Abstract The supramolecular interactions between selegiline hydrochloride (SEL) and β-cyclodextrin (βCD) derivatives were investigated through isothermal titration calorimetry (ITC), nuclear magnetic resonance (NMR), and molecular dynamics (MD) simulations. Thermodynamic analyses revealed spontaneous complex formation in aqueous solution for both βCD and carboxymethyl-β-cyclodextrin (CM-βCD), with CM-βCD exhibiting higher equilibrium constants and more favorable thermodynamic parameters. The complexation process was predominantly entropy-driven, indicating significant contributions from hydrophobic interactions and solvent reorganization. Non-integer stoichiometric values (n ≈ 1.5) obtained by ITC suggested the formation of higher-order supramolecular assemblies. NMR experiments demonstrated strong spatial correlations between the aromatic hydrogens of SEL and the cavity hydrogens of CM-βCD, supporting deep inclusion within the CD cavity. MD simulations corroborated the experimental findings and demonstrated the feasibility of a 3:2 CM-βCD:SEL supramolecular organization. Overall, the combined thermodynamic, spectroscopic, and computational results provide compelling evidence for higher-order, multi-equilibrium cyclodextrin supramolecular assemblies in aqueous solution.
Abstract Torsion sign flips are fundamental discrete events in protein backbone dynamics, yet their geometric triggers remain poorly understood. While traditionally associated with singular critical points, we show that these transitions are primarily driven by local geometric instability, quantified by the Laplacian of a curvature-derived support field, ΔS. Through a large-scale analysis of over 1.3 million residue positions from 3,000 PDB structures, we demonstrate a robust physical law: flip events are significantly depleted in Laplacian-zero (locally balanced) regions and exhibit progressively higher enrichment as the instability magnitude |ΔS| increases. This supports an instability-associated statistical interpretation in which geometric imbalance is correlated with an increased propensity for torsion inversions in the analyzed static structural data set. We further identify a “topological locking” effect, where looplike constraints systematically suppress flips, particularly in high-instability regimes. Multivariate modeling confirms that ΔS provides predictive information inaccessible to standard local descriptors (curvature and torsion). Our results establish a physically interpretable framework linking discrete differential geometry to stochastic backbone dynamics, revealing how local geometric imbalance and global structural constraints are jointly associated with torsion sign-flip propensity in experimentally determined static protein structures.
Abstract In recent years, lithium–sulfur (Li–S) batteries have become attractive owing to their high energy density and higher charge-storage capacity than Li-ion batteries. Despite this, practical applications of Li–S batteries remain limited by several challenges that still need to be resolved, including Li-dendrite formation, which leads to short-circuiting. In the present study, divide-and-conquer density-functional tight-binding molecular dynamics (DC-DFTB-MD) simulations examine neutral cyclo-S8, while lithium polysulfides and polysulfide anions are not included. Increasing LiOTf concentration generally reduces the self-diffusion coefficients because denser, more highly coordinated ionic environments restrict translational motion. The Li+ ion preferentially coordinates with electronegative oxygen atoms of the solvent and triflate anion, whereas nonpolar, neutral S8 interacts mainly through induced-polarization and dispersion forces. These competing coordination environments explain the weak direct Li+–S8 association and the coupled decline of ionic and molecular mobility with salt loading.
