Trypsin mutants for structure-based drug design: expression, refolding and crystallisation

The DINGO dataset: a comprehensive set of data for the SAMPL challenge

Part of the latest SAMPL challenge was to predict how a small fragment library of 500 commercially available compounds would bind to a protein target. In order to assess the modellers' work, a reasonably comprehensive set of data was collected using a number of techniques. These included surface plasmon resonance, isothermal titration calorimetry, protein crystallization and protein crystallography. Using these techniques we could determine the kinetics of fragment binding, the energy of binding, how this affects the ability of the target to crystallize, and when the fragment did bind, the pose or orientation of binding. Both the final data set and all of the raw images have been made available to the community for scrutiny and further work. This overview sets out to give the parameters of the experiments done and what might be done differently for future studies.

Practical aspects of the SAMPL challenge: providing an extensive experimental data set for the modeling community

To provide an experimental basis for a comprehensive molecular modeling evaluation study, 500 fragments from the Maybridge fragment library were soaked into crystals of bovine pancreatic trypsin and the structures determined by X-ray crystallography. The soaking experiments were performed in both single and pooled aliquots to determine if combination of fragments is an appropriate strategy. A further set of data was obtained from co-crystallizing the pooled fragments with the protein. X-ray diffraction data were collected on approximately 1000 crystals at the Australian Synchrotron, and these data were subsequently processed, and the preliminary analysis was performed with a custom software application (Jigsaw), which combines available software packages for structure solution and analysis.

Understanding binding selectivity toward trypsin and factor Xa: the role of aromatic interactions

A congeneric series of four bis-benzamidine inhibitors sharing a dianhydrosugar isosorbide scaffold in common has been studied by crystal structure analysis and enzyme kinetics with respect to their binding to trypsin and factor Xa. Within the series, aromatic interactions are an important determinant for selectivity discrimination among both serine proteases. To study the selectivity-determining features in detail, we used trypsin mutants in which the original binding site is gradually substituted to finally resemble the factor Xa binding pocket. The influence of these mutations has been analyzed on the binding of the closely related inhibitors. We present the crystal structures of the inhibitor complexes obtained by co-crystallizing an "intermediate" trypsin mutant. They could be determined to a resolution of up to 1.2 A, and we measured the inhibitory activity (K(i)) of each ligand against factor Xa, trypsin, and the various mutants. From these data we were able to derive a detailed structure-activity relationship which demonstrates the importance of aromatic interactions in protein-ligand recognition and their role in modulating enzyme selectivity. Pronounced preference is experienced to accommodate the benzamidine anchor with meta topology in the S(1) specificity pocket. One ligand possessing only para topology deviates strongly from the other members of the series and adopts a distinct binding mode addressing the S(1)' site instead of the distal S(3)/S(4) binding pocket.

Engineering protein allostery: 1.05 A resolution structure and enzymatic properties of a Na+-activated trypsin

Some trypsin-like proteases are endowed with Na(+)-dependent allosteric enhancement of catalytic activity, but this important mechanism has been difficult to engineer in other members of the family. Replacement of 19 amino acids in Streptomyces griseus trypsin targeting the active site and the Na(+)-binding site were found necessary to generate efficient Na(+) activation. Remarkably, this property was linked to the acquisition of a new substrate selectivity profile similar to that of factor Xa, a Na(+)-activated protease involved in blood coagulation. The X-ray crystal structure of the mutant trypsin solved to 1.05 A resolution defines the engineered Na(+) site and active site loops in unprecedented detail. The results demonstrate that trypsin can be engineered into an efficient allosteric protease, and that Na(+) activation is interwoven with substrate selectivity in the trypsin scaffold.

Crystal Structures of tRNA-guanine Transglycosylase (TGT) in Complex with Novel and Potent Inhibitors Unravel Pronounced Induced-fit Adaptations and Suggest Dimer Formation Upon Substrate Binding

