Rational design of tryptophan synthases through conformation and correlation-based allosteric networks

ENG- Enzymes are remarkable biological catalysts that operate under mild conditions with high specificity and efficiency. The improvement of enzyme performance is often achieved through experimental methods like Directed Evolution (DE), which can produce highly efficient variants but is costly, time-consuming, and does not often provide an explanation of why certain mutations work better. Computational enzyme design is emerging as a powerful alternative, offering the ability to predict and rationalize the effects of mutations, although it is still developing compared to DE. This thesis focuses on understanding and improving the function of an enzyme complex called Tryptophan Synthase (TrpS), composed of two parts: TrpA and TrpB. These two parts regulate each other through long-range interactions known as allosteric effects, which significantly influence their activity and flexibility. When separated, each part is less efficient, but together they adopt highly active forms. Understanding these communication pathways is key to designing improved enzymes that are efficient even on their own. Advanced computational techniques such as Molecular Dynamics (MD) simulations, Density Functional Theory (DFT) calculations, and correlation-based tools like Shortest Path Map (SPM) were used to reveal how TrpS subunits communicate and change shape during the reaction. In one part of the thesis, TrpA and its natural blueprint ZmBX1 were studied to identify key flexible regions important for enzyme activity. Through rational design involving specific mutations and structural modifications, TrpA variants with much higher efficiency and reduced dependence on TrpB were created. The thesis also explores how TrpB alone can retain high activity by analyzing its dynamic behavior and predicting important mutations using a combination of SPM, AlphaFold2 models, and different force field choices. This work demonstrates how computational approaches can successfully guide the design of enzymes with enhanced properties. Overall, this thesis shows how a deeper understanding of enzyme dynamics and allosteric regulation, combined with powerful computational tools, can lead to the rational design of more efficient enzymes, offering a strong alternative to traditional experimental methods