Biocatalytic reductive aminations with NAD(P)H-dependent enzymes: enzyme discovery, engineering and synthetic applications

Abstract

Chiral amines are pivotal building blocks for the pharmaceutical industry. Asymmetric reductive amination is one of the most efficient and atom economic methodologies for the synthesis of optically active amines. Among the various strategies available, NAD(P)H-dependent amine dehydrogenases (AmDHs) and imine reductases (IREDs) are robust enzymes that are available from various sources and capable of utilizing a broad range of substrates with high activities and stereoselectivities. AmDHs and IREDs operate via similar mechanisms, both involving a carbinolamine intermediate followed by hydride transfer from the co-factor. In addition, both groups catalyze the formation of primary and secondary amines utilizing both organic and inorganic amine donors. In this review, we discuss advances in developing AmDHs and IREDs as biocatalysts and focus on evolutionary history, substrate scope and applications of the enzymes to provide an outlook on emerging industrial biotechnologies of chiral amine production.

---

Bo Yuan

Dr Bo Yuan obtained her doctoral degree in Biochemistry from the Turner Group at University of Manchester in 2011, and then joined Croda Europe Ltd. as a Research Scientist. In 2016 she moved to Xi’an Jiaotong University as a Lecturer. She joined Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences as an Associate Professor in 2021. Her research interests are in protein engineering, asymmetric synthesis and chemoenzymatic cascade reactions.

Dameng Yang

Dr Dameng Yang received his PhD degree in Biochemistry from Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences in 2022 under the guidance of Prof. Zhoutong Sun, and then joined Meihua Holdings Group Co., Ltd as a project leader. His research interests are in synthetic biology, enzyme engineering and metabolic engineering.

Ge Qu

Dr Ge Qu received his doctoral degree in bioinformatics from Adam Mickiewicz University in Poland in 2015. After graduation, he joined the group of Professor Zhoutong Sun at Tianjin Institute of Industrial Biotechnology (TIB), Chinese Academy of Sciences. Currently he is an Associate Professor at TIB and working in the fields of rational enzyme design and biosynthesis.

Nicholas J. Turner

Prof. Nicholas Turner obtained his DPhil in 1985 with Professor Sir Jack Baldwin and from 1985–1987 was a Royal Society Junior Research Fellow at Harvard University with Prof. George Whitesides. He was appointed lecturer in 1987 at Exeter University and moved to Edinburgh in 1995, initially as a Reader and subsequently Professor in 1998. In October 2004 he joined Manchester University as Professor of Chemical Biology. He is Director of the Centre of Excellence in Biocatalysis (CoEBio3) and a Co-Director of SYNBIOCHEM, the BBSRC Synthetic Biology Research Centre. His research interests are in the following fields: use of enzymes as catalysts for organic synthesis; directed evolution of enzymes; protein engineering; biocatalytic retrosynthesis using Computer Assisted Synthesis Planning (CASP) tools; biocatalytic cascade reactions; enzymatic synthesis of therapeutic oligonucleotides.

Zhoutong Sun

Prof. Zhoutong Sun obtained his doctoral degree in microbiology with Prof. Sheng Yang at Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences in 2012, he then moved to Nanyang Technological University in Singapore as a research fellow in metabolic engineering. One year later, he joined the Reetz group as a postdoc at MPI für Kohlenforschung and Marburg University in directed evolution and biocatalysis. In 2016, he became a full professor at Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences. His research interests are in enzyme engineering, metabolic engineering and synthetic biology.

---

1. Introduction

Enantiomerically pure amines are present in numerous pharmaceutical products, agrochemicals, natural products, and functional materials.^1–4 Many asymmetric synthetic methods have been developed to synthesize this class of compounds. Among them, asymmetric reductive amination (ARA) plays a pivotal role in the construction of C–N bonds in pharmaceutical chemistry.⁵ For example, suvorexant (hypnotic),⁶ sitagliptin (antidiabetic),⁷ sertraline (antidepressant),⁸ escitalopram (antidepressant),⁹ levobupivacaine (anesthetic),¹⁰ rivastigmine (dementia treatment)¹¹ (Fig. 1) are highly valuable drugs that have been synthesized employing at least one reductive amination step. Although transition-metal catalyzed ARA has been showcased by a number of successful strategies,¹² it is less developed than imine reduction.^13,14 It is difficult partially due to the inhibition caused by coordination of the amine to the metal catalyst or the incompatibility of transition metal hydrides with ketones.¹⁵ On the other hand, enzymatic ARA can be achieved with benefits such as generally higher stereoselectivities, superior chemo- and regioselectivities, higher turnover rates, and mild reaction conditions.^16,17 Among the biocatalysts that transform various substrates into amines, several classes of enzymes are prominent due to wide applicability, including flavin-dependent monoamine oxidases (MAOs),¹⁸ pyridoxal 5-phosphate (PLP)-dependent ω-transaminases (ATAs),¹⁹ phenylalanine ammonia lyases (PALs)²⁰ and P450 monooxgenases (P450s). Alternatively, NAD(P)H-dependent amine dehydrogenases (AmDHs),^21–25 imine reductases (IREDs)²⁶ and reductive aminases (RedAms)²⁷ are classes of enzymes that utilizes both organic and inorganic ammonia sources^27,28 and a broad range of substrates such as aldehydes, ketones, keto acids or keto esters to form amine products. The catalytic mechanisms of these classes of enzyme are similar, involving the formation of a carbinolamine intermediate with subsequent hydride transfer from the nicotinamide cofactor.^29,30 Although AmDHs traditionally use free ammonia as the amine donor, generating primary amines as the products, it has also been shown that AmDHs can accept organic amine donors and produce secondary amines.^31,32 Similarly, IREDs and RedAms can also use free ammonia as amine donors.³³ In this review, we describe the discovery, evolutionary history, substrate scope and synthetic applications of AmDHs, IREDs and RedAms, in order to provide a comprehensive overview of these classes of enzymes and their applications in synthesis.

Fig. 1 Chiral amine building blocks and drug intermediates synthesized by reductive amination.

2. Amine dehydrogenases (AmDHs)

In 2012, LeuDH was engineered to accept ketones as substrates as reported firstly by Abrahamson et al.³⁴ Following this breakthrough, many efforts have been invested in the identification, engineering, and generation of AmDHs from various sources. In addition, native AmDHs have also been discovered,²⁹ which do not share sequence homology with the engineered AmDHs. In the following sections, we review the discovery of AmDHs from the above-mentioned pathways, and summarize the structural, mechanistic, engineering and synthetic application of AmDHs.

2.1. Structural and mechanistic insights

A variety of AmDHs have been characterized and their crystal structures determined in apo form or in complex with cofactors or substrates. Based on the structural analysis, the catalytic mechanisms of AmDHs have been investigated in-depth. Table 1 lists the crystal structures of experimentally validated LeuDHs, PheDHs, and AmDHs.

Table 1 Structures of selected LeuDHs, PheDHs and AmDHs

Names	Origin	PDB ID	In complex with	Ref.
LeuDHs
WT	Lysinibacillus sphaericus	1LEH	—	35
WT	Sporosarcina psychrophila	3VPX	—	36
WT	Geobacillus stearothermophilus 10	6ACF	—	37
WT	Geobacillus stearothermophilus 10	6ACH	NAD^+	37
WT	Exiguobacterium sibiricum	8GXD	—	38
PheDHs
WT	Rhodococcus sp.	1BXG	NAD^+ and β-phenylpropionate	39
WT	Rhodococcu sp.	1BW9	NAD^+ and phenylpyruvate	39
WT	Rhodococcus sp.	1C1X	NAD^+ and L-3-phenyllactate	40
WT	Rhodococcus sp.	1C1D	NADH and L-phenylalanine	40
AmDHs
WT	Petrotoga mobilis SJ95	6G1H	NAD^+	29
WT	Petrotoga mobilis SJ95	6G1M	NAD^+	29
WT	Cystobacter fuscus	6IAU	NADP^+ and cyclohexylamine	29
W145A	Cystobacter fuscus	7QZN	NAD^+	41
W145A	Cystobacter fuscus	7QZL	NADP^+ and pentylamine	41
WT	Mycolicibacterium smegmatis	6IAQ	NADP^+	29
WT	“MATOUAmDH2” from an eukaryotic organism	7R09	NADP^+ and cyclohexylamine	42

---

As an example, the crystal structure of the native AmDH4 is shown in Fig. 2A. The two dimers are in ‘open’ and ‘closed’ forms. Each of the monomer consists of a Rossmann fold domain at the N-terminal, and a six-strand β-sheet domain at the C-terminal. The structures resemble that of the dihydropicolinate reductase (DHDPRs, PDB ID: 5KT0) and some natural amino acid dehydrogenases such as meso-diaminopimelate dehydrogenases (DAPDHs, PDB ID: 1F06 and 3WBF), but the sequences do not share significant homology. The rossmann fold is well-conserved among the native AmDHs, DHDPRs, and DAPDHs, but the β-sheet domain bears significant differences. The key residues that shape the active site pocket are E102, H197, I198h, H264, Q266, G299, and T303. The structure of the active site of AmDH4 in complex with the substrate (2R,4S)-2,4-diaminopentanoate (2,4-DAP) is shown in Fig. 2B. The key residues that interact with the substrate are R161, H264, H197, E102. The residue E102 is conserved between DAPDHs and AmDH4, which is considered to be involved in the activation of ammonia. Q266 at the bottom of the active site and F168 situated at the flexible lid of the active site are considered to involve in the catalytic events.

Fig. 2 Structures of (A) AmDH4 dimers in complex with NAD⁺ and (B) the active site of AmDH4 in complex with the substrate 2,4-DAP (PDB ID: 6G1M).

The proposed mechanism of the native AmDH catalyzed reductive amination involves several key steps (Fig. 3).^29,31,43 Firstly, ammonia is activated by glutamate (E108 in the case of CfusAmDH) (II), then electrophilic attack takes place on the carbon from the carbonyl moiety, which is stabilized via hydrogen bonds with tyrosine (Y168 for CfusAmDH) at the active site (III). The carbinolamine intermediate is then formed, subsequently, reduction of the carbinolamine by the hydride of the nicotinamide cofactor occurs and then this is dehydrated to form the amine product (IV and V). This mechanism resembles the reductive amination catalyzed by RedAms,³⁰ where both mechanisms involve a tyrosine proton donor for substrate stabilization, and a carbinolamine intermediate followed by hydride transfer from the co-factor NAD(P)H.

Fig. 3 Proposed catalytic mechanisms of CfusAmDHs.²⁹

2.2. Discovery of AmDHs

2.2.1. AmDHs from amino acid dehydrogenases (AADH). Following the first example that converted a LeuDH to AmDH to accept ketones as substrates,³⁴ the Bommarius group reported an engineered PheDH from Bacillus badius to obtain a robust BsAmDH,⁴⁴ which catalyzes the biotransformation of aliphatic and benzylic ketones to amines. 15 amino acid residues in BsLeuDH were mutated by combinatorial active-site saturation test (CAST)⁴⁵ and iterative saturation mutagenesis (ISM)^46,47 to find that K68S/N261L mutations play the key roles in the conversion from AADH to AmDH.

