatoma, mouse C127, baby hamster kidney, and Chinese hamster ovary cells.

Recombinant tissue factor pathway inhibitor (rTFPI) has been expressed in four mammalian expression systems using human SK hepatoma, mouse C127, baby hamster kidney (BHK), and Chinese hamster ovary (CHO) cells as hosts. On sodium dodecyl sulfate polyacrylamide gel electrophoresis, the immunoaffinity purified rTFPIs all show broad bands and the mean molecular weight of SK hepatoma and C127 rTFPIs (M(r) approximately 38,000) appear larger than those of BHK and CHO rTFPIs (M(r) approximately 35,000). All these proteins inhibit factor Xa and appear to bind factor Xa with 1:1 stoichiometry. The ability of these proteins to inhibit tissue factor-induced coagulation in plasma was examined using a prothrombin time assay. The relative activities of SK rTFPI:C127 rTFPI:BHK rTFPI:CHO rTFPI were found to be 28:15:2.1:1. By Western blot using specific antisera against the amino- and carboxy-termini of TFPI as probes, it is found that all the immunoaffinity purified rTFPIs possess approximately equal amounts of the amino terminus, but the C127 and BHK rTFPIs are deficient in carboxy terminus and the CHO rTFPI is essentially devoid of this region of the protein. Mono S chromatography allowed separation of the full-length and the truncated molecules with high and low anticoagulant activities, respectively. The above results suggest that proteolysis of the carboxy terminus of TFPI occurs to different extent when TFPI is expressed in different cells and that the carboxy terminal region of the TFPI molecule is important for the inhibition of tissue factor-induced coagulation.(ABSTRACT TRUNCATED AT 250 WORDS)

Prevention of arterial reocclusion after thrombolysis with recombinant lipoprotein-associated coagulation inhibitor.

BACKGROUND: This study was designed to determine whether arterial reocclusion after thrombolysis can be prevented by lipoprotein-associated coagulation inhibitor (LACI), a physiological inhibitor of tissue factor-induced coagulation mediated by the extrinsic pathway. METHODS AND RESULTS: Thrombosis was induced in femoral arteries of anesthetized dogs with the use of anodal current to elicit extensive vascular injury and formation of platelet-rich thrombi in one artery and with thrombogenic copper wire to elicit fibrin-rich thrombi without appreciable vascular injury in the contralateral artery. Recanalization of both vessels was induced with t-PA (1.7 mg/kg i.v. over 1 hour) and verified with Doppler flow probes. Reocclusion occurred within 2 hours in seven of seven arteries with electrical injury-induced thrombosis and in four of seven arteries with copper wire-induced thrombosis in the absence of LACI. In dogs given infusions of recombinant DNA-produced LACI (225 micrograms/kg over 15 minutes, followed by 4 micrograms/kg/min i.v.) after completion of the infusion of t-PA, no reocclusion occurred during the 2-hour interval of observation in any of the five arteries subjected to electrical injury (p less than 0.001), and cyclic partial occlusions were nearly abolished (0.4 +/- 0.4/hr in LACI-treated dogs compared with 13.7 +/- 5.5/hr in saline-treated dogs, p less than 0.0001). In contrast, reocclusion occurred in two of five arteries with indwelling copper wires, and cyclic partial occlusions were unaffected despite LACI. LACI prolonged the partial thromboplastin time modestly (1.7 +/- 0.2 x baseline) but did not affect platelet counts or aggregation assessed ex vivo. CONCLUSIONS: Inhibition of the extrinsic pathway of coagulation with LACI prevents thrombotic arterial reocclusion after thrombolysis in vessels subjected to extensive vascular injury. Our results demonstrate that activation of the extrinsic pathway plays a critical role in thrombotic reocclusion and that LACI provides a highly targeted approach to facilitate sustained recanalization without directly inhibiting platelets.

Recombinant lipoprotein-associated coagulation inhibitor inhibits tissue thromboplastin-induced intravascular coagulation in the rabbit.

