A reconstituted dilute blood clot lysis assay for the medium throughput screening of thrombolytic compounds.

Antithrombotic and clotting factors have long been targets for drug discovery, necessitating the development of blood assays to determine the efficacy of lead compounds prior to animal testing. We have developed a reconstituted blood clot lysis assay which eliminates the need for on-site donors. The assay uses whole blood stored at 4 degrees C obtained from a local blood bank, diluted 1:10 in phosphate buffer. This blood was supplemented with 125I-labeled fibrinogen and the release of radioactive fibrinopeptides from formed clots was measured. The whole blood used in this assay, which had been stored at 4 degrees C for several days, no longer formed solid or retracting clots. Thus, platelets 5-7 days ex vivo (165 x 10(6) platelets) were added to the whole blood in the presence of thrombin (0.80 IU/ml) to form clots. Solid clots formed within 2 min of thrombin addition and began retracting shortly thereafter. In the absence of any thrombolytic agent, clots fully retracted within 2.5 h and remained stable. Thrombin-stimulated clot formation was completely inhibited by heparin. Clots could be lysed in a dose-dependent fashion in the presence of tissue-type plasminogen activator. Clot lysis could be completely inhibited in a dose-dependent fashion with plasminogen activator inhibitor type 1. To demonstrate the utility of this assay as a screen for thrombolytic agents, a 14-amino-acid PAI-1-inhibitory peptide relieved the PAI-1 effect on tPA in a dose-dependent fashion. These data describe an assay for the screening of potential pro-fibrinolytic agents that target PAI-1 inhibition in a human plasma-based system that is versatile, cost-effective, and physiologically relevant and does not rely on the availability of on-site blood donors.

Exploring structure-activity relationships around the phosphomannose isomerase inhibitor AF14049 via combinatorial synthesis.

Phosphomannose Isomerase (PMI) has been shown by genetic methods to be an essential enzyme in fungal cell wall biosynthesis. The PMI inhibitor AF14049 was discovered as an unanticipated side product from high-throughput library screening against the enzyme from C, albicans. Solid-phase synthetic methods were developed and a series of libraries and discrete analogs synthesized to explore SAR around AF14049.

Encoded combinatorial chemistry: synthesis and screening of a library of highly functionalized pyrrolidines.

The application of a new encoding technology for drug discovery is described. A combinatorial library of mercaptoacyl pyrrolidines has been prepared on a beaded polymeric support. Each polymer bead carries one library constituent in association with an oligomeric "tag," the structure of which is a record of the specific reagents from which that library member was prepared. After the ligands were solubilized, an array of such beads was screened for angiotensin-converting enzyme inhibitory activity, and the structures of active pyrrolidines were deduced by analysis of the associated tags at sub-picomole levels. Several extremely potent enzyme inhibitors were identified, many from multiple beads. The most potent inhibitor was found to have a Ki of 160 pM, approximately 3-fold more active than captopril in the same assay. Direct comparison with iterative deconvolution shows that the encoded screening strategy is a much more efficient means for extracting information from such compound collections, producing more data on a larger number of active structures.

A high-density screening format for encoded combinatorial libraries: assay miniaturization and its application to enzymatic reactions.

A novel, miniaturized high-throughput screening format is described for assay of combinatorial libraries generated on beads. This approach, which is ideally suited to encoded libraries synthesized on beads, utilizes the photolytic cleavage of individual compounds into a high-density well array (>6500 wells within a standard 96-well microtiter plate footprint) with well volumes as low as 0.37 microl. As a model study, an encoded dipeptide library (324 members) acylated with isobutyl succinate was assayed using this format to search for potential inhibitors of matrilysin, a member of the matrix metalloproteinase superfamily. In situ release of compounds from solid support was accomplished by photochemical cleavage after beads and enzyme were distributed to the wells. After the addition of a fluorogenic substrate to the array, the extent of enzyme inhibition and identification of active compounds was quantitated by imaging of the fluorescence emission upon uv irradiation. The structure-activity relationship data generated from the identified inhibitors in this study corroborate previous findings, thus validating the utility of this approach as a means of high-throughput screening of bead-based libraries.

Tissue Factor residue Asp44 regulates catalytic function of the bound proteinase Factor VIIa.

