Discovery of highly potent small molecule kallikrein inhibitors.

Uncontrolled kallikrein activation is involved in diseases such as hereditary angioedema, bacterial septic shock and procedures such as cardiopulmonary bypass. Here we report a series of small molecule compounds that potently inhibit kallikrein activity in vitro. Kinetic studies indicate that some of these compounds are slow binding inhibitors of kallikrein with Ki final less than a nanomolar. The ability of these compounds to inhibit the activity of kallikrein was further confirmed in a plasma model by quantitating the release of bradykinin, an endogenous cleavage product of plasma kallikrein. To understand the inhibitory mechanism of the selected compounds toward kallikrein, the interactions between the selected compounds and kallikrein was explored using molecular modeling based on the information of crystal structures of TF/FVIIa and kallikrein. The information presented in the current study provides an initial approach to develop more selective and therapeutically useful small molecule inhibitors.

Purine nucleoside phosphorylase inhibitor BCX-1777 (Immucillin-H)--a novel potent and orally active immunosuppressive agent.

Patients with purine nucleoside phosphorylase (PNP) deficiency present a selective T-cell immunodeficiency. Inhibitors of PNP are, therefore, of interest as potential T-cell selective immunosuppressive agents. BCX-1777 is a potent inhibitor of PNP from various species including human, mouse, rat, monkey and dog, with IC50 values ranging from 0.48 to 1.57 nM. BCX-1777, in the presence of 2'-deoxyguanosine (dGuo, 3-10 microM), inhibits human lymphocyte proliferation activated by various agents such as interleukin-2 (IL-2), mixed lymphocyte reaction (MLR) and phytohemagglutinin (PHA) (IC50 values < 0.1-0.38 microM). BCX-1777 is a 10-100-fold more potent inhibitor of human lymphocyte proliferation than other known PNP inhibitors like PD141955 and BCX-34. Nucleotide analysis of human lymphocytes indicate that inhibition of proliferation by BCX-1777 correlates with dGTP levels in the cells. BCX-1777 has excellent oral bioavailability (63%) in mice. At a single dose of 10 mg/kg in mice, BCX-1777 elevates dGuo to approximately 5 microM. BCX-1777 was not effective in mouse T-cell models such as delayed type hypersensitivity (DTH) and splenomegaly because mouse T-cells do not accumulate dGTP as do human T-cells. However, in the human peripheral blood lymphocyte severe combined immunodeficiency (hu-PBL-SCID) mouse model, BCX-1777 was effective in prolonging the life span 2-fold or more. This is the first known example of a PNP inhibitor that elevates dGuo in mice similar to the levels observed in PNP-deficient patients. Furthermore, these dGuo levels are also required for in vitro T-cell inhibition by BCX-1777. Thus, BCX-1777 represents a novel class of selective immunosuppressive agents that could have therapeutic utility in various T-cell disorders.

The Arthus reaction in rodents: species-specific requirement of complement.

We induced reverse passive Arthus (RPA) reactions in the skin of rodents and found that the contribution of complement to immune complex-mediated inflammation is species specific. Complement was found to be necessary in rats and guinea pigs but not in C57BL/6J mice. In rats, within 4 h after initiation of an RPA reaction, serum alternative pathway hemolytic titers decreased significantly below basal levels, whereas classical pathway titers were unchanged. Thus the dermal reaction proceeds coincident with systemic activation of complement. The serine protease inhibitor BCX 1470, which blocks the esterolytic and hemolytic activities of the complement enzymes Cls and factor D in vitro, also blocked development of RPA-induced edema in the rat. These data support the proposal that complement-mediated processes are of major importance in the Arthus reaction in rats and guinea pigs, and suggest that BCX 1470 will be useful as an anti-inflammatory agent in diseases where complement activation is known to be detrimental.

Measurement of complex surface deformation by high-speed dynamic phase-stepped digital speckle pattern interferometry.

