How Far Is "Bulk Water" from Interfaces? Depends on the Nature of the Surface and What We Measure

IceCoder: Identification of Ice Phases in Molecular Simulation Using Variational Autoencoder

The identification and classification of different phases of ice within molecular simulations are challenging tasks due to the complex and varied phase space of ice, which includes numerous crystalline and amorphous forms. Traditional order parameters often struggle to differentiate between these phases, especially under the conditions of thermal fluctuations. In this work, we present a novel machine learning-based framework, IceCoder, which combines a variational autoencoder (VAE) with the smooth overlap of atomic position (SOAP) descriptor to classify a large number of ice phases effectively. Our approach compresses high-dimensional SOAP vectors into a two-dimensional latent space using VAE, facilitating the visualization and distinction of various ice phases. We trained the model on a comprehensive data set generated through molecular dynamics simulations and demonstrated its ability to accurately detect various phases of crystalline ice as well as liquid water at the molecular level. IceCoder provides a robust and generalizable tool for tracking ice phase transitions in simulations, overcoming the limitations of traditional methods. This approach may also be generalized to detect polymorphs in other molecular crystals, leading to new insights into the microscopic mechanisms underlying nucleation, growth, and phase transitions while maintaining computational efficiency.

Evaluating long-range orientational ordering of water around proteins: signature of a tug-of-war scenario

Long-range perturbations of water structure and dynamics by biomolecules are of great interest owing to their potential role in biomolecular recognition. In this article, we examined the local and long-range orientational structure of water molecules surrounding proteins with different total charges (+8, 0 and -8), both with and without the presence of a physiological salt environment. A prominent population of in-oriented water molecules was observed in the first hydration shell of the proteins, irrespective of their total charges. Starting from the third hydration layer, water molecules primarily reflected the total charge of the respective protein. This long-range ordering persisted up to the ninth hydration layer without a physiological salt environment and vanished beyond the fifth hydration shell in the presence of a physiological salt environment. Long-range orientational ordering around different types of surface atoms of a protein showed a particularly rich and heterogeneous behaviour. When the surface atom's charge and the protein's total charge were opposite, a clear signature of a tug-of-war was demonstrated in the long-range orientational ordering of water molecules. While water molecules reported the surface atom's charge at shorter distances, at longer distances, water molecules reported the total charge of the protein, with a crossover occurring around 10 Å. This phenomenon persisted even in the presence of a physiological salt environment. Evidence of destructive/constructive superposition of water-mediated orientation waves originating from two individual proteins with similar/opposite total charges was also demonstrated. These results are important for understanding long-range water-mediated recognition phenomena among biomolecules (e.g., protein-protein, protein-ligand, and protein-DNA interactions).

Dynamics of an aqueous suspension of short hyaluronic acid chains near a DPPC bilayer

The synergy between hyaluronic acid (HA) and lipid molecules plays a crucial role in synovial fluids, cell coatings, etc. Diseased cells in cancer and arthritis show changes in HA concentration and chain size, impacting the viscoelastic and mechanical properties of the cells. Although the solution behavior of HA is known in experiments, a molecular-level understanding of the role of HA in the dynamics at the interface of HA-water and the cellular boundary is lacking. Here, we perform atomistic molecular dynamics simulation of short HA chains in an explicit water solvent in the presence of a DPPC bilayer, relevant in pathological cases. We identify a stable interface between HA-water and the bilayer where the water molecules are in contact with the bilayer and the HA chains are located away without any direct contact. Both translation and rotation of the interfacial waters in contact with the lipid bilayer and translation of the HA chains exhibit subdiffusive behavior. The diffusive behavior sets in slightly away from the bilayer, where the diffusion coefficients of water and HA decrease monotonically with increase in HA concentration. On the contrary, the dependence on HA chain size is only marginal due to enhanced chain flexibility as their size increases.

Probing the structure of water in individual living cells

Water regulates or even governs a wide range of biological processes. Despite its fundamental importance, surprisingly little is known about the structure of intracellular water. Herein we employ a Raman micro-spectroscopy technique to uncover the composition, abundance and vibrational spectra of intracellular water in individual living cells. In three different cell types, we show a small but consistent population (~3%) of non-bulk-like water. It exhibits a weakened hydrogen-bonded network and a more disordered tetrahedral structure. We attribute this population to biointerfacial water located in the vicinity of biomolecules. Moreover, our whole-cell modeling suggests that all soluble (globular) proteins inside cells are surrounded by, on average, one full molecular layer (about 2.6 Angstrom) of biointerfacial water. Furthermore, relative invariance of biointerfacial water is observed among different single cells. Overall, our study not only opens up experimental possibilities of interrogating water structure in vivo but also provides insights into water in life.

