Computational analysis of the binding of P1 variants of domain 3 of turkey ovomucoid inhibitor to Streptomyces griseus protease B

1.2A-resolution crystal structures reveal the second tetrahedral intermediates of streptogrisin B (SGPB)

Streptogrisin B (SGPB) has served as one of the models for studying the catalytic activities of serine peptidases. Here we report its native crystal structures at pH 4.2 at a resolution of 1.18A, and at pH 7.3 at a resolution of 1.23A. Unexpectedly, outstanding electron density peaks occurred in the active site and the substrate-binding region of SGPB in the computed maps at both pHs. The densities at pH 4.2 were assigned as a tetrapeptide, Asp-Ala-Ile-Tyr, whereas those at pH 7.3 were assigned as a tyrosine molecule and a leucine molecule existing at equal occupancies in both of the SGPB molecules in the asymmetric unit. Refinement with relaxed geometric restraints resulted in molecular structures representing mixtures of the second tetrahedral intermediates and the enzyme-product complexes of SGPB existing in a pH-dependent equilibrium. Structural comparisons with the complexes of SGPB with turkey ovomucoid third domain (OMTKY3) and its variants have shown that, upon the formation of the tetrahedral intermediate, residues Glu192A to Gly193 of SGPB move towards the alpha-carboxylate O of residue P1 of the bound species, and adjustments in the side-chain conformational angles of His57 and Ser195 of SGPB favor the progression of the catalytic mechanism of SGPB.

The Solution Structure of a Chimeric LEKTI Domain Reveals a Chameleon Sequence

The conversion of an alpha-helical to a beta-strand conformation and the presence of chameleon sequences are fascinating from the perspective that such structural features are implicated in the induction of amyloid-related fatal diseases. In this study, we have determined the solution structure of a chimeric domain (Dom1PI) from the multidomain Kazal-type serine proteinase inhibitor LEKTI using multidimensional NMR spectroscopy. This chimeric protein was constructed to investigate the reasons for differences in the folds of the homologous LEKTI domains 1 and 6 [Lauber, T., et al. (2003) J. Mol. Biol. 328, 205-219]. In Dom1PI, two adjacent phenylalanine residues (F28 and F29) of domain 1 were substituted with proline and isoleucine, respectively, as found in the corresponding P4' and P5' positions of domain 6. The three-dimensional structure of Dom1PI is significantly different from the structure of domain 1 and closely resembles the structure of domain 6, despite the sequence being identical to that of domain 1 except for the two substituted phenylalanine residues and being only 31% identical to the sequence of domain 6. The mutation converted a short 3(10)-helix into an extended loop conformation and parts of the long COOH-terminal alpha-helix of domain 1 into a beta-hairpin structure. The latter conformational change occurs in a sequence stretch distinct from the region containing the substituted residues. Therefore, this switch from an alpha-helical structure to a beta-hairpin structure indicates a chameleon sequence of seven residues. We conclude that the secondary structure of Dom1PI is determined not only by the local protein sequence but also by nonlocal interactions.

Homologous proteins with different folds: the three-dimensional structures of domains 1 and 6 of the multiple Kazal-type inhibitor LEKTI

We have determined the solution structures of recombinant domain 1 and native domain 6 of the multi-domain Kazal-type serine proteinase inhibitor LEKTI using multi-dimensional NMR spectroscopy. While two of the 15 potential inhibitory LEKTI domains contain three disulfide bonds typical of Kazal-type inhibitors, the remaining 13 domains have only two of these disulfide bridges. Therefore, they may represent a novel type of serine proteinase inhibitor. The first and the sixth LEKTI domain, which have been isolated from human blood ultrafiltrate, belong to this group. In spite of sharing the same disulfide pattern and a sequence identity of about 35% from the first to the fourth cysteine, the two proteins show different structures in this region. The three-dimensional structure of domain 6 consists of two helices and a beta-hairpin structure, and closely resembles the three-dimensional fold of classical Kazal-type serine proteinase inhibitors including the inhibitory binding loop. Domain 6 has been shown to be an efficient, but non-permanent serine proteinase inhibitor. The backbone geometry of its canonical loop is not as well defined as the remaining structural elements, providing a possible explanation for its non-permanent inhibitory activity. We conclude that domain 6 belongs to a subfamily of classical Kazal-type inhibitors, as the third disulfide bond and a third beta-strand are missing. The three-dimensional structure of domain 1 shows three helices and a beta-hairpin, but the central part of the structure differs remarkably from that of domain 6. The sequence adopting hairpin structure in domain 6 exhibits helical conformation in domain 1, and none of the residues within the putative P3 to P3' stretch features backbone angles that resemble those of the canonical loop of known proteinase inhibitors. No proteinase has been found to be inhibited by domain 1. We conclude that domain 1 adopts a new protein fold and is no canonical serine proteinase inhibitor.

