Structure-Based Virtual Screening Allows the Identification of Efficient Modulators of E-Cadherin-Mediated Cell-Cell Adhesion.

Cadherins are a large family of transmembrane calcium-dependent cell adhesion proteins that orchestrate adherens junction formation and are crucially involved in tissue morphogenesis. Due to their important role in cancer development and metastasis, cadherins can be considered attractive targets for drug discovery. A recent crystal structure of the complex of a cadherin extracellular portion and a small molecule inhibitor allowed the identification of a druggable interface, thus providing a viable strategy for the design of cadherin dimerization modulators. Here, we report on a structure-based virtual screening approach that led to the identification of efficient and selective modulators of E-cadherin-mediated cell-cell adhesion. Of all the putative inhibitors that were identified and experimentally tested by cell adhesion assays using human pancreatic tumor BxPC-3 cells expressing both E-cadherin and P-cadherin, two compounds turned out to be effective in inhibiting stable cell-cell adhesion at micromolar concentrations. Moreover, at the same concentrations, one of them also showed anti-invasive properties in cell invasion assays. These results will allow further development of novel and selective cadherin-mediated cell-cell adhesion modulators for the treatment of a variety of cadherin-expressing solid tumors and for improving the efficiency of drug delivery across biological barriers.

Multiscale Simulations of a Two-Pore Potassium Channel.

Two-pore domain channels control cell excitability by modulating background potassium currents in response to several physicochemical stimuli. Thanks to the many crystal structures available, the TRAAK channel is one of the most studied, but little is known about its functional dynamics. Here, we explore TRAAK functionality complementing molecular dynamics with Brownian dynamics in a multiscale-modeling framework. We identify potential states of the channel that can prevent ion conduction, and we demonstrate that the suppression of currents is consistent with the presence of lipids inside the cavity.

Mechanistic insights into Pin1 peptidyl-prolyl cis-trans isomerization from umbrella sampling simulations.

The peptidyl-proyl isomerase Pin1 plays a key role in the regulation of phospho(p)-Ser/Thr-Pro proteins, acting as a molecular timer of the cell cycle. After recognition of these motifs, Pin1 catalyzes the rapid cis-trans isomerization of proline amide bonds of substrates, contributing to maintain the equilibrium between the two conformations. Although a great interest has arisen on this enzyme, its catalytic mechanism has long been debated. Here, the cis-trans isomerization of a model peptide system was investigated by means of umbrella sampling simulations in the Pin1-bound and unbound states. We obtained free energy barriers consistent with experimental data, and identified several enzymatic features directly linked to the acceleration of the prolyl bond isomerization. In particular, an enhanced autocatalysis, the stabilization of perturbed ground state conformations, and the substrate binding in a procatalytic conformation were found as main contributions to explain the lowering of the isomerization free energy barrier.

An automated docking protocol for hERG channel blockers.

A docking protocol aimed at obtaining a consistent qualitative and quantitative picture of binding for a series of hERG channel blockers is presented. To overcome the limitations experienced by standard procedures when docking blockers at hERG binding site, we designed a strategy that explicitly takes into account the conformations of the channel, their possible intrinsic symmetry, and the role played by the configurational entropy of ligands. The protocol was developed on a series of congeneric sertindole derivatives, allowing us to satisfactorily explain the structure-activity relationships for this set of blockers. In addition, we show that the performance of structure-based models relying on multiple-receptor conformations statistically increases when the protein conformations are chosen in such a way as to capture relevant structural features at the binding site. The protocol was then successfully applied to a series of structurally unrelated blockers.