Search

Molecular dynamics simulations at 298 and 273 K were carried out in order to explore the ability of calculations to predict NMR spin–lattice relaxation rates of the protons of liquid acetonitrile. The six-site model of Nikitin and Lyubartsev (J. Computat. Chem. 28 (2007), pp. 2020–2026) was used with systems containing up to 13,824 molecules. The effects of employing a cut-off for intermolecular interactions were explored. It was found that the model and procedures used predict intramolecular and intermolecular spin–lattice relaxation rates that are within ∼25% of experimental values at both temperatures.

Nuclear magnetic resonance spectroscopy has evolved in complexity to the extent that it is often useful to carry out computer simulations of experiments as a way to refine experimental conditions, to maximize usable information, or to interpret results reliably. The starting point for such simulations is an appropriate quantum mechanical formulation. This article presents many of the physical ideas that are important in high-resolution NMR spectroscopy and then develops expressions for the contribution of each of these to the Hamiltonian for the system under consideration. The Hamiltonians for the Zeeman interaction, scalar coupling, and the chemical shielding effect are discussed with the aim of explaining the quantum mechanics of NMR in a manner understandable to a nontheoretician. Consequently, much of the discussion is explicit and includes points skimmed over in other presentations in the literature.

The consequences of substitution of fluorine for the para hydrogen of a phenylalanine residue in ribonuclease-S were investigated by conformational energy calculations using the AMBER force field. Both the fluorine-containing protein and the corresponding nonfluorinated material were subjected to conformational adjustment through energy minimization and the minimum energy structures so defined were compared. Fluorine substitution leads to small alterations in many atomic positions in the protein, with adjustments at at sites more than 0.5 nm from the fluorine appearing to be somewhat larger than those within the immediate vicinity of the fluorine. Several atoms proximate to the fluorine atom were observed to move toward the fluorine while others in the same vicinity move away. The greater bulk of the fluorine atom and the strongly different electronic properties of fluorine compared to hydrogen thus appear to be insufficient to cause a consistent, unidirectional change in nearest-neighbor interactions upon introduction of a fluorine atom into a protein structure. Virtually all changes in atomic positions that are predicted by these calculations would be barely detectable by a crystallographic study with a resolution of 0.2 nm.

We describe an exercise intended to introduce students to the proton NMR experiments used to define the molecular conformation of biological molecules in solution. Proton NMR spectra of the octapeptide [sar1] angiotensin II are assigned and nuclear Overhauser effects are then used to develop constraints that define the three-dimensional shape of the peptide. The exercise demonstrates the utility of COSY, TOCSY, and NOESY types of experiments, while underscoring the limitations of these experiments in the determination of structure. The spectroscopic data used for the exercise are made available in several formats so that an instructor can design variations of the exercise to emphasize particular aspects related to data handling or the generation or comparison of derived structures. At least one quarter of biochemistry and a year of organic chemistry are reasonable prerequisites for the exercise, presuming that these courses provide the needed level of familiarity with the basic concepts of proton NMR spectroscopy. Junior- or senior-level chemistry and biochemistry students are probably most benefited from this exercise.

Empirical intermolecular force field parameters for the fluorine atom of the 4-fluorophenyl group have been generated for use in molecular mechanics programs. Partial charges on the ring system were obtained by Mulliken population analysis of wave functions generated by molecular orbital calculations in the STO-3G approximation. Fluorine van der Waals parameters were obtained by non-linear least-squares fitting of crystallographic data following the procedure of Hagler and Lifson. The parameters developed are compared to fluorine parameters reported previously.

Deacylation of unsubstituted and of o-F-, m-F, p-F, α-F, pentafluoro-, p-methyl-, and p-trifluoromethyl-substituted trans-cinnamoyl- α-chymotrypsins has been studied from pH 4 to 8. The deacylation rate constants were found to depend upon the ionization state of a group on the enzyme with an apparent pK in the range 6.3–7.3. The hydrolysis rates of the correspondingly substituted p-nitrophenylcinnamate esters were determined at pH 10.6. Correlation of the data for these model reactions with the corresponding rates of enzyme deacylation suggests that the p-methyl-substituted acylenzyme is about an order of magnitude more reactive than expected while p-trifluoromethyl substitution results in deacylation at a rate 10 times slower than expected. The remaining substituents exert about the anticipated rate effect on deacylation.