Computational design of a thermostable de novo biocatalyst for whole cell biotransformations

Several industrially relevant catalytic strategies have emerged over the last couple of decades, with biocatalysis gaining lots of attention in this respect. However, this type of catalysis would be even more thriving if customizable and stable protein scaffolds would be readily available. Such highly stable protein scaffolds could act as asymmetric reaction chambers in catalyzing a wide range of reactions. In this study, we are detailing the design and experimental characterization of computationally designed de novo proteins with a non-natural fold, which act as biocatalysts. The initial design and several variants of it, catalyze the aldol condensation and the retro-aldol reaction. All designs form a helical barrel structure comprised of six antiparallel straight helices connected by five loops, resulting in an open channel with two accessible cavities. The designs exhibit high thermal stability and an excellent overall fit between measured and calculated scattering profiles from small-angle x-ray scattering experiments. An experimentally determined crystal structure of a surface redesigned variant shows close to perfect agreement to the design model. To highlight the versatility and mutational tolerance of this fold, we rationally designed variants in which the active site is placed at different positions throughout the central channel of the original design. Finally, we showcase that the initial design as well as variants exhibit significant aldolase/retro-aldolase activity in whole cell biotransformations, making them the first de novo biocatalysts to be tested in this form. With a tolerance of up to 20% of organic solvent, these designs hold promise for further utilization and optimization in the field of white biotechnology.