We describe a high-speed digital speckle pattern interferometer incorporating a line-scan camera and a waveguide phase modulator for the measurement of complex deformation (vibration phase and amplitude) at audio acoustic frequencies. Experimental data show continuous phase-stepped recovery of out-of-plane surface deformation in one dimension, obtained at 100 kHz with 2pi/20-rad (0.02-mum) displacement resolution, for surface velocities of 3.2 mm s>(-1) .

Phase-demodulation error of a fiber-optic Fabry-Perot sensor with complex reflection coefficients.

The influence of reflector losses attracts little discussion in standard treatments of the Fabry-Perot interferometer yet may be an important factor contributing to errors in phase-stepped demodulation of fiber optic Fabry-Perot (FFP) sensors. We describe a general transfer function for FFP sensors with complex reflection coefficients and estimate systematic phase errors that arise when the asymmetry of the reflected fringe system is neglected, as is common in the literature. The measured asymmetric response of higher-finesse metal-dielectric FFP constructions corroborates a model that predicts systematic phase errors of 0.06 rad in three-step demodulation of a low-finesse FFP sensor (R = 0.05) with internal reflector losses of 25%.

Structure of 3,4-dichloroisocoumarin-inhibited factor D.

Factor D (D) is a serine protease essential in the activation of the alternative complement pathway. Only a few of the common serine protease inhibitors inhibit D, binding covalently to the serine hydroxyl of the catalytic triad. 3,4-Dichloroisocoumarin (DCI) is a mechanism-based inhibitor which inhibits most serine proteases and many esterases, including D. The structure of the enzyme:inhibitor covalent adduct of D with DCI, DCI:D, to a resolution of 1.8 A is described, which represents the first structural analysis of D with a mechanism-based inhibitor. The side chain of the ring-opened DCI moiety of the protein adduct undergoes chemical modification in the buffered solution, resulting in the formation of an alpha-hydroxy acid moiety through the nucleophilic substitution of both Cl atoms. The inhibited enzyme is similar in overall structure to the native enzyme, as well as to a variety of isocoumarin-inhibited trypsin and porcine pancreatic elastase (PPE) structures, yet notable differences are observed in the active site and binding mode of these small-molecule inhibitors. One region of the active site (residues 189-195) is relatively conserved between factor D, trypsin, and elastase with respect to amino-acid sequence and to conformation. Another region (residues 214-220) reflects the amino-acid substitutions and conformational flexibility between these enzymes. The carbonyl O atom of the DCI moiety was found to be oriented away from the oxyanion hole, which greatly contributes to the stability of the DCI:D adduct. The comparisons of the active sites between native factor D, DCI-inhibited factor D, and various inhibited trypsin and elastase (PPE) molecules are providing the chemical bases directing our design of novel, small-molecule pharmaceutical agents capable of modulating the alternative complement pathway.

Structures of native and complexed complement factor D: implications of the atypical His57 conformation and self-inhibitory loop in the regulation of specific serine protease activity.

Factor D is a serine protease essential for the activation of the alternative pathway of complement. The structures of native factor D and a complex formed with isatoic anhydride inhibitor were determined at resolution of 2.3 and 1.5 A, respectively, in an isomorphous monoclinic crystal form containing one molecule per asymmetric unit. The native structure was compared with structures determined previously in a triclinic cell containing two molecules with different active site conformations. The current structure shows greater similarity with molecule B in the triclinic cell, suggesting that this may be the dominant factor D conformation in solution. The major conformational differences with molecule A in the triclinic cell are located in four regions, three of which are close to the active site and include some of the residues shown to be critical for factor D catalytic activity. The conformational flexibility associated with these regions is proposed to provide a structural basis for the previously proposed substrate-induced reversible conformational changes in factor D. The high-resolution structure of the factor D/isatoic anhydride complex reveals the binding mode of the mechanism-based inhibitor. The higher specificity towards factor D over trypsin and thrombin is based on hydrophobic interactions between the inhibitor benzyl ring and the aliphatic side-chain of Arg218 that is salt bridged with Asp189 at the bottom of the primary specificity (S1) pocket. Comparison of factor D structural variants with other serine protease structures revealed the presence of a unique "self-inhibitory loop". This loop (214-218) dictates the resting-state conformation of factor D by (1) preventing His57 from adopting active tautomer conformation, (2) preventing the P1 to P3 residues of the substrate from forming anti-parallel beta-sheets with the non-specific substrate binding loop, and (3) blocking the accessibility of Asp189 to the positive1y charged P1 residue of the substrate. The conformational switch from resting-state to active-state can only be induced by the single macromolecular substrate, C3b-bound factor B. This self-inhibitory mechanism is highly correlated with the unique functional properties of factor D, which include high specificity toward factor B, low esterolytic activity toward synthetic substrates, and absence of regulation by zymogen and serpin-like or other natural inhibitors in blood.