PheDH and LeuDH have been the first targets as the sources for conversions to AmDHs. Two key mutations with interactions of the substrate side chains have been identified from various PheDHs and LeuDHs. For example, Jiang et al. engineered a PheDH from Bacillus halodurans to utilize a range of benzylic and aliphatic ketone substrates with double mutations (E113D/N276L).⁴⁸ Ye et al. engineered a PheDH from Rhodococcus sp. M4 to an AmDH with a triple mutation K66Q/S149G/N262C.⁴⁹ Furthermore, Liu et al. engineered a thermostable PheDH from Geobacillus kaustophilus to develop a GkAmDH³² with the key double mutations (K78S/N276L). The T_m of GkAmDH is up to 71.5 °C with high activity towards a range of benzylic ketones using inorganic and organic ammonia sources to afford both primary and secondary amine or amino alcohol products. Pushpanath et al. developed a CalAmDH from PheDH from Caldalkalibacillus thermarum, bearing the 2 key mutations (K68S/N266L), with a high T_m up to 83.5 °C.⁵⁰

Although a majority of AmDHs have been converted from LeuDH and PheDH, efforts have also been invested in engineering other classes of amino acid dehydrogenases to generate AmDHs, as exemplified by L-lysine dehydrogenase from Geobacillus stearothermophilus (LysEDH) by Tseliou et al. Benefitting from the thermophile origin of LysEDH, the AmDH variant (LE-AmDH-v1) showed high stability with T_m up to 69 °C, which also retained activity at room temperature, as well as 80% of the activity at 50 °C. A range of aliphatic and aryl ketones were reductively aminated to the corresponding amines in up to 99% ee and conversions (Fig. 4). The F173A mutation was proposed to enlarge the substrate binding pocket that enabled the acceptance of aromatic substrates. For the aliphatic ω-amino acid, the variant afforded (S)-configured product, whereas for all of the other aliphatic and aromatic substrates tested, (R)-configuration was observed.⁵¹ Later, the same variant has been coupled with a NADH oxidase (NOX) in the kinetic resolution of α-chiral amines.⁵² Harnessing the promiscuity of LysEDH, Tseliou et al. performed further engineering on the enzyme to generate dual alcohol dehydrogenase and amine dehydrogenase activity,⁵³ thereby completing a one-enzyme “hydrogen-borrowing” cascade, albeit with low yields (up to 34%). While F173A variant favored the AmDH activity as described, the authors found that the introduction of F173S mutation enhanced the alcohol dehydrogenase activity.

Fig. 4 Biocatalytic reductive amination performed by LE-AmDH variants.

2.2.2. Native AmDHs. The success in the engineering of amino acid dehydrogenases into AmDHs has spurred interest in the identification of native AmDHs. The discovery of a native AmDH from Streptomyces virginiae was reported by Itoh et al., albeit with low stereoselectivity.⁵⁴ Following the initial discovery of a wild type PmAmDH (dubbed AmDH4) from a thermophile Petrotoga mobilis,⁵⁵ Bastard et al. have discovered and characterized five native AmDHs including AmDH4, MicroAmDH from Microbacterium sp., ApauAmDH from Aminomonas paucivorans, MsmeAmDH from Mycobacterium smegmatis and CfusAmDH from Cystobacter fuscus. Detailed catalytic mechanisms have been discussed in Section 2.1. 2029 members using AmDH4, CfusAmDH and MsmeAmDH as templates were modelled by the active site modelling and clustering (ASMC) method.⁵⁶ G1–G5 groups were defined where AmDH4, CfusAmDH and MsmeAmDH were found in G2, G3, and G4 group, respectively.²⁹ The G1 group contained enzymes without (2R,4S)-2,4-diaminopentanoate (2,4-DAP) activity. Substrates with charged groups such as amino acids or 4-oxopentanoic acid (4-OP) like compounds cannot be transformed by enzymes from G3 and G4 groups. G5 mainly contained remote homologues. Caparco et al. performed metagenomic mining, selecting 18 candidates from three databases (IGC, OM-RGC, and MATOUv1).⁵⁷ To further explore the native AmDHs in diversity of environments, the ocean and human genome were screened for potential AmDHs. Among the discovered enzymes, the MATOUAmDH2 was identified with the highest activity towards aldehyde and ketone substrates. Very recently, Bennett et al. determined the structure of MATOUAmDH2 in complex with NADP⁺.⁴² Several variants were constructed including F143A, L144A, L169A, M215A, L180A and T312A to examine the hydrophobic interactions at the active site. Slight increase in k_cat/K_m from 0.21 to 0.27 S⁻¹ mM⁻¹ was obtained for the variant M215A, which was confirmed by the biotransformations of cyclohexanone with ammonia.

2.3. The history of evolution for AmDHs

The pioneering work for the generation of AmDHs has sparked research interests in the protein engineering of AmDHs. Although these enzymes are important as they allow the reductive amination to utilize ketone substrates with many merits including generating only water as the by-product and using cheap inorganic amine donors, they suffer from narrow substrate scope and relatively low stability. Therefore, protein engineering of AmDHs has been carried out on the purpose of expanding the substrate scope and improving stability while maintaining or increasing activity of the AmDH.

The AmDH structures contain an independent substrate-binding domain and cofactor-binding domains. Domain shuffling has been firstly applied in the engineering of AmDHs generated from LeuDH and PheDH by the Bommarius group in 2014. The final variant cFL1-AmDH consisted of the ketone/amine binding domain of F-AmDH from Rhodococcus sp. M4 and cofactor-binding domain from L-AmDH from Bacillus sphaericus. The variant exhibited expanded substrate scope to benzylic carbonyls and higher thermostability.⁵⁸ The same group also performed the reactions in a biphasic system using heptane and ammonium formate buffer, enabling the hydrophobic substrates to be accepted with improved enzymatic activity.⁵⁹ Later in 2018, this variant was used by Löwe et al. as the model enzyme to investigate the substrate scope, synthetic processes and solvent tolerance in detail.⁶⁰ Recently, Li et al. continued with domain engineering on AmDH engineered from a PheDH.⁶¹ The resulting variant cFLF-AmDH combined the coenzyme-binding domain of L-AmDH from Bacillus stereothermophilus and the core structure of F-AmDH from Bacillus stereothermophilus with lower coenzyme affinity than L-AmDH. Therefore, cFLF exhibited 2-fold increase in NADH binding affinity and 4.4-fold increase in k_cat/K_m comparing with F-AmDH.

Co-factors and the recycling systems drastically affect the economic viability for the applications of AmDHs. To achieve high atom efficiency and efficient use of cofactors, the Turner group has developed a dual-enzyme hydrogen-borrowing process, which has utilized a range of primary and secondary alcohols for the synthesis of corresponding amines (Fig. 5).⁶² AmDH was applied with alcohol dehydrogenases (ADHs from Aromatoleum sp., Lactobacillus sp., or Bacillus sp.) to achieve a redox-neutral, self-sufficient process. The amine products were synthesized with up to 99% ee and 96% conversions in (R)-configuration starting from either (R)-, (S)- or racemic alcohols. This concept has also been reported by Chen et al, who introduced 2 knowledge-based key residues (K77S/N270L) to EsLeuDH from Exiguobactertium sibiricum and considerably improved activities for a variety of ketones. The EsAmDH was then coupled with ADHs to achieve the self-sufficient process. In preparative scale, 2-pentanamine was obtained in >90% conversions starting from 2-pentanol.⁶³ The similar approach has been applied to a variety of cascade reactions involving several different classes of enzymes (with ADHs) such as transaminases,⁶⁴ reductive aminase,⁶⁵ and ene-reductases.⁶⁶ Other than the redox neutral cascades, two reductive enzymatic processes can also be performed in one-pot employing ene-reductases (old yellow enzyme, OYE) and AmDHs to convert α,β-unsaturated enones to amines.⁶⁷

Fig. 5 Hydrogen-borrowing self-sufficient cascades.

Early examples of AmDHs exhibited narrow substrate scope, and efforts were invested to expand the scope for bulky, aliphatic and aromatic substrates. In 2018, Chen et al. synthesized bulky aliphatic amines via an engineered LfAmDH from Lysinibacillus fusiformis from LeuDH with up to 400 mM substrate loading. The same key mutations (K68S/N261L) that were responsible for altering the substrate specificities were introduced to three amine dehydrogenases (EsAmDH from Exiguobacterium sibiricum, LfAmDH from Lysinibacillus fusiformis, and BspAmDH from Bacillus sphaericus) and the best one was identified as the LfAmDH. They identified key mutations A113G and T134G that enlarged the volume of the substrate-binding cavity, which accommodated the large aliphatic substrates well, giving excellent enantio-selectivity (>99% ee, (R)-selectivity) with >99% conversions.⁶⁸ This variant has greatly expanded the substrate scope of AmDHs. Later the same group further engineered the variant to expand the substrate scope to include α-hydroxy ketones. Two variants (LfAmDH-M₀ and LfAmDH-M₃) carrying further mutations (K68T/N261L and K68T/N261L/A113G/T134G, respectively) afforded the (S)-α-hydroxyl amines with excellent ee (>99%%) and conversions (>70%).⁶⁹ Therefore, both (R)- and (S)-products can be obtained with the two enantio-complementary variants provided by above studies (Fig. 6).

Fig. 6 Reductive amination by AmDHs affording both (R)- and (S)-configured amines.

There have been growing interests in the protein engineering of AmDHs to expand the substrate scope to aliphatic ketones⁶⁸ or aromatic ketones,⁷⁰ and improving the thermostability.⁷¹ Later in 2021, Wang et al. reported a further engineered variant of the GkAmDH from Geobacillus kaustophilus.⁷² By combining a semi-rational design strategy and directed evolution (error-prone PCR and DNA shuffling), the authors identified further key residues that rendered up to 110-fold increase in activity comparing to the starting variant for the synthesis of bulky aromatic ketone substrates. The substrate scope has also been extended to methyl ketones with alkyl chains, remote functional groups, or heterocycles, as well as ethyl and propyl ketones with alkyl chains. Furthermore, the authors employed the variant M3 and M8 with high activity in preparative synthesis of several pharmaceutically useful building blocks (dilevalol and medroxalol), both afforded good to high yields and excellent enantio-selectivity (Fig. 7).

Fig. 7 Engineering GkAmDH to catalyze the synthesis of intermediates of dilevalol and medroxalol.

Knaus et al. investigated and expanded the substrate scope of AmDHs by testing 3 previously reported variants, namely Bb-PhAmDH, Rs-PhAmDH, and Chi-AmDH.⁷³ Upon optimization, (R)-configured bulky-bulky ketones were tested with high activity and selectivity (up to >99% conversions and ee) when catalyzed by Chi-AmDH and Rs-PhAmDH. Formate dehydrogenase (FDH) was used to improve the atom economy and carbonate was produced as the sole by-product.

