Modellgestützte Untersuchung und Reduktion von Selektivitätsbeschränkungen bei der enzymkatalysierten Synthese chiraler Hydroxyketone aus prochiralen Diketonen
Zusammenfassung der Projektergebnisse
The project aimed for a systematic description and evaluation of thermodynamic, kinetic and molecular effects influencing the regio- and stereoselectivity in alcohol dehydrogenases (ADH) catalysed diketone reduction reactions. Based on these results, a rational approach should be developed to reduce the selectivity constraints by molecular or protein engineering. As model system the reduction of 2,3-pentanedione by the carbonyl reductase 2 from Candida parapsilosis (CPCR2) was applied. For this reaction and for a selection of 19 structurally different diketone compounds, substrate specificity, kinetic parameters, reaction products and stereoselectivity were investigated. The results indicated a significant dependency of all parameters to the structure of the substrates. Due to the close mashed selection of substrates, observed influences could be attributed to certain structural features of the compounds. Based on the model system we applied modelling techniques to gain understanding in the mechanisms behind the enzymatic system. Therefore we established a tool for data preprocessing, model discrimination and parameter estimation. Herein, we show the application of the incremental method for model discrimination. Based on the Akaike information criterion, we were able to discriminate between a huge variety of candidate models. This procedure was broken down in the identification of structural models and later on including inhibition and irreversibilities into the structure. Subsequent parameter estimation and sensitivity analysis proved to be a valuable approach to uncover macroscopic phenomena like irreversible reactions. Moreover, modelling of the half reactions is sufficient to establish a model network (shown in 6.4). This model network can be used to improve the product yield or other measures with the help of optimal experimental design methods, which were also implemented. The investigation showed that most cyclic diketones were not converted at all. Thus, based on the novel crystal structure of CPCR2, created in collaboration with Gideon Grogan (University of York), a mutational analysis was performed to increase the substrate specificity of CPCR2. With three neighbouring residues in the active site a mutational hot spot could be identified where the conversion of single cyclic and bulky substrates could be improved. For those α- and βcyclic diketones, still not converted by CPCR2 a model enzyme capable of these reactions should be identified. With ThaADH from Thauera aromatica such an enzyme was detected by screening of metabolic pathways. The characterization yielded a preference for the desired α- and β-cyclic diketones. Applying a homology model of ThaADH structural differences to CPCR2 were detected, that could explain the divergent substrate spectra.
Projektbezogene Publikationen (Auswahl)
- Investigation of structural determinants for substrate specificity in zinc dependent alcohol dehydrogenases – Biocat 2014. Hamburg (Germany), 31.08.-04.09.2014
Loderer C, Ansorge-Schumacher MB
- Investigation of structural determinants for substrate specificity in zinc dependent alcohol dehydrogenases – VAAM Jahrestagung 2014. Dresden (Germany), 05.10.-08.10.2014
Loderer C, Ansorge-Schumacher MB
- Kinetic parameter estimation based on mechanistic modeling of Candida parapsilosis carbonyl reductase (CPCR), Biocat2014, Hamburg , 31.08. – 04.09.2014
Wagner D, Loderer C, Ansorge-Schumacher M, Spieß, AC
- Structure of NADH-dependent carbonyl reductase (CPCR2) from Candida parapsilosis provides insight into mutations that improve catalytic properties. ChemCatChem, 6(4): 1103–1111 (2014)
Man H, Loderer C, Ansorge-Schumacher MB, Grogan G
(Siehe online unter https://doi.org/10.1002/cctc.201300788) - Carbonyl reductase of Candida parapsilosis–Stability analysis and stabilization strategy. Journal of Molecular Catalysis B: Enzymatic, 112:45–53 (2015)
Grosch JH, Loderer C, Jestel T, Ansorge-Schumacher M, Spieß AC
(Siehe online unter https://doi.org/10.1016/j.molcatb.2014.12.001)
