Search

Skin barrier function is localized in its outermost layer, the stratum corneum (SC), which is comprised of corneocyte cells embedded in an extracellular lipid matrix containing ceramides (CERs), cholesterol (CHOL), and free fatty acids (FFAs). The unique structure and composition of this lipid matrix are important for skin barrier function. In this study, experiments and molecular dynamics simulation were combined to investigate the structural properties and phase behavior of mixtures containing nonhydroxy sphingosine CER (CER NS), CHOL, and FFA. X-ray scattering for mixtures with varying CHOL levels revealed the presence of the 5.4 nm short periodicity phase in the presence of CHOL. Bilayers in coarse-grained multilayer simulations of the same compositions contained domains with thicknesses of approximately 5.3 and 5.8 nm that are associated with elevated levels, respectively, of CER sphingosine chains with CHOL, and CER acyl chains with FFA chains. The prevalence of the thicker domain increased with decreasing CHOL content. This might correspond to a phase with ∼5.8 nm spacing observed by x-rays (other details unknown) in mixtures with lower CHOL content. Scissoring and stretching frequencies from Fourier transform infrared spectroscopy (FTIR) also indicate interaction between FFA and CER acyl chains and little interaction between CER acyl and CER sphingosine chains, which requires CER molecules to adopt a predominantly extended conformation. In the simulated systems, neighbor preferences of extended CER chains align more closely with the FTIR observations than those of CERs with hairpin ceramide chains. Both FTIR and atomistic simulations of reverse mapped multilayer membranes detect a hexagonal to fluid phase transition between 65 and 80°C. These results demonstrate the utility of a collaborative experimental and simulation effort in gaining a more comprehensive understanding of SC lipid membranes.

We have developed a multi-input E(n) equivariant graph convolution-based model designed for the prediction of chemical properties that result from the interaction of heterogeneous molecular structures. By incorporating spatial features and constraining the functions learned from these features to be equivariant to E(n) symmetries, the interactional-equivariant graph neural network (IEGNN) can efficiently learn from the 3D structure of multiple molecules. To verify the IEGNN’s capability to learn interactional properties, we tested the model’s performance on three molecular data sets, two of which are curated in this study and made publicly available for future interactional model benchmarking. To enable the loading of these data sets, an open-source data structure based on the PyTorch Geometric library for batch loading multigraph data points is also created. Finally, the IEGNN’s performance on a data set consisting of an unknown interactional relationship (the frictional properties resulting between monolayers with variable composition) is examined. The IEGNN model developed was found to have the lowest mean absolute percent error for the predicted tribological properties of four of the six data sets when compared to previous methods.

Molecular dynamics simulations of mixtures of the ceramide nonhydroxy-sphingosine (NS), cholesterol, and a free fatty acid are performed to gain molecular-level understanding of the structure of the lipids found in the stratum corneum layer of skin. A new coarse-grained force field for cholesterol was developed using the multistate iterative Boltzmann inversion (MS-IBI) method. The coarse-grained cholesterol force field is compatible with previously developed coarse-grained force fields for ceramide NS, free fatty acids, and water and validated against atomistic simulations of these lipids using the CHARMM force field. Self-assembly simulations of multilayer structures using these coarse-grained force fields are performed, revealing that a large fraction of the ceramides adopt extended conformations, which cannot occur in the single bilayer in water structures typically studied using molecular simulation. Cholesterol fluidizes the membrane by promoting packing defects, and an increase in cholesterol content is found to reduce the bilayer thickness due to an increase in interdigitation of the C24lipid tails, consistent with experimental observations. Using a reverse-mapping procedure, a self-assembled coarse-grained multilayer system is used to construct an equivalent structure with atomistic resolution. Simulations of this atomistic structure are found to closely agree with experimentally derived neutron scattering length density profiles. Significant interlayer hydrogen bonding is observed in the inner layers of the atomistic multilayer structure that are not found in the outer layers in contact with water or in equivalent bilayer structures. This work highlights the importance of simulating multilayer structures, as compared to the more commonly studied bilayer systems, to enable more appropriate comparisons with multilayer experimental membranes. These results also provide validation of the efficacy of the MS-IBI derived coarse-grained force fields and the framework for multiscale simulation.

The structural properties of two- and three-component gel-phase bilayers were studied using molecular dynamics simulations. The bilayers contain distearoylphosphatidylcholine (DSPC) phospholipids mixed with alcohols and/or fatty acids of varying tail lengths, with carbon chain lengths of 12, 16, and 24 studied. Changes in both headgroup chemistry and tail length are found to affect the balance between steric repulsion and van der Waals attraction within the bilayers, manifesting in different bilayer structural properties. Lipid components are found to be located at different depths within the bilayer depending on both chain length and headgroup chemistry. The highest bilayer ordering and lowest area per tail are found in systems with medium-length tails. While longer tails can enhance van der Waals attractions, the increased tail-length asymmetry is found to induce disorder and reduce tail packing. Bulkier headgroups further increase steric repulsion, as reflected in increased component offsets and reduced tail packing. These findings help explain how bilayer composition affects the structure of gel-phase bilayers.

We demonstrate how the recently developed Python-based Molecular Simulation and Design Framework (MoSDeF) can be used to perform molecular dynamics screening of functionalized monolayer films, focusing on tribological effectiveness. MoSDeF is an open-source package that allows for the programmatic construction and parametrization of soft matter systems and enables TRUE (transferable, reproducible, usable by others, and extensible) simulations. The MoSDeF-enabled screening identifies several film chemistries that simultaneously show low coefficients of friction and adhesion. We additionally develop a Python library that utilizes the RDKit cheminformatics library and the scikit-learn machine learning library that allows for the development of predictive models for the tribology of functionalized monolayer films and use this model to extract information on terminal group characteristics that most influence tribology, based on the screening data.