Abstract Development of precise, non-invasive, and dynamic methods for detecting amyloid-β (Aβ) aggregates is of vital importance for the early diagnosis and therapeutic evaluation of Alzheimer’s disease (AD). Recently, the Akalumine (AkaL)-based firefly bioluminescent probe has emerged as a promising tool for deep-tissue Aβ imaging. However, despite the successful experimental application of this probe, an understanding that links its microscopic mechanisms to the observed bioluminescence (BL) intensity, kinetics, and imaging performance is still lacking. In this work, a multiscale theoretical investigation combining classical molecular dynamics (cMD) simulations, absolute binding free energy (ABFE) calculations, quantum mechanics/molecular mechanics (QM/MM), and QM/MM metadynamics simulations was performed to comprehensively elucidate the mechanism of this bioluminescent system. Our cMD and ABFE results reveal that Aβ acts as a reservoir capable of storing AkaL via hydrophobic interactions, thereby dynamically modulating the concentration of free AkaL and the evolution of BL intensity. Changes in BL intensity can serve as an indicator of the concentration and distribution of Aβ. Furthermore, the complete BL process of AkaL catalyzed by firefly luciferase was systematically studied. The initial adenylation follows the SN2 mechanism. The water molecule was found to play a crucial catalytic role in the subsequent deprotonation by mediating proton transfer, significantly lowering the energy barrier. Then, oxygenation occurs via a triplet-to-singlet intersystem crossing (ISC) process to generate the dioxetanone intermediate. The decomposition of the dioxetanone intermediate follows the gradually reversible charge-transfer initiated luminescence (GRCTIL) mechanism that generates the light emitter Oxy-AkaL in the S1 state. The calculated emission wavelength is in reasonable agreement with the experimental BL wavelength. Overall, this work provides theoretical insights into the mechanism of firefly BL-based AD detection and offers guidance for the rational design of future bioluminescent probes for AD.
Abstract Passive membrane permeation depends on both solute partitioning and coordinate-dependent mobility, but their separate and spatial-coupling contributions are rarely quantified within a common resistance framework. CNFD (6b,7-dihydro-5H-cyclopenta[b]naphtho[2,1-d]furan-5,6(9aH)-dione), a semisynthetic naphthoquinone with low-micromolar cytotoxicity and tumor-growth inhibition in preclinical models, was examined in a four-component model lipid bilayer. Established simulated tempering-enhanced umbrella sampling, WHAM, the Hummer positional-autocorrelation formalism, and the inhomogeneous solubility-diffusion model were combined. Three independent replicas yielded favorable interfacial partitioning (−4.66 kJ mol–1) and a positive membrane-center free energy (+12.36 kJ mol–1). Controlled profile substitutions showed that PMF heterogeneity and position-dependent diffusion amplified CNFD resistance by factors of 10.54 and 14.89, respectively, whereas their residual spatial-coupling factor was 1.134. The PMF maximum and diffusion minimum were separated by 2.14 nm. The primary intrinsic Peff was 9.09 × 10–2 cm s–1. Applying the same analysis to β-lapachone and plumbagin identified contrasting resistance regimes, with plumbagin combining a favorable center PMF with much lower central resistance localization. These results show how spatially separated free-energy and mobility landscapes govern intrinsic lipid-phase permeation.
Abstract A series of equilibrium and nonequilibrium molecular dynamics simulations were conducted to investigate the Fickian diffusion of dissolved carbon dioxide under kaolinite-slit confinement. The relationship between the chemical potential of the dissolved carbon dioxide and its molar fraction was revealed. The Fickian diffusion coefficients were also computed and were found to increase linearly with the molar fraction. The adsorption behavior of the dissolved carbon dioxide as a solute was found to be associated with the accumulation and preferential orientation of the solvent water. The radial distribution functions of the water oxygen in various regions were analyzed, revealing the spatial inhomogeneity of the solvent structure. The inhomogeneity of dissipation, quantified by local mobility, was found to be proportional to the first peak of the radial distribution function of solvent water.
Abstract DNA, the fundamental genetic material in living organisms, is of great scientific interest due to its biological role and interactions with small molecules. Ultraviolet (UV) radiation poses a substantial environmental risk by inducing DNA damage, potentially leading to mutagenesis and skin cancers. The study investigated purpurin interaction with ct-DNA and its protective effect against UV-induced DNA damage. UV exposure caused significant DNA damage, indicated by 36% reduced absorbance; however, purpurin significantly lessened this damage to only 19%. In addition, Raman spectra showed no significant changes in purpurin-treated DNA, indicating that the DNA’s native structure was preserved, suggesting that purpurin is able to protect the DNA against UV-induced DNA damage. UV–vis and cyclic voltammetry studies indicated purpurin-ct-DNA complex formation with a binding constant of 104 M–1. Thermodynamic analysis shows negative ΔH [-(39.91 ± 0.77) kJ mol–1] and ΔS [-(51.68 ± 2.70) J mol–1], indicating hydrogen bonding and van der Waals forces, with negative ΔG confirming spontaneity. Groove binding is the preferred binding mode as evaluated by various experimental techniques. Circular dichroism (CD) and Fourier-transform infrared (FT-IR) spectra suggested that purpurin binding does not alter DNA’s structural integrity. Molecular dynamics (MD) simulation further supported the structural stability and flexibility of DNA. These findings provide mechanistic insight into the interaction of purpurin with DNA and its potential role in maintaining DNA structural integrity under UV irradiation.