Interaction of MRI Contrast Agent [Gd(DOTA)]- with Lipid Membranes: A Molecular Dynamics Study

Contrast agents are important imaging probes in clinical MRI, allowing the identification of anatomic changes that otherwise would not be possible. Intensive research on the development of new contrast agents is being made to image specific pathological markers or sense local biochemical changes. The most widely used MRI contrast agents are based on gadolinium(III) complexes. Due to their very high charge density, they have low permeability through tight biological barriers such as the blood-brain barrier, hampering their application in the diagnosis of neurological disorders. In this study, we explore the interaction between the widely used contrast agent [Gd(DOTA)]- (Dotarem) and POPC lipid bilayers by means of molecular dynamics simulations. This metal complex is a standard reference where several chemical modifications have been introduced to improve key properties such as bioavailability and targeting. The simulations unveil detailed insights into the agent's interaction with the lipid bilayer, offering perspectives beyond experimental methods. Various properties, including the impact on global and local bilayer properties, were analyzed. As expected, the results indicate a low partition coefficient (KP) and high permeation barrier for this reference compound. Nevertheless, favorable interactions are established with the membrane leading to moderately long residence times. While coordination of one inner-sphere water molecule is maintained for the membrane-associated chelate, the physical-chemical attributes of [Gd(DOTA)]- as a MRI contrast agent are affected. Namely, increases in the rotational correlation times and in the residence time of the inner-sphere water are observed, with the former expected to significantly increase the water proton relaxivity. This work establishes a reference framework for the use of simulations to guide the rational design of new contrast agents with improved relaxivity and bioavailability and for the development of liposome-based formulations for use as imaging probes or theranostic agents.

Determination of the Thickness of Interfacial Water by Time-Resolved Sum-Frequency Generation Vibrational Spectroscopy

The physics and chemistry of a charged interface are governed by the structure of the electrical double layer (EDL). Determination of the interfacial water thickness (diw) of the charged interface is crucial to quantitatively describe the EDL structure, but it can be utilized with very scarce experimental methods. Here, we propose and verify that the vibrational relaxation time (T1) of the OH stretching mode at 3200 cm-1, obtained by time-resolved sum frequency generation vibrational spectroscopy with ssp polarizations, provides an effective tool to determine diw. By investigating the T1 values at the SiO2/NaCl solution interface, we established a time-space (T1-diw) relationship. We find that water has a T1 lifetime of ≥0.5 ps for diw ≤ 3 Å, while it displays bulk-like dynamics with T1 ≤ 0.2 ps for diw ≥ 9 Å. T1 decreases as diw increases from ∼3 Å to 9 Å. The hydration water at the DPPG lipid bilayer and LK15β protein interfaces has a thickness of ≥9 Å and shows a bulk-like feature. The time-space relationship will provide a novel tool to pattern the interfacial topography and heterogeneity in Ångstrom-depth resolution by imaging the T1 values.

Interplay of Hydropathy and Heterogeneous Diffusion in the Molecular Transport within Lamellar Lipid Mesophases

Lipid mesophases are being intensively studied as potential candidates for drug-delivery purposes. Extensive experimental characterization has unveiled a wide palette of release features depending on the nature of the host lipids and of the guest molecule, as well as on the environmental conditions. However, only a few simulation works have addressed the matter, which hampers a solid rationalization of the richness of outcomes observed in experiments. Particularly, to date, there are no theoretical works addressing the impact of hydropathy on the transport of a molecule within lipid mesophases, despite the significant fraction of hydrophobic molecules among currently-available drugs. Similarly, the high heterogeneity of water mobility in the nanoscopic channels within lipid mesophases has also been neglected. To fill this gap, we introduce here a minimal model to account for these features in a lamellar geometry, and systematically study the role played by hydropathy and water-mobility heterogeneity by Brownian-dynamics simulations. We unveil a fine interplay between the presence of free-energy barriers, the affinity of the drug for the lipids, and the reduced mobility of water in determining the net molecular transport. More in general, our work is an instance of how multiscale simulations can be fruitfully employed to assist experiments in release systems based on lipid mesophases.

Lipid/water interface of galactolipid bilayers in different lyotropic liquid-crystalline phases

In this study, carried out using computational methods, the organisation of the lipid/water interface of bilayers composed of galactolipids with both α-linolenoyl acyl chains is analysed and compared in three different lyotropic liquid-crystalline phases. These systems include the monogalactosyldiglyceride (MGDG) and digalactosyldiglyceride (DGDG) bilayers in the lamellar phase, the MGDG double bilayer during stalk phase formation and the inverse hexagonal MGDG phase. For each system, lipid-water and direct and water-mediated lipid-lipid interactions between the lipids of one bilayer leaflet and those of two apposing leaflets at the onset of new phase (stalk) formation, are identified. A network of interactions between DGDG molecules and its topological properties are derived and compared to those for the MGDG bilayer.

Spatially Resolved Hydration Thermodynamics in Biomolecular Systems

Water is essential for the structure, dynamics, energetics, and thus the function of biomolecules. It is a formidable challenge to elicit, in microscopic detail, the role of the solvation-related driving forces of biomolecular processes, such as the enthalpy and entropy contributions to the underlying free-energy landscape. In this Perspective, we discuss recent developments and applications of computational methods that provide a spatially resolved map of hydration thermodynamics in biomolecular systems and thus yield atomic-level insights to guide the interpretation of experimental observations. An emphasis is on the challenge of quantifying the hydration entropy, which requires characterization of both the motions of the biomolecules and of the water molecules in their surrounding.