Abstract A comprehensive understanding of cyclodextrin-chiral drug interactions is crucial for efficient enantiomer separation. The chiral recognition mechanism of β-cyclodextrin (β-CD) toward propranolol was investigated using a multi-scale framework integrating molecular dynamics (MD), umbrella sampling (US) and density functional theory (DFT) calculations. The MD analysis showed that among the four binding modes, the R-type enantiomer adopting the isopropyl insertion mode (RIpr@CD) has the best structural stability. Subsequently, US simulations and the MM-PBSA showed that RIpr@CD achieved the optimal thermodynamic stability, with the most negative PMF (−34.84 kJ/mol) and MM-PBSA (−18.58 kJ/mol) values. Furthermore, the independent gradient model based on Hirshfeld partition (IGMH) analysis and electrostatic potential (ESP) mapping confirm that the optimal binding of RIpr@CD stems from the formation of more continuous van der Waals (vdW) interactions and stronger hydrogen bonds (H-bonds). These findings provide a fundamental understanding of the chiral recognition mechanisms mediated by various forces and establish a robust methodological framework for the rational design of efficient chiral separation systems.
Abstract Charge transfer (CT) is the delocalization of electron density from a donor to an acceptor. It is a crucial component of noncovalent interactions that significantly influences hydrogen bonding, ion solvation, and the structure of aqueous and biomolecular systems. Because direct experimental measurement of CT is challenging, theoretical methods rooted in quantum mechanics (QM) are essential. However, defining CT uniquely within QM and separating it from polarization remains a major methodological hurdle. This review surveys the theoretical methodologies developed to quantify CT, focusing primarily on water and ion–water dimers. We systematically examine various energy decomposition analysis (EDA) schemes, spanning variational methods and perturbation-theory approaches like symmetry-adapted perturbation theory (SAPT). We discuss the recommended levels of theory and basis sets for reliable CT estimates, giving particular attention to basis set dependence and convergence. Furthermore, we explore how these insights inform explicit CT energy expressions for classical molecular mechanics (MM) and molecular dynamics (MD) simulations, and evaluate efforts to incorporate CT into polarizable force fields and emerging machine learning models. Across representative systems, CT typically accounts for 5 to 20% of the total interaction energy, with larger contributions in strongly interacting ion–water complexes. Finally, we highlight the inherently nonadditive nature of CT, which is vital for accurately modeling complex condensed-phase phenomena, and outline the ongoing challenges in constructing transferable, physically grounded CT models for large-scale molecular simulations.
Abstract O6-methylguanine (O6-meG) is a highly mutagenic DNA lesion, yet the structural signals that trigger its repair remain unclear. We performed molecular dynamics simulations of a library of O6-meG:C and G:C base paired dsDNA and introduced a statistical framework to identify key characteristics that distinguish the lesion from undamaged DNA. The simulations indicate significantly increased bending flexibility of O6-meG containing strands and show that the structural signature is localized to the lesion and not spread over flanking residues. Both the lesion and its base pair are disrupted in a manner that is largely independent of sequence, creating detectable structural signatures on both strands. This bilateral detectability may provide repair enzymes with redundant structural cues. By identifying key features from a large pool of descriptors, the employed analysis protocol may be more widely useful.
Abstract Exon 19 mutations in the epidermal growth factor receptor (EGFR) kinase domain, including the common ΔE746-A750 deletion and the less frequent L747P substitution, are clinically important drivers of non-small cell lung cancer (NSCLC). Although these mutations occur within the β3-αC region, how they alter EGFR conformational dynamics and influence the response to the covalent inhibitor afatinib remains incompletely understood. Here, we investigated mutation-dependent changes in EGFR conformational ensembles, dual-mode of afatinib binding, and Cys797-mediated covalent reaction energetics in wild-type, ΔE746-A750, and L747P kinase domains. We find that both mutations preferentially stabilize active-like kinase conformations but do so through distinct structural mechanisms. The ΔE746-A750 deletion produces greater P-loop and ATP-pocket conformational heterogeneity, whereas L747P restricts αC-helix and activation-loop dynamics and more strongly stabilizes the active-like ensemble. Afatinib binding disrupts the K745-E762 interaction and alters the αC-helix conformation while the pre-formed covalent state samples αC-out configurations that increase accessibility of an αC-adjacent pocket that is less accessible in the apo state. Alchemical calculations further reveal mutation- and protonation-state-dependent differences in the conditional noncovalent binding free energy of afatinib. QM/MM calculations of Cys797-afatinib bond formation reveal mutation-dependent differences in the calculated reaction barrier, with L747P showing a lower barrier than ΔE746-A750 within the employed QM/MM model. Together, these results show that EGFR exon 19 mutations reshape the conformational landscape and local structural environment of the kinase, thereby influencing afatinib binding and the energetic environment of covalent bond formation. These findings provide a structural framework for understanding how distinct EGFR mutations can produce different responses to covalent kinase inhibition.
Abstract The rational construction of discrete supramolecular architectures with controlled topological complexity remains a significant challenge in coordination-driven self-assembly. Herein, we report a modular multicomponent self-assembly strategy for the construction of a homologous series of perylene diimide (PDI)-based fluorescent metallacycles with progressively increasing structural complexity. By simply varying the geometry and connectivity of auxiliary carboxylate ligands, discrete monocyclic, bicyclic, and tricyclic metallacycles were successfully assembled and characterized by nuclear magnetic resonance spectroscopy, electrospray ionization mass spectrometry, and single-crystal X-ray diffraction. Photophysical and frontier-orbital analyses reveal that these metallacycles retain strong PDI-centered absorption and high fluorescence quantum yields of 69.5–81.4%, while the tricyclic metallacycle displays a distinct donor–acceptor-type orbital separation. Furthermore, these metallacycles exhibited efficient fluorescence sensing toward picric acid, with the tricyclic metallacycle showing the highest sensitivity with a detection limit of 2.3 μmol/L. This work demonstrates that topological engineering through multicomponent coordination-driven self-assembly provides an effective strategy for developing structurally sophisticated fluorescent supramolecular sensors.