Abstract This study combined in situ experiments and molecular simulations to investigate the heating behavior of sodium chloride (NaCl) solutions, up to 1 M, irradiated by microwaves. MW-induced evaporation rate, before reaching boiling, of pure water was obtained at 2.6 × 10–4 g/(s·cm2), which was more than 10 times the water evaporation under nonheating conditions. This evaporation rate increased with rising NaCl concentrations, along with higher temperatures in the outer water layer. At a NaCl concentration of 1 M, the MW-induced evaporation rate increased by 110%, compared with pure water. Notably, the heating rates varied with both NaCl concentration and distance from the air/water surface. Within 1 mm of the surface, the heating rate of a 1 M NaCl solution increased by 90%, compared with pure water. However, the heating rate was almost independent of NaCl concentration at 5 mm from the surface. Molecular dynamics simulations were conducted on a 9 nm-thick water slab under an oscillating electric field of 1 V/nm to enable a qualitative comparison. Despite differences in field amplitude and physical scale, the simulations show a trend similar to the experimental observations, in which both heating and evaporation rates increase with NaCl concentration. In particular, the obtained evaporation rates before boiling were in the same order of magnitude. The simulation provides molecular insights into the impacts of NaCl and microwaves on water dynamics, including diffusion (prior to boiling) and water–water interactions. The combined experimental and simulation findings provide useful parameters for optimizing microwave-assisted evaporation.
Abstract Collagen is the primary constituent of the supramolecular fibers found in connective tissues. To be able to predict collagen’s properties, we need rigorous characterization at the molecular scale. In this work, we focus on collagen-mimetic peptides (CMPs) that are composed of short sequences of (PPG) tripeptides folded into a triple helix. We use molecular dynamics simulations with the AMOEBA polarizable force field to model these CMPs at physiological temperature. We show that AMOEBA captures length-dependent fraying of the triple helix, consistent with experimental observations. We also introduce new metrics to quantify the deformation of the triple helix and hydration dynamics. This enables a systematic quantification of CMPs under a variety of conditions, which was missing in the field. We apply our approach to mutated CMPs and CMPs in the presence of d-glucose. We find that translational diffusion anisotropy is a molecular signature of CMP structural integrity.
Abstract Tracking the progression of the bone resorption process is important for clinical research and development of pharmaceuticals for bone-related diseases. Although fluorescence imaging is found to be one of the powerful tools for studying bone, conventional fluorescence microscopy relies on extremely high excitation powers that induce photobleaching and phototoxic effects on biological samples, preventing long-term observations. To overcome these limitations, we propose a quantum light-based approach using entangled two-photon fluorescence microscopy (ETPFM). We demonstrate ETPFM for imaging fixed demineralized unstained and Hoechst- and DAPI-stained bone sections at photon fluxes nearly 6 orders of magnitude lower than classical two-photon excitation. This quantum-correlated light preserves tissue integrity, eliminates photobleaching, and reveals microstructures with high contrast. Endogenous fluorescence is detected without exogenous dyes, and osteoclast cells are clearly resolved in the stained bone sections. ETPFM thus provides a minimally invasive, quantum-enhanced approach for long-term imaging of bone microenvironments.