Structure of diisopropyl fluorophosphate-inhibited factor D.

Factor D (D) is a serine protease, crucial for the activation of the alternative complement pathway. Only a limited number of general serine protease inhibitors are known to inhibit D, most of which covalently bind to the serine hydroxyl of the catalytic triad. The structure of the first enzyme:inhibitor covalent adduct of D with diisopropyl fluorophosphate (DIP:D) to a resolution of 2.4 A is described. The inhibited enzyme is similar in overall structure to the native enzyme and to trypsin, yet exhibits notable differences in the active site. One region of the active site is conserved between D and trypsin with respect to amino-acid sequence and to conformation. Another reflects the amino-acid substitutions and conformational flexibility between these enzymes. The active-site histidine residue is observed in the gauche+ conformation, not the normal gauche- orientation seen in the classic catalytic triad arrangement required for enzymatic activity in serine proteases. Comparisons of the active sites between native D, the DIP:D adduct, and DIP-inhibited trypsin have provided fundamental insights currently being employed in the design of novel small-molecule pharmaceutical agents capable of modulating the alternative complement pathway.

Structure of human factor D. A complement system protein at 2.0 A resolution.

Factor D, an essential enzyme for the activation of the alternative pathway of the complement system, belongs to the serine protease superfamily. The crystal structure of the enzyme was solved by a combination of multiple isomorphous replacement and molecular replacement methods. The present model was refined to an R-factor of 18.8% using 23,681 observed reflections between 7.5 and 2.0 A resolution, with a root-mean-square deviation from standard bond lengths of 0.016 A. The two non-crystallographically related molecules in the triclinic unit cell have distinctive active site conformations. The protein has the general structural fold of a serine protease, but there are several unique amino acid substitutions resulting in significant alterations in the critical loops responsible for catalysis and substrate specificity in serine proteases. Factor D is the first complement serine protease whose three-dimensional structure has been determined.

Modelling of the serine-proteinase fold by X-ray and neutron scattering and sedimentation analyses: occurrence of the fold in factor D of the complement system.

Solution scattering is a powerful means of determining the overall arrangement of domains in the multidomain proteins of complement. the serine-proteinase domain is central to all proteolytic events during complement activation. As models of this domain, bovine beta-trypsin, trypsinogen, alpha-chymotrypsin and chymotrypsinogen A were studied by neutron and X-ray synchrotron solution scattering. At pH 7, all the X-ray and neutron M(r) values corresponded to monomeric proteins. The X-ray radii of gyration, RG, of beta-trypsin, trypsinogen, alpha-chymotrypsin and chymotrypsinogen A (measured in positive solute-solvent contrasts) were 1.59 nm, 1.78 nm, 1.71 nm and 1.76 nm (+/- 0.05-0.11 nm) in that order. Neutron contrast variation showed that the RG at infinite contrast, RC, for these four proteins were 1.57 nm, 1.70 nm, 1.67 nm and 1.78 nm (+/- 0.03 nm) in that same order. The radial inhomogeneity of neutron-scattering density, alpha, was positive at (5-13) x 10(-5), and corresponds to the preponderance of hydrophilic residues near the protein surface. On trypsinogen activation, a small reduction in the RG value of 0.13 +/- 0.07 nm was just detectable, while the RG of chymotrypsinogen A was unchanged after activation. The RC and alpha values of the four proteins can be calculated by using crystallographic co-ordinates. The reduced RG of beta-trypsin relative to trypsinogen was explained in terms of the removal of the extended N-terminal hexapeptide of trypsinogen. The full X-ray and neutron-scattering curves in positive and negative contrasts agreed well with scattering curves calculated from crystallographic coordinates to a nominal structural resolution of 4.5 nm, provided that the internal structure was considered in neutron modelling, and that the hydration was considered in X-ray modelling. Sedimentation-coefficient data also provide information on the disposition of domains in multidomain proteins. It was found that the hydrated X-ray sphere model could be directly utilized to calculate sedimentation coefficients. X-ray scattering on factor D showed from its RG of 1.78 nm that this is monomeric and very similar in structure to beta-trypsin. The X-ray-scattering curve of factor D was readily modelled using the beta-trypsin crystal structure after allowance for sequence changes. The success of these modellings provides a basis for the constrained modelling of solution scattering data for the multidomain proteins of complement.

