Expression and functional characterization of two natural heparin-binding site variants of antithrombin.

Essentials Heparin-binding site (HBS) variants of antithrombin (AT) are associated with thrombosis risk. HSB variants have, in general, normal progressive inhibitory activity but reduced heparin affinity. Thrombosis in HSB carriers has been primarily attributed to the loss of heparin cofactor activity. Results here demonstrate that HSB variants of AT also lack anti-inflammatory signaling functions. SUMMARY: Background Several heparin-binding site (HBS) variants of antithrombin (AT) have been identified that predispose carriers to a higher incidence of thrombosis. Thrombosis in carriers of HBS variants has been primarily attributed to a loss in their heparin-dependent anticoagulant function. Objective The objective of this study was to determine whether HSB mutations affect the anti-inflammatory functions of variants. Methods Two HBS variants of AT (AT-I7N and AT-L99F), which are known to be associated with a higher incidence of thrombosis, were expressed in mammalian cells and purified to homogeneity. These variants were characterized by kinetic assays followed by analysis of their activities in established cellular and/or in vivo inflammatory models. The possible effects of mutations on AT structure were also evaluated by molecular modeling. Results The results indicated that, whereas progressive inhibitory activities of variants were minimally affected, their heparin affinity and inhibitory activity in the presence of heparin were markedly decreased. Unlike wild-type AT, neither AT variant was capable of inhibiting activation of nuclear factor-κB or downregulation of expression of cell adhesion molecules in response to lipopolysaccharide (LPS). Similarly, neither variant elicited barrier protective activity in response to LPS. Structural analysis suggested that the L99F substitution locally destabilizes AT structure. Conclusions It is concluded that the L99F mutation of AT is associated with destabilization of the serpin structure, and that the loss of anti-inflammatory signaling function of the HBS variants may also contribute to enhanced thrombosis in carriers of HBS mutations.

Combining bioinformatics, chemoinformatics and experimental approaches to design chemical probes: Applications in the field of blood coagulation.

Bioinformatics and chemoinformatics approaches contribute to the discovery of novel targets, chemical probes, hits, leads and medicinal drugs. A vast repertoire of computational methods has indeed been reported over the years and in this review, I will briefly introduce some concepts and approaches, namely the analysis of potential therapeutic target binding pockets, the preparation of compound collections and virtual screening. An example of application is provided for two proteins acting in the blood coagulation system. Overall, in silico methods have been shown to improve R and D productivity in both, academic settings and in the private sector, if they are integrated in a rational manner with experimental approaches. However, integration of tools and pluridisciplinarity are seldom achieved. Efforts should be done in this direction as pluridisciplinarity and a true acknowledgment of all the contributing actors along the value chain could enhance innovation and reduce skyrocketing costs.

Theileria parasites secrete a prolyl isomerase to maintain host leukocyte transformation.

Infectious agents develop intricate mechanisms to interact with host cell pathways and hijack their genetic and epigenetic machinery to change host cell phenotypic states. Among the Apicomplexa phylum of obligate intracellular parasites, which cause veterinary and human diseases, Theileria is the only genus that transforms its mammalian host cells. Theileria infection of bovine leukocytes induces proliferative and invasive phenotypes associated with activated signalling pathways, notably JNK and AP-1 (ref. 2). The transformed phenotypes are reversed by treatment with the theilericidal drug buparvaquone. We used comparative genomics to identify a homologue of the peptidyl-prolyl isomerase PIN1 in T. annulata (TaPIN1) that is secreted into the host cell and modulates oncogenic signalling pathways. Here we show that TaPIN1 is a bona fide prolyl isomerase and that it interacts with the host ubiquitin ligase FBW7, leading to its degradation and subsequent stabilization of c-JUN, which promotes transformation. We performed in vitro and in silico analysis and in vivo zebrafish xenograft experiments to demonstrate that TaPIN1 is directly inhibited by the anti-parasite drug buparvaquone (and other known PIN1 inhibitors) and is mutated in a drug-resistant strain. Prolyl isomerization is thus a conserved mechanism that is important in cancer and is used by Theileria parasites to manipulate host oncogenic signalling.