Computational analysis of binding of P1 variants to trypsin

The binding of P1 variants of bovine pancreatic trypsin inhibitor (BPTI) to trypsin has been investigated by means of molecular dynamics simulations. The specific interaction formed between the amino acid at the primary binding (P1) position of the binding loop of BPTI and the specificity pocket of trypsin was estimated by use of the linear interaction energy (LIE) method. Calculations for 13 of the naturally occurring amino acids at the P1 position were carried out, and the results obtained were found to correlate well with the experimental binding free energies. The LIE calculations rank the majority of the 13 variants correctly according to the experimental association energies and the mean error between calculated and experimental binding free energies is only 0.38 kcal/mole, excluding the Glu and Asp variants, which are associated with some uncertainties regarding protonation and the possible presence of counter-ions. The three-dimensional structures of the complex with three of the P1 variants (Asn, Tyr, and Ser) included in this study have not at present been solved by any experimental techniques and, therefore, were modeled on the basis of experimental data from P1 variants of similar size. Average structures were calculated from the MD simulations, from which specific interactions explaining the broad variation in association energies were identified. The present study also shows that explicit treatment of the complex water-mediated hydrogen bonding network at the protein-protein interface is of crucial importance for obtaining reliable binding free energies. The successful reproduction of relative binding energies shows that this type of methodology can be very useful as an aid in rational design and redesign of biologically active macromolecules.

Design, docking, and evaluation of multiple libraries against multiple targets

We present a general approach to the design, docking, and virtual screening of multiple combinatorial libraries against a family of proteins. The method consists of three main stages: docking the scaffold, selecting the best substituents at each site of diversity, and comparing the resultant molecules within and between the libraries. The core "divide-and-conquer" algorithm for side-chain selection, developed from an earlier version (Sun et al., J Comp Aided Mol Design 1998;12:597-604), provides a way to explore large lists of substituents with linear rather than combinatorial time dependence. We have applied our method to three combinatorial libraries and three serine proteases: trypsin, chymotrypsin, and elastase. We show that the scaffold docking procedure, in conjunction with a novel vector-based orientation filter, reproduces crystallographic binding modes. In addition, the free-energy-based scoring procedure (Zou et al., J Am Chem Soc 1999;121:8033-8043) is able to reproduce experimental binding data for P1 mutants of macromolecular protease inhibitors. Finally, we show that our method discriminates between a peptide library and virtual libraries built on benzodiazepine and tetrahydroisoquinolinone scaffolds. Implications of the docking results for library design are explored.

Predicting the reactivity of proteins from their sequence alone: Kazal family of protein inhibitors of serine proteinases

An additivity-based sequence to reactivity algorithm for the interaction of members of the Kazal family of protein inhibitors with six selected serine proteinases is described. Ten consensus variable contact positions in the inhibitor were identified, and the 19 possible variants at each of these positions were expressed. The free energies of interaction of these variants and the wild type were measured. For an additive system, this data set allows for the calculation of all possible sequences, subject to some restrictions. The algorithm was extensively tested. It is exceptionally fast so that all possible sequences can be predicted. The strongest, the most specific possible, and the least specific inhibitors were designed, and an evolutionary problem was solved.

Predicting the reactivity of proteins from their sequence alone: Kazal family of protein inhibitors of serine proteinases

An additivity-based sequence to reactivity algorithm for the interaction of members of the Kazal family of protein inhibitors with six selected serine proteinases is described. Ten consensus variable contact positions in the inhibitor were identified, and the 19 possible variants at each of these positions were expressed. The free energies of interaction of these variants and the wild type were measured. For an additive system, this data set allows for the calculation of all possible sequences, subject to some restrictions. The algorithm was extensively tested. It is exceptionally fast so that all possible sequences can be predicted. The strongest, the most specific possible, and the least specific inhibitors were designed, and an evolutionary problem was solved.

What can the structures of enzyme-inhibitor complexes tell us about the structures of enzyme substrate complexes?

Proteinases perform many beneficial functions that are essential to life, but they are also dangerous and must be controlled. Here we focus on one of the control mechanisms: the ubiquitous presence of protein proteinase inhibitors. We deal only with a subset of these: the standard mechanism, canonical protein inhibitors of serine proteinases. Each of the inhibitory domains of such inhibitors has one reactive site peptide bond, which serves all the cognate enzymes as a substrate. The reactive site peptide bond is in a combining loop which has an identical conformation in all inhibitors and in all enzyme-inhibitor complexes. There are at least 18 families of such inhibitors. They all share the conformation of the combining loops but each has its own global three-dimensional structure. Many three-dimensional structures of enzyme-inhibitor complexes were determined. They are frequently used to predict the conformation of substrates in very short-lived enzyme-substrate transition state complexes. Turkey ovomucoid third domain and eglin c have a Leu residue at P(1). In complexes with chymotrypsin, these P(1) Leu residues assume the same conformation. The relative free energies of binding of P(1) Leu (relative to either P(1) Gly or P(1) Ala) are within experimental error, the same for complexes of turkey ovomucoid third domain, eglin c, P(1) Leu variant of bovine pancreatic trypsin inhibitor and of a substrate with chymotrypsin. Therefore, the P(1) Leu conformation in transition state complexes is predictable. In contrast, the conformation of P(1) Lys(+) is strikingly different in the complexes of Lys(18) turkey ovomucoid third domain and of bovine pancreatic trypsin inhibitor with chymotrypsin. The relative free energies of binding are also quite different. Yet, the relative free energies of binding are nearly identical for Lys(+) in turkey ovomucoid third domain and in a substrate, thus allowing us to know the structure of the latter. Similar reasoning is applied to a few other systems.