Molecular Dynamics Simulation-Driven Focused Virtual Screening and Experimental Validation of Inhibitors for MTDH-SND1 Protein-Protein Interaction.

Protein-protein interactions (PPIs), in general, are attractive yet challenging drug targets. As a typical PPI, MTDH-SND1 interaction has recently been reported to be a promising drug target to malignant breast cancer and other cancer types. However, the lack of well-defined deep pockets on the MTDH-SND1 interface makes it a tough target for rational drug discovery attempts. To address this issue, in this study, a long time-scale molecular dynamics (MD) simulation-driven focused screening strategy was proposed and reported. A total of 12 virtual hits were purchased and tested in SPR assay, yielding 10 SND1 binders with micromolar or less affinities. As an example, compound L5, the second best hit with a KD of 2.64 µM, was further assayed in MDA-MB-231 breast cancer cells, showing an antiproliferation IC50 value of 57 µM in a CCK8 assay with a dampened interruption between MTDH and SND1 proteins detected by immunofluorescence colocalization imaging. As the most potent small molecule inhibitor in the class so far, our preliminary study combining molecular dynamics simulation and in vitro cellular functional evidence indicates L5 could serve as a lead compound for future optimization or pharmacologic studies, and the MD-driven focused screening strategy could be useful for other PPI drug discovery attempts.

General Charge Transfer Dipole Model for AMOEBA-Like Force Fields.

The development of highly accurate force fields is always an importance aspect in molecular modeling. In this work, we introduce a general damping-based charge transfer dipole (D-CTD) model to describe the charge transfer energy and the corresponding charge flow for H, C, N, O, P, S, F, Cl, and Br elements in common bio-organic systems. Then, two effective schemes to evaluate the charge flow from the corresponding induced dipole moment between the interacting molecules were also proposed and discussed. The potential applicability of the D-CTD model in ion-containing systems was also demonstrated in a series of ion-water complexes including Li+, Na+, K+, Mg2+, Ca2+, Fe2+, Zn2+, Pt2+, F-, Cl-, Br-, and I- ions. In general, the D-CTD model demonstrated good accuracy and good transferability in both charge transfer energy and the corresponding charge flow for a wide range of model systems. By distinguishing the intermolecular charge redistribution (charge transfer) under the influence of an external electric field from the accompanying intramolecular charge redistribution (polarization), the D-CTD model is theoretically consistent with current induced dipole-based polarizable dipole models and hence can be easily implemented and parameterized. Along with our previous work in charge penetration-corrected electrostatics, a bottom-up approach constructed water model was also proposed and demonstrated. The structure-maker and structure-breaker roles of cations and anions were also correctly reproduced using Na+, K+, Cl-, and I- ions in the new water model, respectively. This work demonstrates a cost-effective approach to describe the charge transfer phenomena. The water and ion models also show the feasibility of a modulated development approach for future force fields.