Abstract
Cryptic pockets are visible in ligand-bound protein structures but are occluded in unbound structures. Utilizing these pockets in fragment-based drug-design provides an attractive option for proteins not tractable by classical binding sites. However, owing to their hidden nature, they are difficult to identify. Here, we show that small glycols find cryptic pockets on diverse set of proteins. Initial crystallography experiments serendipitously revealed the ability of ethylene glycol, a small glycol, to identify a cryptic pocket on W6A mutant of RBSX protein (RBSX-W6A). Explicit-solvent molecular dynamics (MD) simulations of RBSX-W6A with exposed-state of the cryptic pocket (ethylene glycol removed) revealed closure of the pocket reiterating that cryptic pockets in general prefer to stay in closed-state in absence of the ligands. Also, no change in the pocket was observed for simulations of RBSX-W6A with occluded-state of the cryptic pocket, suggesting that water molecules are not able to open the cryptic pocket. “Cryptic-pocket finding” potential of small glycols was then supported and generalized through additional crystallography experiments, explicit-cosolvent MD simulations, protein dataset construction and analysis. The cryptic pocket on RBSX-W6A was found again upon repeating the crystallography experiments with another small glycol, propylene glycol. Use of ethylene glycol as probe molecule in cosolvent MD simulations led to the enhanced sampling of the exposed-state of experimentally observed cryptic sites on test set of two proteins (Niemann-Pick C2, Interleukin-2). Further, analyses of protein structures with validated cryptic sites showed that ethylene glycol molecules binds to sites on proteins (G-actin, Myosin II, Bcl-xL, Glutamate receptor 2) which become apparent upon binding of biologically relevant ligands. Our study thus suggests potential application of the small glycols in experimental and computational fragment-based approaches to identify cryptic pockets in apparently undruggable and/or difficult targets, making these proteins amenable to drug-design strategies.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Databases: The atomic coordinates and structure factors of all the three structures have been deposited by us in Protein Data Bank (PDB) and are available under the accession codes 5EFD, 5XC0 and 5XC1.
Conflicts of interest: There are no conflicts of interest to declare.