Abstract Despite the promise of alkali/urea aqueous solutions as green solvents, the atomistic-scale mechanism of chitin dissolution remains elusive. In contrast to the ability to dissolve cellulose, the order of alkali effectiveness for dissolving chitin is potassium hydroxide (KOH) as the best and lithium hydroxide (LiOH) as the worst. Employing chitotriose as a model compound and utilizing neutron total scattering coupled with Empirical Potential Structure Refinement (EPSR) as the principal analytical tools, we compared its KOH/urea and LiOH/urea aqueous solutions and elucidated the intrinsic mechanism of chitotriose dissolution via its all-atom structure in solution. The results show that, compared with Li+, K+ preferentially enriches around the oxygen atoms of the amide group and hydroxyl group in the periphery of chitotriose molecules, triggering the disruption of its intermolecular hydrogen-bond network in its crystal state. Then, urea molecules play an indirect role via cation bridging. This “cation-dominated, urea-assisted” mechanism elucidates the synergistic dissolution principle of the alkali/urea system. Taking into account that the dissolution of biomacromolecules requires prior swelling, we proposed that the dissolution of chitin in alkali-urea aqueous solution first requires the disruption of intermolecular hydrogen bonds in the chitin crystals during the swelling process, followed by dissolution through van der Waals interactions.
Abstract Human CYP2E1 oxidizes both eicosapentaenoic acid (EPA, an ω–3 polyunsaturated fatty acid [PUFA]) and arachidonic acid (AA, an ω–6 PUFA) to bioactive lipid mediators, yet the mechanistic basis for its regio- and enantioselective preferences remains incompletely understood. Here, we use a combined molecular dynamics (MD) and hybrid quantum mechanics/molecular mechanics (QM/MM) strategy to compare the oxidation of EPA and AA by human CYP2E1 at atomistic resolution. MD simulations indicate that EPA and AA occupy essentially the same CYP2E1 cavity but are stabilized through distinct interaction patterns. Compared with AA, EPA more frequently forms a hydrogen bond with A108 at the channel entrance and exhibits more uniform sampling of near-attack geometries in the ω–1, ω–2, and ω–3 regions. QM/MM free energy profiles show that ω–1S hydroxylation has the lowest barrier for both substrates, but that the barriers for the ω–2/ω–3 epoxidation pathways are close to that for ω–1S hydroxylation in EPA, rendering these pathways kinetically competitive. Energy decomposition analysis indicates that regioselectivity is largely determined by the QM energy term (ΔEQM), whereas enantioselectivity at ω–1 reflects small differences in packing and entropic cost between the pro-R and pro-S transition states in the confined active site. These results provide a coherent mechanistic picture of CYP2E1-catalyzed oxidation of ω–6 and ω–3 PUFAs and identify specific active-site residues as promising targets for tuning regio- and stereoselectivity in lipid oxidation.
Methanol-water mixtures find use in many applications, particularly catalytic energy conversion processes. Their importance has motivated numerous computational studies, most of which employed molecular dynamics based on classical force fields. These enable simulations of large systems on long time scales but do not reliably describe reactive dynamics involving bond breaking and bond formation. In contrast, ab initio molecular dynamics (AIMD) based on density functional theory (DFT) is generally more reliable for such applications but has a high computational cost, which discourages systematic studies of alcohol-water mixtures. To remedy this, we trained a machine learning interatomic potential capable of probing the properties of aqueous methanol mixtures at the DFT level using the SCAN functional. Our results show that SCAN qualitatively reproduces multiple key experimental features arising from the amphiphilic nature of methanol, including density, diffusion coefficients, X-ray structure factors, and Kirkwood-Buff integrals. We also find that structural correlations between water molecules are somewhat overestimated, leading to a stronger preferential association than that predicted by experiments. However, increasing the temperature by 30 K mitigates this effect and also recovers the correct mobilities of both methanol and water. These results indicate that SCAN provides an accurate description of methanol-water mixtures, making it a reliable choice for investigating the reactive dynamics in such systems.
