Since the mid 1990s, more and more companies have turned to fragment-based lead discovery as an alternative to high-throughput screening to identify new lead compounds. The method aims to identify very small, low molecular weight, molecules that bind to the target protein with low affinity (high µM to mM). When the target protein has a well-defined binding site, growing the fragment can be an effective strategy to gain greater potency but when the target protein has neighboring shallow pockets, linking two or more weakly binding fragments may be a better approach. In the latter situation, the best possible outcome is that the binding affinity of individual fragments will not be compromised by the linker and that overall affinity will be improved by reducing rotational and translation entropy.
Often, however, the linked fragments bind differently to the individual fragments, and a study published in the journal Nature Chemical Biology has now shown that the nature of the tether may be as important as the individual fragments in determining overall affinity. The authors explored the energetic and structural effects of rigid and flexible linkers on the binding of a fragment-based inhibitor of human uracil DNA glycosylase and found that the flexibility and strain of a given linker can have a significant impact on binding affinity, even when the individual fragments are optimally positioned. The study explored the effect of variable length amine and oxime linkages between a uracil substrate fragment and random library fragments and provided unambiguous evidence that the tether is not just a passive agent for presenting the individual fragments. The linkers do not interact directly with the enzyme and differences in binding affinities were attributed largely to the conformational preferences of a given linker and how well it presents each fragment to the appropriate binding site. The observed effects were not apparent from an inspection of the structures and emphasize the importance of linker optimisation in fragment-based lead discovery.