The coagulation pathways are initiated by the cell-surface receptor Tissue Factor (TF), which binds the serine proteinase coagulation Factor VIIa (VIIa), resulting in enhanced catalytic function, both amidolytic, towards small pseudo-substrates, and proteolytic, towards macromolecular substrates. Here we implicate Asp44 in TF as a ligand-interactive residue that, in contrast with previously characterized binding residues, is involved in the enhancement of VIIa catalytic function. Whereas charge neutralization by replacement of Asp44 with Asn did not reduce function of human TF, the exchange by Ala resulted in mutants with 8-fold reduced affinity for binding of VIIa. Enhancement of VIIa amidolytic function by TF Ala44 was reduced by 20-25% relative to wild-type and support of proteolytic function was diminished 6-fold indicating that this cofactor residue is significantly enhancing proteolysis of the macromolecular substrate by VIIa. Replacement of Asp44 by Glu, Thr, and Arg exhibited a less severe phenotype with an approx. 4-fold reduced affinity for VIIa and a 2-3 fold diminished activation of Factor X. The improved activity of these mutants as compared with the Ala replacement is consistent with functional importance of an extended side chain at this position. The specific influence of the Asp44 exchange on catalytic function of the TF x VIIa complex indicates fine specificity of the TF ligand interface in mediating receptor and cofactor function.

Energetic contributions and topographical organization of ligand binding residues of tissue factor.

Tissue factor is the cellular receptor and macromolecular enzymatic cofactor for the serine protease coagulation factor VIIa. The ligand binding extracellular domain of tissue factor consists of two structural modules which fold similar to fibronectin type III modules, consistent with the classification of tissue factor as a member of the class 2 cytokine receptor family. On the basis of the three-dimensional structure, we here analyze the importance of tissue factor residues for binding of ligand by scanning alanine mutagenesis. The identified significant binding contacts account for as much as 80% of the calculated total free energy of ligand binding. Most residues with energetic contributions to ligand binding are well exposed to solvent, and the area for ligand interaction extends from the cleft formed by the two structural modules (residues Lys20, Ile22, Lys48, Asp58, Arg135, Phe140) to the convex-shaped edge of the three- and four-stranded sheets characterized by a patch of surface-exposed hydrophobic side chains in the amino-terminal module (residues Gln37, Asp44, Trp45, Phe76, Tyr78). The binding residues are dispersed over an extended surface area, indicating adaptation to the recognition of specific structural modules of the macromolecular ligand factor VIIa. This analysis provides detailed insight into the three-dimensional organization of the ligand docking structure of the initiating cofactor for the coagulation pathways.

The alkylating properties of chlorambucil.

Previous work has indicated an aziridinium ion mechanism in the hydrolysis of chlorambucil, and the present work on the alkylation of nucleophiles fully supports this mechanism. This mechanism forms the basis for understanding the kinetics of alkylation reactions because their rates are limited by the rate of formation of the aziridinium ion and the alkylation reaction competes with the hydrolytic reaction. We have measured alpha N, where alpha N(N) is the rate of reaction of the aziridinium ion with a nucleophile N relative to its reaction with water for several nucleophiles that are related to those found in proteins. The alpha values for hydroxide ion and some other bases are greater than 10(3), but the effective values at pH 7.5 are much smaller because there is little base present. The kinetic equations show that it is very difficult to alkylate a nucleophile extensively at pH 7.5 before chlorambucil has hydrolyzed. Therefore, it is not clear why angiotensin-converting enzyme is completely inhibited by low concentrations of chlorambucil. On the other hand, damage to DNA is easily understood.

Key ligand interface residues in tissue factor contribute independently to factor VIIa binding.

Scanning alanine mutagenesis of the cell surface protease receptor tissue factor suggested importance of residues Lys20, Ile22, Asp58, Arg135, and Phe140 for binding of ligand, the serine protease coagulation factor VIIa. Ligand binding by single alanine replacement mutants was characterized by functional assays which concordantly demonstrated a calculated 1-2.5 kcal/mol reduction in free energy of binding as a result of each of the mutations. Catalytic and proteolytic function appeared to be not impaired by the residue replacements, indicating that these residues are not specifically required for the catalytic enhancement of VIIa produced by the assembly with tissue factor. Multiple mutations were further combined in one mutant protein to assess whether these residues provide independent contacts with the ligand VIIa. The Lys20/Asp58 and the Arg135/Phe140 residue pairs did not independently contribute to the binding of ligand. In contrast, the combination with Ile22 consistently produced a further decrease in affinity for VIIa, demonstrating that this residue acts as an independent contact site for the ligand VIIa. The total contribution of the five residues to the free energy of binding of VIIa at 37 degrees C was calculated to be 5.4 kcal/mol representing approximately one-third of the total binding energy.

Mutational mapping of functional residues in tissue factor: identification of factor VII recognition determinants in both structural modules of the predicted cytokine receptor homology domain.

Alanine scanning mutagenesis of tissue factor, the initiating receptor and cofactor molecule for the coagulation pathways, was used to define residue side chains with functional contributions. Approximately half of the residues were exchanged, and several stretches of functional residues throughout the entire extracellular domain were identified which contributed to overall coagulant function. Mutants were further characterized with respect to their affinity for binding of ligand, providing evidence that identified functional sequence spans are involved in ligand interaction. The tissue factor extracellular domain is suggested to adopt the folding pattern of the cytokine receptor homology unit, which is typically composed of two seven-beta-strand modules. Evaluation of the mutational analysis within this structural context suggests that functionally important residues are spatially proximate and clustered at the boundary of the predicted beta-strand modules. Residues contributing to ligand binding by tissue factor were identified in positions corresponding to ligand interactive residues in the growth hormone receptor and contact residues of other cytokine receptors, consistent with a conserved structural region for ligand interaction throughout the cytokine receptor family.