Recently Kong et al. engineered an AmDH from Jeotgalicoccus aerolatus (JaAmDH) to achieve enhanced activities for aryl ketones and heteroaryl ketones.⁷⁴ The authors identified three AmDHs with the best one being JaAmDH. They introduced the preserved K68 and N261 amino acids for converting the amino acid dehydrogenase (AADH) activity to AmDH activity. Further engineering of JaAmDH has significantly improved specific activity towards ketones bearing heterocycles moiety, which were previously found difficult to be utilized by AmDHs. Key mutations including K67S/N260L/E113V/V291G were introduced to create the variants M3₃, with k_cat/K_m improving up to 46 folds for the model substrate 1-(6-methylpyridin-2-yl)ethan-1-one. Up to 18 heteroaryl ketones and 12 aryl ketones were subjected to reductive amination and ee were all above 99% with conversions ranging from 74% to 99% (Fig. 8). Preparative scale reactions on 200–1000 mM substrate concentration afforded conversions and ee both up to 99%.

Fig. 8 Reductive amination of aryl and heteroaryl ketones by AmDHs.

In addition to the AmDHs converted from LeuDHs and PheDHs, native AmDHs have also been targeted for protein engineering, which has enabled the production of industrial relevant compounds in high substrate loadings. Cai et al. engineered PmAmDH (or AmDH4) by directed evolution using techniques including error-prone PCR and DNA shuffling, and the k_cat/K_m was elevated by 18 folds compared with the wild type AmDH with key mutations I80T/P224S/E296G, while the T_m of the variant reached 60 °C. Reductive amination was performed on levulinic acid in preparative scales at a 0.5 M (58 g L⁻¹) substrate concentration, and 97% conversions was achieved in 24 h with >99% ee.⁷⁵

Desymmetrization strategies have been firstly applied on 1,3-cyclodiketones using AmDHs very recently. The reductive amination was carried out to stereoselectively aminate one of the ketone groups to produce 2,2-disubstituted β-amino ketones containing a quaternary carbon center. A library of 51 AmDHs were screened, and LsPheDH from Lysinibacillus sphaericus bearing the key mutations K79S/N277L was selected. Saturation libraries for residues within 5 Å of the substrate and NADH were constructed and screened, and subsequently the combinatorial libraries were constructed based on the positive hits. Finally, 5 variants were employed in the biotransformation of a range of substrates with expanded scope, namely W0 (LsPheDH K79S/N277L), W2 (T125A), W3 (N145L), W5 (T125A/N145L), and W8 (L51A/T125A/N145L/S157P). Using these variants, up to 97% conversion, 99% de and 99% ee have been achieved. Preparative-scale reactions were also performed with good yields (60–85%), and by conducting a two-step reaction followed by Buchwald–Hartwig cross coupling, a cyclopenta[b]hydroquinoline derivative was obtained in up to 55% yield (Fig. 9).⁷⁶

Fig. 9 Desymmetric reductive amination of 2,2-disubstituted-1,3-ketones by AmDHs.

The Sun group conducted data mining, discovery and protein engineering of AmDHs on purpose of identifying new enzymes and improving enzymatic activity towards hydroxyl ketone substrates.^77,78 They developed a high throughput assay with fluorescent probe to quantitatively determine the ketones for AmDH library screenings.⁷⁹ Five AmDHs have been discovered, namely GsAmDH from Geobacillus stearothermophilus, BsAmDH from Bacillus stearothermophilus, LsAmDH from Lysinibacillus sphaericus CBAM5, SpAmDH from Sporosarcina psychrophile, and TiAmDH from Thermoactinomyces intermedius, which exhibited significant activity towards 1-hydroxybutan-2-one after the introduction of the two key residues responsible of converting AADH to AmDH. 99% conversion and 99% ee were obtained by GsAmDH on the model substrate. In addition, good to excellent conversions and ee were obtained for a range of aliphatic hydroxyl ketone and benzylic hydroxyl ketone substrates. Later, site-directed saturation mutagenesis (SSM), CAST/ISM strategies were employed to further engineering the SpAmDH and GsAmDH for several rounds to achieve up to 4.2 fold improved activity towards hydroxyl ketones, with >99% ee and 99% conversions (Fig. 10).^78,80

Fig. 10 Expanded substrate scope by engineered AmDHs.

2.4. Applications of AmDHs

Immobilization of enzymes represents a practical technique which possesses several advantages comparing to free enzymes such as improved thermostability and higher organic solvent tolerance.^81,82 Immobilization often enables the biocatalysts to be reused for a number of batches and with continuous flow processes, the overall economy could be considerably improved. AmDHs have been employed in several studies involving enzyme immobilization on various supports and flow processes. For example, in 2017, Liu et al. immobilized an AmDH (crude enzyme) engineered from PheDH from Rhodococcus sp. M4 on magnetic nanoparticles (MNP) (Fig. 11).⁸³ Additionally, they also immobilized a glucose dehydrogenase (GDH) used for cofactor recycling systems, which reduced the cost of the overall reaction while retaining the productivity. Ren et al. coated the AmDH by polyethylenimine (PEI), and then the AmDH-PEI hybrid was entrapped by titania to form the AmDH-PEI-Ti nanoparticles. The immobilized enzyme displayed improved thermostability, and can be used for 6 cycles while retaining 50% of the initial activity.⁸⁴ In 2018, Caparco et al. investigated the effect of peptide linkers of leucine zipper-AmDH/FDH fusion proteins on the catalytic efficiency of the immobilized enzymes on inorganic supraparticles.⁸⁵ Later in 2020, the authors immobilized an AmDH and a formate dehydrogenase with fluorescent model proteins and a leucine zipper binding domain onto the calcium-phosphate–protein supraparticles, and successfully converted ketone substrates to amines.⁸⁶

Fig. 11 Highly specific affinity attachment-enabled immobilization of his-tagged AmDH from cell-free extract on Ni-NTA-MNP.

Recently, continuous flow was attempted using a fixed-bed reactor packed with immobilized Ch1-AmDH and FDH. The Nuvia® IMAC resin from Bio-Rad was used as the carrier, and amination of 5-methyl-2-hexanone was performed with up to 443 g L⁻¹ day⁻¹ productivity (Fig. 12).⁸⁷

Fig. 12 Block flow diagram of the continuous flow packed bed reactor system.

Synthetic biology utilizes enzymes in metabolic pathways to construct whole-cell systems with several enzymatic routes to targeted products. In 2018, Yu et al. constructed a C–H amination cascade in vivo, cascading P450 BM3 (19A12) catalyzed C–H hydroxylation, alcohol dehydrogenase (ADH) ScCR catalyzed oxidation of the alcohol, and EsAmDH catalyzed amination in one-pot (Fig. 13).⁸⁸ Non-interfering cofactor recycling systems have been employed and eventually a one-pot two-step cascade reaction has been established. The atom economy reached 100% while the reaction could be performed at scale by whole-cell. Houwman et al. has constructed an ADH with AmDH in resting E. coli cells implementing the “hydrogen-borrowing” concept in vivo.⁸⁹ Liu et al. coupled whole-cell with cell-free systems, and the cofactor recycling system was enhanced by separation.⁹⁰ A NADH oxidase (NOX) was coupled with an ADH to obtain full conversion from racemic alcohols to ketones, and a GDH with AmDH were used for the reductive amination of the ketones, as a result, a TTN for NADH recycling of 1410 was achieved, which was reported as the highest TTN thus far.

Fig. 13 One-pot two-step strategy to produce cyclohexylamine. Adapted with permission from ref. 88 with modifications. Copyright 2018 Elsevier Inc.

Cascading AmDH with organic/metal catalysts represents major opportunities in organic synthesis for the production of important drug intermediates. However, it is difficult to identify suitable catalysts and conditions for the two previously incompatible systems to work in synergy in aqueous phase. The Turner group reported a one-pot reaction with a Pd catalyst and AmDH to produce chiral N-arylamines.⁹¹ A Buchwald–Hartwig cross-coupling (BHA) reaction was performed with the addition of a surfactant DL-a-tocopherol methoxypolyethylene glycol succinate (TPGS-750-M) to improve the water miscibility of the catalyst and accelerate the rates of the coupling reaction in enzymatic compatible conditions (Fig. 14). In the end, up to 90% conversions of the chiral N-arylamines were obtained. Additionally, multi-enzyme cascades have also been alternative methodologies, for example, Chen et al. reported a trienzymatic cascade involving a ω-transaminase, an AmDH and a FDH.⁹²

Fig. 14 Cascade of AmDHs catalyzed RA and Buchwald–Hartwig cross-coupling (BHA).

In summary, AmDHs have exhibited enormous potential in the synthetic applications of chiral amines. The key advantage of AmDHs is the ability to utilize inorganic ammonia donors producing only water as the side products, which brings environmental and economic benefits in the industrial applications. In addition, investigations have also been carried out to explore the synthesis of secondary amines catalyzed by AmDHs, although the scope is very limited and requires further efforts to enhance the ability to produce chiral secondary amines. On the other hand, in the following section, we introduce imine reductases (IREDs) which advances in the synthesis of secondary and tertiary chiral amines.

3. Imine reductases and reductive aminases (IREDs and RedAms)

Imine reductases (IREDs) have been firstly discovered by Mitsukura et al. in 2010.⁹³ Both (R)- and (S)-selective IREDs were identified by screening against a large number of bacteria, actinomycetes, and fungi. In the following decade, IREDs have sparked significant research interests and enormous efforts have been invested in the engineering of IREDs to find applications across the industrial sectors. IREDs are NAD(P)H-dependent oxidoreductases that are capable of catalyzing reductive amination not only of cyclic imines, but also between ketone substrates and amine donors to produce primary, secondary and tertiary amines.^{25,26,94–100} Hereafter, we summarize the structures, mechanisms, evolutionary history and applications of IREDs.