1,2,4-Oxadiazoles identified by virtual screening and their non-covalent inhibition of the human 20S proteasome.

Although several constitutive proteasome inhibitors have been reported these recent years, potent organic, noncovalent and readily available inhibitors are still poorly documented. Here we used a structure- and ligand-based in silico approach to identify commercially available 1,2,4-oxadiazole derivatives as non-covalent human 20S proteasome inhibitors. Their optimization led to the newly synthesized compound 4h that is a mixed proteasomal inhibitor of the chymotrypsin- like activity (K(i) of 26,1 nM and K'(i) of 7.5 nM) which is in addition selective versus the challenging cathepsin B and calpain proteases. Molecular modelling studies corroborated the mechanism of inhibition and suggest an unusual binding of the inhibitor within the S5 binding pocket (ß6 subunit). The cellular effects of our compounds validate their utility as potential pharmacological agents for anti-cancer pre-clinical studies.

A leap into the chemical space of protein-protein interaction inhibitors.

Protein-protein interactions (PPI) are involved in vital cellular processes and are therefore associated to a growing number of diseases. But working with them as therapeutic targets comes with some major hurdles that require substantial mutations from our way to design drugs on historical targets such as enzymes and G-Protein Coupled Receptor (GPCR). Among the numerous ways we could improve our methodologies to maximize the potential of developing new chemical entities on PPI targets, is the fundamental question of what type of compounds should we use to identify the first hits and among which chemical space should we navigate to optimize them to the drug candidate stage. In this review article, we cover different aspects on PPI but with the aim to gain some insights into the specific nature of the chemical space of PPI inhibitors. We describe the work of different groups to highlight such properties and discuss their respective approach. We finally discuss a case study in which we describe the properties of a set of 115 PPI inhibitors that we compare to a reference set of 1730 enzyme inhibitors. This case study highlights interesting properties such as the unfortunate price that still needs to be paid by PPI inhibitors in terms of molecular weight, hydrophobicity, and aromaticity in order to reach a critical level of activity. But it also shows that not all PPI targets are equivalent, and that some PPI targets can demonstrate a better druggability by illustrating the better drug likeness of their associated inhibitors.

AMMOS software: method and application.

Recent advances in computational sciences enabled extensive use of in silico methods in projects at the interface between chemistry and biology. Among them virtual ligand screening, a modern set of approaches, facilitates hit identification and lead optimization in drug discovery programs. Most of these approaches require the preparation of the libraries containing small organic molecules to be screened or a refinement of the virtual screening results. Here we present an overview of the open source AMMOS software, which is a platform performing an automatic procedure that allows for a structural generation and optimization of drug-like molecules in compound collections, as well as a structural refinement of protein-ligand complexes to assist in silico screening exercises.

Defining the structure of membrane-bound human blood coagulation factor Va.

BACKGROUND: Blood coagulation factor (F) Va is the essential protein cofactor to the serine protease FXa. Factor Va stimulates the thrombin-to-prothrombin conversion by the prothrombinase complex, by at least five orders of magnitude. Factor Va binds with very high affinity to phosphatidylserine containing phospholipid membranes, which allows the visualization of its membrane-bound state by transmission electron microscopy (EM). METHODS: In this paper we present an averaged three-dimensional structure of FVa molecules attached to phosphatidylserine containing lipid tubes, as determined by EM and single particle analysis. The low-resolution FVa three-dimensional structure is compared with the available atomic models for FVa. RESULTS: The experimental data are combined with the most suitable atomic model and a membrane-bound FVaEM model is proposed that best fits the protein density defined by EM. In the FVaEM model, the C1 and C2 membrane-binding domains are juxtaposed onto the membrane surface and the model geometries indicate a deeper insertion of both C domains into the lipid bilayer than has been previously suggested. CONCLUSION: The present structure is a first step towards a higher-resolution experimental structure of a human FVa molecule in its membrane-bound conformation, allowing the visualization of individual domains within FVa and its association with the membrane.

