Protein engineering of P450 enzymes for sustainable synthesis.

Vlada B. Urlacher

Heinrich Heine University Düsseldorf, Germany. vlada.urlacher@hhu.de

The sustainable synthesis of high-value compounds depends on the development of efficient catalysts, including enzymes. Cytochrome P450 monooxygenases show remarkable selectivity and activity toward their native substrates, making them a promising catalyst. However, extending their use to non-natural reactions or non-physiological substrates is challenging, especially when precise regio- and stereoselective oxidation of complex molecules containing multiple chemically equivalent hydroxylation sites is required.

Over the past two decades, protein engineering has evolved from directed evolution to more sophisticated strategies. In particular, recent developments in computer-assisted design and AI-driven approaches have proven to be invaluable for tailoring cytochrome P450 properties. For instance, we used Hamiltonian Replica Exchange Molecular Dynamics (H-REMD) to analyze substrate binding poses in P450 BM3 mutants.1 This analysis revealed positions in the active site that stabilized productive substrate orientations while destabilizing unproductive ones. These findings led to the development of highly regioselective enzyme variants.

Although site-specific mutagenesis can improve enzyme selectivity, it often results in reduced activity or stability or insufficient protein expression. To address this trade-off, we employed the PROSS algorithm, which predicts stabilizing mutations without compromising enzyme selectivity.2 When applied to CYP107Z mutants, PROSS-based modifications significantly increased soluble protein expression, achieving a six-fold increase over the parental variant, while maintaining activity and selectivity.

Additionally, P450-catalyzed reactions suffer from other limitations, including the lack of identified or compatible redox partner proteins, an imbalanced ratio of P450 enzymes to their redox partners and low substrate solubility in water. These aspects will also be discussed in the talk.

1. Petrović D. et al. J. Chem. Inf. Model. 2018, 58, 848-858.

2. Goldenzweig, A. et al. Mol. Cell. 2016, 63, 337-346.