Lipoprotein-associated coagulation inhibitor produces feed-back inhibition of tissue factor (tissue thromboplastin)-induced coagulation in the presence of factor Xa Recombinant lipoprotein-associated coagulation inhibitor (rLACI) was tested for its ability to modify thromboplastin-induced intravascular coagulation in a rabbit model that allows monitoring of iodine-125 fibrin accumulation/disappearance in the lung and sampling of blood for the measurement of coagulation parameters. Infusion of thromboplastin into the rabbit caused a rapid increase of radioactivity over the lungs, possibly due to the accumulation of 125I fibrin in the lungs, followed by a rapid decline of radioactivity, suggestive of removal of fibrin from the lungs. Thromboplastin also caused a rapid decrease of systemic fibrinogen that was accompanied by a lengthening of the activated partial thromboplastin time and prothrombin time. The effect of coinfusion of rLACI with thromboplastin or bolus injection of rLACI before thromboplastin infusion was studied. At a high dose of rLACI (800 micrograms/kg body weight), the thromboplastin-induced radioactivity increase in the lungs and the systemic fibrinogen decrease were completely suppressed. The activated partial thromboplastin time and prothrombin time of the plasma samples lengthened, possibly due to the presence of thromboplastin in circulation. The thromboplastin-induced radioactivity increase over the lungs was not completely suppressed by lower doses of rLACI (135 to 270 micrograms/kg body weight), but these doses of rLACI prevented systemic fibrinogen decrease. At a bolus dose of 23 micrograms/kg body weight, rLACI provided 50% protection of the fibrinogen consumption (fibrinogen decreased to 82% compared with 65% in rabbits treated with thromboplastin alone). These results show that rLACI is effective in the inhibition of thromboplastin-induced coagulation in vivo.

Immunoaffinity purification and characterization of lipoprotein-associated coagulation inhibitors from Hep G2 hepatoma, Chang liver, and SK hepatoma cells. A comparative study.

A polyclonal antibody against a synthetic peptide corresponding to amino acids 3-25 of mature lipoprotein-associated coagulation inhibitor (LACI) was raised in rabbits. The antibody was used to study the production of LACI by Hep G2 hepatoma, Chang liver, and SK hepatoma cells, and to purify LACI from the culture media. By using an amidolytic assay for factor Xa, it was found that the culture media from these liver-derived cell lines contain inhibitors of factor Xa. In Hep G2 hepatoma culture medium, approximately 50% of Xa inhibitory activity was due to LACI. In the Chang liver and SK hepatoma culture media over 95% of the Xa inhibitory activity was due to LACI. The LACIs were purified from these media by immunoaffinity chromatography on an anti-LACI-lg-Sepharose 4B column and preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The purified LA-CIs varied in molecular weight depending on whether the media were concentrated before chromatography. An Mr approximately 38,000 LACI was obtained by chromatography of unconcentrated media. Chromatography of concentrated media yielded a LACI of Mr approximately 35,000 with the same amino-terminal sequence, suggesting partial proteolysis in the carboxyl-terminal region. In addition, an Mr approximately 25,000 form of LACI was also present. The purified Mr approximately 38,000 and approximately 35,000 LACI species from the above cells possess similar specific activities when measured by an anti-Xa/amidolysis assay. To study the role of LACI in the control of coagulation, pooled human plasma was depleted of LACI antigen by immunoaffinity absorption and reconstituted with varying amounts of purified LACI to examine the effect on tissue factor (TF)-induced coagulation. LACI depletion shortens the time of TF-induced clotting of plasma and the clotting time is linearly related to the LACI concentration after reconstitution. These results suggest that LACI plays an important role in limiting TF-induced coagulation in human plasma. Comparison of the potencies of various purified LACIs in the prolongation of TF-induced coagulation revealed that LA-CIs from different sources are not equivalent. The plasma LACI, SK hepatoma LACI, and Chang liver LACI are approximately 7-, 6-7, and 1.3-fold higher in specific activity than Hep G2 hepatoma LACI in the TF-induced clotting assay when compared on an anti-Xa/amidolysis unit basis, suggesting possible differences in post-translational modification of these LA-CIs.

Crystallization of a chymotrypsin inhibitor from Erythrina caffra seeds.

Crystals of a chymotrypsin inhibitor from Erythrina caffra seeds have been grown out of lithium sulfate, by the hanging drop method of vapor diffusion. The crystals belong to the rhombohedral space group R32, with a = 67.2 A and alpha = 99.4 degrees, and diffract to 3 A resolution.

Affinity purification of active plasminogen activator inhibitor-1 (PAI-1) using immobilized anhydrourokinase. Demonstration of the binding, stabilization, and activation of PAI-1 by vitronectin.

