Dr Matthew Quesne

Introduction

My PhD was under the supervision of Dr Sam de Visser in Manchester, where I specialised in modelling dioxygenase enzyme reaction mechanisms. I then moved to Poland for a 2-year PostDoc in the group of Dr Tomasz Borowski, looking at metal dependant enzyme catalysed reactions. My second PostDoc focused on modelling heterogenous catalysts for CO₂ utilisation. During my time at the UK catalysis Hub, my research focused on the translation of elements from enzymatic catalysts into mesoporous material catalysts (amongst other projects). My group takes an interdisciplinary approach to catalytically upgrading greenhouse gases, such as CO₂ and methane, into industrially relevant platform chemicals.

We study natural catalysts, which are ahead of us by millions of years of evolution. We use thermodynamical descriptors such as active site basicity or ionisation potentials to computationally screen synthetic alternatives.

We work with experimental colleagues to predict the effect of point mutations on enzyme function. We use a wide range of computational techniques including: Quantum Mechanics Molecular Mechanics (QM/MM), Molecular Dynamics (MD) and Density Functional Theory (DFT). Currently, my group is helping to develop the ChemShell QM/MM code to improve its functionality for studying enzymatic reaction mechanisms.

Current major projects include:

• Computational screening of covalent polymers for direct methane conversion to methanol, informed by cytoplasmic soluble methane monooxygenase.
• Computational screening of metal exchanged zeolites, informed by bioengineered P450, heme dioxygenases.
• Multiscale modelling of proteins immobilised on support materials, using multiscale techniques.
• Computational screening of nonnative functionality in synthetic enzyme polymorphs.

Detailed research programme

Computational chemistry techniques are diverse, this is because depending on the chemical problem to be studied, there is a large amount of different system sizes and time scales involved. In general, there will always be a trade-off between the size of a model and how accurate the computational method. The same trade-off is needed for timescales, therefore, a large model evolving over a long time can only be realistically calculated by a lower level of theory. In our group we use a multiscale approach that treats a small amount of a catalyst (where the electrons move) at a higher level of theory then the majority of the system. We then take high quality thermodynamical data to screen modelled changes in a catalyst to predict effects on reactivity.

Cytoplasmic soluble methane monooxygenase is one of the two direct methane conversion polymorphs able to efficiently and selectively oxidise methane. My group is interested in using QM/MM to calculate the thermodynamical properties of this enzyme’s active site. We will use a valence bond approach to transform this computational aided understanding into descriptors to screen a new class of synthetic covalent polymers for potential reactivity.

We also apply QM/MM techniques to study the effects of the introduction of non-canonical amino acids into first and second coordination spheres of enzyme active sites. We use QM/MM techniques to generate 2D potential energy surfaces and to reproduce changes of rate-limiting barrier hights. We also used state-of-the-art modelling to rationalise the thermodynamic/kinetic reasons for mechanistic change.