3.1. Structures and mechanistic insights into IREDs and RedAms

The first X-ray crystal structure of IRED from Streptomyces kanamyceticus was reported by Rodriguez-Mata et al.¹⁰¹ Subsequently, many crystal structures of IREDs from various origins were resolved as presented in Table 2.^102–105

Table 2 Structures of selected IREDs

Names	Origin	PDB ID	In complex with	Ref.
SkIRED	Streptomyces kanamyceticus	3ZHB	NADP^+	101
SkIRED	Streptomyces sp. GF3546	4OQY	NADPH	103
SaIRED	Streptomyces aurantiacus	4OQZ	—	103
NhIRED	Nocardiopsis halophila	4D3S	n-Octyl β-D-glucoside (OBDG)	104
BcIRED	Bacillus cereus BAG3X2	4D3F	NADP^+
AoIRED	Amycolatopsis orientalis	5A9T/5FWN	NADP^+/	102
			(R)-Methyltetrahydroisoquinoline
AtIRED	Aspergillus terreus	5OJL	NADPH4 and dibenz[c,e]azepine
AtIRED	Aspergillus terreus	6EOD	NADP(H)	106
AtIRED	Aspergillus terreus	6EOH	NADPH and ethyl levulinate	30
AtIRED	Aspergillus terreus	6EOI	NADPH and ethyl-5-oxohexanoate
AtIRED	Aspergillus terreus	6H7P	NADPH4, cyclohexanone and allylamine
AspRedAm	Aspergillus oryzae	5G6R/5G6S	NADPH and (R)-rasagiline
SrIRED	Streptosporangium roseum	5OCM	NADP^+ and 2,2,2-trifluoroacetophenone hydrate	28
MsIRED	Myxococcus stipitatus	6TO4	NADP^+	107
MsIRED	Myxococcus stipitatus	6TOE	NAD^+	108
IR-G36-M5	Actinoalloteichus hymeniacidonis	7WNW	NADP^+	109
EneIRED	Pseudomonas putida	7A3W	NADPH	110
EneIRED-07	Micromonospora sp. Rc5	7OSN	NADP^+	111
A208N variant	Ensifer adhaerens	8A5Z	NADP^+	112

---

As an example, the crystal structure of AtRedAm in complex with NADPH is shown in Fig. 15A.³⁰ The structure consists of two monomers with a typical domain-swapping feature that exists in most of the IREDs and RedAms. Each monomer is constructed with a Rossmann domain at the N-terminal and a helical bundle at the C-terminal. The active site is shaped by the N-terminal domain of one monomer and C-terminal domain of the other monomer, which exists in ‘open’ (apo-) and ‘closed’ (bound with NADPH) form. The proposed catalytic residues are Y183, D175, and N98. A hydrogen bond is formed between the ketone carbonyl group of the substrate ethyl levulinate and the hydroxyl group from Y183, and the carbon atom of the carbonyl group is placed within 4.5 Å from the C4 atom of NADPH (Fig. 15B). This distance is not sufficient for efficient hydride transfer. The quaternary complex of cyclohexanone, allylamine, the redox-inactive NADPH4 were resolved, and the crystal structure is shown in Fig. 15C. Y183 is coordinated with a water molecule that interact with carbonyl group of cyclohexanone. The allylamine is observed near D175 and N98, with interactions between the amine group and the side chains of the residues. This structure of the quaternary complex provides evidence for both ketone and amine binding, which leads to the imine formation and reduction.

Fig. 15 Crystal structures of AtIRED. (A) AtIRED in complex with NADPH; (B) the active site of AtIRED with the substrate ethyl levulinate; (C) the quaternary complex of cyclohexanone and allylamine. (PDB ID: 6H7P).

The catalytic mechanisms have been determined to start with ketone binding to Y177, and deprotonation of the amine by D169 (II). Then the carbonyl group of the ketone is attached by the deprotonated amine again and forms a carbinolamine intermediate (III). Subsequently, the hydroxyl group of the intermediate is protonated, and water as a leaving group is eliminated while Y177 deprotonates the amine. Finally the prochiral iminium ion intermediate is obtained (IV), which is then reduced by the NADPH hydride to yield the final amine product (V) (Fig. 16).³⁰

Fig. 16 The hypothetical mechanisms for RedAms.³⁰

3.2. Discovery, history of evolution and substrate scope of IREDs

The discovery of IREDs was by Nagasawa et al. in 2010, where 2-methyl-1-pyrroline, a cyclic imine, was reduced by the whole-cell strains Streptomyces sp. GF3587 and 3546. The genes encoding them were subjected to heterologous expression in E. coli and employed in the biotransformations to afford both (R)- and (S)-configured amine products.^93,113–115 The synthetic potential of Streptomyces sp. GF3546 has been explored and the substrate scope has been expanded to a variety of cyclic imines.¹¹⁶ Scheller et al. identified three new IREDs ((R)-SrIRED from Streptosporangium roseum DSM 43021, (R)-StIRED from Streptomyces turgidiscabies and (S)-PeIRED from Paenibacillus elgii),¹¹⁷ the substrate scope was expanded for the (R)-configured IRED from Streptomyces sp. GF3587 with further mutagenesis performed and variants D172A and D172L generated with moderate improvements (Table 3, entry 1).¹¹⁸ Following this line of work, IRED toolbox has been expanded further by screening IRED candidates and characterized a library of 20 IREDs (Table 3, entry 2).^104,119,120

Table 3 Substrate scope and conditions for reductive amination catalyzed by IREDs

Entry	Enzymes	Sources	Substrates^a	Products	Product information^b	pH	T/°C	T m/°C	Ref.
a : conversions ≥ 80%; : 40% < conversions < 80%; : conversions ≤ 40%; : ee ≥ 80%; : 40% < ee < 80%; : ee ≤ 40%. b R/S: the absolute configuration of the products; conv.: conversions.
1	IREDs	Streptomyces sp.			R or S, 4–99% conv., 8–98% ee	7.0	30		118
2	IR-20/IR-23	Streptomyces tsukubaensis/Streptomyces vidiochromogenes			R or S, up to 99% conv., up to 99% ee	7.0	30		119
3	R-IRED-Sr/R-IRED_Ms/S-IRED_Pe	Streptosporangium roseum/Myxococcus stipitatus/Paenibacillus elgii			R or S, 39–99% conv., up to 87% isolated yields up to 94% de, 90–99% ee	7.0	25		130
4	IREDs	Various sources			R or S, up to 99% conv., up to 99% ee	7.0	30		129
5	AspRedAm	Aspergillus oryzae			R or S, up to 99% conv., up to 97% ee	7.0	30		65
6	IRED14/IR-sip	Nocardia cyriacigeorgica GUH-2/Streptomyces ipomoeae 91-03			R or S, 71–94% conv., 52–90% ee	9.5	30		128
7	IRED5/8/24	Mycobacterium smegmatis/Glycomyces tenuis/Cupriavidus sp. HPC(L)			S, 82–98% conv., 96–99% ee	7.0	30		122
8	CfIRED	Cystobacter ferrugineus			R or S, up to 92% conv., up to 99% ee	8.0	30		146
9	AtRedAm/AdRedAm	Aspergillus terreus/Ajellomyces dermatitidis			R or S, up to 96% conv., up to 96% ee	9.0	30		30
10	IRED40/IRED30	Sandaracinus amylolyticus/Sciscionella marina			R or S, 47–95% isolated yields, 94–99% ee	9.0	30		147
11	IREDs	Various sources			R or S, up to 99% conv., up to 99% ee	7.5/6.0	30		148
12	MsIRED	Mycobacterium smegmatis			21–77% conv., 0% ee	7.0	30		124
13	IRED12/IRED5	Nocardia cyriacigeorgica GUH-2 /Cupriavidus sp. HPC(L)			R or S, 9–99% conv., 26–99% ee	7.4	30		125
14	IRED-E/IRED-D	Nocardiopsis alba/Mesorhizobium sp. L2C089B000			R or S, 2–99% conv., 94–99% de	8.0	30		149
15	IR46-M3	Saccharothrix espanaensis			1R, 2S, 95.6% conv., 99.9% ee, 84.4% isolated yields,	4.6	30	70.0	145
16	IR-Sip	Streptomyces ipomoeae 91-03			R or S, 86–95% conv., 28–95% ee, 3–81% isolate yields	9.5	30		28
17	Artificial IREDs				R or S, up to 100% conv., up to 96% ee				150
18	SpRedAm	Streptomyces purpureus			S, 96–96.5% conv., up to 99.5% dr, up to 80% isolated yields	7.0	25–30		151
19	NfRedAm/NfisRedAm	Neosartorya fumigata/Neosartorya fischeri			R or S, up to 97% conv., up to 97% ee	9.0	30	50–54	33
20	AdRedAm/NfRedAm/NfisRedAm	Ajellomyces dermatitidis/Neosartorya fumigate/Neosartorya fischeri			S, 34–84% conv., 85–99% ee	9.0	30		127
21	pIRs	Various sources			20–99% conv.	8.0	25		132
22	IR1-Y194F/D232H and IR25	Leishmania major/Micromonospora echinaurantiaca			R or S, up to 99% conv., 93–99% ee, 31–86% isolated yields	7.5	30	59	136
23	IR45 and variants	Streptomyces aurantiacus			S, 75–100% conv., 99% ee	7.0	30		152
24	ScIR/SvIR	Streptomyces clavuligerus/Streptomyces viridochromogenes			R or S, >99% conv., > 99% ee, 60–80% isolated yields	7.0	30		137
25	pIRs	Various sources			R or S, up to 99% conv., up to 99% ee, 31–91% isolated yields	7.0–8.0	30–40	53.5	133
26	pIR-88/pIR-202 and others	Various sources			R or S, 18–99% conv., 36–99% ee, 21–91% isolated yields	7.0	30		134
27	pIRs	Various sources			R or S, up to 99% conv., up to 99% ee, 27–80% isolated yields	7.5	30		153
28	IRED45	Streptomyces aurantiacus			S, 58–99% conv., 77–99% ee	7.0	30		154
29	ScIRED-R3-V4	Streptomyces clavuligerus			R, up to 99% conv., up to 99% ee, 82.5% isolated yield	6.0	30	49.8	138
30	IREDs	Various sources			R or S, 50–99% conv., 45–99% ee, 21–77% isolated yields	7.0	30		155
31	EneIRED	Pseudomonas sp.			R or S, up to 99% conv., up to 99% ee, up to 99 : 1 dr	9.0	25		110
32	IREDs	Various sources			R or S, 24–99% conv., up to 99 : 1 dr, 60–72% isolated yields	7.5	25		156
33	IR-36 and others	Mesorhizobium sp. 1M-11			R or S, 4–99% conv., 23–99% ee, 41–84% isolated yields	7.5–8.0	25		135
34	IR-G36-M5	Actinoalloteichus hymeniacidonis			R, up to 99% conv., up to 99% ee, 73–84% isolated yields	7.0	30	63.5	109
35	IR88/IR86/IR51/IR64	Various sources			R or S, 57–99% yields, 98–99% ee	7.5	25		141
36	IR-G02/IR-G21/IR-G35/IRED2	Various sources			R or S, up to 99% conv., up to 99% ee, 39% isolated yield	7.0	30	52.0	139
37	PcIRED M3	Penicillium camemberti			R or S, 4–99% conv., up to 99% ee, 84.7% isolated yield	7.0	30	51.5	142
38	IREDs	Various sources			R or S, up to 98% conv., up to 99% ee	7.0	30		157
39	AtIRED	Amycolatopsis thermoflava			S, 24–99% conv., 97–99% ee, 30–87% isolated yields	7.5	30		158
40	R-IRED	Streptomyces sp. GF3587			R, 11–99% conv., 20–99% ee	7.8	30		159

---

Although the majority of IREDs are (R)-selective, the Grogan and Turner groups reported the structure of an (S)-IRED from Amycolatopsis orientalis (AoIRED).¹⁰²AoIRED exhibited (S)-selectivity towards a range of cyclic imines with high activity, although interestingly (R)-configured products were observed upon prolonged storage of the enzyme or minor changes in the structure of the substrate. Based on crystal structures in complex with the cofactor NADPH or the substrate (R)-1-methyl-1,2,3,4-tetrahydroisoquinoline ((R)-MTQ), site-directed mutagenesis on the residues Y179, N241, and N171 was performed, and kinetic properties were investigated. Recently, Prejanò et al. conducted density functional theory (DFT) calculations to further investigate the mechanisms involved in the inversion of stereoselectivity of AoIRED.¹²¹ The transition state models were constructed with the substrate 1-methyl-3,4-dihydroisoquinoline, and revealed that the energy barrier for the (S)-selective path is lower than the (R)-selective path by 3.2 kcal mol⁻¹, which correlated well with the experimental data where the product was obtained in 81% ee. In particular, the variants Y179A and N171A were introduced to the models, and the energies for (R)- pathways were both lower than (S)-pathways in the variants, which correlated well with the enantio-selectivity switch of the experimental data.

