A coarse-grained model for assisting the investigation of structure and dynamics of large nucleic acids by ion mobility spectrometry-mass spectrometry.

Ion Mobility Spectrometry-Mass Spectrometry (IMS-MS) is a rapidly emerging tool for the investigation of nucleic acid structure and dynamics. IMS-MS determinations can provide valuable information regarding alternative topologies, folding intermediates, and conformational heterogeneities, which are not readily accessible to other analytical techniques. The leading strategies for data interpretation rely on computational and experimental approaches to correctly assign experimental observations to putative structures. A very effective strategy involves the application of molecular dynamics (MD) simulations to predict the structure of the analyte molecule, calculate its collision cross section (CCS), and then compare this computational value with the corresponding experimental data. While this approach works well for small nucleic acid species, analyzing larger nucleic acids of biological interest is hampered by the computational cost associated with capturing their extensive structure and dynamics in all-atom detail. In this report, we describe the implementation of a coarse graining (CG) approach to reduce the cost of the computational methods employed in the data interpretation workflow. Our framework employs a five-bead model to accurately represent each nucleotide in the nucleic acid structure. The beads are appropriately parameterized to enable the direct calculation of CCS values from CG models, thus affording the ability to pursue the analysis of larger, highly dynamic constructs. The validity of this approach was successfully confirmed by the excellent correlation between the CCS values obtained in parallel by all-atom and CG workflows.

Separation of the Stern and diffuse layer in coarse-grained models: the cases of phosphatidyl serine, phosphatidic acid, and PIP2 monolayers.

We present here a method to separate the Stern and diffuse layer in general systems into two regions that can be analyzed separately. The Stern layer can be described in terms of Bjerrum pairing and the diffuse layer in terms of Poisson-Boltzmann theory (monovalent) or strong coupling theory plus a slowly decaying tail (divalent). We consider three anionic phospholipids: phosphatidyl serine, phosphatidic acid, and phosphatidylinositol(4,5)bisphosphate (PIP2), which we describe within a minimal coarse-grained model as a function of ionic concentration. The case of mixed lipid systems is also considered, which shows a high level of binding cooperativity as a function of PIP2 localization. Implications for existing experimental systems of lipid heterogeneities are also discussed.

General solution to the electric double layer with discrete interfacial charges.

We provide extensive molecular dynamics simulations of counterion and coion distributions near an impenetrable plane with fixed discrete charges. The numerical results are described by an explicit solution that distinguishes the plasma (√(A(c))/σ>3) and the binding regime (√(A(c))/σ<3) where σ is the ion diameter and A(c) = ∣e∕ν∣ (ν is the surface charge density). In the plasma regime, the solution consists of a product of two functions that can be computed from simpler models and reveals that the effects of the discreteness of the charge extends over large distances from the plane. The solution in the binding regime consists of a Stern layer of width σ and a diffuse layer, but contrary to standard approaches, the strong correlations between ions within the Stern layer and the diffuse layer require a description in terms of a "displaced" diffuse layer. The solution is found to describe electrolytes of any valence at all concentrations investigated (up to 0.4M) and includes the case of additional specific interactions such as van der Waals attraction and other generalizations. We discuss some open questions.

Electrostatic correlations at the Stern layer: physics or chemistry?

We introduce a minimal free energy describing the interaction of charged groups and counterions including both classical electrostatic and specific interactions. The predictions of the model are compared against the standard model for describing ions next to charged interfaces, consisting of Poisson-Boltzmann theory with additional constants describing ion binding, which are specific to the counterion and the interfacial charge ("chemical binding"). It is shown that the "chemical" model can be appropriately described by an underlying "physical" model over several decades in concentration, but the extracted binding constants are not uniquely defined, as they differ depending on the particular observable quantity being studied. It is also shown that electrostatic correlations for divalent (or higher valence) ions enhance the surface charge by increasing deprotonation, an effect not properly accounted within chemical models. The charged phospholipid phosphatidylserine is analyzed as a concrete example with good agreement with experimental results. We conclude with a detailed discussion on the limitations of chemical or physical models for describing the rich phenomenology of charged interfaces in aqueous media and its relevance to different systems with a particular emphasis on phospholipids.