Biocatalysis of poorly water-soluble substrates

Research output: ResearchPh.D. thesis – Annual report year: 2018


View graph of relations

Biocatalysis has grown in industrial application to the point where there are now hundreds of established enzyme-catalysed processes spanning various industries. However, many compounds of interest in such chemical syntheses are poorly water-soluble. If these compounds are to be used as substrates for enzyme-catalysed reactions they will form heterogeneous reaction mixtures because water is the predominant solvent for biocatalysis. With the introduction of multiple phases comes new challenges for experimental assessments and process performance. A chief challenge lies in identifying and addressing the extent of substrate mass transfer limitations into the aqueous phase where biocatalysis takes place.
This thesis aims to explore systematic methods for handling poorly water-soluble substrates in reaction evaluations, which are crucial for biocatalytic reaction design. A standardised method of scoping reactions and gauging their performance is developed and presented. The method involves comparing reaction trajectories performed under varying conditions (enzyme and substrate concentrations) and can distinguish process ‘bottlenecks’ of enzyme kinetics, enzyme stability or non-enzyme related limitations. The method was shown to enable the qualitative (and to an extent quantitative) identification of substrate mass transfer limitations through time-course measurements of substrate conversion.
Major advances in protein engineering techniques have facilitated the ‘tailoring’ of enzymes for specific process requirements (e.g. improved thermostability, solvent tolerance, substrate scope/affinity, and catalytic activity). However, the investment in research effort, time and costs for protein engineering are disproportionate to the number of realised industrial processes. Furthermore, this thesis suggests this is due to a disconnect in the earliest stages of reaction design and conceptualisation where the required process performance is typically not considered or understood. Therefore, process performance requirements are suggested as a series of metrics in this thesis such that they may be used to guide reaction assessments and for benchmarking. These metric requirements are based on the value-class of the product (i.e. process margins) and cost of biocatalyst formulation, and include: reaction conversion, reaction yield, yield on co-substrate (where appropriate), selectivity, product concentration, productivity/space-time yield, specific productivity and biocatalyst yield. These metrics were shown to be appropriate and informative in two case studies.
These two cases studies involved biocatalysis of poorly water-soluble substrates and were chosen in order to demonstrate the application of the proposed reaction assessment and subsequent benchmarking with the process performance metrics.
The first case, which is the enantioselective desymmetrisation of a diester by recombinant pig liver esterase, highlights the importance of quantifying mass transfer limitations when gauging process metrics (especially biocatalyst yield). The required process metrics for a high-value product (100 € kg-1) were all satisfied with the exception of biocatalyst yield. In order to increase the biocatalyst yield to the required range of 10-40 g gbiocat-1, the exceptional metrics of space-time yield (20 g L-1 h-1) or product concentration (74 g L-1) could be reduced by an order of magnitude and still satisfy target process performance metrics. The second case study features the biocatalytic Baeyer-Villiger oxidation of a macrocyclic ketone to its corresponding lactone, and resulted in an even more complex solid-liquid-gas reaction medium. Here, the necessity and usefulness of alternative, online, analytical methods involving oxygen mass balances in the gas-phase and activity assays by sacrificial sampling are demonstrated. The use of co-solvents to improve substrate mass transfer in this case is also presented. Due to the lower value-class of macrocyclic lactones (5-20 € kg-1) more stringent process metrics are required when compared with the first case study. Oxygen transfer was found to limit the rate of reaction, and as a result all process performance metrics fell short by an order of magnitude. However, the rate of oxygen transfer can still be improved through scale-up, sparging pure oxygen instead of air, or operating the reactor under pressure, which may improve performance metrics sufficiently.
Based on the two case studies, a generic and pragmatic procedure for conceptualising biocatalytic reactions is proposed with a special focus towards those involving poorly water-soluble substrates. This comprehensive route for development includes the precise order and types of experiments required to design a reaction from initial considerations up to the point where later-stage process development tools and directed improvement strategies can be applied in addressing the challenges of poorly water-soluble substrates in industrial biocatalytic processes.
Original languageEnglish
Place of PublicationKgs. Lyngby
PublisherTechnical University of Denmark (DTU)
Number of pages138
StatePublished - 2018
Download as:
Download as PDF
Select render style:
Download as HTML
Select render style:
Download as Word
Select render style:

ID: 159267055