Crystallization and preliminary X-ray investigation of factor D of human complement.

Human factor D, an essential enzyme of the alternative pathway of complement activation, has been crystallized. Crystals were grown by vapor diffusion using polyethylene glycol 6000 and NaCl as precipitants. The factor D crystals are triclinic and the space group is P1 with unit cell dimensions a = 40.8 A, b = 64.7 A, c = 40.3 A, alpha = 101.0 degrees, beta = 109.7 degrees, gamma = 74.3 degrees. The unit cell contains two molecules of factor D related by a non-crystallographic 2-fold axis. The crystals grow to dimensions of 0.8 mm x 0.5 mm x 0.2 mm within five days, are stable in the X-ray beam and diffract beyond 2.5 A.

Depression of humoral responses and phagocytic functions in vivo and in vitro by fish oil and eicosapentanoic acid.

Previous studies have demonstrated that eicosapentanoic acid (EPA) has anti-inflammatory properties in both humans and experimental animals and may also depress humoral immunity in experimental animals. Our investigations showed that the addition of eicosapentanoic acid to human peripheral blood mononuclear cell cultures inhibited B cell responses to mitogenic stimulation and depressed the expression of interleukin 2 receptors in pokeweed mitogen-stimulated lymphocytes. Neutrophils were also affected in their ability to release the contents of primary and secondary granules, particularly when stimulated with antigen-antibody complexes. Similar depressions of B cell responses and neutrophil functions were observed in a normal volunteer who ingested 6 g/day of a commercially available fish oil extract (equivalent to 2.1 g of EPA/day) during a 6-week period. Phagocytosis, enzymatic release, circulating immunoglobulin levels, and the response to tetanus toxoid both in vivo and in vitro were depressed during ingestion of fish oil. Most parameters showed a trend toward normalization 6 weeks after the suspension of fish oil supplementation. These effects of fish oil extracts and EPA on phagocytosis and humoral responses may be advantageously used in the therapy of chronic inflammatory diseases and autoimmune diseases but could be a cause for concern when these compounds are used for longer periods of time and with minimal medical supervision for the prophylaxis of atherosclerosis.

Human endothelial cell damage induced by interactions between polymorphonuclear leukocytes and immune complex-coated erythrocytes.

We have examined, in vitro, the effects of autologous erythrocytes (RBC) coated with soluble immune complexes (RBC-IC) on the interactions of polymorphonuclear leukocytes (PMN) with human endothelial cell (EC) cultures. RBC-IC were prepared by incubating human RBC with soluble immune complexes, prepared with keyhole limpet hemocyanin (KLH) and rabbit anti-KLH in antigen excess, using normal human serum as a complement source. Several parameters believed to be related to EC-PMN interactions, including adherence of PMN to EC, detachment of EC after incubation with media harvested from RBC-IC-stimulated PMN, and the release of 2-deoxy-D-[3H]glucose from damaged EC, were investigated. The proportion of PMN adhering to EC increased in direct proportion to the number of RBC-IC used to stimulate PMN. Media harvested from PMN stimulated with RBC, RBC incubated with complement, RBC-IC, or opsonized zymosan caused variable degrees of detachment of cultured EC. The percentage of EC detached was directly related to the intensity of the PMN degranulation as measured by lysozyme release with a correlation coefficient of 0.935, comparing the natural log of the lysozyme released to the percentage EC detachment. Finally, RBC-IC added to PMN and EC induced PMN-mediated EC damage as measured by the release of 2-deoxy-D-[3H]glucose from the labeled EC. The release of 2-deoxy-D-[3H]glucose from the labeled EC was directly related to the number of RBC-IC used to stimulate the PMN. These data indicate that RBC-IC are able to stimulate the type of interactions between PMN and EC that are believed to cause EC damage and may play a role in the initiation or exacerbation of inflammatory vascular lesions.

