Adrian Mulholland

University of Bristol

Multiscale simulation for protein design and enzyme evolution: from drug resistance to biocatalysis

Adrian Mulholland*

University of Bristol

* Presenting author

Combined quantum mechanics/molecular mechanics (QM/MM) methods have proved widely useful in modelling of reactions in proteins (identifying mechanisms and catalytic interactions) in protein-ligand binding and enzyme inhibition, in spectroscopy and other electronic structure calculations on biomolecules. Recent examples of QM/MM modelling of enzyme-catalysed reactions include the mechanism of choline cleavage in the glycyl radical enzyme choline trimethylamine-lyase (CutC) [1] and identifying the interactions that distinguish between substrates (oxytocin) and inhibitors (angiotensin IV) of insulin-regulated aminopeptidase (IRAP) [2]. QM/MM simulations have revealed mechanisms of bacterial enzymes that break down beta-lactams, causing antibiotic resistance [3]. QM/MM simulations also reveal determinants of their catalytic activity. They show that the active site electric field is a crucial factor in catalysis, e.g. in determining the ability of class A beta-lactamases to hydrolyse carbapenems [4]. Electric fields are optimized in natural enzymes for specific catalytic activities and provide a guide for enzyme design [5]. Electric field calculations and molecular dynamics simulations can be included in enzyme design to accelerate enzyme engineering and design [6]. Correlated ab initio methods, which can give chemically accurate productions of reaction barriers, can be applied, e.g. ab initio QM embedded in DFT, all within a MM environment in multiscale calculations, to overcome limitations of DFT [7]. QM/MM simulations can generate data to train efficient machine learning models of reactivity, which can provide accurate, efficient models for reaction simulations. QM/MM calculations can also investigate and predict electronic properties of proteins such as spectra [8] and can be included in practical protein design and engineering workflows to optimize such properties e.g. in the design and development of proteins as photovoltaics and for biomolecular electronics [9].

References

[1] Hanževački M. et al. J. Phys. Chem. B 129 9322 (2025)

[2] Hanževački M. et al. J. Chem. Inf. Model. 65 6261 (2025)

[3] Beer M. et al. ACS Infectious Diseases 8 1521 (2022)

[4] Jabeen H. et al. ACS Catalysis 14 7166 (2024)

[5] Suardíaz R. et al. Angew. Chem. Int. Ed. 64 e202515743 (2025)

[6] Merlicek L.P. et al. Angew. Chem. Int. Ed. 64 e202507031 (2025)

[7] Ranaghan K.E. et al. J. Chem. Inf. Model. 59 2063 (2019)

[8] Pfalzgraf HE et al. Chemical Science 16 16266-16283 (2025)

[9] Hutchins GH et al. Proceedings of the National Academy of Sciences USA 120 e2306046120 (2023)

All Sessions by Adrian Mulholland

11:00 - 11:30
FUNCHAL 500 anos

Invited Lecture 33 - Adrian Mulholland

Multiscale simulation for protein design and enzyme evolution: from drug resistance to biocatalysis

Scroll to Top