Human Hep G2 hepatoma and HT 1080 fibrosarcoma cells were cultured in large scale under conditions which allowed enhanced secretion of plasminogen activator inhibitor-1 (PAI-1). A modified urokinase was obtained by reacting urokinase with phenylmethylsulfonyl fluoride followed by alkali treatment. The resulting product, called anhydrourokinase, was found to reversibly bind the PAI-1 when immobilized on cyanogen bromide-activated Sepharose 4B beads. Using this affinity absorbent, we have purified PAI-1 from the cell-conditioned media. A number of differences have been observed during Hep G2 and HT 1080 PAI purification. 1) The PAI activity in Hep G2 medium concentrate is more stable, and the concentrate depleted of active PAI-1 showed spontaneous regeneration of PAI-1 activity. In contrast, the PAI activity in HT 1080 medium concentrate declines rapidly on standing. 2) Hep G2 PAI-1 invariably copurified with an adhesive protein, vitronectin or its NH2-terminal fragment, while pure HT 1080 PAI-1 alone was obtained by affinity purification on anhydrourokinase-Sepharose 4B. 3) Based on specific activity measurement and complex formation analysis using a sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis technique, the purified Hep G2 PAI-1 appears completely active while the HT 1080 PAI-1 is only one-fourth as active. SDS was found to exert dual effects on purified PAI-1s. SDS treatment partially inactivated a fully active Hep G2 PAI-1 and a moderately active HT 1080 PAI-1 but partially activated an HT 1080 PAI-1 whose activity had previously been allowed to decay to a very low level. Purified vitronectin was found to enhance and stabilize the PAI-1 activity of the partially active HT 1080 PAI-1. It is concluded that fully active PAI-1 in association with vitronectin can be isolated by anhydrourokinase-Sepharose 4B chromatography and that vitronectin is a binding protein for PAI-1 which activates and stabilizes PAI-1.

Eicosanoid synthesis by rabbit hydronephrotic cortical interstitial cells in culture.

Rabbit hydronephrotic cortical interstitial cells in primary culture were labeled with [1-14C]arachidonic acid and the eicosanoids released after stimulation with bradykinin or A23187 were studied by reverse-phase high performance liquid chromatography. The major arachidonic acid metabolite formed was prostaglandin (PG)E2, comprising more than 30% of the total radioactivity released. 12-Hydroxyheptadecatrienoic acid, probably representing spontaneous breakdown of the cyclic endoperoxides PGG2 and/or PGH2, made up 10 to 15% of the radioactivity released. Other cyclooxygenase products that were released included PGF2 alpha, PGD2, 6-keto PGF1 alpha and only minute amounts of thromboxane B2. Small quantities of the lipoxygenase products 15-, 12- and 5-hydroxyeicosatetraenoic acids (HETEs) as well as leukotrienes (LT)B4, LTC4 and LTD4 were also identified. Significantly larger quantities of 15- and 5-HETEs were recovered at 2 to 5 min than after longer incubations with A23187, suggesting esterification of these HETEs into cellular phospholipids. The data indicate that interstitial cells of the hydronephrotic kidney synthesize a variety of cyclooxygenase and lipoxygenase products of arachidonic acid, which may contribute to the pathophysiology of hydronephrosis. Moreover, it is suggested that PGG2 and/or PGH2 that are released from these cells may be metabolized further by adjacent kidney cells or circulating blood elements to other eicosanoid products, thus increasing the diversity of eicosanoids synthesized in the hydronephrotic kidney.

Renal reduced nicotinamide adenine dinucleotide phosphate:cytochrome c reductase-mediated metabolism of the carcinogen N-[4-(5-nitro-2-furyl)-2-thiazolyl]acetamide.

N-[4-(5-Nitro-2-furyl)-2-thiazolyl]acetamide (NFTA) metabolism was examined in vitro using microsomes prepared from rat liver and renal cortex and from rabbit liver and renal cortex and outer and inner medulla. NFTA nitroreduction was observed with each tissue. Three mol of NADPH were used per mol of NFTA reduced. Substrate and inhibitor specificity suggested that the microsomal nitroreduction was due to NADPH:cytochrome c reductase. Metabolite(s) formed bound to protein, RNA, DNA, and synthetic polyribonucleotides. Maximum covalent binding was seen with polyguanylic acid. A guanosine-NFTA adduct was isolated. Binding was inhibited by sulfhydryl compounds and vitamin E. The [14C]NFTA:glutathione or [3H]glutathione:NFTA conjugates obtained from microsomal incubations showed identical chromatographic properties as the product obtained by the reaction of synthetic N-hydroxy-NFTA with [3H]glutathione. Structures of synthetic N-hydroxy-NFTA and the microsomal reduction product 1-[4-(2-acetylaminothiazolyl)]-3-cyano-1-propanone were established by mass spectrometry. The latter reduction product did not bind macromolecules. These results suggest that renal NADPH:cytochrome c reductase reduces NFTA to an N-hydroxy-NFTA intermediate that binds nucleophilic sites on macromolecules.

Metabolic activation of the carcinogen N-[4-(5-nitro-2-furyl)-2-thiazolyl]acetamide by prostaglandin H synthase.