Angiotensin converting enzyme: substrate inhibition.

Phosphate, borate, and Tris inhibit angiotensin converting enzyme (ACE), but HEPES buffer is inert. Measurements of substrate inhibition were made in HEPES buffer at pH 7.0 and 25 degrees C and 37 degrees C. Substrate inhibition was marked and goes to completion. A new equation for substrate inhibitions enables one, under favorable circumstances, to determine whether there is cooperativity in the binding of substrate to the inhibitory and active sites. Cooperativity does occur with ACE using Hipp-His-Leu as substrate. The kinetic parameters were measured (Km = 0.21 mM, K* = 0.65 mM at 37 degrees C). The enzyme concentration (1.94 X 10(-8) M) was determined by titration with lisinopril so that kcat (5 X 10(3) at 37 degrees C) could be determined. Using this value and the molecular weight the specific activity of ACE was calculated for different common buffers. The specific activity in HEPES calculated from Vmax was 33.7 units/mg at 37 degrees C.

Purification of bovine angiotensin converting enzyme.

A change has been made in the commonly used lisinopril affinity gel procedure for purifying angiotensin converting enzyme. The new method greatly decreases the time required and greatly increases the yield of pure enzyme. All of the enzyme in various bovine tissues was extracted with 0.5% triton X-100 and applied to the affinity column; 70% was trapped and all of the trapped enzyme was released as the apoenzyme by EDTA. The holoenzyme was recovered by dialysis against zinc containing buffer. The turnover numbers were precisely the same for enzyme from lung, atrium, kidney, striatum and blood. The tissue concentrations of ACE were very different but the final specific activities were the same.

The binding of zinc to angiotensin-converting enzyme.

The equilibrium constant for the dissociation of zinc ion from angiotensin-converting enzyme (ACE) was measured using zinc ion buffers of zinc chloride and nitrilotriacetic acid (NTA). The dissociation constant is 6.4 X 10(-10) M. The fraction of active enzyme at equilibrium is independent of the presence of substrate which indicates that hippuryl-histidylleucine binds equally well to the holoenzyme and apoenzyme. The rate constant for the dissociation of zinc from ACE was measured as 0.68 min-1 for the free enzyme; the rate constant for the enzyme substrate complex was roughly 0.18 min-1. The association of zinc ion and ACE is very fast; the rate constant is 1.06 X 10(9) M-1 min-1. Ethylenediaminetetraacetic acid (EDTA) and NTA rapidly remove zinc from ACE with rate constants of 1.27 X 10(3) and 2.2 X 10(3) M-1 min-1. The equilibrium constant for the reaction of NTA with ACE was measured as 4.6 X 10(-2) and was calculated for EDTA as 3.8 X 10(3).

Hybridization and partial cDNA sequence analyses of bovine lung angiotensin I-converting enzyme.

The mRNA encoding angiotensin I-converting enzyme, a zinc-metallo dipeptidyl carboxyhydrolase, has been identified in extracts prepared from bovine lung tissue. Bovine lung poly(A) + mRNAs were subjected to electrophoresis and northern blot hybridization analysis using a radiolabeled synthetic 24-deoxyoligonucleotide probe complementary to eight codons for amino acids at the active-site of the enzyme (Harris, R.B. & Wilson, I.B., J. Biol. Chem. 260, 2208-2211, 1985). This amino acid sequence contains the catalytic glutamic acid residue. A single RNA species (approximately equal to 4 kb) was detected which is 1 kb larger than predicted from the molecular weight of the enzyme. The excess nucleic acid composition may be due to leader and/or trailer sequences or the RNA may encode a high molecular weight precursor form of the enzyme. We have cloned an EcoR1-HindIII digest fragment (1400 bp) of the duplex cDNA derived from the bovine lung converting enzyme poly(A) + mRNA and also Bal31 deletion fragments generated from the 1400 bp clone. Several of the Bal31 clones contain the active-site sequence codons of the enzyme and the complete cDNA sequence of one of these (72 bp) has been determined. We found the amino acid sequence at the active site to be -Phe-Thr-Glu-Leu-Ala-Asn-Ser-, containing the catalytic Glu residue. This sequence is identical with the sequence that we previously determined by manual Edman degradation analysis of the appropriate active-site peptide except that we now find Asn instead of Asp. We have sequenced 670 bp of the 1400 bp clone but have not yet overlapped the active-site sequence.(ABSTRACT TRUNCATED AT 250 WORDS)