Development of a Mechanism and Structure-Guided Methodology for De Novo Enzyme Design

点击翻译按钮获取中文摘要

Project Summary/Abstract Naturally enzymes are characterized by their ability to accelerate chemical reactions by orders of magnitude and with far greater specificity than those observed in aqueous solution. However, these catalytic properties have yet to be realized in synthetic chemical or biological systems. While there have been many recent advances in the de novo design of proteins, achieving the exquisite control of individual atoms and functional groups necessary for enzyme catalysis remains a long-standing challenge and has relied on directed evolution and high-throughput screening to improve upon designs. The first principles design of an active site capable of substrate binding and catalysis with rates and affinities similar to those found in natural enzymes would represent a breakthrough in our understanding of structure-function relationships and the origins of protein (dys)function. A recent advance in the de novo design of small-molecule-binding proteins demonstrates the ability to position non-covalent interactions, such as hydrogen bonds from a protein to ligand functional groups, with sub-Å accuracy. Based on this methodology, it’s hypothesized that if non-covalent interactions can be rationally designed for binding, then they can be directed towards achieving de novo enzyme design through preferential TS-stabilization to achieve fast reaction rates. In order to test this hypothesis, enzymes capable of the model Kemp elimination reaction will be de novo designed using a bottom-up approach to install a general base and tune its reactivity, assemble an active site capable of substrate- and TS-analog-binding, and develop a negative design strategy for preferential TS-stabilization. These fundamental insights will be directed towards the first de novo metallo-β-lactamase, a model system for studying antibiotic resistance and protein evolution with human health implications. Catalytic turnover of the large and highly polar β-lactam antibiotics provides a sensitive test of our ability to design non-covalent interactions. This will be achieved by expanding upon the size, asymmetry, and topology of designable protein scaffolds using a bioinformatic and function-guided design strategy for multi-domain proteins, de novo design of a Zn2+- and substrate-binding site, and development of de novo methods for preferential transition-state stabilization. The proposed investigations will be achieved using both computational and experimental methods, starting from in silico approaches to benchmark the folding and binding of reaction intermediates in designed proteins. Promising designs will then be expressed, purified, and characterized in terms of their binding affinity and catalytic rate constants using isothermal calorimetry and optical spectroscopies. These functional studies will be complemented by structural characterization using X-ray crystallography and NMR spectroscopy to confirm the accuracy of the design methodology. Successful de novo design of functional enzymes would represent a breakthrough in our understanding of structure-function relationships and the role of non-covalent interaction in complex protein function.