Periodic Reporting for period 3 - BIO-H-BORROW (Biocatalytic Amine Synthesis via Hydrogen Borrowing)
Reporting period: 2020-06-01 to 2021-11-30
Nature generates amines (e.g. alkaloids, amino acids) via multienzyme biosynthetic pathways and in doing so creates some of the most complex natural products that have been discovered to date (e.g. morphine, vinblastine, reserpine). These biosynthetic pathways are truly remarkable in the way in which many enzymes are able to collectively process substrates in an orchestrated assembly line manner with no use of protecting groups and minimal side reactions. Despite our much deeper understanding of how these biosynthetic pathways operate, in terms of the substrates and enzymes involved, only recently have scientists begun to assemble these pathways in host systems for eventual scale-up and production of opiates.
Most synthetic amines are produced using abiological methods, particularly where cost and scale of operation are important. The majority of these chemical processes for amine production involve the use of high temperatures, pressures and expensive metal catalysts. In addition, the precursors for amine manufacture are largely derived from the petrochemical industry and hence are non-sustainable. Clearly a major challenge going forward is to try to recreate these cascade processes, both in vitro and in vivo, particularly for the production of non-natural synthetic amines.
The overall aim of the BIO-H-BORROW project is to develop a new biocatalytic approach for amine synthesis in which alcohols are used as universal substrates, ammonia is the source of the amine in the product, and the only by-product of the reaction is water. Initial work focused on characterisation and engineering of suitable enzymes and development of supporting analytical methodologies.
Most synthetic amines are produced using abiological methods, particularly where cost and scale of operation are important. The majority of these chemical processes for amine production involve the use of high temperatures, pressures and expensive metal catalysts. In addition, the precursors for amine manufacture are largely derived from the petrochemical industry and hence are non-sustainable. Clearly a major challenge going forward is to try to recreate these cascade processes, both in vitro and in vivo, particularly for the production of non-natural synthetic amines.
The overall aim of the BIO-H-BORROW project is to develop a new biocatalytic approach for amine synthesis in which alcohols are used as universal substrates, ammonia is the source of the amine in the product, and the only by-product of the reaction is water. Initial work focused on characterisation and engineering of suitable enzymes and development of supporting analytical methodologies.
Initial work has focused on characterisation and engineering of suitable enzymes and the development of supporting analytical methodologies. AmDH variants were engineered for activity towards a novel panel of substrates and used in analytical and preparative reactions, demonstrating the versatility of these enzymes for production of bulky and functionalized chiral amines for the first time.
A diverse panel of 384 novel putative RedAm homologues has been identified, based on sequence homology to known RedAm enzymes. All enzymes within the panel have been expressed and produced as dried lysate powder, with the full panel now available in microtitre plate format for simplified screening using a liquid phase colorimetric assay developed within the project. These plates have been screened towards a broad range of substrates and an interesting multifunctional enzyme has been identified enzyme that is able to catalyse three distinct chemical reactions within a single active-site, namely imine formation, conjugate (C=C) reduction and imine reduction. Additionally, a solid phase colorimetric screen is currently being developed to increase the screening throughput for both AmDH and RedAm mutant libraries, and to test for variants or homologues with improved thermostability.
Application development has focused on cofactor utilisation and reaction configuration. Active site engineering for altered cofactor specificity in an exemplary ADH has now been accomplished, providing a suitably active and cofactor-compatible variant to pair with an AmDH for hydrogen borrowing cascades. Application of this new variant in the two-enzyme hydrogen-borrowing cascade has provided an initial demonstration of the conversion of racemic alcohols to chiral amines which has been extended to a continuous flow system in which the two enzymes are co-immobilised and where conversion to product was maintained at a steady rate over more than 20 hours. Additional cascades consisting of a RedAm and an oxidoreductase have also been established. Here, an alcohol oxidase or a carboxylic acid reductase has been paired with a RedAm in order to produce secondary amines from primary alcohols and carboxylic acids, respectively. These additional cascade variations highlight new biocatalytic systems for amine production from a larger range of substrates, and importantly provide an expansion to access secondary amine products.
A diverse panel of 384 novel putative RedAm homologues has been identified, based on sequence homology to known RedAm enzymes. All enzymes within the panel have been expressed and produced as dried lysate powder, with the full panel now available in microtitre plate format for simplified screening using a liquid phase colorimetric assay developed within the project. These plates have been screened towards a broad range of substrates and an interesting multifunctional enzyme has been identified enzyme that is able to catalyse three distinct chemical reactions within a single active-site, namely imine formation, conjugate (C=C) reduction and imine reduction. Additionally, a solid phase colorimetric screen is currently being developed to increase the screening throughput for both AmDH and RedAm mutant libraries, and to test for variants or homologues with improved thermostability.
Application development has focused on cofactor utilisation and reaction configuration. Active site engineering for altered cofactor specificity in an exemplary ADH has now been accomplished, providing a suitably active and cofactor-compatible variant to pair with an AmDH for hydrogen borrowing cascades. Application of this new variant in the two-enzyme hydrogen-borrowing cascade has provided an initial demonstration of the conversion of racemic alcohols to chiral amines which has been extended to a continuous flow system in which the two enzymes are co-immobilised and where conversion to product was maintained at a steady rate over more than 20 hours. Additional cascades consisting of a RedAm and an oxidoreductase have also been established. Here, an alcohol oxidase or a carboxylic acid reductase has been paired with a RedAm in order to produce secondary amines from primary alcohols and carboxylic acids, respectively. These additional cascade variations highlight new biocatalytic systems for amine production from a larger range of substrates, and importantly provide an expansion to access secondary amine products.
Prior to commencing work on this ERC advanced grant Dr. R. HEATH worked on the development of a Choline Oxidase enzyme. This researcher was funded by European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 613849 supporting the project BIOOX. Choline oxidase, in conjunction with RedAm, was initially used as a comparison for the Hydrogen Borrowing technology and has highlighted the limitations of the current state of the art. While the Hydrogen Borrowing system allows for redox neutral amine synthesis, the reversibility of both the ADH and AspRedAm enzyme catalysed reactions leads to conversions that ultimately are limited by reaction thermodynamics. This leads to a requirement for a large excess of amine substrate and is more pronounced in the synthesis of secondary amines, where the equilibrium is less favourable. These issues are not obvious at analytical reaction scale reaction but present difficulties when the reaction is scaled. Choline oxidase catalyses an irreversible alcohol oxidation reaction and as such there are no issues with the thermodynamic equilibrium position between the alcohol and carbonyl intermediate and thus significantly higher conversions can be achieved. We have demonstrated a cascade involving a variant of Choline Oxidase (AcCO6) coupled with AspRedAm by screening a panel of eight alcohols with five different amines. From this screening 39 out of 40 reactions showed activity and over 75% had >99% conversion under test conditions (https://doi.org/10.1021/jacs.8b11561). We have also shown (data not published) that the alcohol oxidase works with diol substrates. We thus propose to include oxidases as well in our investigations for oxidation of the target alcohols. Concurrently, we are researching alternative nicotinamide recycling methods in order to drive alcohol oxidation and/or imine reduction.