The bacterial tRNA-guanine transglycosylase (TGT) is a tRNA modifying enzyme catalyzing the exchange of guanine 34 by the modified base preQ1. The enzyme is involved in the infection pathway of Shigella, causing bacterial dysentery. As no crystal structure of the Shigella enzyme is available the homologous Zymomonas mobilis TGT was used for structure-based drug design resulting in new, potent, lin-benzoguanine-based inhibitors. Thorough kinetic studies show size-dependent inhibition of these compounds resulting in either a competitive or non-competitive blocking of the base exchange reaction in the low micromolar range. Four crystal structures of TGT-inhibitor complexes were determined with a resolution of 1.58-2.1 A. These structures give insight into the structural flexibility of TGT necessary to perform catalysis. In three of the structures molecular rearrangements are observed that match with conformational changes also noticed upon tRNA substrate binding. Several water molecules are involved in these rearrangement processes. Two of them demonstrate the structural and catalytic importance of water molecules during TGT base exchange reaction. In the fourth crystal structure the inhibitor unexpectedly interferes with protein contact formation and crystal packing. In all presently known TGT crystal structures the enzyme forms tightly associated homodimers internally related by crystallographic symmetry. Upon binding of the fourth inhibitor the dimer interface, however, becomes partially disordered. This result prompted further analyses to investigate the relevance of dimer formation for the functional protein. Consultation of the available TGT structures and sequences from different species revealed structural and functional conservation across the contacting residues. This suggests that bacterial and eukaryotic TGT could possibly act as homodimers in catalysis. It is hypothesized that one unit of the dimer performs the catalytic reaction whereas the second is required to recognize and properly orient the bound tRNA for the catalytic reaction.

Understanding Protein–Ligand Interactions: The Price of Protein Flexibility

In order to design selective, high-affinity ligands to a target protein, it is advantageous to understand the structural determinants for protein-ligand complex formation at the atomic level. In a model system, we have successively mapped the factor Xa binding site onto trypsin, showing that certain mutations influence both protein structure and inhibitor specificity. Our previous studies have shown that introduction of the 172SSFI175 sequence of factor Xa into rat or bovine trypsin results in the destabilisation of the intermediate helix with burial of Phe174 (the down conformation). Surface exposure of the latter residue (the up conformation) is critical for the correct formation of the aromatic box found in factor Xa-ligand complexes. In the present study, we investigate the influence of aromatic residues in position 174. Replacement with the bulky tryptophan (SSWI) shows reduced affinity for benzamidine-based inhibitors (1) and (4), whereas removal of the side-chain (alanine, SSAI) or exchange with a hydrophilic residue (arginine, SSRI) leads to a significant loss in affinity for all inhibitors studied. The variants could be crystallised in the presence of different inhibitors in multiple crystal forms. Structural characterisation of the variants revealed three different conformations of the intermediate helix and 175 loop in SSAI (down, up and super-up), as well as a complete disorder of this region in one crystal form of SSRI, suggesting that the compromised affinity of these variants is related to conformational flexibility. The influence of Glu217, peripheral to the ligand-binding site in factor Xa, was investigated. Introduction of Glu217 into trypsin variants containing the SSFI sequence exhibited enhanced affinity for the factor Xa ligands (2) and (3). The crystal structures of these variants also exhibited the down and super-up conformations, the latter of which could be converted to up upon soaking and binding of inhibitor (2). The improved affinity of the Glu217-containing variants appears to be due to a shift towards the up conformation. Thus, the reduction in affinity caused by conformational variability of the protein target can be partially or wholly offset by compensatory binding to the up conformation. The insights provided by these studies will be helpful in improving our understanding of ligand binding for the drug design process.

ZZ made EZ: influence of inhibitor configuration on enzyme selectivity

Selectivity of drug targeting is necessary in order to forestall undesired side-effects. Here, we examine the structural grounds for the configuration-dependent selectivity of 2,7-bis(4-amidinobenzylidene)-cycloheptan-1-one (1) for factor Xa and trypsin: Previous studies showed that factor Xa is preferentially inhibited by the (Z,Z) configuration isomer of (1), whilst trypsin binds equally well to both (E,Z) and (Z,Z) forms. Using engineered trypsin variants, we find similar overall binding modes for the (E,Z) and (Z,Z) isomers. Minor changes in van der Waals' contacts to Tyr99 (Leu in trypsin) explain the differential inhibition of factor Xa. We note differences in the experimental electron densities observed from co-crystallisation and soaking experiments: while the co-crystallisation of (1) with variants containing Tyr99 (Leu99) reveal the exclusive presence of the (Z,Z) ((E,Z)) configurations respectively, soaking experiments with either variant result in mixtures of (E,Z), (Z,Z) and (E,E). This discrepancy arises presumably from differences in the spatial (packing considerations) or chemical (crystallisation conditions) microenvironments. The results presented here represent an extreme example of the problems that face structure-based drug design, in particular the dangers inherent in relying on a single crystal structure for interpreting protein-ligand interactions.