Quantum biochemistry description of PI3Kα enzyme bound to selective inhibitors

Abstract

The PI3K class I is composed of four PI3K isoforms that serve as regulatory enzymes governing cellular metabolism, proliferation, and survival. The hyperactivation of PI3Kα is observed in various types of cancer and is linked to poor prognosis. Unfortunately, the development inhibitors selectively targeting one of the isoforms remains challenging, with only few agents in clinical use. The main difficulty arises from the high conservation among residues at the ATP-binding pocket across isoforms, which also serves as target pocket for inhibitors. In this work, molecular dynamics and quantum calculations were performed to investigate the molecular features guiding the binding of selective inhibitors, alpelisib and GDC-0326, into the ATP-binding pocket of PI3Kα. While molecular dynamics allowed crystallographic coordinates to relax, the interaction eergy between each amino acid residues and inhibitors was obtained by combining the Molecular Fractionation with Conjugated Caps scheme with Density Functional Theory calculations. In addition, the atomic charge of ligands in the bound and unbound (free) was calculated. Results indicated that the most relevant residues for the binding of alpelisib are Ile932, Glu859, Val851, Val850, Tyr836, Met922, Ile800, and Ile848, while the most important residues for the binding of GDC-0326 are Ile848, Ile800, Ile932, Gln859, Glu849, and Met922. In addition, residues Trp780, Ile800, Tyr836, Ile848, Gln859 Val850, Val851, Ile932 and Met922 are common hotspots for both inhibitors. Overall, the results from this work contribute to improving the understanding of the molecular mechanisms controlling selectivity and highlight important interactions to be considered during the rational design of new agents.

Communicated by Ramaswamy H. Sarma

Keywords:

• Alpelisib
• GDC-0326
• PI3Kα-inhibitors
• phosphoinositide-3-kinase
• molecular fragmentation with conjugated caps
• quantum calculation

Author contributions

FJFS performed the calculations and analysis. FJFS, FFNA, and FLSO prepared figures and tables. GZ supervised the work and performed analysis. GZ and JVC wrote the initial manuscript. All authors discussed the results, modified the manuscript, and approved the definitive version.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by grant 437373/2018-5 from the National Council for Scientific and Technological Development [CNPq] (to GZ). FJFS and FFNA fellowships were supported by CAPES. FLS fellowship was supported by a UFC/PIBIC program (PIBIC 2019/2020 - Edital N° 7/2019).