Generation of phenotypically distinct macrophage-hepatoma hybrid clones.

By fusion of C3H/HeJ splenic adherent mononuclear cells enriched for macrophages with HPRT-deficient C57L/J HH- hepatoma cells, we have generated six macrophage-hepatoma hybrid clones. The hybrid nature of isolated clones was demonstrated by karyotypic analysis. The hybrid clones were screened for macrophage properties by assaying the presence of two enzymes: nonspecific esterase and lysozyme. Three of six hybrids expressed higher amount of Ia antigen and less amount of FcR; the other three hybrids expressed higher amounts of Fcr, but no Ia antigen. Phagocytosis of serum-opsonized beads is positively correlated with FcR expression, while the proliferation of antigen-primed lymphocytes is only induced by antigen-pulsed hybrids expressing Ia antigen. One hybrid clone (MH3-1) secreted significantly higher level of PGE2 and also expressed Ia antigen with higher ability of antigen-presentation. The data suggest that the cell hybridization can segregate macrophage-featured phenotypes into different hybrid clones which perform distinct functions. It may facilitate the study on the relationship of macrophage functions and the relationship between the functions and defined cell structure.

Autologous red blood cells potentiate antibody synthesis by unfractionated human mononuclear cell cultures.

We have tried to determine the most favourable conditions for the in vitro induction of specific antibody (Ab) responses to tetanus toxoid (TT) and keyhole limpet haemocyanin (KLH). Human peripheral blood mononuclear cells (PBMNC) were obtained from normal volunteers and stimulated with PWM, TT, KLH, and mixtures of PWM and antigens in the presence or absence of autologous red blood cells (RBC) (1:50 ratio of PBMNC/RBC). The cultures were harvested on day 11; immunoglobulins were determined immunonephelometrically and Ab levels by ELISA with human antibodies used for calibration. While anti-TT responses were easy to induce with PBMNC from recently boosted individuals, the production of anti-TT from PBMNC obtained from non-recently boosted individuals was only possible when PBMNC were stimulated with TT and PWM in the presence of autologous RBC. Similarly, anti-KLH responses were easier to induce with PBMNC from an immune donor; maximal response was observed after stimulation with PWM + KLH in the presence of autologous RBC. Stimulation of primary anti-KLH responses with PBMNC from non-immune donors was only successful when the cells were stimulated with KLH + PWM in the presence of autologous RBC. The potentiation of human B-cell responses with autologous RBC can be abrogated by pretreatment of PBMNC with anti-CD2 antibodies and is associated with increased expression of IL-2 receptors and increased production of gamma interferon (IFN-gamma). However, addition of IFN-gamma in different doses and at different times to PWM-stimulated PBMNC cultures was not as effective as addition of RBC in enhancing the production of immunoglobulin and antibody.

Peripheral blood mononuclear leukocytes release a mediator(s) that induces phagocytosis of C-reactive protein-coated cells by polymorphonuclear leukocytes.

