Principal Investigator Jill Gready Project u52

Division of Biochemistry and Molecular Biology, Machine VP

John Curtin School of Medical Research

Co-Investigator Alain-Dominique Gorse

Division of Biochemistry and Molecular Biology, John Curtin School of Medical Research

Enzyme-ligand Docking and Design

The work is part of a drug design and development program originated at the University of Sydney and moved to JCSMR in May '95 on formation of the Computational Molecular Biology and Drug Design Group. The project consists mostly of molecular dynamics studies of docking of designed pteridine analogues into dihydrofolate reductase (DHFR). DHFR is a key enzyme in nucleic acid biosynthesis and a target for cytotoxic drugs. The overall goal of the current program is development of new DHFR inhibitors for possible use as anticancer or antimicrobial agents. An x-ray structure of a human DHFR ternary complex is being used for starting co-ordinates. Experiments to follow up predictions will be undertaken within a synthesis, enzymology and other biological testing effort in the Group and a collaboration for x-ray studies.

What are the basic questions addressed?

The "docking" problem is related to the location, the conformation and the relative affinity of bound ligands In the present project, the second and third aspects of the docking problem are investigated using energy-driven searching methods based on Molecular Dynamics (MD) simulations. The aim of this study is to determine the orientations of the deazapterin ring and the conformational preferences of groups appended to the deazapterin ring in a set of substituted deazapterin cations docked into the DHFR binding site.

What are the results to date and future of the work?

From the previous results, conditions of simulations for modelling the ligand in the DHFR active site have been refined. The methodology is based on the simulated annealing technique which involves "heating" and "cooling" the complex during the MD simulations in order to increase the efficiency of producing conformational transitions by overcoming larger potential energy barriers, and to relax towards conformations of low energies at the desired temperature.

Five binding sites for the 8-substituents have been considered. In the first two, the side chains of DHFR have been kept as they are in the x-ray structure, and modifications in the ligand positions, orientations and conformations have been undertaken in order to optimise the binding geometry, the van der Waals contact, while filling of the available space. In the last three, adjustments were also made in side-chain conformations of the enzyme residues. Using a new method based on standard thermodynamic cycles and on linear approximation of polar and non-polar free energy contributions from MD averages, binding affinities of the different ligands in each binding site are now being compared with the experimental dissociation constants. This will allow definition of structure-activity relationships, which will then lead to the design of new inhibitors of the DHFR.

At present a qualitative agreement between experimental and theoretical binding affinity is obtained, but the main limitations for a quantitative agreement are the treatment of the long-range electrostatic interactions, the use of a dynamic zone which does not allow the full flexibility of the DHFR molecule as in a periodic box, and the simulation time which should be long enough to reach a stable complex

What computational techniques are used and why is a supercomputer required?

The new AMBER 4.1 suite of programs vectorised for the VP2200 is suitable for such a quantitative drug-design project.


Molecular dynamics simulations of the docking of substituted N5-deazapterins to dihydrofolate reductase A.D. Gorse and J.E. Gready Prot. Eng., in press