Enzymes are widely used in the production of pharmaceuticals, specialty chemicals, cosmetics and food products. By enabling chemical reactions under milder conditions, they can reduce energy use and improve the efficiency and environmental performance of industrial processes.
The ZenBio project, led by researchers at the Fraunhofer Institute for Cell Therapy and Immunology’s Bio-analytics and Bioprocesses branch in Potsdam, aims to accelerate the development of enzymes adapted to industrial conditions. The platform combines cell-free protein synthesis, laboratory automation and artificial intelligence to generate and analyze large numbers of enzyme variants.
Designing custom enzymes remains difficult because changes in genetic sequences can affect their three-dimensional structure and function in complex ways. Experimental information about these relationships is also limited. The ZenBio team plans to address this gap by producing many variants and recording how they perform under different conditions, creating a data set for AI-assisted design.
The system uses genetic instructions for an enzyme together with a cell extract containing the components required to produce it. This allows the proteins to be made directly in the extract, rather than inside living cells. Compared with cell-based production, the method can reduce waiting times and may help researchers work with variants that are difficult to produce or fold correctly in cells.
The cell-free approach also supports miniaturization and precise control of the reaction conditions. A robotic pipetting station is intended to produce and examine up to 1,500 enzyme variants in parallel. The researchers expect to determine within a few hours whether variants can tolerate factors such as temperature, pH changes or organic solvents.
The resulting measurements will be used by the AI platform to identify links between enzyme sequences and their properties, then suggest variants for subsequent design cycles. The process is intended to progressively tailor enzymes to industrial requirements, including higher temperatures, altered pH levels and exposure to solvents.
The project has completed the first stage of its automated workflow. The team will now continue miniaturizing and automating the platform and test it with an initial enzyme class.
Sources: Phys.org.