Reaction kinetics identification for the analysis of reaction mechanisms of ThDP-dependent enzymes
Final Report Abstract
The key outcome of this project is a kinetic modelling framework for the branched reaction networks occurring in ThDP-dependent enzyme catalysis. We have designed and analysed a workflow that supports designing, performing, and analysing meaningful kinetic experiments. The key arguments in order of appearance in the new workflow are: First, the kinetic analysis relies on progress curve analysis, requiring relatively few individual experiments – however, with many more concentration analyses, which simultaneously provide information on rates, substrate affinity and inhibition, enzyme inactivation and reaction equilibria. Second, the set of initial progress curve experiments should be planned as Box Behnken type fractional factorial design, with the highest concentrations being the technically relevant or feasible concentrations. Third, these progress curves can be analysed simultaneously according to a set of alternative kinetic models. Fourth, we have come up with a set of new mechanistic rate equations that cover branched reaction pathways and have shown that all of these rate equations are structurally identifiable. Fifth, we can test a priori whether all kinetic parameters in a model are identifiable. Sixth, we can analyse a range of competing model simultaneously since the parameter estimation procedure is computationally cheap, and thereby, decide for the best model in terms of fit quality (residual sum of squares - RSS) and information content (Akaike’s information criterion - AIC). Seventh, we also obtain information on the accuracy of the parameter estimates by analysing the variance-covariance-matrix. Eighth, we can use sensitivitybased optimal experimental design (OED) to improve the parameter quality and E-optimal OED to reduce parameter correlations, if we are not yet satisfied with the parameter quality. We have validated the modelling framework for a range of ThDP-enzyme-catalysed reactions, namely the self-carboligation of benzaldehyde to (R)-benzoin by the benzaldehyde lyase from Pseudomonas fluorescens (PfBAL), the cross-carboligation of propanal and benzaldehyde to hydroxylbutyrophenone by PfBAL, and the carboligation of pyruvate and benzaldehyde to (S)-phenylacetylcarbinol using Escherichia coli MenD. In all cases, excellent model fits were obtained, and very good parameter accuracies if sufficient experimental data were available, namely in case of the PfBAL-catalysed reactions. We could clearly show that the rate of substrate-dependent inactivation can be distinguished for different substrates. We could also discriminate the influence of alternative substrates in the same self-carboligation with respect to their affinity and the substrate inactivation.
Publications
-
2010. Systematic determination of intrinsic reaction parameters in enzyme immobilizates. Chem. Eng. Sci. 65:2491–2499
Zavrel M, Michalik C, Schwendt T, Schmidt T, Ansorge-Schumacher M, Janzen C, Marquardt W, Büchs J, Spiess AC
-
2016. BioCatNet: a database system for the integration of enzyme sequences and biocatalytic experiments. ChemBioChem 17:2093–2098
Buchholz PCF, Vogel C, Reusch W, Pohl M, Rother D, Spieß AC, Pleiss J
-
2017. How graphical analysis helps interpreting optimal experimental designs for nonlinear enzyme kinetic models. AIChE J. 63:4870–4880
Ohs R, Wendlandt J, Spiess AC
-
2017. Thermodynamic activity-based intrinsic enzyme kinetic sheds light on enzyme-solvent interactions. Biotechnol. Prog. 33:96–103
Grosch J-H, Wagner D, Nistelkas V, Spieß AC
-
2018. Simultaneous Identification of Reaction and Inactivation Kinetics of an Enzyme-catalyzed Carboligation. Biotechnol. Prog. 34:1081–1092
Ohs R, Leipnitz M. Schöpping M, Spiess AC
-
2019. Progress Curve Analysis Within BioCatNet: Comparing Kinetic Models for Enzyme-Catalyzed Self-Ligation. Biotechnol. J. 14:1800183
Buchholz PCF, Ohs R, Spieß AC, Pleiss J
