Development of density functional methods for iron-sulfur clusters : application to the elusive structure of the iron-molybdenum cofactor of nitrogenase

Abstract

Iron-sulfur proteins are found in all forms of life, performing key functions in many of the most important biological processes, such as nitrogen fixation, respiration, DNA repair, and photosynthesis. The functional centers of these metalloproteins are iron-sulfur clusters, which are efficient electron-transfer agents, and provide a range of other functions, from catalysis to iron sensing. The main focus of this dissertation is to eliminate the structural uncertainties of one of the most complex iron-sulfur clusters, the iron-molybdenum cofactor of nitrogenase. The long-term goal of nitrogenase research is to understand all of the molecular-level details that enable biological nitrogen fixation, which may lead to alternatives to the fossil fuel-reliant, high temperature and pressure Haber-Bosch process for industrial nitrogen fixation. The research presented here is all based on computational chemistry, using primarily density functional theory with comprehensive experimental support from the literature. Computational chemistry can help explain experimental findings, and it can be used to probe structural and mechanistic details that cannot be observed through experiment; however, methods for theoretical modeling of iron-sulfur clusters must be developed before accurate results can be obtained. In this work, the best available methods (functionals, basis sets, and population analyses) are developed by comparing calculated properties of a series of model iron-sulfur complexes with experimental reference data. The newly recommended methods are then applied to a protein-embedded cluster, and the effects of the network of hydrogen bonds, dipoles, and electrostatic interactions are systematically probed. In addition to new insights regarding the role of the protein in tuning iron-sulfur cluster properties, an algorithm is derived for constructing computational models that capture all significant environmental effects. Future work will benefit from the foundations developed in these first two studies. The final work is a computational evaluation of the uncertain composition, charge, magnetic structure, and protonation state of the iron-molybdenum cofactor, based on a series of criteria defined by all available experimental data. The results show that the [Mo IV -2Fe II -5Fe III -9S ²- -C ⁴-] composition with hydroxyl-protonated homocitrate ligand is favored over the [Mo IV -4Fe II -3Fe III -9S ²- -N ³- ] composition preferred in the literature, adding new context to the many proposed mechanisms for biological nitrogen fixation.