Also targeting cyclic imines, Zumbrägel et al. expanded the substrate scope of IREDs to 3-thiazolidine derivatives with sulfur-containing heterocycles. A pH-indicator assay was developed where color changes were correlated to the pH changes from the 3-thiazolines to the products. High enantioselectivity (>99% ee) and conversions (>96%) have been obtained for reduction of 3-thiazolines and 2H-1,4-benzothiazines, especially sulfur-containing spiro-heterocyclic amines were synthesized in high stereoselectivity (Table 3, entries 7, 12 and 13). Co-expression of IRED and GDH in E. coli enabled in situ cofactor recycling and preparative synthesis of a (S)-3-thiazolidine derivative at a 18 g L⁻¹ substrate loading, 78% isolated yields and 99% ee.^122–124 In addition to the sulfur-containing cyclic imines, the same group has also investigated N and O containing heterocycles such as 2H-1,4-benzoxazines.¹²⁵

The discovery and expansion of the IREDs toolbox have provided important tools to the construction of chiral amine products. However, the formation of the imine intermediate has not been described by then. Therefore, initially IREDs were believed to primarily catalyze the reduction of cyclic imines. Nevertheless, it was soon to be found that intermolecular reductive amination could also be achieved and IREDs have been capable of catalyzing the coupling of the ketones with amine nucleophiles. Huber et al. firstly described an intermolecular reductive amination on ketones and in methylammonium buffer with IREDs from Streptomyces sp. GF3546 and S. aurantiacus with low conversions.¹⁰³ Going forward, Scheller et al. employed a previously reported (R)-SrIRED in the amination of a variety of aldehydes and ketones.¹²⁶ Primary and secondary amine products have been obtained using ammonia, methylamine or aniline as amine donors, albeit moderate conversions.

Aleku et al. reported a reductive aminase from Aspergillus oryzae (AspRedAm) that can catalyze both the formation of the imines and the imine reduction to produce chiral amines.²⁸ A broad range of 32 ketone acceptors with 19 different amine donors have been tested for activity of the AspRedAm. From 1 eq. to 50 eq. of amine donors were used, and high conversions and ee have been achieved. With the crystal structure of RedAm, 6 residues including N93, D169, Y177, W210, M239 and Q240 were pointed out to be the guidance for distinctive RedAm activities. Two more RedAms were identified from Aspergillus terreus (AtRedAm) and Ajellomyces dermatitidis (AdRedAm) which contained the equivalent six residues. The studies indicate that the coupling of the ketone substrate with the amine donor follows an ordered sequential Ter Bi mechanism. For the first time, further evidence demonstrated that AspRedAm can catalyze imine formation. Later their crystal structures have been established in complex with cofactors and substrates and the catalytic mechanism by AtRedAm was proposed.^30,106 The substrate scope of this enzyme has been expanded to fluorine-containing arylkontones.¹²⁷

IREDs can be used in the synthesis of a number of pharmaceuticals and their amine-containing intermediates. One of the early examples was reported by Matzel et al, who screened a panel of 39 IREDs targeting the pharmaceutically relevant (R)-rasagiline in preparative scale with high yields (up to 81%) and ee (up to 90%) (Table 3, entry 6). Tertiary amines have also been synthesized in preparative scale using N-methyl propargylamine as the amine donor, albeit long reaction time (7 days) and excessive amine utilization (40–100 eq.).¹²⁸

Although the reductive amination has been shown to proceed with high efficiency, the previously used large excess of amine donors used (up to 100 eq.) prohibits the applications in industry. Therefore, identifying more powerful biocatalysts, reducing the excess use of amine and improving the activity and stereoselectivity have become a focus in the following investigations of IREDs. Indeed, Roiban et al. expanded the application of the reaction to the synthesis of a range of secondary and tertiary amines including anilines and heteroaromatic amines with only 1.0–1.1 eq. of amine donors (Table 3, entry 4).¹²⁹ Using 6.25 mM (1.25 eq.) 1,2-dicarbonyl substrates and 5 mM 1,2-diamines (Table 3, entry 3), Borlinghaus et al. reported a double reductive amination, leading to the formation of piperazine derivatives, which were key intermediates for pharmaceuticals such as Indinavir or Mirtazapin. Both (R)-IREDs and (S)-IREDs were tested, and MsIRED from Myxococcus stipitatus was identified with good reactivity.¹³⁰ Later, key residues have been recognized to be responsible for changing the stereo-preference of MsIRED from Myxococcus stipitatus. The same group constructed SSM libraries at the regions 240–246. Beneficial mutations were recombined, and the final variant containing five mutations (A241V/H242Y/N243D/V244Y/A245L) exhibited a switch of enantio-selectivity from (R)-configuration (99% ee) to (S)-configuration with 91% ee.¹⁰⁸

The IREDs toolbox was further expanded by mining and characterizing novel IREDs using tools such as BLAST. Several panels of bacterial IREDs have been identified to catalyze the formation of primary and secondary amines.¹³¹ The discovery and characterization of 80 putative IREDs based on search with the sequence identity >30% with AspRedAm were reported by Montgomery et al.¹³² The sequence-structure analysis was performed using a SmartScaffold approach. It was revealed that (R)-selective IREDs have the AspRedAm-equivalent P121, F176, M/I210, and S243 residues, and AspRedAm equivalent T121, M/L176, F/W210, and A243 for (S)-selective IREDs. They concluded that the IRED sequence–activity relationships are dependent heavily on the substrates, and IREDs known to catalyze the reductive amination of cyclic imines are also capable of catalyzing the intermolecular reductive aminations (Table 3, entry 21).

The IREDs catalyzed synthesis of chiral amine intermediates have been significantly advanced by screening of a metagenomic panel of 384 enzymes including the use of a colorimetric high throughput (HTS) method.¹³³ The substrate scope has been greatly expanded to include the synthesis of β-amino esters from β-keto esters via a dynamic kinetic resolution (DKR) process at preparative scales, affording high yields and ee (Table 3, entry 25). A wide range of acetophenone derivatives, dimethylamine derivatives, β-keto esters derivatives were demonstrated to be synthesized by the IREDs catalyzed reductive amination. Later, the same metagenomic panel were screened towards the synthesis of intermediates for the drug Rotigotine used for Parkinsons disease therapy (Table 3, entry 26), which reduced the synthetic steps from 5 to 3.¹³⁴ Further screening has identified IREDs from the same panel to synthesize N-alkyl-α-amino esters from α-ketoesters (Table 3, entry 33).¹³⁵ The above work did not require engineering efforts but afford products in near-perfect enantio-selectivity and high conversions, which demonstrated the powerful potential of the metagenomic enzyme discovery and high-throughput screening.

With the rapid expansion of the IRED toolbox, applications in the synthesis of valuable pharmaceuticals have been developed and exhibited promising potential in upscaled productions of important drugs. For example, Xu et al. attempted the IREDs catalyzed synthesis of 1,4-diazepanes via intramolecular reductive amination. (R)-selective LmIRED from Leishmania major and (S)-selective MeIRED from Micromonospora echinaurantiaca have been identified from screening a panel of 48 IREDs. The Y194F/D232H variant from LmIRED exhibited synergistic effect with the highest catalytic efficiency and highest selectivity for a range of substrates (Table 3, entry 22). Preparative scale biotransformations at 100 mM substrate concentration were performed affording excellent ee (99%) for the synthesis of a Suvorexant intermediate.¹³⁶

Chiral 2-aryl-substituted pyrrolidines are important building blocks for a number of potential pharmaceutical candidates including CDK8 inhibitor MSC2530818, tropomyosin receptor kinase inhibitor Larotrectinib (LOXO-101), etc. The Zheng group screened a panel of IREDs towards 2-aryl-substituted pyrrolines and obtained two best performing IREDs: (R)-selective ScIR from Streptomyces clavuligerus and (S)-selective SvIR from Streptomyces viridochromogenes. They developed preparative scale (100 mL) processes for both (R)- and (S)-IREDs to afford conversions >99% with up to 79.7% isolated yields and >99% ee at up to 40 mM substrate loadings (Table 3, entry 24).¹³⁷ Recently, the same group further engineered ScIR for 3 rounds, and with the support of crystal structure of WT ScIRED, they have been able to select 10 residues in proximity with the substrate and constructed SSM libraries.¹³⁸ The variant ScIRED-R1-V1 was generated, resulting in a 10-fold increase in activity. For the second round, all 27 sites were selected from 4–8 Å in the second shell. The resulting variants ScIRED-R2-V2 displayed over 70-fold improvement in activity over the WT enzyme, and ScIRED-R2-V3 displayed higher thermostability. Finally, based on B-factor analysis, the best variant ScIRED-R2-V4 was obtained which was able to transform 80 g L⁻¹ substrate, affording >99.5% ee and >99% conversion (Table 3, entry 29). In the upscaled synthesis of Larotrectinib intermediate (R)-2-(2,5-difluorophenyl) pyrrolidine on 20 L scale, 1.32 kg of the product was obtained in 4.5 h with 82.5% isolated yield, >98.5% purity and >99% ee, which resulted in a 352 g L⁻¹ day⁻¹ space-time-yield (STY).

Similarly, targeting N-containing heterocycles, Zhang et al., screened a panel of 86 enzymes against N-Boc-3-piperidinone with benzyl amine as the amine donor. IR-G36 was identified with the highest (R)-selective product (78% ee), albeit only 8% conversion. Upon examining the crystal structure and docking structures, we used site-saturation mutagenesis (SSM) and CAST to perform several rounds of engineering on IR-G36. The final variant IR-G36-M5 bearing mutations at various positions (N121T/S260F/F207I/M238A/M264H/A267H/L197I/N271K/H189A/G145K) showed 5.4 °C improvement in T_m, and 4193-fold in k_cat/K_m. The crystal structure of the variant IR-G36-M5 in complex with NADP⁺ was obtained, and additional hydrogen bonds and salt bridges were identified, which provided insights to the stability of the variant. Further preparative scale biotransformations afforded N-boc protected azacycloalkylamines in ketone loadings up to 47 g L⁻¹ and up to 99% conversions and ee (Table 3, entry 34).¹⁰⁹ Later, bulky amines as donors have been included into the scope (Table 3, entry 36).¹³⁹

