In silico prediction of substrate properties for ABC-multidrug transporters

Area of Science:

• Pharmacology
• Computational Chemistry
• Drug Discovery

Background:

• ATP-binding cassette (ABC)-transporters, including ABCB1, ABCC1, and ABCG2, are crucial in cancer multi-drug resistance (MDR).
• These transporters significantly influence drug pharmacokinetics, affecting bioavailability, distribution, and blood-brain barrier penetration.

Purpose of the Study:

• To review computational methods for predicting ABC-transporter substrate properties of drug-like compounds.
• To discuss the development of models for selectivity profiling of ABCB1 and ABCG2 ligands.

Main Methods:

• Overview of diverse computational approaches, including linear discriminant analysis, pharmacophore modeling, and machine learning algorithms.
• Evaluation of model performance within specific chemical spaces.

Main Results:

• Various computational models demonstrate satisfactory predictive performance for defined chemical spaces.
• General applicability across the entire drug-like chemical space remains a challenge for current models.
• Initial efforts in selectivity profiling for ABCB1 and ABCG2 are presented.

Conclusions:

• Computational tools are advancing for predicting ABC-transporter interactions, aiding drug development.
• Further validation is required to ensure generalizability of these predictive models.
• These approaches can facilitate ADMET profiling and guide the design of new therapeutics.