DynlogPA Novel High-Throughput LC−MS Assay Quantifying Dynamic Lipophilicity and Chameleonicity of bRo5 Macrocyclic Peptides

Abstract

The landscape of drug discovery has expanded beyond the traditional chemical space into the domain of large, flexible molecules, including beyond rule-of-five (bRo5) macrocyclic peptides (MCPs). Molecular chameleonicity, the capacity to dynamically adapt the conformation to the environment, balancing aqueous solubility and membrane permeability, has the potential to improve the oral bioavailability of these compounds. However, we lack experimental methods that quantify chameleonicity in a high-throughput manner. To address this need, we introduce DynlogP, a novel high-throughput, fully automated assay based on reverse-phase LC−MS that can quantify chameleonicity with minimal compound input. DynlogP measures calibrated lipophilicity under conditions promoting either solvent hydrogen bonding (log P₇₀) or intramolecular hydrogen bonding (log P₉₅), providing a quantitative measure of chameleonicity (Δlog P₉₅₋₇₀) that is validated with NMR and molecular dynamics simulations. We profiled hundreds of molecules with DynlogP and demonstrate that chameleonicity is a common feature of MCPs. Machine learning models trained with DynlogP data allow for the prospective prediction of chameleonicity. Adoption and further refinement of DynlogP have the potential to empower compound prioritization and the design of orally bioavailable MCP-based therapeutics.