Large panel screenings and further engineering efforts have enabled a number of target pharmaceuticals to be synthesized by IREDs. Zhan et al. developed the IRED catalyzed synthesis of a key intermediate to the tyrosine kinase inhibitor Tofacitinib, which has been a successful drug commercialized in more than 50 countries.¹⁴⁰ A panel of 175 IREDs were screened against the racemic ketone substrate rac-1-benzyl-4-methylpiperidin-3-one to obtain (3R,4R)-1-benzyl-N,4-dimethylpiperidin-3-amine bearing two consecutive stereocenters. 43 active enzymes were identified. Further exploration and optimization of amine donors, substrate concentrations led to the identification of two IREDs with the highest activity and enantio-selectivity. Preparative reactions were carried out using 100 mM rac-1-benzyl-4-methylpiperidin-3-one at 25 mL scale afforded both (3R,4R)-1-benzyl-N,4-dimethylpiperidin-3-amine or (3S,4S)-1-benzyl-N,4-dimethylpiperidin-3-amine in high isolated yields (83% and 91%), excellent stereoselectivity (97% and >99% ee values) and diastereoselectivity (>99 : 1 dr). The same group reported later another application of IREDs in the synthesis of enantiocomplementary fused-ring tetrahydroisoquinoline (THIQs) and tetrahydro β-carboline (THβCs). A panel of 104 IREDs were screened and high stereoselectivity (up to >99% ee) were obtained. A number of fused-ring THIQ and THβCs were synthesized at preparative scales including natural alkaloids with stereo-complementary IREDs in high yields (up to 85%) (Table 3, entry 35).¹⁴¹

Very recently, the Zheng group discovered an IRED from Penicillium camemberti (PcIRED). Targeting the calcimimetic drug Cinacalcet, ISM was performed on this enzyme. With three rounds of engineering combining site-directed saturation mutagenesis and combination of beneficial mutations, a variant M3 was obtained with 488-fold increase in activity and 11 °C improvement in melting temperature over the wild-type enzyme. 100 mL scale synthesis of cinacalcet was performed with 500 mM (R)-1-(1-naphthyl)ethylamine and 1.2 eq. of the ketone substrate 3-[3-(trifluoromethyl)phenyl]propanal, and the final isolated yield reached 84.7% with >99% ee. In particular, broad scope for the bulky ketone and amine nucleophiles were well tolerated by the variants, and high cis/trans selectivity was obtained towards a range of 3-phenyl-cyclobutanone or 4-phenyl-cyclohexanone derivatives (Table 3, entry 37).¹⁴²

It has been established that IREDs catalyze both the imine formation and asymmetric reduction of the imines. However, there is more evidence towards the latter than the former. Gilio et al. characterized the imine formation as well as the reductive amination by IREDs and showed that larger amine donors may undergo spontaneous formation of imines which were then recruited from solution and reductively aminated by IREDs. Site saturation mutagenesis library of a knowledge-based site A208 has led to an A208N variant with superior activity towards bulky amines such as isoindoline or octahydrocyclopenta(c)pyrrole coupled with cyclohexanone. Preparative scale biotransformations with 1.2 molar eq. of amines afforded up to 93% isolated yields.¹¹²

With the use of large data sets, machine learning models have been trained and combined with directed evolution to explore the sequence space of the enzymes more efficiently. Ma et al. conducted deep mutational scanning (DMS), error-prone PCR, structure-guided machine-directed evolution and low N engineering to IRED-88 that was identified from genome mining. The authors demonstrated that machine-learning combined with directed evolution allowed the screening of a smaller library to obtain comparable activities and stereoselectivities to the traditional strategies. Machine-learning models can be helpful when the throughput of the screening is limited due to cost, scale, or time and a combination use of the methodologies might be necessary depending on the available resources.¹⁴³

Most recently, a team at Manchester established the IREDFisher workflow by designing an in silico streamline with sequence analysis, template-based homology modelling, and molecular docking. The final scoring algorithm consist of Autodock Vina scoring, key catalytic hydride transfer distance and iminium intermediate-stabilizing residues. Experimental validation results from IREDFisher-selected panels outperformed random selection in hits with conversions over 50%. In addition, user-friendly web interface (https://enzymeevolver.com/IREDFisher) was provided which enabled users with no computational experiences to exploit the tools with convenience.¹⁴⁴

Since the discovery of IREDs and the advancement of high throughput screening, directed evolution and rational design methodologies, extensive research efforts have been invested in expanding the substrate scope of IREDs. Herein a selection of substrate structures from various literature precedence are listed in Table 3. IREDs from various sources have been identified, and both discovery and engineering of IREDs have enabled the biotransformation of a large number of substrates carrying various moieties. In particular, the thermostability of IREDs has been improved dramatically. For example, in the work reported by Schober et al, the melting point (T_m) of the IRED was improved from 41.3 °C to 70.0 °C through 3 rounds of mutagenesis.¹⁴⁵ Therefore, T_m of the various wild-type IREDs or their variants are included in Table 3. The pH and temperature are also critical when designing the enzymatic reactions, and generally IREDs prefer slightly basic conditions (pH 7.0–9.0) and room temperature (30 °C). These conditions are also summarized in Table 3.

3.3. Towards the applications of IREDs

3.3.1. IREDs in cascades.
3.3.1.1. Multi-enzymatic cascades. One-pot multi-enzymatic approaches as well as whole-cell biocatalysis have received increasing research attention as they fulfil the industrial needs for the synthesis of structurally diverse molecules. In particular, chemoenzymatic methodologies are important since they harness the benefits of selectivity of biocatalysis and highly diverse reactivity of organic/metal catalysis. Hereafter we discuss the applications of IREDs in cascades with either multi-enzymatic biocatalysis or chemo-catalysis for the synthesis of industrially relevant building blocks.

Piperidines and pyrrolidines are core structures for a variety of biologically active molecules. Using multi-enzymatic cascades, the synthesis of chiral piperidines or pyrrolidines could be achieved using carboxylic acid reductase (CAR), ω-transaminase (ω-TA), and imine reductase (IRED) in one-pot.¹⁶¹ On the other hand, whole cells are often important sources of the enzymes for biotransformations. Targeting the synthesis of chiral substituted piperidines, the Flitsch group designed a four-enzyme, four-step cascade in whole cells starting from keto acids using a bioretrosynthetic approach. Carboxylic acid reductase, ω-transaminase and IREDs have been constructed in the whole-cell system. Under optimized reaction conditions, the 4-methyl-2-phenylpiperidine was obtained in >98% de and 93% conversions, with isolated yield of 59% on preparative scale (Fig. 17).¹⁶⁰

Fig. 17 The whole-cell enzymatic cascades achieved by the design–build–test workflow. Adapted with permission from ref.160. Copyright 2017 American Chemical Society.

IREDs have also been employed in a cascade with P450 monooxygenases to produce secondary amines from cyclo-alkanes. Formate dehydrogenase from Candida boidinii (CboFDH) was used for the cofactor regeneration of P450, and a variant of ADH from Thermoanaerobacter ethanolicus (TeSADH W110A) was used to construct a hydrogen-borrowing cascade for IREDs.¹⁶² Further cascades of IREDs have been built with ene-reductases (EREDs) for the synthesis the 2-substituted saturated amine heterocycles with 2 stereocenters,¹⁶³ and galactose oxidase (GOase) to afford L-3-N-aminopiperidines and L-3-N-aminoacepanes using GOase variants M_3–5 and F2^164,165 to oxidize the aminoalcohol, followed by spontaneous cyclization to form an imine intermediate, which was subsequently reduced by AdRedAm.¹⁶⁶

Yang et al. screened and engineered IREDs for the production of a series of tetrahydroisoquinoline alkaloids (THIQAs) from dihydroisoquinoline (DHIQ) precursors (Table 3, entry 23). The IRED variant with improved enzymatic activity was then combined with a coclaurine N-methyltransferase (CNMT) from Coptis japonica, and a glucose dehydrogenase (GDH) either in a single cell (pACYCDuet and pET28a) or one-pot cascade where IRED and GDH were co-expressed and then added to the CNMT culture. The biosynthetic pathway led to full conversions in 1–3 days and isolated yields of the final THIQAs reached 93% and 96%.¹⁵² Later, the same group reported the applications of IREDs and CNMT cascades in the synthesis of chiral N-methylated 1-aryl-tetrahydro-β-carboline (THβC), which is a key intermediate for drugs such as tadalafil that are used for the treatment of erectile dysfunction (Table 3, entry 28).¹⁵⁴ Key active-site residues were identified and mutated by analyzing the interactions of the bulky substrate 1-phenyl-DHβC with the IRED, and further combinations were constructed with the beneficial mutations. The final triple variant IR45-L228′A/M250′L/E251′M was able to reduce most of the substituted 1-phenyl-DHβCs in >99% ee and conversions, and preparative scale biotransformations were also performed in 1 L scale to achieve full conversions. Targeting THβCs and THIQs, further discovery and engineering work has revealed the powerful potential of IREDs to catalyze this class of stereoselective synthesis of the sterically demanding molecules (Table 3, entries 38–40).^157–159

In 2022, Li et al. reported the two-step sequence employing the EREDs and IREDs for the production of N-substituted γ-amino esters and γ-lactams using cyclopropylamine as the amine partner (Table 3, entry 31). Screening efforts were invested in identification of the most active ERED and IRED towards α,β-unsaturated γ-ketoesters substrates such as methyl 3-oxocyclohex-1-ene-1-carboxylate and ethyl (E)-4-oxo-2-propylpent-2-enoate. The carbonyl reduction was the first step and then upon completion the IRED catalyzed reductive amination was initiated as the second step. All of the four diastereomers of the γ-amino esters and two enantiomers of the γ-lactams were obtained using various combinations of (R)- and (S)-selective EREDs and IREDs in 60–72% yields and >99% ee or dr on preparative scales.¹⁵⁶

3.3.1.2. Chemo-enzymatic cascades. Chemoenzymatic approaches have proved to be potentially valuable methods for the synthesis of complex target molecules. Azepanes and benzazepines are core motifs of both medicinal and pesticidal active molecules. Zawodny et al. utilized IREDs in the reduction of heterocyclic 7-membered rings containing an imine moiety, followed by treatment of the amine reduction products with base (LDA in a 4 : 1 mixture of Et₂O and DMPU), enabling aryl migration to take place, thereby generating enantiopure α-functionalized azepine-derived ureas in good yields. This is an excellent example of chemoenzymatic synthesis of heterocyclic tertiary amines (Table 3, entry 8), taking advantage of the enantioselectivity of enzymes to prepare steric challenging amine products (Fig. 18).¹⁴⁶

Fig. 18 IREDs-catalyzed reductive amination followed by organolithium mediated α-functionalization of azepine-derived ureas.

