The specific recognition of target molecules by proteins and nucleic acids is fundamental to all life processes. However, our current knowledge ofthe relationship between the structure ofthese macromolecules and the affinity oflig-and binding is very poor, a major limitation in structure based ligand design. In recent years the number of known structures of protein-ligand complexes has grown significantly, but for many of these the interaction has not been well characterised thermodynamically. Individual protein-ligand systems are

* To whom correspondence should be addressed. E-mail: [email protected]

in any case of limited use for developing an empirical model of the thermodynamics of protein-ligand interactions. Series of related ligands binding to a protein may be more instructive, since the structural and thermodynamic differences can (in theory) be correlated. Many of the series of protein-ligand complexes studied to date are ofj ust a small series of compounds, the ligand not being changed in a systematic fashion. Where structures of a large number of enzyme-inhibitor complexes have been solved, the ligands tend to be rather varied molecules developed as part of a medicinal chemistry programme. In such cases the mode of binding may vary considerably, making it difficult to determine the contributions of individual interactions to binding. The priority of a commercial drug discovery project is to determine whether and how tightly a given ligand has bound to the target. Finally, comparison of the thermodynamics of complexes from a variety of laboratories and systems is often impossible because the binding assays, and the conditions used, are so varied.

What is needed is a macromolecule that can form a very large variety of high affinity ligand complexes in which both the macromolecule and ligand adopt essentially invariant conformations. This will allow the design of a large number of ligands in order to remove or introduce new interactions with the host molecule and to measure the effect of these changes on the affinity of binding.

0 0

Post a comment