(T1230-05-25) Applying Conformational Dynamics in the Evolution of Enzymes for Biocatalysis

Purpose: Enzymes enable unique natural and specific functionalization which are difficult to achieve by conventional chemical processes. However, to meet industrial and chemical requirements for enzyme to catalyze reactions that usually do not occur in nature, they often need to be reshaped. That will involve for instance changes of multiple residues along the whole sequence of the protein. In consequence, this will affect the conformational flexibility of the enzyme which plays an important role in its function and regulation. For example, this will modify the substrate binding, allosteric regulation, or product release. With the aim to adapt enzymes to meet requirements for the development of high performance specific biocatalytic processes, we have established a robust methodology platform that cover sequence mining, automated in vitro screening, and dynamic computational modeling. We have applied this methodology with the aim to identify an enzyme from the family of Ene‑Reductase (ERED) able to catalyze an oxidation reaction in chemical process. This way, the new biocatalytic process, could allow the reduction of the number of steps compared to a classical chemical process and will avoid the use of expensive transition metal.

Methods: Several ERED candidates were screened for their potential activity in such an oxidation reaction but none of the enzymes tested were able to catalyze the desaturation of our target substrate. Among these candidates, one enzyme from Geobacillus kaustophilus was described for its ability to desaturate several simple cyclic ketones into the corresponding enones and testosterone into boldenone. Furthermore, a specific mutant was designed to exhibit an increased redox potential, enabling therefore a better conversion of testosterone. The testosterone was the closest substrate to our target and notwithstanding a close structure, the mutated ERED from Geobacillus kaustophilus did not display any oxidation activity on our substrate. To understand further the difference in activity between our substrate and testosterone and to design the mutant oxidizing our specific substrate, we applied a conformational dynamic approach (shortest path map) developed by Silvia Osuna.

Results: Both proteins, wild-type (WT) and mutant were modelled to identify their conformational landscape in inactive, unbound state. The simulations on the WT protein have revealed a higher flexibility, populating different conformations along the conformational landscape, while the mutant form seems to have a reduced number of conformations. Focusing on the mutant, the binding conformation was analyzed with both substrates. We found the distance of binding of the testosterone in the active pocket highly compatible with a catalytic activity. On the contrary our specific substrate displays an unstable binding in the active site. The strong stability of the testosterone in the active site will explain its efficient conversion. Furthermore, we identified the key residues involved in the stabilization of the testosterone as well as residues involved in the stabilization of the cofactor.

Conclusion: Using these preliminary data, we have identified the shortest path map for the mutant. The analysis of these data combined with Rosetta enzyme design protocol and short molecular dynamic simulation will help to understand the correlation between the catalytic residues and other parts of the protein. Hence, this will give us some insights on how to redesign the active site to stabilize the specific substrate. At that stage several point mutations have been identified and will be simulated as well as tested in vitro to alter the shape of the active site, to determine, firstly, if that will allow to accommodate the specific substrate, and secondly if its conversion will be performed.