Frog: a FRee Online druG 3D conformation generator.

In silico screening methods based on the 3D structures of the ligands or of the proteins have become an essential tool to facilitate the drug discovery process. To achieve such process, the 3D structures of the small chemical compounds have to be generated. In addition, for ligand-based screening computations or hierarchical structure-based screening projects involving a rigid-body docking step, it is necessary to generate multi-conformer 3D models for each input ligand to increase the efficiency of the search. However, most academic or commercial compound collections are delivered in 1D SMILES (simplified molecular input line entry system) format or in 2D SDF (structure data file), highlighting the need for free 1D/2D to 3D structure generators. Frog is an on-line service aimed at generating 3D conformations for drug-like compounds starting from their 1D or 2D descriptions. Given the atomic constitution of the molecules and connectivity information, Frog can identify the different unambiguous isomers corresponding to each compound, and generate single or multiple low-to-medium energy 3D conformations, using an assembly process that does not presently consider ring flexibility. Tests show that Frog is able to generate bioactive conformations close to those observed in crystallographic complexes. Frog can be accessed at http://bioserv.rpbs.jussieu.fr/Frog.html.

Molecular models of the procoagulant factor VIIIa-factor IXa complex.

BACKGROUND: Formation of the intrinsic tenase complex is an essential event in the procoagulant reactions that lead to clot formation. The tenase complex is formed when the activated serine protease, Factor IXa (FIXa), and its cofactor Factor VIIIa (FVIIIa) assemble on a phospholipid surface to proteolytically convert the zymogen Factor X (FX) into its active form FXa. The physiological relevance of the tenase complex is evident in hemophilia A or B patients who present with bleeding disorders. OBJECTIVES: The purpose of this study was to establish three-dimensional (3D) models of the FVIIIa-FIXa complex. METHODS: First, we built two new theoretical models of FVIIIa via homology modeling, inter-domain docking and loop simulation algorithms as well as a model for FIXa. This was followed by pseudo-Brownian protein-protein docking in internal coordinates with the ICM (Internal Coordinates Mechanics) program between the two FVIIIa and the FIXa structures. RESULTS: Ten representative models of this complex are presented based on agreements with known experimental data and according to structural criteria. CONCLUSIONS: These novel 3D models will help guide future site directed mutagenesis aimed at improving the functionality of FVIIIa and/or FIXa and will contribute to a better understanding of the role of this macromolecular complex in the blood coagulation cascade.

RPBS: a web resource for structural bioinformatics.

RPBS (Ressource Parisienne en Bioinformatique Structurale) is a resource dedicated primarily to structural bioinformatics. It is the result of a joint effort by several teams to set up an interface that offers original and powerful methods in the field. As an illustration, we focus here on three such methods uniquely available at RPBS: AUTOMAT for sequence databank scanning, YAKUSA for structure databank scanning and WLOOP for homology loop modelling. The RPBS server can be accessed at http://bioserv.rpbs.jussieu.fr/ and the specific services at http://bioserv.rpbs.jussieu.fr/SpecificServices.html.

A critical role for Gly25 in the B chain of human thrombin.

