Lipinski’s rule-of-five (Ro5) is considered the general guideline for cell-permeable, oral drug discovery. In some cases, efficient binding to a target is only achieved with a ligand which is beyond the Ro5 (bRo5), however. Analysis of the targets associated with a select set of bRo5 ligands revealed that many of these targets are also bound by Ro5 ligands, with greater or equal binding affinity. Due to the cost and time needed to test protein-ligand interactions experimentally, a computational method to determine when a bRo5 drug is advantageous for a target, over a Ro5 drug, is very appealing. To this end, the binding hot spots of each target were studied using the FTMap server, which distributes small organic probe molecules of different size, shape, and polarity on the surface of the protein to be studied, finds the most favorable positions for each probe type, clusters the probes, ranks the clusters on the basis of their average energy, and ultimately defines regions that bind several different probe clusters as the predicted binding hot spots. Next, all ligands which bind to the targets with known structure (ranging from 1 to 246 ligands) were analyzed relative to the FTMap-identified binding hot spots. The ligands bind the target’s hot spots in a variety of different ways. We identified the targets that have ligands with molecular weight and binding affinity which both positively correlate with the quantity of hot spots near the ligand – indicating a bRo5 ligand may be advantageous for the target. In these cases, larger ligands generally reach more hot spots and result in stronger measured binding affinities. The binding hot spots of such targets are generally located in elongated regions that make such extension favorable. Notably, there are some cases where molecular weight and/or binding affinity of the ligands do not correlate with the quantity of hot spots near the ligand, thus a bRo5 ligand may not be advantageous for these targets. Ligand based chemical tractability analysis using high affinity ligand datasets from ChEMBL are in line with these findings. A case-by case evaluation determines why extending to distant hot spots is not advantageous for binding affinity in such cases. For example, increasing the molecular weight of ligands is generally unfavorable if the hot spots are located in a compact cluster, and thus improving efficacy has to rely on optimizing protein-ligand interactions while retaining the size of the compound.

Computational study of proteins that bind beyond rule-of-five ligands

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.