We studied the phagocytosis by human polymorphonuclear leukocytes (PMN) of sheep erythrocytes (E) passively sensitized with pneumococcal C-polysaccharide (E-PnC), E-PnC coated with C-reactive protein (E-PnC-CRP), and E coated with rabbit antisheep E IgG (E-IgG). PMN isolated from the blood of normal individuals failed to ingest either E-PnC or E-PnC-CRP; however, after incubation with supernatants from stimulated peripheral blood mononuclear leukocytes, glass-adherent PMN ingested E-PnC-CRP with a mean phagocytic index of 47.8 +/- 14.9 (means +/- SD, n = 9) and E-PnC to a lesser extent with a phagocytic index of 9.6 +/- 2.9 (mean +/- SD, n = 4). We also observed a statistically significant increase in the ingestion of E-IgG by lymphokine-stimulated PMN with phagocytic indices of 85.2 +/- 21.2 (mean +/- SD, n = 12) for unstimulated PMN and 158 +/- 37.1 (mean +/- SD, n = 9) for stimulated PMN. The best conditions for stimulating release of this phagocytosis-promoting mediator included exposure to phytohemagglutinin (PHA) in the presence of monocytes that had ingested IgG-coated sheep erythrocytes. The induction of phagocytosis of E-PnC-CRP was rapid, reaching a maximal level after stimulation of the adherent PMN with the conditioned media for 30 min. The factor(s) responsible for the induction of the ingestion of E-PnC-CRP was less than 10,000 daltons and was heat stable (56 degrees C for 45 min). These data are similar to earlier results obtained with PMN activated by 12-0-tetradecanoyl-phorbol-13-acetate (PMA), an important contrast being that in the current studies PMN were activated by a soluble factor released from stimulated mononuclear cells under conditions simulating those in vivo.

Treatment of acute serum sickness by plasma ultrafiltration.

The clinical value of a scaled-down prototype of an extracorporeal plasma ultrafiltration system for the treatment of acute serum sickness in rabbits was examined. The system uses two filters: the primary separates red cells from plasma, and the secondary filter excludes high molecular weight proteins from plasma. In vitro and in vivo experiments showed that the secondary filters rejected substantially more IgM (80-90%) than IgG (10-30%) or albumin (10%) and totally rejected immune complexes (IC) prepared in vitro. Two groups of rabbits were submitted to either sham filtration with the primary filter only (n = 5), receiving back the remixed components of their blood, or to the complete ultrafiltration protocol (n = 7), with removal of high molecular weight proteins and IC. Several parameters were studied longitudinally, such as circulating IC, which appeared to rise more slowly in animals whose blood was ultrafiltered, and total proteinuria, which appeared to remain at lower levels in the same animals. Histologic examination of the kidneys, collected after killing, showed evidence of glomerular IC deposition in three of five sham-treated animals (a similar frequency to that observed in a separate group of five rabbits with acute serum sickness), while one of six treated animals had evidence of glomerular deposition of IC. These observations are tentative because of the small number of animals in each group, but are encouraging. Further studies with larger groups of animals are needed to determine whether the observed effects are reproducible and to better characterize the factors directly related to the removal of circulating IC.

Inhibition of platelet-activating factor by rabbit C-reactive protein.

Incubation of 0.5 nM platelet-activating factor (PAF) with 123 micrograms/ml C-reactive protein (CRP) for 60 min at room temperature in the presence of 2 mM calcium resulted in total inhibition of the platelet-aggregating capacity of 0.5 nM PAF. There was no inhibition of platelet aggregation if PAF and CRP were added to the platelets simultaneously, without prior incubation of the mixture, or if calcium was not present during the incubation of CRP and PAF. The inhibitory effect of CRP (123 micrograms/ml) was dependent upon the time of incubation with PAF (1 nM). We observed 0, 42, and 85% inhibition of platelet aggregation with incubations of 0, 20, and 120 min, respectively. The inhibitory effect of CRP was also concentration dependent. After incubations of 20 min at room temperature with 0.5 nM PAF there was no inhibition of PAF-induced platelet aggregation using 12.3 micrograms/ml CRP with the inhibition increasing to 73% using 123 micrograms/ml CRP. Pretreatment of the platelets with CRP (123 micrograms/ml) for 60 min before addition of PAF did not affect the induction of platelet aggregation by PAF. These observations indicate that CRP inhibits PAF and strongly suggest that binding of CRP to PAF, presumably to the phosphocholine moiety of PAF, is required for this inhibition.