Nature employs metalloproteins to mediate chemical transformations with impressive rates and selectivities. The efficiency of metalloproteins has been attributed in part to the control of the local environments surrounding their functional active metal center(s). This control is often achieved by non-covalent interactions including hydrogen bonds (H-bonds) from amino acid residues and extended H-bonding networks that include water molecules. In synthetic systems, it has been difficult to establish similar noncovalent interactions with control and predictability. In this dissertation, an approach has been used to model the active sites of metalloproteins by immobilizing metal complexes within a protein host. Using biotin-Streptavidin (Sav) technology, artificial metalloproteins (ArMs) have been designed that model several important features in both the primary and secondary coordination sphere that are seen in native metalloproteins.
Chapter 2 describes the development of FeII and FeIII artificial proteins that model key structural aspects in mononuclear Fe dioxygenases. Structural characterization showed that there was coordination from two nitrogen atom donors from the synthetic ligand and one oxygen atom donor from a nearby glutamate amino acid residue that models the 2-His-1-carboxylate facial triad found in mononuclear nonheme Fe enzymes. Additionally, the water molecules that complete the coordination sphere of the Fe center participated in H-bonding, which is also seen in native metalloproteins.
Chapter 3 discusses the efforts to bind an α-ketoglutarate analog, phenylpyruvate, and dioxygen to the FeII ArM from Chapter 2. A combination of spectroscopic and crystallographic techniques supported the docking of phenylpyruvate near the Fe active site through a pi-stacking interaction with the synthetic ligand. Further structural characterizations from multiple crystals showed that dioxygen can coordinate to the Fe center in different conformations. Mössbauer studies supported the formation of an FeIII–peroxido species within Sav.
Chapter 4 details the design of di-FeIII ArMs. An optical screen was designed to determine the proper match of synthetic ligand and Sav variant to form a di-Fe active site. The optical screen utilized endogenous phenolate coordination to produce a blue color to determine the correct match. Spectroscopic and crystallographic techniques supported phenolate coordination to the Fe center and the formation of a di-FeIII center with a µ–oxido/hydroxido bridging ligand within Sav. Additionally, studies with exogenous ligands, acetate, azide, isothiocyante, and cyanide, showed the Fe···Fe distance ranged between 3.7-4.1 Å and are comparable with native di-Fe proteins.