No abstract

Aliphatic N-oxides as cosolvents with water play an important role in stabilizing and destabilizing the structure of biopolymers such as cellulose and proteins. To allow for detailed microscopic investigations, an empirical force field to be used in molecular simulations is developed for two N-oxide species, N,N,N-trimethylamine-N-oxide (TMAO) and N-methylmorpholine-N-oxide (NMMO). The intra-and intermolecular force field is parametrized mainly on the basis of quantum-chemical calculations and is tested against available experimental spectroscopic, crystallographic, and liquid state data. Special emphasis is put on the identification of transferable potential terms in order to guide future parametrization of other species. By construction, the force field is compatible with widely used potential functions for proteins and carbohydrates. With the resulting parameter set, molecular dynamics simulations are carried out on binary mixtures of water and N-oxides, revealing structural features and the influence of intramolecular N-oxide flexibility. Limitations and possible extensions of the presented models are also discussed.

A new method for display and analysis of lipophilic/hydrophilic properties on molecular surfaces is presented. The present approach is based on the concept of Crippen and coworkers that the overall hydrophobicity of a molecule (measured as the logarithm of the partition coefficient in an octanol/water system) can be obtained as a superposition of single atom contributions. It is also based on the concept of molecular lipophilicity potentials (MLP) first introduced by Audry and coworkers in order to establish a 3D lipophilicity potential profile in the molecular environment. Instead of using a l/r- or an exponential distance law between the atomic coordinates and a point on the molecular surface, a new distance dependency is introduced for the calculation of an MLP-value on the solvent-accessible surface of the molecule. In the present formalism the Crippen values (introduced for atoms in their characteristic structural environment) are 'projected' onto the van der Waals surface of the molecule by a special weighting procedure. This guarantees that only those atomic fragments contribute significantly to the surface values that are in the close neighbourhood of the surface point. This procedure not only works for small molecules but also allows the characterization of the surfaces of biological macromolecules by means of local lipophilicity. Lipophilic and hydrophilic domains can be recognized by visual inspection of computer-generated images or by computational procedures using fuzzy logic strategies. Local hydrophobicities on different molecular surfaces can be quantitatively compared on the basis of the present approach.

An empirical potential energy function for fluor-and for hydroxyapatite is formulated and parametrised. The parameter optimisation involves a hierarchy of reference data and techniques comprising of quantum-chemical calculations for Coulomb interactions and intramolecular contributions, as well as experimental data and molecular dynamics simulations for the remaining nonbonded parameters. For fluorapatite both a flexible and a rigid phosphate model are derived, while for hydroxyapatite only the rigid variant is determined. Simulations with the final models reproduce the experimental crystal parameters within less than 1% deviation for a wide range of temperatures between 73 and 1273 K. In the case of flexible fluorapatite the computed and the experimental infrared spectra at 300 K agree excellently.