Extremely large differences in DFT energies for nitrogenase models.

Area of Science:

• Biochemistry
• Computational Chemistry
• Enzymology

Background:

• Nitrogenase is crucial for converting atmospheric nitrogen (N2) into ammonia, a process vital for life.
• The enzyme's active site features a complex molybdenum-iron (MoFe) cofactor.
• Previous computational studies using density-functional theory (DFT) lack consensus on key intermediates, such as the E4 state.

Purpose of the Study:

• To investigate the reasons for the lack of consensus in DFT studies of the nitrogenase active site.
• To evaluate the performance of different DFT methods in accurately describing protonation states and energies.
• To compare computational findings with experimental data for the E4 intermediate.

Main Methods:

• Application of various DFT functionals to model the nitrogenase active site.
• Analysis of relative energies for different protonation states of the MoFe cofactor.
• Assessment of structural parameters, including metal-metal distances and oxidation states.
• Correlation of energy calculations with Hartree-Fock exchange content.

Main Results:

• Significant variations (up to 600 kJ mol-1) in relative energies between different DFT methods for various protonation states.
• Proton binding to sulfur/carbon as protons versus binding to metals as hydrides alters metal oxidation states and distances.
• Non-hybrid DFT and TPSSh best reproduce the resting active site structure, while B3LYP and PBE0 excel at H2 dissociation energies.
• No tested DFT method identifies the experimentally suggested E4 structure with bridging hydrides as the lowest energy state.

Conclusions:

• The choice of DFT method critically impacts the calculated energetics and structures of the nitrogenase active site, particularly concerning protonation states.
• Understanding the limitations of current DFT methods is essential for accurate modeling of nitrogenase function.
• Further development of computational approaches is needed to reconcile theoretical predictions with experimental observations of key intermediates.