Kinetic Characterization of 100 Glycoside Hydrolase Mutants Enables the Discovery of Structural Features Correlated with Kinetic Constants

Abstract

The use of computational modeling algorithms to guide the design of novel enzyme catalysts is a rapidly growing field. Force-field based methods have now been used to engineer both enzyme specificity and activity. However, the proportion of designed mutants with the intended function is often less than ten percent. One potential reason for this is that current force-field based approaches are trained on indirect measures of function rather than direct correlation to experimentally-determined functional effects of mutations. We hypothesize that this is partially due to the lack of data sets for which a large panel of enzyme variants has been produced, purified, and kinetically characterized. Here we report the kcat and KM values of 100 purified mutants of a glycoside hydrolase enzyme. We demonstrate the utility of this data set by using machine learning to train a new algorithm that enables prediction of each kinetic parameter based on readily-modeled structural features. The generated dataset and analyses carried out in this study not only provide insight into how this enzyme functions, they also provide a clear path forward for the improvement of computational enzyme redesign algorithms.

Thermal stability and kinetic constants for 129 variants of a family 1 glycoside hydrolase reveal that enzyme activity and stability can be separately designed

“…Michaelis-Menten parameters for each mutant are determined as described previously [10]. For previously characterized mutants, kinetic constants are drawn from the publication.…”

“…The data set of protein expression, thermal stability, k cat , K M , and k cat /K M enabled us to evaluate the performance of these three current force-field–based approaches to modeling stability changes caused by mutations, building on previous work where we evaluated the ability of Rosetta to predict changes in kinetic constants for this model system [10]. Similar to the original study, we found only a weak correlation (PCC <0.3) between predicted and observed stability for each of these established protocols.…”

Design to Data for mutants of β-glucosidase B from Paenibacillus polymyxa: Q22T, W123R, F155G, Y169M, W438D, V401A

“…Foldit Standalone 7 was used to design and model six variants of BglB as previously described. 6 The six mutational changes were scored by the Rosetta energy function and a total system energy score (TSE) is given. All variants were chosen with no greater than five TSE change between the wild type and variant scores to ensure proper folding of the enzyme.…”

“…However, machine learning analysis indicated that algorithms, which predict function based on calculated structural features, could be developed with more training data. 6 Here, six single amino acid variants of BglB are quantitatively characterized and compared with computational predictions through a design-build-test methodology. The mutants are first designed and visualized using Foldit, 7 and predictions for enzyme stability are quantified through a measure of total system energy score (TSE).…”

Repurposing citizen science games as software tools for professional scientists

“…An early prototype of Foldit Standalone was used to iterate on enzymes designed by Foldit players [13]. While still in development, Foldit Standalone had already begun to play a critical role in both research and education [5, 41]. Students have used this tool to re-engineer the reaction specificity of proteases for the development of therapeutics for Anthrax [49] and Celiac Disease [19].…”