Bend and Snap: Computationally Exploring the Helix Bias for Classifying Abl Kinase Allosteric Modulators

Description

Leite and colleagues combined three computational approaches to sort drug candidates that switch the Abl kinase enzyme on or off through a side pocket. That distinction matters for leukemia and breast cancer drug design. Compounds binding the pocket called the myristoyl pocket either bend a regulatory helix to inhibit the enzyme or leave it straighter to activate it. Computer classification of the two classes has been unreliable.

Leite benchmarked molecular docking, structure co-folding prediction, and full molecular dynamics simulations against known compounds. Co-folding predicted the helix bending correctly and separated true modulators from decoys far better than docking, with discrimination scores approaching perfect for inhibitors. Docking still revealed stabilizing fluorine contacts deep in the pocket, while dynamics showed the helix sampling many shapes and linked its bending to long-range restraint of the activation loop. The combined protocol gives a validated route to function-specific Abl drug design.

Catalog record

Source: https://europepmc.org/article/MED/PMC13509006