Phosphorylation of amyloid-β (Aβ40) at Ser8 and Ser26 exerts opposing effects on fibril formation: Ser8 phosphorylation promotes aggregation, whereas Ser26 phosphorylation strongly inhibits it. Using replica exchange with solute tempering (REST2) simulations and coarse-grained modeling, we reveal the atomic-level mechanisms underlying these effects. Ser26 phosphorylation forms a highly stable pS26-K28 salt bridge that competes with and disrupts the native E22/D23-K28 interaction required for hairpin formation and fibrillization, yielding a compact, globular conformation that is aggregation-incompetent. In contrast, Ser8 phosphorylation stabilizes the hairpin structure with a preformed N-terminal attachment (the S* structure) via pS8-K16/K28 salt bridges, lowering the entropic barrier for N-terminal alignment. This mechanism aligns with experimental observations that pS8 fibrils gain stability due to N-terminal incorporation into the fibril core, thereby promoting fibrillization. Together, these results demonstrate that monomeric conformational landscapes directly encode aggregation propensity, providing a mechanistic framework for understanding how post-translational modifications modulate amyloid assembly pathways.
Carbon nanotubes (CNTs) enable ultrafast water flow, but the underlying mechanism remains debated. Here, using Deep Potential molecular dynamics simulations with quantum mechanical accuracy, we revisit enhanced water transport in CNTs by analyzing interfacial friction and confined water properties. It is found that the interfacial friction coefficient is primarily governed by the free-energy barrier amplitude. The viscosity of confined water shows discontinuous changes in channels below 1.62 nm in diameter but varies nearly continuously in wider ones. For larger channels, viscosity can be approximated by a weighted average of interfacial and bulk values, allowing accurate prediction of flow enhancement within a continuum fluid mechanics framework. Across all channel sizes, variations in structural order parameters and average hydrogen-bond number correlate with viscosity changes, indicating that altered water structure drives viscosity modification. This work reveals the coupling between confined water structure, viscosity, and nanofluidic transport, providing insights for designing advanced nanofluidic devices.
Multiple myeloma (MM) is a malignant blood cancer marked by severe bone destruction and immune microenvironment disruption. Yet relapse and drug resistance remain major clinical challenges. Tanshinone IIA (TIIA) exhibits potent antitumor activity, but its molecular targets and mechanisms in MM remain unclear. Here, we systematically elucidate TIIA's pharmacological mechanisms in MM using network pharmacology, transcriptomics, molecular docking, molecular dynamics (MD) simulations, and in vitro cellular experiments. Histone deacetylase 6 (HDAC6) was identified as a key prognostic target in MM via integrative bioinformatics─combining cross-database analysis, machine learning (LASSO, SVM-RFE, random forest), and survival validation in the MMRF-CoMMpass cohort. Molecular docking and MD simulations showed stable binding of TIIA to HDAC6. In vitro, TIIA directly inhibited HDAC6 enzymatic activity and selectively killed U266 and RPMI 8226 myeloma cells in a dose-dependent manner, with minimal toxicity to normal cells. Additionally, gene set enrichment analysis (GSEA) and single-sample GSEA (ssGSEA) immune profiling using the LM22 signature suggested that HDAC6 participates in microenvironmental remodeling by modulating cell adhesion-mediated resistance and orchestrating an immunosuppressive niche characterized by monocyte depletion. This study highlights for the first time the critical role of HDAC6 in TIIA-mediated antimyeloma activity and provides novel mechanistic insights and potential targeted therapeutic strategies for the treatment of MM.
The hydration layer at the polymer-water interface has been proposed as a key determinant of the antithrombogenicity of polymer materials in medical devices. To investigate how the structure and diffusivity of water influence antithrombogenicity, molecular dynamics (MD) simulations were performed on two benchmark antithrombogenic polymers: poly(2-methoxyethyl acrylate) (PMEA) and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC). Although these polymers have markedly different side-chain structures, both are known to exhibit excellent antithrombogenic properties. The results reveal that PMEA-water systems exhibit a significantly higher ratio of intermediate water to bound water than PMPC-water systems. In contrast, the structured water layer in PMPC-water systems shows lower diffusivity and higher activation energy than that in PMEA-water systems. These differences in structure and diffusive properties indicate that the two polymers achieve bioinertness through fundamentally distinct hydration structures and dynamics. The antithrombogenicity of PMEA is governed by a quantitative mechanism, in which the abundance of intermediate water forms a voluminous physical cushion. In contrast, the antithrombogenicity of PMPC is attributed to a qualitative mechanism, forming an energetically robust hydration shield through structural stabilization of the hydration layer. These findings suggest that both the abundance and energetic stability of structured water are critical factors governing the antithrombogenicity of polymer materials, providing a fundamental framework for the rational design of next-generation biomaterials.
