A Linear Interaction Energy Model for Cavitand Host–Guest Binding Affinities

Abstract

Host-guest systems provide excellent models to explore molecular recognition in solution along with relevant technological applications from drug carriers to chemosensors. Here, we present a linear interaction energy (LIE) model to predict the binding affinity in host-guests with remarkable efficiency and predictive power. Using four host families, including cucurbiturils, octa acids, and β-cyclodextrin, and a large set (49) of chemically diverse guests, we demonstrate that binding-affinity predictions with a root mean square error <1.5 kcal/mol from experiments can be obtained with a few nanoseconds of molecular dynamics. The parameters of the LIE model are shown to be transferable among host-guest families, and the quality of the predictions is essentially force-field independent. Inclusion of the strain energy of the host in the bound state appears to be critically important to improve the quality of the predictions, particularly when the host and the guest have comparable sizes. Unsuccesful predictions for 28 additional highly charged and bulky guests to cucurbit[7]uril indicate future directions for improvement.

Assessment of host–guest molecular encapsulation of eugenol using β-cyclodextrin

“…In contrast, the E vdw increases with increasing temperature , thus indicating the presence of repulsion forces caused by the instability of the complex. Recently, a study demonstrated that the binding affinity in host–guest systems including β-cyclodextrin may be estimated with a root mean square error <1.5 kcal mol −1 from the experimental results using the LIE method ( Montalvo-Acosta et al, 2018 ), which is closely related to the error of <1 kcal mol −1 found in the experimental values obtained in the prediction of the relative binding affinity for a vast range of protein–ligand systems ( Gapsys et al, 2020 ).…”

Recent Developments in Free Energy Calculations for Drug Discovery

“…The performance of LIE in host-guest systems is also evaluated on 4 host families (cucurbiturils, octa acids, β-cyclodextrin) with an array of 49 chemically diverse guests. The base LIE is modified to include host strain energy, and parameters are found to be transferable between the different host systems, notably resulting in binding predictions with RMSE below 1.5 kcal/mol through only a few nanoseconds of simulation ( Montalvo-Acosta et al, 2018 ). Ngo et al estimate HIV-1 protease inhibitor binding affinities with a modified LIE that includes a polar interaction term obtained from PBE, training on 22 samples and testing on a set of 11 ligands demonstrates good performance with 1.25 kcal/mol RMSE and 0.83 Pearson correlation ( Ngo et al, 2020a ).…”

ChemFlow─From 2D Chemical Libraries to Protein–Ligand Binding Free Energies

“…A significant outcome of this analysis was that empirical scoring, which is used to rank the docking poses, can be theoretically improved via end-point free energy approaches such as MM-PBSA or LIE that explicitly account for configurational sampling and ligand desolvation upon binding. Although computationally more intensive, “free energy rescoring” of docking results is becoming increasingly more accessible, particularly after the advent of commodity GPU computing. , Recently, end-point approaches were shown to provide accurate binding free energy predictions in host–guest systems as compared to isothermal titration calorimetry (ITC) results. , …”

“…19,20 Recently, end-point approaches were shown to provide accurate binding free energy predictions in host− guest systems as compared to isothermal titration calorimetry (ITC) results. 21,22 ■ IMPLEMENTATION An efficient implementation of end-point free energy methods for virtual screening requires a significant degree of automation to prepare, dock, and rescore many ligand poses along with sufficient computer resources to run molecular dynamics (MD) simulations started from the docking results. These intrinsic difficulties have so far hindered the establishment of fully automated tools for virtual screening with free energy rescoring.…”

“… 19 , 20 Recently, end-point approaches were shown to provide accurate binding free energy predictions in host–guest systems as compared to isothermal titration calorimetry (ITC) results. 21 , 22 …”