Inline quantification of NADH und NADPH under process conditions
Final Report Abstract
Nicotine amide dinucleotide cofactors are of paramount importance for the industrial utilisation of the enzyme class of oxidoreductases. For their rational use a knowledge gap exists between predictions based on initial rate measurements and their application in processes. The main goal of the project is the online quantification of cofactors under process conditions by application of fluorescence spectrometry allowing monitoring of the reduced form of the cofactors at µmol L^-1 which are common in their application. Two techniques were established for quantitative NADPH detection: microplate fluorescence measurement and fiber optical fluorescence measurement. Both were validated in the range of 1 to 100 µmol L^-1 NADPH. Stable calibration factors were obtained for routine use under process conditions. As a model reaction the enzymatic reduction of 2-octanone to (R)-2-octanol via alcohol dehydrogenase from Lactobacillus brevis (LbADH) was thoroughly investigated. A monophasic approach was developed with the addition of a ionic liquid to the aqueous buffer phase serving as solubiliser and allowing to work at much higher substrate concentrations. An enzyme coupled cofactor regeneration with glucose dehydrogenase from Bacillus spec (GDH) and glucose as hydride donor (cosubstrate) was used for better utilization of the cofactor (NADP+/NADPH). Beside routine batch experiments, a continuous process was realised. In a configuration of a cascade of two enzyme membrane reactors the process could be demonstrated for more than 1000 hours with LbADH turnover numbers of more than 4×10^7 and space-time yield of up to 34 g L^−1 d^−1 with 99.9% enantioselectivity. Furthermore, downstream processing via adsorption of the alcohols was included, allowing 90% recycle of the aqueous buffer. Thus, the E-factor (amount of waste stream per product) was reduced to 13. For both batch and continuous synthesis a mathematical model with Michaelis-Menten type kinetics has been developed and implemented for process optimisation. As quantitative data about cofactor concentrations under process conditions is unavailable, major part of the work was the cofactor stability measurements for the model enzymatic process. For this, NADPH was identified as the most critical and was focused on. The measurements validate findings from literature, for example deteriorative influence of phosphate, (acetate), elevated temperatures, UV-radiation, and H+-ions. Whereas, Mg2+ -ions, low temperatures, OH–-ions, and the ionic liquid TEGO IL K5 do increase half-life. Unprecedented, higher stability of NADPH at higher initial NADPH concentrations was observed. In general, the complexity of our findings, in combination with data derived from literature, support our recommendation to measure cofactor stabilities for the conditions which will be applied in the intended process as quantitative literature data is sparse and possibly not applicable. Fulfilling the main goal of the project, online monitoring of cofactor concentrations has been demonstrated for the model reaction in batch and continuous mode. Comparison with model predictions as well as with offline measurements has been used as a tool for process optimisation. In batch mode experiments were performed with both multiplate reader and FS-probe detection of NADPH. In all cases reaction times of up to 10 hours were achieved. In this case the implemented mathematical model was able to describe the substrate conversion rather well, whereas deviations between predicted and measured NADPH concentrations were observed, especially for the start of the reaction, presumably due to diffusion effects. For the continuous experiments we demonstrated an application with integrated NADPH monitoring via the FS-probe in a simplified set-up for reaction times of over 3 months. Via offline measurements with the multiplate fluorescence reader, it was confirmed that the fiber optical FS-probe delivers the “true” cofactor concentration. Again the conversion coincided well with the model prediction while the NADPH concentration in the reactor deviated strongly from the prediction, underlining the importance of in-situ measurements. Further on, the membrane inside the enzyme membrane reactor set-up plays an important role in the process. Directly connected with it pressure changes or leaks affect the cofactor concentration immediately. Enzyme stability and productivity is the other major factor. In summary we may say that online NADPH monitoring via FS spectroscopy is possible and it brings further insight into the enzymatic process investigated. It is a useful tool which can be implemented to optimize the process in real time and compensate for diverse effects of pressure or enzyme loss, e.g., by membrane leaching. A limitation is the condition that the substrate and further components of the reaction mixture should not be fluorescent species, as a correction factor should be implemented, which might not be experimentally accessible in each case. Degradation effects due to light source intensity should be taken into consideration as well. Nevertheless, we think that the findings as a result of our work are essential for the rational usage of cofactors in reductive in vitro biotransformations.
Publications
- Reaction Engineering of Biocatalytic Enantioselective Reduction: A Case Study for Aliphatic Ketones, Org. Process Res. Dev. 17/8 (2013) 1027–1035
Susanne Leuchs, Joana Lima-Ramos, Lasse Greiner, Naweed Al-Haque, Pär Tufvesson and John M. Woodley
(See online at https://doi.org/10.1021/op400117t) - “Enantioselective reduction of sparingly water-soluble ketones: continuous process and recycle of the aqueous buffer system”, Green Chem. 15 (2013) 167-176
Susanne Leuchs, Shukrallah Na'amnieh and Lasse Greiner
(See online at https://doi.org/10.1039/c2gc36558h) - „Continuous Enzymatic Synthesis of Enantiopure Aliphatic (R)-2-Alcohols” (DECHEMA Forschungsinstitut Forschungsberichte Nr. 2), Shaker Verlag ISBN 978-3-8440-2249-0 (2013), Dissertation
Susanne Leuchs
