Understanding and Modulating Metalloenzymes with Unnatural Amino Acids, Non-Native Metal Ions, and Non-Native Metallocofactors
Evan N. Mirts,Ambika Bhagi-Damodaran,Yi Lu
DOI: https://doi.org/10.1021/acs.accounts.9b00011
IF: 24.466
2019-01-01
Accounts of Chemical Research
Abstract:Metalloproteins set the gold standard for performing important functions, including catalyzing demanding reactions under mild conditions. Designing artificial metalloenzymes (ArMs) to catalyze abiological reactions has been a major endeavor for many years, but most ArM activities are far below those of native enzymes, making them unsuitable for most pratical applications. A critical step to advance the field is to fundamentally understand what it takes to not only confer but also fine-tune ArM activities so they match those of native enzymes. Indeed, only once we can freely modulate ArM activity to rival (or surpass!) natural enzymes can the potential of ArMs be fully realized. A key to unlocking ArM potential is the observation that one metal primary coordination sphere can display a range of functions and levels of activity, leading to the realization that secondary coordination sphere (SCS) interactions are critically important. However, SCS interactions are numerous, long-range, and weak, making them very difficult to reproduce in ArMs. Furthermore, natural enzymes are tied to a small set of biologically available functional moieties from canonical amino acids and physiologically available metal ions and metallocofactors, severely limiting the chemical space available to probe and tune ArMs. In this Account, we summarize the use of unnatural amino acids (UAAs) and non-native metal ions and metallocofactors by our group and our collaborators to probe and modulate ArM functions. We incorporated isostructural UAAs in a type 1 copper (T1Cu) protein azurin to provide conclusive evidence that axial ligand hydrophobicity is a major determinant of T1Cu redunction potential ( E°'). Closely related work from other groups are also discussed. We also probed the role of protein backbone interactions that cannot be altered by standard mutagenesis by replacing the peptide bond with an ester linkage. We used insight gained from these studies to tune the E°' of azurin across the entire physiological range, the broadest range ever achieved in a single metalloprotein. Introducing UAA analogues of Tyr into ArM models of heme-copper oxidase (HCO) revealed a linear relationship between p Ka, E°', and activity. We also substituted non-native hemes and non-native metal ions for their native equivalents in these models to resolve several issues that were intractable in native HCOs and the closely related nitric oxide reductases, such as their roles in modulating substrate affinity, electron transfer rate, and activity. We incorporated abiological cofactors such as ferrocene and Mn(salen) into azurin and myoglobin, respectively, to stabilize these inorganic and organometallic compounds in water, confer abiological functions, tune their E°' and activity through SCS interactions, and show that the approach to metallocofactor anchoring and orientation can tune enantioselectivity and alter function. Replacing Cu in azurin with non-native Fe or Ni can impart novel activities, such as superoxide reduction and C-C bond formation. While progress was made, we have identified only a small fraction of the interactions that can be generally applied to ArMs to fine-tune their functions. Because SCS interactions are subtle and heavily interconnected, it has been difficult to characterize their effects quantitatively. It is vital to develop spectroscopic and computational techniques to detect and quantify their effects in both resting states and catalytic intermediates.