BASF, University of Graz develop computer model to boost enzyme performance

In Austria, chemical giant BASF, the University of Graz, and the Austrian Research Centre of Industrial Biotechnology have developed a new, computer-assisted model to optimize efficiency of enzyme-based production processes and, in turn, enable new biocatalytic production processes to be scaled up faster from the lab to industrial manufacturing.

BASF  uses enzymes to make products such as vitamins, flavorings or ingredients for cosmetics and detergents. However, enzymes are very sensitive and stop working properly if, for example, the temperature is too high. “They are then no longer correctly folded and lose their three-dimensional structure, which means no further catalytic reactions can take place in their active center,” said Dr. Stefan Seemayer, global head of computational protein engineering at BASF.  Enzymes also cannot function optimally when temperatures are too low, producing lower volumes of the desired product. The activity of the biocatalysts can also be influenced by substances contained in the reaction medium, such as solvents. 

“In order to get the largest possible amounts of the desired product, we need to find the optimal point for the enzyme, where both the reaction temperature and the solvent concentration result in the highest possible activity,” Seemayer said.

In the past, determining this optimal combination of temperature and solvent concentration was a complex process involving many laboratory experiments. Researchers from BASF, acib and the University of Graz have now developed a regression model as an extension of conventional biochemical models. The model makes it significantly easier to determine the optimal combination. Only a few preliminary lab tests, such as determining the unfolding curve of the enzyme, are necessary. The obtained data are entered into the computer model, which then computes the optimal combination of reaction temperature and solvent concentration for the best-possible enzyme performance. With this new method, different enzymes can be compared more easily with each other, and their performance can be optimized.