We have recently identified (Akhavan S et al., Thromb Haemost 2000; 84: 989-97) a patient with a mild bleeding diathesis associated to an homozygous mutation in the thrombin B chain (Gly25Ser, chymotrypsinogen numbering, i.e. position 330 in human prothrombin numbering). Transient transfection of wild-type prothrombin (FII-WT) and mutant prothrombin (designated FII-G25(330)S) cDNA in COS-7 cells showed a mild reduction (50%) in FII-G25(330)S production. Recombinant proteins, stably expressed in Chinese hamster ovary cells, were isolated and activated by Taipan snake or Echis carinatus venoms. We show that the G25(330)S mutation results in a decrease in the rate of prothrombin proteolytic activation. The mutation also significantly decreases (i) the catalytic activity of thrombin with a 9-fold reduction in catalytic efficiency of the mutant toward S-2238; (ii) the interaction with benzamidine; (iii) the rate of inhibition by TLCK and antithrombin; and (iv) the rate of hydrolysis of macromolecular substrates (fibrinogen, protein C). In contrast, exosite I does not appear to be affected by the molecular defect. These results, together with molecular modeling and dynamics, indicate that Gly25(330) is important for proper expression and probably proper folding of prothrombin, and also plays a critical role in both the alignment of the catalytic triad and the flexibility of one of the activation segments of prothrombin.

Theoretical and experimental study of the D2194G mutation in the C2 domain of coagulation factor V.

Coagulation factor V (FV) is a large plasma glycoprotein with functions in both the pro- and anticoagulant pathways. In carriers of the so-called R2-FV haplotype, the FV D2194G mutation, in the C2 membrane-binding domain, is associated with low expression levels, suggesting a potential folding/stability problem. To analyze the molecular mechanisms potentially responsible for this in vitro phenotype, we used molecular dynamics (MD) and continuum electrostatic calculations. Implicit solvent simulations were performed on the x-ray structure of the wild-type C2 domain and on a model of the D2194G mutant. Because D2194 is located next to a disulfide bond (S-S bond), MD calculations were also performed on S-S bond depleted structures. D2194 is part of a salt-bridge network and investigations of the stabilizing/destabilizing role of these ionic interactions were carried out. Five mutant FV molecules were created and the expression levels measured with the aim of assessing the tolerance to amino acid changes in this region of molecule. Analysis of the MD trajectories indicated increased flexibility in some areas and energetic comparisons suggested overall destabilization of the structure due to the D2194G mutation. This substitution causes electrostatic destabilization of the domain by approximately 3 kcal/mol. Together these effects likely explain the lowered expression levels in R2-FV carriers.

Molecular recognition in the protein C anticoagulant pathway.

The protein C (PC) anticoagulant system provides specific and efficient control of blood coagulation. The system comprises circulating or membrane-bound protein components that take part in complicated multimolecular protein complexes being assembled on specific cellular phospholipid membranes. Each of the participating proteins is composed of multiple domains, many of which are known at the level of their three-dimensional structures. The key component of the PC system, the vitamin K-dependent PC, circulates in blood as zymogen to an anticoagulant serine protease. Activation is achieved on the surface of endothelial cells by thrombin bound to the membrane protein thrombomodulin. The endothelial PC receptor binds the Gla domain of PC and stimulates the activation. Activated PC (APC) modulates the activity of blood coagulation by specific proteolytic cleavages of a limited number of peptide bonds in factor (F)VIIIa and FVa, cofactors in the activation of FX and prothrombin, respectively. These reactions occur on the surface of negatively charged phospholipid membranes and are stimulated by the vitamin K-dependent protein S. Regulation of FVIIIa activity by APC is stimulated not only by protein S but also by FV, which, like thrombin, is a Janus-faced protein with both pro- and anticoagulant potential. However, whereas the properties of thrombin are modulated by protein-protein interactions, the specificity of FV function is governed by proteolysis by pro- or anti-coagulant enzymes. The molecular recognition of the PC system is beginning to be unravelled and provides insights into a fascinating and intricate molecular scenario.

Functional analysis of the EGF-like domain mutations Pro55Ser and Pro55Leu, which cause mild hemophilia B.

