Computation informed selection of parameters for protein radical EPR spectra simulation.

Many important enzymatic reactions are characterised by free radical intermediates that can be detected experimentally. EPR spectroscopy can be used for monitoring formation of free radical states in enzymes. This provides vital information about the kinetic mechanism of the reactions. To uncover the molecular mechanism of the enzyme, it is necessary to know which part of the enzyme is involved in the electron transfer - where on the enzyme the radical is formed. This question is often addressed by site directed mutagenesis methods. However, the conclusions drawn from such studies are often ambiguous. A complementary method will be presented that allows radical site assignment on the basis of the EPR spectrum lineshape. Computer simulation of the lineshape requires a large number of parameters to be explicitly specified. These parameters are intimately linked to the radical conformation and its immediate microenvironment. Thus, provided the 3D structure of the enzyme is known, an experimental spectrum can be related to a specific amino acid residue in the enzyme. Unfortunately, the credibility of such approaches is weakened by the large number of input parameters in the EPR spectral simulation procedure. This difficulty can be overcome by finding relationships between these parameters. These relationships can be used for diminishing the dimension of the input parameters space. We have performed Density Functional Theory (DFT) calculations of EPR parameters of model tyrosine and tryptophan (neutral) radicals for an array of the residues’ conformations, each of which is involved in a hydrogen bond of variable strength. Calculated EPR parameters are thus presented as dependences on two variables. This opens the possibility to formulate continuous functions that allow one to significantly diminish the input parameters space dimension. Successful derivation of such functions might lead to a method when an experimental EPR spectrum of a protein radical can be confidently used to pinpoint the radical location in a protein, if the protein structure is known.