The multitorsional potentials that account for the coupling between the conformational states of amino-acid residues at the coarse-grained level developed and implemented in the UNRES force field in our earlier work have been revised to cover proteins containing d-amino-acid residues by expressing the phase shifts of the consecutive Cα···Cα···Cα···Cα backbone virtual-bond dihedral angles as sums of single-residue-type contributions. Initial multitorsional-potential parameters were determined by the maximum-likelihood method, based on the statistics from the Protein Data Bank. Subsequently, UNRES with the new multitorsional potentials was calibrated by using a set of nine proteins with different secondary and tertiary structures. The UNRES force field with the revised multitorsional potentials showed improved performance in modeling the structures of α and α + β proteins and produced good models of proteins containing both l- and d-amino-acid residues. The results suggest that the along-chain coupling of local conformational states at the coarse-grained level extends to polypeptide chains composed of heterochiral amino-acid residues.
Predicting coupled transport and thermodynamic properties of supercooled ionic liquids (ILs) is critical for low-temperature energy applications, yet sparse data and the absence of low-temperature phenomenological relations make the prediction of these properties challenging. To address this challenge, we develop a physics-informed multitask neural network to predict viscosity (η), specific heat capacity (Cp), and density (ρ) of pure ILs, IL mixtures, and IL solutions. The η branch is constrained by the Vogel-Fulcher-Tammann (VFT) equation, while Cp and ρ are trained with lower-order physics-guided losses because their phenomenological relations in the supercooled regime are absent. We show that the multitask model captures nonlinear Cp and ρ behavior through cross-learning from η, unlike independently trained single-task models. By implementing a VFT-type correction to the interaction potential energy, we show quantitatively that the temperature dependence of Cp can be explained by the loss of configurational entropy due to kinetic limitations. The multitask model was then used to screen candidates with optimum η, Cp, and ρ for application as low-temperature heat transfer fluids, and the structural features governing these properties were identified by explainable AI algorithms and quantum chemical surface charge density calculations.
Tryptophan (TRP) self-assembly is responsible for the development of several neurodegenerative disorders, such as hypertryptophanemia, for which effective therapeutics are still lacking. Although several experimental studies have demonstrated varying inhibitory efficacies of different functional moiety-rich compounds, the mechanistic role of individual functional groups while inhibiting the spontaneous self-assembly is still elusive. We have performed molecular dynamics simulations to explore different kinds of interspecies interactions between TRP and distinct functional groups present in experimentally studied aromatic and nonaromatic compounds with varying concentrations: tannic acid (TA), which contains both aromatic and hydroxyl moieties, and polyols, namely, sorbitol and mannitol, which contain no aromatic ring but only hydroxyl moieties. TA reduces TRP self-assembly more effectively than polyols, a difference not attributable solely to molecular size or multivalency. Quantitative analyses of different TRP-TA interactions, including π-π stacking between aromatic rings and H-bonding between respective hydroxyl moieties, reveal that lifetimes of π-π stacking are much higher than the H-bond lifetimes. Notably, the number and strength of π-π stacking interactions dominate over the H-bonding interaction in TA-mediated inhibition, although a large number of hydroxyl groups are present in TA. Consequently, TRP exhibits longer residence times in the vicinity of TA than polyols. Furthermore, TA weakens π-π stacking between TRP molecules, reducing TRP aggregation. The computed TRP-TA binding energy may therefore serve as a guide for evaluating the efficacies of other chemical compounds, demonstrating the importance of the aromatic ring in inhibiting the early stages of TRP oligomerization.