We studied the functional role of two mutations, Pro55Ser and Pro55Leu, located in the N-terminal Epidermal Growth Factor-like domain (EGF1) of coagulation factor (F) IX. Both mutations cause mild hemophilia B with habitual FIX coagulant activities of 10-12% and FIX antigen levels of 50%. We found that activation by FVIIa/TF and FXIa was normal for FIXPro55Ser, but resulted in proteolysis of FIXPro55Leu at Arg318-Ser319 with a concomitant loss of amidolytic activity, suggesting intramolecular communication between EGF1 and the serine protease domain in FIX. This was further supported by experiments using an anti-EGF1 monoclonal antibody. Activation of FX by FIXaPro55Ser was impaired in both the presence and the absence of phospholipid or FVIIIa, indicating that Pro55 is not directly involved in binding to FVIIIa. We also studied the effect of the two Pro55 mutations on Ca2+ affinity and found only small changes. Thus, the Pro55Ser mutation causes hemophilia primarily through to an impaired ability to activate FX whereas at least in vitro the Pro55Leu defect interferes with the activation of FIX.

Proposed lipocalin fold for apolipoprotein M based on bioinformatics and site-directed mutagenesis.

Apolipoprotein M (apoM) is a novel apolipoprotein that is predominantly present in high-density lipoprotein. Sensitive sequence searches, threading and comparative model building experiments revealed apoM to be structurally related to the lipocalin protein family. In a 3D model, characterized by an eight-stranded anti-parallel beta-barrel, a segment including Asn135 could adopt a closed or open conformation. Using site-directed mutagenesis, we demonstrated Asn135 in wild-type apoM to be glycosylated, suggesting that the segment is solvent exposed. ApoM displays two strong acidic patches of potential functional importance, one around the N-terminus and the other next to the opening of the beta-barrel.

Structural and energetic characteristics of the heparin-binding site in antithrombotic protein C.

Human activated protein C (APC) is a key component of a natural anticoagulant system that regulates blood coagulation. In vivo, the catalytic activity of APC is regulated by two serpins, alpha1-antitrypsin and the protein C inhibitor (PCI), the inhibition by the latter being stimulated by heparin. We have identified a heparin-binding site in the serine protease domain of APC and characterized the energetic basis of the interaction with heparin. According to the counter-ion condensation theory, the binding of heparin to APC is 66% ionic in nature and comprises four to six net ionic interactions. To localize the heparin-binding site, five recombinant APC variants containing amino acid exchanges in loops 37, 60, and 70 (chymotrypsinogen numbering) were created. As demonstrated by surface plasmon resonance, reduction of the electropositive character of loops 37 and 60 resulted in complete loss of heparin binding. The functional consequence was loss in heparin-induced stimulation of APC inhibition by PCI, whereas the PCI-induced APC inhibition in the absence of heparin was enhanced. Presumably, the former observations were due to the inability of heparin to bridge some APC mutants to PCI, whereas the increased inhibition of certain APC variants by PCI in the absence of heparin was due to reduced repulsion between the enzymes and the serpin. The heparin-binding site of APC was also shown to interact with heparan sulfate, albeit with lower affinity. In conclusion, we have characterized and spatially localized the functionally important heparin/heparan sulfate-binding site of APC.

Secondary substrate-binding exosite in the serine protease domain of activated protein C important for cleavage at Arg-506 but not at Arg-306 in factor Va.

Proteolytic inactivation of activated factor V (FVa) by activated protein C (APC) is a key reaction in the regulation of hemostasis. We now demonstrate the importance of a positive cluster in loop 37 of the serine protease (SP) domain of APC for the degradation of FVa. Lysine residues in APC at positions 37, 38, and 39 form a secondary binding site for FVa, which is important for cleavage of FVa at Arg-506 while having no effect on Arg-306 cleavage. In contrast, topological neighbors Lys-62, Lys-63, and Arg-74 in APC appear of minor importance in FVa degradation. This demonstrates that secondary binding exosites of APC specifically guide the proteolytic action of APC, resulting in a more favorable degradation of the 506-507 peptide bond as compared with the 306-307 bond.

