Multi-scale enzyme modelling for SynBio: optimizing biocatalysts for selective synthesis of bioactive compounds

It is becoming increasingly popular to use the powerful principles present in nature to our advantage. A key example is the extraordinary ability of organisms to make molecules with high specificity (pure, potentially complex molecules are obtained) and efficiency (little energy is used). Nature uses enzymes, proteins that act as catalysts to promote chemical reactions, to achieve this. These enzymes typically work under mild conditions. Enzymes are already used in industry to help make molecules that we require in cost-efficient, comparatively green and sustainable processes. Nature has not, however, provided us with an enzyme to suit the production of every desired molecule; typically, enzymes only catalyze specific chemical reactions with specific starting materials. But the process of evolution teaches us that enzymes are malleable for engineering different properties. For example, making small changes (mutations) in specific amino acids (the building blocks of proteins) of enzymes can allow these enzymes to accept different substrates and thereby catalyze the formation of new, desired molecules.
Even though it is possible to determine the positions of atoms in an enzyme with great detail (e.g. using X-ray crystallography), the full effects of making changes to amino acids are not evident. This limits researchers in assessing what the (beneficial or non-beneficial) effects of such mutations are. For example, changing a single amino acid can affect how efficient an enzyme works, by changing how well the starting material binds, how efficient the starting material is converted, and how stable the enzyme is. I have been at the forefront of developing and employing methods that combine quantum mechanics and standard (Newtonian) mechanics to simulate chemical reactions in enzymes. With these methods, computer simulations can be used to assess the different possible effects of amino acid changes. The proposed research will develop efficient protocols to do this, and enhance the methods so that they can be applied to more complex enzyme systems and reactions.
The enzyme systems under investigation are examples of enzymes that can make many different and potentially valuable molecules. Anti-influenza drugs, anti-cancer compounds, antibiotics as well as popular natural flavour compounds are examples of molecules that could be produced with the help of these enzymes. By gaining knowledge on how these enzymes work and developing protocols to modify them, we can obtain ways to produce new beneficial compounds (such as drugs) in a cost-efficient and sustainable way.

Multi-scale (and especially QM/MM) modelling has become an established technique to investigate reactions in enzymes (as evidenced by the 2014 Nobel Prize in Chemistry). The time is now opportune to take the next step and apply multi-scale simulation to 'real world' problems. The proposed research will 1) develop biomolecular simulation methodologies and protocols to help optimize enzyme biocatalysts; 2) obtain new insight into enzyme catalysis and (in particular) enzyme (stereo)selectivity; 3) use the developed protocols and insight to modify and optimize enzymes (in collaboration with experimentalists) to obtain efficient and highly specific biocatalysts for the biosynthesis of natural product analogues. Initially, state-of-the-art QM/MM reaction modelling and binding free energy simulation techniques will be applied to develop efficient protocols for screening kinetic parameters (kcat and KM, respectively) of a well-studied sialic-acid producing aldolase. Subsequently, similar screening protocols will be developed (and verified) for more complex systems, that are attractive due to their biosynthetic power and versatility for the production of valuable natural products. In order to confidentially predict kinetic parameters for these systems, several step changes in simulation are required. Specifically, chemically complex key reaction steps in a well-characterized fungal terpene synthase (investigated to learn how to modify terpene biosynthesis) will require the use of new computational approaches (e.g. DFT-embedded coupled-cluster calculations). Further, understanding and modifying the biologically complex ketoreductase (KR) activities of two different polyketide synthase (PKS) systems will require the combination of QM/MM and protein-docking (e.g. for modelling the reactive complex in a well-studied PKS II KR) as well as QM/MM and homology & coarse-grained modelling (for the modelling of KR domains in a multi-enzyme trans-AT type I PKS).

The proposed work will lead to new, detailed insights into enzymes that can be used as biocatalysts, and the development of 'proof-of-priniciple' engineered enzymes that are optimized for the production of valuable chemicals and potential novel drug compounds. In the process, computational tools, methods and protocols to help engineering of other enzymes and proteins will be developed. These advances in computer simulation and screening of enzyme variants will facilitate basic biochemical and synthetic biology research (e.g. by helping experimental scientists better understand and predict the effect of mutations on structure, dynamics and catalysis) as well as applied commercial research (e.g. by aiding re-design of enzymes to obtain novel biocatalysts). New examples of engineered biocatalysts and the methods to obtain them will be of high interest to the biotech industry, due to their ability to enhance the cost-effectiveness and sustainability of synthesising chemical compounds.

Apart from these short-term benefits to a wide range of academic and industrial researchers, the proposed work has significant potential impact in the medium and long term.
In the medium term, the methods and protocols developed can significantly reduce the time and resources spent to optimize biocatalysts for particular reactions. This has direct economic and environmental benefits: biocatalysts can be brought to market sooner, with reduced use of chemicals and energy in its development process. UK-based SMEs (and larger companies) are active in the field of biocatalyst optimization and sales, and the national economy can therefore benefit significantly.
In the medium to long term, the application (through commercial availability) of biocatalysts optimised by using the protocols developed in the work can have enormous economic and environmental benefits by allowing the sustainable production of desired molecules, including fine-chemicals, drugs and biofuels. Biocatalysts are already starting to transform our current chemical industry by improvements in the methods, cost-effectiveness, safety, health, and environmental impact of the processes involved, and these impacts will be extended by the availability of additional biocatalysts. Specifically, the biocatalytic systems studied in this research are those that can be employed in the (sustainable) production of novel anti-viral, antibiotic and anti-cancer compounds, which has obvious significant long-term public health benefits in addition to the economic benefits.

Finally, the research programme will help bring together and enhance two research communities that are strong within the UK: biomolecular simulation and chemical biology/enzymology. The UK should capitalise on the strengths of these communities to catch up and, in future, overtake efforts in other countries in the area of biocatalyst (re)design. The close collaboration with experimental groups, and the training and knowledge exchange activities planned, will help develop a new generation of cross-disciplinary researchers that are well-equipped for research and development in synthetic biology, but also in other areas that require a cross-disciplinary outlook (both in academia and industry). The related professional skills acquired by the researchers involved (significant IT, communication, analytical thinking and time management skills) will make them valuable for many/all employment sectors. This impact on UK 'human capital' will further contribute to the longer term benefits.