It has been demonstrated that N-[4-(5-nitro-2-furyl)-2-thiazolyl]acetamide (NFTA), when fed with the diet, causes transitional carcinomas in rats. An important step in the mechanism of NFTA-induced carcinogenesis is endogenous metabolic activation to an ultimate carcinogen. We have proposed that the enzyme complex prostaglandin H synthase (PHS) is involved in the activation of certain renal and urinary tract carcinogens. This proposal was assessed by examining the activation of the 5-nitrofuran renal carcinogen NFTA and its deacetylated analogue 2-amino-4-(5-nitro-2-furyl)thiazole (ANFT) by PHS. Ram seminal vesicular and rabbit renal inner medullary microsomes were used as a source of PHS. Both NFTA and ANFT were activated by PHS to bind microsomal protein. Both microsomal preparations activated ANFT to bind DNA. However, only ram seminal vesicular microsomes activated NFTA to bind DNA. The rate of ANFT binding to macromolecules was considerably greater than NFTA with both microsomal preparations. Although activated ANFT was shown to bind several different homopolynucleotides, a preference for binding polyguanylic acid was demonstrated. Glutathione inhibition of carcinogen binding to macromolecules was shown to be due to the formation of a thioether conjugate. Deacetylation of NFTA was demonstrated in both tissues with deacetylation significantly exceeding acetylation of ANFT to NFTA in the kidney. Thus, renal PHS activation of both NFTA and ANFT was demonstrated with the rate of ANFT activation being considerably greater than NFTA. The conversion of NFTA to ANFT by intact tissue suggests that ANFT may contribute to NFTA renal carcinogenesis.

Comparative effects of prostaglandin H synthase-catalyzed binding of two 5-nitrofuran urinary bladder carcinogens.

Understanding the role of prostaglandin H synthase (PHS) in the carcinogenic process and the metabolic steps involved in the activation of carcinogens will facilitate experiments using pharmacological agents to prevent carcinogenesis. This study assesses the relative amounts of PHS-catalyzed binding of the urinary tract carcinogens [2-14C]-N-[4-(5-nitro-2-furyl)-2-thiazolyl] formamide (FANFT) and [2-14C]-2-amino-4-(5-nitro-2-furyl) thiazole (ANFT). Binding to protein and nucleic acid was assessed using PHS prepared from ram seminal vesicle, dog bladder transitional epithelium and rabbit renal inner medulla. PHS-catalyzed binding of ANFT was significantly greater than FANFT in each tissue. Substrate and inhibitor experiments were consistent with the prostaglandin hydroperoxidase activity of PHS catalyzing the binding of FANFT and ANFT. Oxygen was required for metabolism with arachidonic acid but not with peroxide as cosubstrate. The amount of PHS-catalyzed ANFT binding to protein was at least 4-fold greater than FANFT. Whereas a significant amount of FANFT was bound to protein, no FANFT binding to DNA could be detected. By contrast, PHS catalyzed the binding of ANFT to both protein and DNA. A PHS-catalyzed metabolite of ANFT was tentatively identified as the 4-keto analog by mass spectral analysis. The lower rate of PHS-catalyzed metabolism of FANFT compared to ANFT and the lack of detectable FANFT binding to DNA suggest that the metabolic steps involved in the initiation of FANFT-induced bladder cancer include 1) deformylation of FANFT to ANFT, 2) PHS-catalyzed activation of ANFT and 3) binding of an activated ANFT metabolite(s) to DNA.

Microsomal nitroreductase activity of rabbit kidney and bladder: implications in 5-nitrofuran-induced toxicity.

Reductive metabolism of the aromatic nitro group of 5-nitrofurans is thought to be an important step in the mechanism of their toxicity. Microsomal nitroreductase activity with p-nitrobenzoic acid and N-[4-(5-nitro-2-furyl)-2-thiazolyl[formamide (FANFT) as substrates was assessed in the renal cortex, outer medulla, inner medulla, bladder transitional epithelial and non-epithelial bladder tissue. Cortex and transitional epithelial tissue contained the most p-nitrobenzoic acid reductase activity. However, FANFT reductase activity was similar in all areas tested except nonepithelial bladder tissue, which was 10% of the others. FANFT reduction was inhibited by oxygen, but not by carbon monoxide, allopurinol or aspirin and required NADPH. These results are consistent with NADPH-cytochrome c reductase catalyzed FANFT reduction. In medullary microsomes, the apparent Km and Vmax were 0.125 mM and 0.84 nmol/mg of protein per min, respectively. Transitional epithelial microsomes incorporated approximately 1 and 10% of the total [2-14C]FANFT metabolized into t-RNA and trichloroacetic acid-precipitable material, respectively. Two products of FANFT reduction were demonstrated by high-pressure liquid chromatography. One product was reversibly oxidized to FANFT and the other was tentatively identified by mass spectral analysis as an open chain nitrile. In view of the relatively low oxygen tension in the renal inner medulla and bladder mucosa, these results suggest that medullary and transitional epithelial nitro-reductases may be involved in the pathogenesis of 5-nitrofuran toxicity.