Further chemoenzymatic approaches were developed with IREDs with ammonia borane for the deracemization of racemic amines. The oxidation of amines in presence of NADP⁺ was exploited in combination with imine reduction by ammonia borane and cofactor regeneration by NADPH oxidase (NOX) using oxygen. This system was applied to a broad range of cyclic and acyclic substrates and represents a useful addition to the synthetic toolbox of chiral amines starting from racemic amines.¹⁶⁷

Electrical energy has been extensively used in the cofactor regeneration of oxidoreductases to take advantage of sustainable and clean energy sources. Al-Shameri et al. used molecular H₂ for NAD⁺-reducing hydrogenase (SH) coupled with putrescine oxidase catalyzed oxidation and IRED catalyzed reductive amination for the synthesis of piperidine derivatives from diamine substrates. A gas-selective permeable membrane was used to facilitate the transfer of gases and enzymes were immobilized on commercially available carriers. The system was easily scaled up to 300 mL and the immobilized enzymes can be reused up to six times.¹⁶⁸

Dihydropinidines are a class of piperidines with antifeedant activity. The Kroutil group reported the synthesis of dihydropinidines using both chemo- and bio-cascades involving IREDs on multigram scale. The steps are: (1) Lewis acid catalyzed Michael addition; (2) pig liver esterase (PLE) catalyzed hydrolysis followed by acid-promoted decarboxylation to afford diketone nonane-2,6-dione; (3) two stereo-complementary transaminases from Arthrobacter (ArR-TA and ArS-TA) catalyzed transamination of the diketone nonane-2,6-dione followed by spontaneous cyclization to the imine intermediate; (4) imine reduction of 4 by screening a panel of 14 IREDs and identification of highly active (R)- and (S)-selective IREDs. Finally, implementation of the TA and IRED catalyzed reactions enabled a one-pot concurrent cascade reaction on preparative scale, affording >99% ee and 57% overall isolated yields for the final product (2R,6S)-dihydropinidine (Fig. 19).¹⁴⁹ Similarly, using the TA-IRED cascade or Pt/C catalyzed hydrogenation in a continuous flow reactor, tri-substituted piperidines with three chiral centres have been obtained in high diastereomeric ratio (dr) (>99 : 1) and good yield (up to 73%) on preparative scale.¹⁶⁹

Fig. 19 Chemo-enzymatic approaches to the synthesis of dihydropinidine.

Very recently, the deracemization methodology to combine MAO-N, IREDs and metal catalyzed allylation has been achieved in one-pot to produce a range of C(1)-substituted THIQs. The allylBPin intermediate was generated by Yb(OTf)₃ catalyzed allylboration to form racemic THIQs, subsequently biocatalytic deracemization afforded the C(1)-substituted THIQs in high yields and good to high ee. The system could also be scaled up with up to 64% isolated yield and 98% ee (Table 3, entry 40).¹⁵⁹

3.3.2. Multifunctional enzymes for reductive amination. IREDs are promiscuous enzymes which were initially shown to catalyze fluoro-substituted ketone reductions by Lenz et al.¹⁷⁰ Recently, the Turner group reported a multifunctional enzyme, identified from a metagenomic IRED library, which they named EneIRED. They showed that conjugate reduction (CR) and reductive amination (RA) could be achieved by one enzyme through CR–RA dual cycles. The substrate scope included a range of enals and enones (both linear and cyclic) with various amine donors, affording up to 99% ee, 99% dr and 99% conversions (Table 3, entry 31). Mechanistic investigations revealed a stepwise double-hydride transfer mechanism confirmed by isotopic labelling experiments (Fig. 20). The hydride transfer was shown to occur at C-1 and C-3 of the unsaturated carbonyl substrate. The α,β-unsaturated carbonyl V and amine forms the intermediate VI, while upon hydride transfer the carbonyl reduction takes place and forms the complex VII. With the carbonyl VIII being released into solution, a further complex IX was formed with amine and VIII. Finally, the complex IX undergoes reductive amination to yield the optically active product X. X-ray crystallography of the EneIRED in complex with NADP⁺ were performed and the structure was revealed. The EneIRED possess a Y177 at the top of the active site in common with other IREDs. However, an additional Y181 was also discovered to point to the active site towards the cosubstrate binding cleft. Further investigation of the role of Y177 and Y181 were conducted and showed that Y177 is responsible for steric constraint for controlling the face for hydride delivery in both CR and RA steps, and Y181 plays an more important role for CR and not RA.¹¹⁰

Fig. 20 Proposed catalytic cycle of the EneIRED catalyzed carbonyl reduction and reductive amination.

As discussed in previous sections, piperidines are often target products for applications of IREDs. However, most of the approaches employ multi-enzymatic cascades starting from ketoacids,^160,161 diamines¹⁷¹ or amino alcohols.¹⁶⁶ Alternatively, pyridines could be obtained by mild and naturally occurring pathways and serve as a source through dearomatization by quaternization-activation of the pyridine nitrogen. Very recently, the Turner group reported a chemoenzymatic cascade whereby pyridines were subjected to chemical reduction to produce tetrahydropyridines (THP), followed by biocatalytic amine oxidation by a 6-hydroxy-D-nicotine oxidase (6-HDNO) variant and conjugate reduction of the C=C and C=N bonds with the multifunctional EneIRED (Fig. 21). The chemo-enzymatic cascade could be applied to the synthesis of both (R)- and (S)-configured products on preparative scale by using EneIREDs with complementary stereoselectivity. For example, the antipsychotic drug Preclamol was obtained in both (R)-(+)- and (S)-(−)-conformation with 96% ee and >50% yields. This system was also applied to the synthesis of other APIs such as OSU6162 and Niraparib in high ee.¹¹¹

Fig. 21 Chemo-enzymatic dearomatization of pyridines for the synthesis of chiral 3- and 3,4-substituted piperidines.

In addition to multifunctional IREDs, short-chain dehydrogenases (SDR), whose native activity is carbonyl reduction, were also reported to catalyze reductive amination reactions. For example, SDRs from Narcissus pseudonarcissus, Zephyranthes treatiae, and Amaryllidacea Zephyranthes treatiae were all reported to reduce imines.^172,173 Furthermore, the β-hydroxyacid dehydrogenases from Thermocrinus albus (TaβHADs) were also discovered to reduce cyclic imines to the corresponding piperidine products.¹⁰⁷ Structural investigations revealed both common and different features for the βHADs and IREDs.¹⁷⁴

3.3.3. Large-scale industrial applications. The first major industrial application of IREDs was reported by scientists at GlaxoSmithKline (GSK) in 2019 in the synthesis of a lysine-specific demethylase-1 (LSD1) inhibitor GSK2879552 (Table 3, entry 15).¹⁴⁵ After initial panel screening, 95% conversions have been achieved for the chiral amine product with only 1.1 equivalents of the amine donor. The IRED from Saccharothrix espanaensis was identified as the best performing wild type enzyme. Subsequently, 3 rounds of directed evolution were performed by construction of site saturation mutagenesis libraries, recombination of beneficial mutations using gradually increased substrate loadings (up to 24 g L⁻¹). Finally, the variant M3 with 21 mutations compared to wild type was constructed, and 20 L scale production was conducted at 20.1 g L⁻¹ substrate loadings to afford 84.4% isolated yield, 99.7% ee. The TON achieved with the final variant was >38 000 fold increased compared to wild type enzyme. This work not only demonstrated the potential of IREDs in commercial productions at low amine excess, high substrate loadings and large scales, but also showed the robustness of the enzyme that could withstand a buffer pH at 4.2, and preincubation at 40 °C.

Researchers at Pfizer employed IREDs to perform the reductive amination with methylamine of a key ketone intermediate for the drug abrocitinib to directly obtain the intermediate isopropyl (1S,3S)-3-(methylamino)cyclobutane-1-carboxylate on a 230 kg commercial plant scale production (Table 3, entry 18).¹⁵¹ The drug abrocitinib is a Janus kinase 1 (JAK1) inhibitor and is in development to treat atopic dermatitis. Screening a panel of wild-type IRED enzymes and 3 rounds of further engineering with single site saturation mutagenesis (SSM) and recombination of beneficial mutations yielded a final variant IR021-R3-V6, bearing N131H/A170C/F180M/G217D mutations that locate mostly at the substrate or cofactor binding sites. Pilot plant and plant productions were carried out on 65 kg and 230 kg scales, and the conversions and ee were up to 92.5% and >99.5%. Results in various scales were shown to be perfectly in accordance without the decline that normally would be observed for scale-up reactions.

Another example was also demonstrated by researchers at Pfizer where they showcased a multi-step, chemobiocatalytic synthesis of a CDK 2/4/6 Inhibitor, which was developed as an oncology candidate. A key step to the intermediate is the reductive amination of the ketone 2-hydroxy-2-methylcyclopentanone to the corresponding amine (1R)-2-(benzylamino)-1-methylcyclopentanol.¹⁷⁵ A panel of 88 IREDs were screened, and 98 further SSM libraries were constructed and screened. Finally, a variant IR007-143 achieved 43% conversion at a 50 g L⁻¹ substrate concentration and 2.5 wt% enzyme loading.

Process aspects of using IREDs in downstream processing (DSP) have been further investigated to showcase the applicational potentials.¹⁷⁶ Employing IREDs in continuous flow process would enable the combinations of incompatible processes in one streamline. For example, the Turner group used a multi-point injection reactor for inline mixing and packed bed systems for various enzymes including transaminases, galactose oxidases, and IREDs to perform the cascade reactions in a single continuous flow system.¹⁷⁷

3.3.4. Cofactor regenerations. Hydrogen-borrowing cascades are effective methods to combine both oxidation and reduction steps in one-pot with high atom economy. Following the first report of the hydrogen-borrowing concepts utilizing AmDHs and ADHs,^62,63 the Turner group reported a hydrogen-borrowing process using ADH-150 with AspRedAm, which offers advantages in the production of a broad range of optically active amines from chiral secondary alcohols.⁶⁵

Initially, the cofactors for IREDs had been strictly limited to NADPH, which limited the possibilities of using a variety of cofactor recycling systems, since the availability of NADPH regenerating systems is more rare and the cost of NADH is lower.¹⁷⁸ Borlinghaus et al. reported a CSR-SALAD (namely “cofactor specificity reversal-structural analysis and library design”) strategy to semi-rationally construct mutant libraries with the guidance of crystal structures with the NADPH binding pocket and coordinating residues to the phosphate moiety at the NADPH molecule. The final variant V10 gave a TON towards NADH of 38 min⁻¹, comparing to a lower NADPH TON of 0.6 min⁻¹, demonstrating that the cofactor preference has been successfully switched. Other research regarding the recycling of NADPH has also been reported, including H₂-driven cofactor recycling.¹⁷⁹ Vidal et al. reported an approach using CSR-SALAD to predict residues crucial for cofactor preference, and then artificial selection platform was built to identify highly active variants with switched cofactor preference.¹⁸⁰ They manipulated the E. coli genes to restrict the growth under anaerobic conditions, therefore, the restoration of growth can be linked to the presence of a variant that is active for the un-nature substrate. MsIRED was used as one of the model enzymes, and a higher activity was obtained with NAD⁺ preference instead of NADP⁺ with just a single round of mutagenesis.

IREDs are mostly NADPH dependent enzymes, and traditional cofactor recycling systems using glucose and GDH have been coupled with IREDs to enable efficient biocatalytic reactions. However, glucose as a co-substrate affords gluconic acid as the product with very poor atom economy. Alternatively, cyanobacteria have been shown to utilize water splitting for electron supply for reductive reaction, offering good atom economy and ease for product extraction. Büchsenschütz et al. integrated the genes encoding IREDs into the cyanobacterium Synechocystis sp. PCC 6803 under the control of the light-inducible promotor P_psbA2.¹⁸¹ Initially low conversions were obtained (<36%), however, upon promotor optimization, the substrate 6-methyl-2,3,4,5-tetrahydropyridine can be reduced in 83% conversion to form the corresponding amine, and up to 99% ee were obtained for the substrates.

