Molecular Recognition of Proteins and Ligand Design

点击翻译按钮获取中文摘要

Project Summary/Abstract The purpose of the research program is to develop improved computational methods for the simulation of biomolecular systems and to apply them to discover new drugs for treatment of human diseases, especially HIV/AIDS, inflammatory diseases, and cancer. The approach combines technology for computer-aided molecular design, synthetic organic chemistry, biological assaying, and structural biology, i.e., crystallographic determination of structures of the designed molecules bound to their protein targets by X-ray diffraction. The PI’s group develops and applies widely used force fields, which are at the heart of biomolecular modeling, and methods for computing free energy changes in solution. Discovery of initial active compounds (“hits”) is facilitated by virtual screening and by de novo design with the ligand-growing program BOMB. Optimization of the hits to yield potent, drug-like inhibitors is then guided by free-energy perturbation (FEP) calculations using Monte Carlo (MC) statistical mechanics and molecular dynamics (MD) simulations for the inhibitors and protein-inhibitor complexes in water. The viability of the approach has been well established through the discovery of numerous potent inhibitors for multiple proteins. It serves as a model for efficient drug discovery that is applicable to the pursuit of remedies for numerous diseases. The principal biomolecular targets are now macrophage migration inhibitory factor (MIF) and JAK2 kinase. Disruption of the cytokine signaling of MIF has known potential for treatment of inflammatory diseases and cancer, while reversal of the activating effect of the V617F mutation for JAK2 kinase is expected to provide remedies for the majority of myeloproliferative disorders. Organic molecules are being designed, synthesized, and tested to achieve these therapeutic goals. For MIF, substantial progress has been made with the discovery of compounds in two chemical series that bind extraordinarily tightly to the protein and inhibit the growth of prostate cancer cells. Lead optimization is also well along for JAK2 for which desired, selective binding to the pseudokinase JH2 domain instead of the JH1 kinase domain has been achieved. Additional exploration of these and new chemical series is planned to provide multiple, structurally diverse compounds that are suitable for preclinical development. For both targets, the progress and interpretation of activity data are greatly enhanced by the acquisition of high-resolution crystal structures of many protein-inhibitor complexes. In addition, there continue to be numerous technical advances for the force fields and for computing free energies of binding for the complexes, which are used to guide the selection of molecules to synthesize and test. The research program is particularly notable for the close interplay of state-of-the-art computation and experiment in one laboratory. The immediate feedback on the success of the computational predictions provides an important driving force to seek ever improving methodology.