Accelerating the drug discovery process requires predictive computational protocols capable of reducing or simplifying the synthetic and/or combinatorial challenge. Docking-based virtual screening methods have been developed and successfully applied to a number of pharmaceutical targets. In this review, we first present the current status of docking and scoring methods, with exhaustive lists of these. We next discuss reported comparative studies, outlining criteria for their interpretation. In the final section, we describe some of the remaining developments that would potentially lead to a universally applicable docking/scoring method.

No abstract

We report the development and validation of a novel suite of programs, FITTED 1.0, for the docking of flexible ligands into flexible proteins. This docking tool is unique in that it can deal with both the flexibility of macromolecules (side chains and main chains) and the presence of bridging water molecules while treating protein/ligand complexes as realistically dynamic systems. This software relies on a genetic algorithm to account for the flexibility of the two molecules as well as the location of bridging water molecules. In addition, FITTED 1.0 features a novel application of a switching function to retain or displace key water molecules from the protein-ligand complexes. Two independent modules, ProCESS and SMART, were developed to set up the proteins and the ligands prior to the docking stage. Validation of the accuracy of the software was achieved via the application of FITTED 1.0 to the docking of inhibitors of HIV-1 protease, thymidine kinase, trypsin, factor Xa, and MMP to their respective proteins.

In this contribution, we report that a self-assembled platinum molecular square [Pt(en)(4,4'-dipyridyl)]4 can act as an efficient G-quadruplex binder and telomerase inhibitor. Molecular modeling studies show that the square arrangement of the four bipyridyl ligands, the highly electropositive nature of the overall complex, as well as hydrogen bonding interactions between the ethylenediamine ligands and phosphates of the DNA backbone all contribute to the observed strong binding affinity to the G-quadruplex. Through thermal denaturation studies with duplex and quadruplex FRET probes and enzymatic assays, we demonstrate that this platinum square strongly binds to G-quadruplexes and can act as an inhibitor of telomerase. This study thus shows the potential of supramolecular self-assembly to readily generate scaffolds of unique geometries for effective targeting of G-quadruplexes and for the ultimate development of selective antitumor therapies.

Several modifications and additions to Fitted1.5 led to the development of Fitted2.6. Among the novel implementations are a matching algorithm-enhanced genetic algorithm and a ring conformational search algorithm. With these various optimizations, we also hoped to remove the biases and to develop a docking program that would provide results (i.e., poses) as independent as possible to the input ligand and protein conformations and used parameters, although keeping the options to provide additional experimental information. These biases were investigated within Fitted2.6 along with FlexX, GOLD, Glide, and Surflex. The input ligand conformation was found to have a major impact on the program accuracy as drops as large as 10-50% were observed with all the programs but Fitted. This comparative study also demonstrates that the accuracy of Fitted is similar to that of other widely used programs. We have also demonstrated that protein flexibility, displaceable water molecules, and ring conformational search algorithms, three of the main Fitted features, significantly increased its accuracy. Finally, we also proposed potential modifications to the available programs to further improve their accuracy in binding mode prediction.

Although much effort has been devoted to the development of programs suited for the docking of ligands to proteins, much less progress has been achieved in the nucleic acid field. We have developed a unique approach for docking aminoglycosides to RNA considering the flexibility of these macromolecules using conformational ensembles and accounting for the role of the first hydration shell. This concept, successfully implemented in AutoDock, relies on the computation of the intermolecular interaction energy that accounts for the presence of dynamically bound water molecules to the RNA. As an application, a set of 11 aminoglycosides was docked with an average root-mean-square deviation (RMSD) of 1.41 A to be compared with an average RMSD of 3.25 A when the original AutoDock protocol was used.

No abstract

Class II lanthipeptide synthetases (LanM enzymes) catalyze the multistep post-translational modification of genetically encoded precursor peptides into macrocyclic (often antimicrobial) lanthipeptides. The reaction sequence involves dehydration of serine/threonine residues, followed by intramolecular addition of cysteine thiols onto the nascent dehydration sites to construct thioether bridges. LanMs utilize two separate active sites in an iterative yet highly coordinated manner to maintain a remarkable level of regio-and stereochemical control over the multistep maturation. The mechanisms underlying this biosynthetic fidelity remain enigmatic. We recently demonstrated that proper function of the haloduracin β synthetase (HalM2) requires dynamic structural elements scattered across the surface of the enzyme. Here, we perform kinetic simulations, structural analysis of reaction intermediates, hydrogen−deuterium exchange mass spectrometry studies, and molecular dynamics simulations to investigate the contributions of these dynamic HalM2 structural elements to biosynthetic efficiency and fidelity. Our studies demonstrate that a large, conserved loop (HalM2 residues P349−P405) plays essential roles in defining the precursor peptide binding site, facilitating efficient peptide dehydration, and guiding the order of thioether ring formation. Moreover, mutations near the interface of the HalM2 dehydratase and cyclase domains perturb cyclization fidelity and result in aberrant thioether topologies that cannot be corrected by the wild type enzyme, suggesting an element of kinetic control in the normal cyclization sequence. Overall, this work provides the most comprehensive correlation of the structural and functional properties of a LanM enzyme reported to date and should inform mechanistic studies of the biosynthesis of other ribosomally synthesized and post-translationally modified peptide natural products.