Three-dimensional model of the SHBG-like region of anticoagulant protein S: new structure-function insights.

Protein S (PS) is a vitamin K-dependent glycoprotein that consists of several modules including a C-terminal sex hormone-binding globulin (SHBG)-like domain that has been subdivided into two laminin LG-type domains. The SHBG-like region of PS is known to bind to a complement regulator molecule, C4b-binding protein (C4BP), coagulation factor Va (FVa) and receptor tyrosine kinases. Inherited PS deficiency has been associated with thromboembolic disease. Yet, study of the mechanisms by which the SHBG-like region of PS serves its essential functions has so far been hampered because of the lack of structural information. Recently, the three-dimensional (3D) structure of LG domains from plasma SHBG, laminin and neurexin have been reported and were found related to the pentraxin family. We used these X-ray structures to build homology models of the SHBG-like region of human PS. We then analyzed previously reported experimental/clinical data in the light of the predicted structures. A potential calcium-binding site is found in the first LG domain of PS and D292 could play a role in this process. This region is close to the interface between the two LG domains and is also surrounded by segments that have been suggested by synthetic peptide studies to be important for C4BP or FVa binding. The 39 point mutations linked to PS deficiencies or reported as neutral variants were rationalized in the 3D structure. Proteins 2001;43:203-216.

Screening the molecular surface of human anticoagulant protein C: a search for interaction sites.

Protein C (PC), a 62 kDa multi-modular zymogen, is activated to an anticoagulant serine protease (activated PC or APC) by thrombin bound to thrombomodulin on the surface of endothelial cells. PC/APC interacts with many proteins and the characterisation of these interactions is not trivial. However, molecular modelling methods help to study these complex biological processes and provide basis for rational experimental design and interpretation of the results. PC/APC consists of a Gla domain followed by two EGF modules and a serine protease domain. In this report, we present two structural models for full-length APC and two equivalent models for full-length PC, based on the X-ray structures of Gla-domainless APC and of known serine protease zymogens. The overall elongated shape of the models is further cross-validated using size exclusion chromatography which allows evaluation of the Stokes radius (rs for PC = 33.15 A; rs for APC = 34.19 A), frictional ratio and axial ratio. We then propose potential binding sites at the surface of PC/APC using surface hydrophobicity as a determinant of the preferred sites of intermolecular recognition. Most of the predicted binding sites are consistent with previously reported experimental data, while some clusters highlight new regions that should be involved in protein-protein interactions.

Localization of a hydrophobic binding site for anticoagulant protein S on the beta -chain of complement regulator C4b-binding protein.

C4b-binding protein (C4BP) is a plasma glycoprotein involved in regulation of the complement system. C4BP consists of seven alpha-chains and one unique beta-chain, all constructed of repeating complement control protein (CCP) modules. The beta-chain, made up of three CCPs, binds tightly to vitamin K-dependent protein S, a cofactor to anticoagulant activated protein C. When bound to C4BP, protein S loses its activated protein C cofactor function. In this study, we have mutated potentially important amino acids located at the surface of CCP1 of the beta-chain to probe the protein S-C4BP interaction. The substitutions were designed after analysis of a homology-based three-dimensional structure of the beta-chain and were L27T/F45Q, I16S/V18S, V31T/I33N, I16S/V18S/V31T/I33N, L38S/V39S, and K41E/K42E. The mutants were expressed in a prokaryotic system, purified using an N-terminal His-tag, refolded using an oxido-shuffling system, and tested in several assays for their ability to bind protein S. Our data define Ile(16), Val(18), Val(31), and Ile(33) as crucial for protein S binding, with secondary effects from Leu(38) and Val(39). In addition, Lys(41) and Lys(42) contribute slightly to the interaction. Our results further confirm that surface hydrophobicity analysis may be used to identify ligand recognition sites.