Decoding PI3K isoform selectivity through an ensemble-based framework integrating correlated motions, allosteric...

Area of Science:

• Computational chemistry
• Structural biology
• Pharmacology

Background:

• Developing selective phosphoinositide 3-kinase (PI3K) inhibitors for cancer therapy is difficult due to conserved ATP-binding sites.
• Broad PI3K inhibition causes significant toxicities, necessitating isoform-specific targeting.

Purpose of the Study:

• To establish a computational framework for identifying molecular determinants of PI3K isoform selectivity.
• To guide the rational design of novel, isoform-selective PI3K inhibitors.

Main Methods:

• Utilized molecular dynamics (MD) simulations, residue-network analysis, and hydration thermodynamics.
• Analyzed 100-ns all-atom simulations of PI3K isoforms (α, β, γ, δ).
• Applied network analysis and hydrogen bond residence time analysis to identify key residues and water interactions.

Main Results:

• Identified distinct 'dynamic signatures' rather than static pocket differences across PI3K isoforms.
• Discovered isoform-specific supercritical hubs acting as allosteric communication bottlenecks.
• Found PI3Kγ to be the most druggable, driven by water displacement and hydration thermodynamics, independent of pocket size.

Conclusions:

• PI3K isoform selectivity is governed by dynamic ensembles, allosteric pathways, and water-mediated interactions.
• The computational framework provides a hypothesis to design PI3K inhibitors targeting dynamic features for improved selectivity and reduced toxicity.