2019 Theses Doctoral

Engineering Approaches to Control Activity and Selectivity of Enzymes for Multi-Step Catalysis

Abdallah, Walaa Khaled

Enzymes are desirable catalysts as they may exhibit high activity, high selectivity, and may be easily engineered. Additionally, enzymes can be mass-produced recombinantly making them a potentially less expensive option than their organic or inorganic counterparts. As a result, they are being used more in industrial applications making their relevance ubiquitous. In this work, various engineering approaches were developed to control the activity and selectivity of enzymes for multi-step catalysis. Unlike nature, many industrial processes require multiple steps to produce the desired product, which is both timely and expensive. Through the use of enzymes, biosynthesis can be used to develop efficient multi-step catalytic cascades.
The majority of this work focused on engineering a hyperthermophilic enzyme from the aldo-keto reductase (AKR) superfamily, alcohol dehydrogenase D (AdhD) from Pyrococcus furiosus, to develop approaches to control activity and selectivity. As the AKR superfamily contains many unifying characteristics, such as a conserved catalytic tetrad, (α/β)8-barrel quaternary fold, conserved cofactor binding pocket, and varying substrate loops, the approaches developed here can be applied to many enzymes. AKR members participate in a broad range of redox reactions, such as those involving aldehydes, hydrocarbons, xenobiotics, and many more, and are necessary in physiological processes in all living systems, making these enzymes industrially relevant. AdhD in particular can oxidize alcohols or reduce aldehydes/ketones in the presence of NAD(P)(H). Furthermore, the tools utilized here are modular and can be used to develop pathways with enzymes from different superfamilies’ to expand their current capabilities.
In our initial engineering efforts, AdhD cofactor selectivity was broadened or reversed through site directed mutations or insertions in substrate loop B, on the back side of the cofactor binding pocket. To further examine how substrate loops affect cofactor selectivity, allosteric control was added to AdhD through the insertion of a calcium-dependent repeat-in-toxin domain from Bordetella pertussis. Through the chimeric protein, β-AdhD, we demonstrated that the addition of calcium shifts cofactor selectivity in real-time, reminiscent of a protein dimmer. Our next focus shifted towards unnatural amino acid incorporation to add an extra level of selectivity to AdhD. This was done by merging the properties of AdhD and an organic catalyst, TEMPO, for selective alcohol oxidation. We also demonstrated the ability to impart enzymatic selectivity onto an organic catalyst. This was done both in solution and in AdhD hydrogels for added functionality. The next study focused on increasing catalytic efficiency while retaining AdhD structure by engineering the microenvironment of AdhD with supercharged superfolder GFP (sfGFP). The complex interplay between salt, pH, and protein charge was studied and it was determined that catalysis is a function of protein charge, which can affect apparent local ionic strength. The final study focused on utilizing the previous tools examined to engineer substrate channeling in a multi-step cascade with hexokinase II (HK2), sfGFP, and glucose-6-phosphate dehydrogenase (G6PD).
In conclusion, we have utilized a myriad of tools to develop engineering approaches to regulate AdhD activity and selectivity. These tools were then extended to engineer substrate channeling in a three-enzyme system. These approaches are modular and provide a foundation for the development of multi-step catalytic cascades.