Cardiotoxins (CTs) from snake venoms are a family of homologous highly basic proteins that have extended hydrophobic patterns on their molecular surfaces. CTs are folded into three β-structured loops stabilized by four disulfide bridges. Being well-structured in aqueous solution, most of these proteins are membrane-active, although the exact molecular mechanisms of CT-induced cell damage are still poorly understood. To elucidate the structure–function relationships in CTs, a detailed knowledge of their spatial organization and local conformational dynamics is required. Protein domain motions can be either derived from a set of experimental structures or generated via molecular dynamics (MD). At the same time, traditional clustering algorithms in the Cartesian coordinate space often fail to properly take into account the local large-scale dihedral angle transitions that occur in MD simulations. This is because such perturbations are usually offset by changes in the neighboring dihedrals, thus preserving the overall protein fold. States with a “locally perturbed” backbone were found in experimental 3D models of some globular proteins and have been shown to be functionally meaningful. In this work, the possibility of large-scale dihedral angle transitions in the course of long-term MD in explicit water was explored for three CTs with different membrane activities: CT 1, 2 (Naja oxiana) and CT A3 (Naja atra). Analysis of the MD-derived distributions of backbone torsion angles revealed several important common and specific features in the structural/dynamic behavior of these proteins. First, large-amplitude transitions were detected in some residues located in the functionally important loop I region. The K5/L6 pair of residues was found to induce a perturbation of the hydrophobic patterns on the molecular surface of CTs—reversible breaking of a large nonpolar zone (“bottom”) into two smaller ones and their subsequent association. Second, the characteristic sizes of these patterns perfectly coincided with the dimensions of the nonpolar zones on the surfaces of model two-component (zwitterionic/anionic) membranes. Taken together with experimental data on the CT-induced leakage of fluorescent dye from such membranes, these results allowed us to formulate a two-stage mechanism of CT–membrane binding. The principal finding of this study is that even local conformational dynamics of CTs can seriously affect their functional activity via a tuning of the membrane binding site − specific “hot spots” (like the K5/L6 pair) in the protein structure.
Molecular surfaces are one of the key players in processes of bimolecular recognition and interaction. Nowadays, state of the art methods exist for visualizing molecule surface and surface distributed properties in three-dimensional space. However, such visual information could only be analyzed by human eye and therefore prompt to be biased and onerous in case of large sets of objects. Here we present a method to create 2D projections or ”earth maps” of whole protein surface – protein surface topography (PST). Representing complex molecule surfaces as an array of data gives the advantage of simple and pictorial visualization of surface properties. PST can be used to easy visualize conformational changes between different states of molecules, perform group analysis, and reveal common patterns or dissimilarities. It is useful tool to add to docking experiments, illustrating complementary features between ligand and receptor surfaces.
Atomistic aspects of the structural organization, dynamics, and functioning of hydrated lipid bilayers - model cell membranes - are primarily governed by the fine balance of intermolecular interactions between all constituents of these systems. Besides the hydrophobic effect, which shapes the overall skeleton of lipid membranes, very important contribution to their behavior is made by hydrogen bonds (H-bonds) between lipid head groups. The latter determine such crucial phenomena in cell membranes, like dynamic ultra-nanodomain organization, hydration, fine-tuning of microscopic physico-chemical properties that allow the membrane to adapt quickly when binding/insertion external agents (proteins, etc.) The characteristics of such H-bonds (strength, spatial localization, etc.) dramatically depend on the local polarity properties of the lipid-water environment. In this work, we calculated free energies of H-bonded complexes between typical donor (NH3+, NH, OH) and acceptor (C=O, OH, COO-, COOH) groups of lipids in vacuo and in a set of explicit solvents with dielectric constants (ε) from 1 to 78.3, which mimic membrane environment at different depth. This was done using Monte Carlo simulations and an assessment of the corresponding Potential of Mean Force profiles. The strongest H-bonded complexes were observed in the nonpolar environment and their strength increased sharply with decreasing ε below 17. When ε changed, the largest free energy gain (> 10.8 kcal/mol) was observed for pairs of acceptors C=O and O(H) with donor NH3+. The complexation of the same acceptors with NH-donor in this range of ε was rather less sensitive to the environmental polarity: by ~1.5 kcal/mol. Dielectric-dependent interactions of polar lipid groups with water were evaluated as well. The results explain the delicate balance that determines the unique pattern of H-bonds for a particular lipid bilayer. Understanding the factors that regulate the propensity to H-bonding in lipid bilayers provides a fundamental basis for the rational design of new membrane nano-objects with predefined properties.
This study is dedicated to the introduction of a novel method that automatically extracts potential structural alerts from a data set of molecules. These triggering structures can be further used for knowledge discovery and classification purposes. Computation of the structural alerts results from an implementation of a sophisticated workflow that integrates a graph mining tool guided by growth rate and stability. The growth rate is a well-established measurement of contrast between classes. Moreover, the extracted patterns correspond to formal concepts; the most robust patterns, named the stable emerging patterns (SEPs), can then be identified thanks to their stability, a new notion originating from the domain of formal concept analysis. All of these elements are explained in the paper from the point of view of computation. The method was applied to a molecular data set on mutagenicity. The experimental results demonstrate its efficiency: it automatically outputs a manageable number of structural patterns that are strongly related to mutagenicity. Moreover, a part of the resulting structures corresponds to already known structural alerts. Finally, an in-depth chemical analysis relying on these structures demonstrates how the method can initiate promising processes of chemical knowledge discovery. © 2015 American Chemical Society.