Quantum refinement for improvement of metalloenzyme crystal structures

University Of Calif-Lawrenc Berkeley Lab, Berkeley CA

Investigators

Linked publications & trials

Abstract

Project Summary Iron metalloproteins are a large class of enzymes that are involved in many chemistries essential to human health. In many of these proteins, iron sulfur bonds play an important functional role, both in interactions between the iron active site and a sulfur-containing substrate and in [Fe-S] clusters that serve as electron transfer centers. Such enzymes are involved in the biosynthesis of antibiotics, sulfur metabolism, and cellular respiration. Understanding these enzymes? mechanisms of catalysis under physiologically relevant conditions is thus essential for answering critical health-related questions. Recently, X-ray free electron lasers (XFELs) have emerged as a way to collect time-resolved X-ray diffraction (XRD) data on enzyme crystals at room temperature without causing radiation damage. This has made it possible to follow catalytic reactions in real time under physiological conditions, providing significant insight into mechanism. However, an important issue with the analysis of XFEL XRD data still needs to be resolved. Refinement is an important step in solving the structure of a protein based on the electron density map obtained from XRD. Standard crystallographic refinement procedures use stereochemical restraints to aid in the structure solution and ensure that the final structure is chemically reasonable. These restraints are often not accurate for metal-ligand bonds, which can bias the refinement results, especially for structures solved to medium resolution (~1.6-2.4 Å). Further, these stereochemical restraints do not accurately reflect potential changes in the electron density around the metal site. Such subtle changes in charge distribution can lead to structural changes that are mechanistically important; thus, refinement that more accurately includes these effects is essential, especially for highly covalent moieties such as Fe-S bonds and clusters. Quantum mechanical (QM) calculations can more accurately describe the electron density of covalent metal-ligand bonds and clusters, thus providing an attractive supplement to standard refinement procedures. In this proposal, I will be developing and calibrating a quantum refinement method for integration as a module into the crystallography structure analysis software platform PHENIX. The method will use density functional theory (DFT) calculations on the metal active site and immediately surrounding ligands to obtain the gradient used in the crystallographic refinement procedure, and will be benchmarked on model iron- sulfur proteins. I will then apply the method to time-resolved XFEL XRD data on two metalloproteins, isopeninicillin N synthase (IPNS) and the O2-tolerant membrane-bound [Ni-Fe] hydrogenase (MBH) collected during their O2 reactions. Quantum refinement of these structures will couple the subtle changes in electronic structure during catalysis to changes in structure, giving more accurate structures and elucidating the mechanism of isopenicillin N biosynthesis in IPNS and the mechanism of O2-tolerance in MBH. The development of this method will have important implications for understanding further metalloenzyme mechanisms in the future.