Method for Including the Dynamic Fluctuations of a Protein in Computer-Aided Drug Design

Abstract

We have recently presented a new pharmacophore design method that allows for the incorporation of the inherent flexibility of a target active site. The flexibility of the enzymatic system is described by collecting many conformations of the uncomplexed protein; this ensemble of conformational states can come from a molecular dynamics (MD) simulation, multiple crystal structures, or many NMR structures. Binding sites for functional groups that complement the active site are determined through multiple-copy calculations. These calculations are conducted for each protein conformation, providing a large collection of potential binding sites. The Cartesian coordinates from each protein conformation are overlaid through RMS fitting of essential catalytic residues, and the pharmacophore model is described by binding regions that are conserved over many protein conformations. Previously, we developed a "dynamic" pharmacophore model for HIV-1 integrase using 11 conformations of the protein from an MD simulation; the MUSIC procedure was used to calculate binding positions for methanol molecules in each configuration of the active site. Here we present "static" pharmacophore models developed with a single conformation of the protein from two new crystal structures (standard protocol for multiple-copy methods). The static models do not perform as well as the previous dynamic model in fitting known inhibitors of HIV-1 integrase. To test the applicability of the dynamic pharmacophore method and the assumption that any reliable source of protein conformations is applicable, we have now developed a second dynamic pharmacophore model based on the two crystal structures also used for the development of the static models. Though the dynamic model based on the two crystal structures does not fit as many known inhibitors as the previous dynamic model, it is a significant improvement over the static models. Even better performance is expected with the addition of new crystal structures as they become available. However, it is notable that using only two structures leads to great improvement in the models.

Combining docking and molecular dynamic simulations in drug design

“…Even the use of just two crystal-structure models produced improved results over single-structure models. 103 McCammon and co-workers 104 introduced the so-called relaxed-complex scheme, which takes into account the possibility that a ligand may bind to only a few conformations of the receptor. A long MD simulation of the apo receptor is first conducted to sample extensively its conformational space, followed by the rapid docking of mini-libraries of candidate inhibitors against a large ensemble of snapshots.…”

Relation between native ensembles and experimental structures of proteins

“…An increasing amount of experimental evidence, such as the recognition of dissimilar ligands (47) and the enhancement of antibody affinity and specificity by making the unbound state more similar to the bound state (5), supports such a model. The use of ''dynamic'' pharmacophore models in such calculations has already led to improved results (48); these models have been derived from either MD (48) or several crystal structures chosen in a similar way to the HSP ensembles (49). Hence, HSP ensembles or ensemble-refined experimental structures (30) represent a possible alternative to MD simulations for the purpose of ensemble generation.…”

Electron transfer in bis(hydrazines), a critical test for application of the Marcus model

“…The coupling may be introduced, in fortunate cases, as a local coupling between the connecting atomic orbitals of donor and bridge and bridge and acceptor, multiplied by a transfer capability function that is a property of the bridge. 7 Nelsen's group has shown that the rate constants for intramolecular electron transfer within several aromatic-bridged bis(hydrazine) radical cations are accurately predicted using the simple Marcus twostate model with electron transfer parameters derived from their optical spectra. 8 However, for one case (that having 9,10-anthracenediyl as the bridge), intramolecular electron transfer was much faster than predicted from the two-state model, and the optical spectrum was anomalous.…”