The road to computationally designed artificial metalloenzymes

Abstract

Artificial metalloenzymes (ArMs) enable the introduction of organometallic cofactors into protein scaffolds for new-to-nature chemistry. Traditionally, these have been constructed from a handful of scaffolds capable of binding a range of cofactors and substrates. However, in comparison to natural enzymes, this can result in a lack of specificity, rate enhancement and stability. Recent advances in computational binder and enzyme design, and deeper insights into how natural (metallo)enzymes exploit their scaffolds to enhance active site performance, have greatly expanded the toolbox available to ArM designers. In this perspective, we outline some of these advances, and provide insight into how they can be leveraged to create more efficient and robust ArMs.