Rapid Model-Based Optimization of a Two-Enzyme System for Continuous Reductive Amination in Flow

Abstract

Enzymes are increasingly combined into multienzyme systems for cost and productivity benefits. Further advantages can be gained through the use of immobilized enzymes, allowing continuous biotransformations in flow. However, the optimization of such multienzyme systems is challenging, particularly where immobilized enzymes are used. Here, we meet this challenge using both mechanistic and empirical modeling to optimize a reaction involving a reductive aminase and a glucose dehydrogenase for continuous biocatalytic reductive amination in flow. Crucially, the construction of the mechanistic model was achieved quickly, with only a few important parameters determined experimentally, and ensemble modeling used to facilitate the use of estimates or literature values. Upon reaching the limits of the mechanistic model's capabilities, we show that solution space can be further explored using a definitive screening design to generate an empirical model of the reaction, using the mechanistic model's prediction as a starting point. We demonstrate the use of this empirical model to design optimal processes for either high productivity or to minimize necessary cofactor and cosubstrate concentrations. Our results demonstrate that the synergistic use of both mechanistic and empirical modeling offers a route for rapid optimization of multienzyme systems of immobilized enzymes in flow with minimal experimental effort.

Development of Continuous Flow Systems to Access Secondary Amines Through Previously Incompatible Biocatalytic Cascades**

“…For the generation of secondary amines such as 3 from aldehyde 1 , RedAm and glucose dehydrogenase (GDH) combinations can be used and have also been demonstrated in flow systems previously. [ 14 , 31 , 32 ] Here, the reductive aminase from Ajellomyces dermatitidis ( Ad RedAm, 200 mg, 10 wt %) and the GDH from Bacillus megaterium ( Bs GDH, 10 mg, 10 wt %) were immobilized and both loaded into a packed bed reactor. Pleasingly, this also gave full conversion of the in situ generated hydrocinnamaldehyde 1 to the corresponding N ‐allyl amine 3 at steady state which was maintained for four hours (STY: 2.1 g L −1 h −1 ) (Table 1 ).…”

“…For the generation of secondary amines such as 3 from aldehyde 1 , RedAm and glucose dehydrogenase (GDH) combinations can be used and have also been demonstrated in flow systems previously [14, 31, 32] . Here, the reductive aminase from Ajellomyces dermatitidis ( Ad RedAm, 200 mg, 10 wt %) and the GDH from Bacillus megaterium ( Bs GDH, 10 mg, 10 wt %) were immobilized and both loaded into a packed bed reactor.…”

Selective oxidation of alkyl and aryl glyceryl monoethers catalysed by an engineered and immobilised glycerol dehydrogenase

“…In the last years, the immobilization of engineered enzyme variants has allowed their application under industrially relevant conditions. 16,[33][34][35] To that aim, selective immobilization protocols that assure the activity and increase the stability of the enzymes upon the attachment to the solid carrier are pretended.…”

The rise of continuous flow biocatalysis – fundamentals, very recent developments and future perspectives

“…Finnigan et al combined both mechanistic and empirical modeling to optimize a two-enzyme system for continuous reductive amination. 212 The multi-dimensional optimization shows a clear advantage compared to traditional time consuming one-factor at a time (OFAT) approaches. In addition, the availability of these models can be extended by modern public platforms such as github, and they can be operated with cloud-based Python notebooks.…”