How does one really assess druggability of a particular target?
Besides the possession of clear disease relevance, druggability is another important factor in target selection within drug discovery. Druggability in the context of drug discovery is often reduced to the ability or the potential to be able to discover small-molecule binders to the particular target.
In terms of assessing this property, there are a few different methodologies, of which the first is the “druggability-by-association” criteria. This is based on close homology of the query target to other proteins that have been previously drugged, for example receptors within the same family. It is quite an easy to understand assessment, but using this criteria severely reduces druggable target space. The second method is guided by structure. Here the target’s binding site (orthosteric or allosteric) is assessed and judged to either have or not possess the requisite features that would allow for the discovery of small-molecule binders. Typically deep, enclosed pockets are ideally suited for ligand identification, as they offer a low dielectric environment for the often lipophilic small-molecules. In other words, a non-polar cavity with a few specific polar amino acids for ligand recognition lends itself to hit identification by a variety of methods, eg. many aminergic GPCRs such as the histamine, dopamine, and adrenergic receptors. On the other hand, cavities that are near the protein surface (solvent-exposed) and quite shallow constitute very hard targets. Often screens of large libraries of compounds, containing hundreds of thousands of molecules, yield no hits, eg. protein-protein interfaces. Such sites are labeled undruggable.
The advent of the FBLD paradigm, together with the development of highly sensitive screening techniques (NMR, SPR), have allowed for the discovery of low-affinity starting points to some of these targets. The idea is based on discovering binding hotspots on the cavity and once a small scaffold that utilizes the hotspot is identified, this can then be optimized to form higher affinity leads using a structure-guided approach. We shall strive to work on such challenging targets using the FBLD approach in order to make the undruggable, druggable!