3.3.5 Artificial metalloenzymes. Artificial metalloenzymes (ArMs) represent complementary biocatalysts to imine reductases, and indeed reductive amination was reported by ArMs even before IREDs.^150,182 ArMs possess advantages such as the use of formate as a hydride source instead of the more expensive NADPH cofactors. The Ward group employed a biotin–streptavidin (Sav) technology to perform directed evolution on an artificial transfer hydrogenase (ATHase) with [Cp*Ir(biot-p-L)Cl] (Fig. 22). Screening of cell-free extracts towards cyclic imines has enabled reductive amination to afford amine products with high ee.^183,184 They employed a screening method they developed with DiAm (1,1-azobis N,N-dimethylformamide) which allows the use of unpurified cell-free extract, and subsequently ISM was conducted on residues within 10 Å of the iridium center combined with rational engineering. Final variants S112A-N118P-K121A-S122M and S112R-N118P-K121A-S122M-L124Y afforded (R)- and (S)-amine products in 95% and 86% ee, respectively.

Fig. 22 Reduction of a cyclic imine using ATHase based on biotin–streptavidin technology.

4. Future perspectives

In this review, we have focused on NAD(P)H-dependent enzymes, namely AmDHs and IREDs (or RedAms) that are capable of catalyzing the reductive amination of ketones utilizing both inorganic and organic amines to produce primary, secondary and tertiary chiral amines. The phylogenetic tree has been constructed against a variety of AmDHs, IREDs and RedAms (Fig. S1, ESI†), which shows that the sequence homology between IREDs and RedAms are more prominent than that with AmDHs. The importance of enzymatic reductive amination has been clearly established for the synthetic applications of chiral amines. In terms of the discovery of novel enzymes, developing advanced enzyme toolbox, novel design of synthetic routes and large-scale industrial applications, both AmDHs and IREDs are among the most robust enzymes which have attracted considerable attention in recent years. Going forward, there remain considerable opportunities and challenges for the industrialization of the enzymes. Some priorities for the future development of AmDHs and IREDs are: (1) synthetic biology enables the manufacturing of high-value products from sustainable feedstock, and applications of engineered enzymes in construction of non-nature pathways could be valuable for industrialization of the enzymes. (2) Artificial intelligence (AI) is drawing attention across the globe, including within the biocatalysis community.¹⁸⁵ There are already reports of using machine learning methods in the directed evolution of IREDs,¹⁴³ and Alphafold2 has now been used routinely to predict protein structures. In the future, more in-depth applications of AI-assisted protein engineering could be expected. (3) High throughput screening has been a bottleneck for engineering of many enzymes. Currently the HTS method reported for IREDs¹³³ utilizes the reverse reaction of IREDs (oxidation of amines) which limits its applications. (4) Continuous flow systems enable the reuse and recycling of biocatalysts and dramatically reduce the cost of the enzymes in industrial application, thereby implying the vital roles for enzyme immobilization techniques. Especially for enzymes require co-factors and regeneration systems, co-immobilization of enzymes require further in-depth investigations. (5) For AmDHs, the substrate scope of ketone substrates and amine nucleophiles are still relatively narrow, with only a few reported examples showing that AmDHs can accept secondary amine donors. Future engineering work could focus on the development of AmDHs for the synthesis of more structurally diverse products.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (No. 2019YFA0905100 and 2021YFA0910400), the National Natural Science Foundation of China (No. 32171462 and 32171268), Tianjin Synthetic Biotechnology Innovation Capacity Improvement Project (No. TSBICIP-CXRC-040), the Natural Science Foundation of Tianjin (No. 21JCJQJC00110).

References

1. D. Ghislieri and N. J. Turner, Top. Catal., 2014, 57, 284–300.
2. C. E. Paul, M. Rodríguez-Mata, E. Busto, I. Lavandera, V. Gotor-Fernández, V. Gotor, S. García-Cerrada, J. Mendiola, Ó. de Frutos and I. Collado, Org. Process Res. Dev., 2014, 18, 788–792.
3. D. Wu, P. Cai, X. Zhao and Y. Pan, Org. Lett., 2017, 19, 5018–5021.
4. M. Höhne and U. T. Bornscheuer, ChemCatChem, 2009, 1, 42–51.
5. O. I. Afanasyev, E. Kuchuk, D. L. Usanov and D. Chusov, Chem. Rev., 2019, 119, 11857–11911.
6. N. A. Strotman, C. A. Baxter, K. M. J. Brands, E. Cleator, S. W. Krska, R. A. Reamer, D. J. Wallace and T. J. Wright, J. Am. Chem. Soc., 2011, 133, 8362–8371.
7. D. Q. Pham, A. Nogid and R. Plakogiannis, Am. J. Health-Syst. Pharm., 2008, 65, 521–531.
8. B. L. S. H. Berman and C. Salzman, Ment. Hosp., 1992, 43, 671–672.
9. W. J. Burke, Expert Opin. Invest. Drugs, 2002, 11, 1477–1486.
10. M. S. A. G. M. Keating, Drugs, 2010, 70, 761–791.
11. G. G. Abhilash Desai, Expert Opin. Pharmacother., 2001, 2, 653–666.
12. Q. Yin, Y. Shi, J. Wang and X. Zhang, Chem. Soc. Rev., 2020, 49, 6141.
13. Y. Tian, L. a Hu, Y.-Z. Wang, X. Zhang and Q. Yin, Org. Chem. Front., 2021, 8, 2328–2342.
14. T. Irrgang and R. Kempe, Chem. Rev., 2020, 120, 9583–9674.
15. X. Tan, S. Gao, W. Zeng, S. Xin, Q. Yin and X. Zhang, J. Am. Chem. Soc., 2018, 140, 2024–2027.
16. S. Simic, E. Zukic, L. Schmermund, K. Faber, C. K. Winkler and W. Kroutil, Chem. Rev., 2022, 122, 1052–1126.
17. H. Kohls, F. Steffen-Munsberg and M. Hohne, Curr. Opin. Chem. Biol., 2014, 19, 180–192.
18. V. F. Batista, J. L. Galman, D. C. G. A. Pinto, A. M. S. Silva and N. J. Turner, ACS Catal., 2018, 8, 11889–11907.
19. D. Koszelewski, K. Tauber, K. Faber and W. Kroutil, Trends Biotechnol., 2010, 28, 324–332.
20. F. Parmeggiani, N. J. Weise, S. T. Ahmed and N. J. Turner, Chem. Rev., 2018, 118, 73–118.
21. J. Liu, W. Kong, J. Bai, Y. Li, L. Dong, L. Zhou, Y. Liu, J. Gao, R. T. Bradshaw Allen, N. J. Turner and Y. Jiang, Chem. Catal., 2022, 2, 1–27.
22. M. Sharma, J. Mangas-Sanchez, N. J. Turner and G. Grogan, Adv. Synth. Catal., 2017, 359, 2011–2025.
23. M. D. Patil, G. Grogan, A. Bommarius and H. Yun, ACS Catal., 2018, 8, 10985–11015.
24. J. J. Sangster, J. R. Marshall, N. J. Turner and J. Mangas-Sanchez, ChemBioChem, 2022, 23, e202100464.
25. G. Grogan, Curr. Opin. Chem. Biol., 2018, 43, 15–22.
26. J. Mangas-Sanchez, S. P. France, S. L. Montgomery, G. A. Aleku, H. Man, M. Sharma, J. I. Ramsden, G. Grogan and N. J. Turner, Curr. Opin. Chem. Biol., 2017, 37, 19–25.
27. G. A. Aleku, S. P. France, H. Man, J. Mangas-Sanchez, S. L. Montgomery, M. Sharma, F. Leipold, S. Hussain, G. Grogan and N. J. Turner, Nat. Chem., 2017, 9, 961–969.
28. P. Matzel, S. Wenske, S. Merdivan, S. Guenther and M. Hoehne, ChemCatChem, 2019, 11, 4281–4285.
29. O. Mayol, K. Bastard, L. Beloti, A. Frese, J. P. Turkenburg, J.-L. Petit, A. Mariage, A. Debard, V. Pellouin, A. Perret, V. de Berardinis, A. Zaparucha, G. Grogan and C. Vergne-Vaxelaire, Nat. Catal., 2019, 2, 324–333.
30. M. Sharma, J. Mangas-Sanchez, S. P. France, G. A. Aleku, S. L. Montgomery, J. I. Ramsden, N. J. Turner and G. Grogan, ACS Catal., 2018, 8, 11534–11541.
31. V. Tseliou, M. F. Masman, W. Bohmer, T. Knaus and F. G. Mutti, ChemBioChem, 2019, 20, 800–812.
32. L. Liu, D.-H. Wang, F.-F. Chen, Z.-J. Zhang, Q. Chen, J.-H. Xu, Z.-L. Wang and G.-W. Zheng, Catal. Sci. Technol., 2020, 10, 2353–2358.
33. J. Mangas-Sanchez, M. Sharma, S. C. Cosgrove, J. I. Ramsden, J. R. Marshall, T. W. Thorpe, R. B. Palmer, G. Grogan and N. J. Turner, Chem. Sci., 2020, 11, 5052–5057.
34. M. J. Abrahamson, E. Vázquez-Figueroa, N. B. Woodall, J. C. Moore and A. S. Bommarius, Angew. Chem., Int. Ed., 2012, 51, 3969.
35. P. J. Baker, A. P. Turnbull, S. E. Sedelnikova, T. J. Stillman and D. W. Rice, Structure, 1995, 3, 693–705.
36. Y. Zhao, T. Wakamatsu, K. Doi, H. Sakuraba and T. Ohshima, J. Mol. Catal. B: Enzym., 2012, 83, 65–72.
37. H. Yamaguchi, A. Kamegawa, K. Nakata, T. Kashiwagi, T. Mizukoshi, Y. Fujiyoshi and K. Tani, J. Struct. Biol., 2019, 205, 11–21.
38. T. Wu, Y. Wang, N. Zhang, D. Yin, Y. Xu, Y. Nie and X. Mu, ACS Catal., 2023, 13, 158–168.
39. J. L. Vanhooke, J. B. Thoden, N. M. W. Brunhuber, J. S. Blanchard and H. M. Holden, Biochemistry, 1999, 38, 2326–2339.
40. N. M. W. Brunhuber, J. B. Thoden, J. S. Blanchard and J. L. Vanhooke, Biochemistry, 2000, 39, 9174–9187.
41. L. Ducrot, M. Bennett, G. André-Leroux, E. Elisée, S. Marynberg, A. Fossey-Jouenne, A. Zaparucha, G. Grogan and C. Vergne-Vaxelaire, ChemCatChem, 2022, 14, e202200880.
42. M. Bennett, L. Ducrot, C. Vergne-Vaxelaire and G. Grogan, ChemBioChem, 2022, 23, e202200136.
43. R. D. Franklin, J. A. Whitley, J. M. Robbins and A. S. Bommarius, Chem. Eng. J., 2019, 369, 634–640.
44. M. J. Abrahamson, J. W. Wong and A. S. Bommarius, Adv. Synth. Catal., 2013, 355, 1780–1786.
45. M. T. Reetz, Angew. Chem., Int. Ed., 2011, 50, 138–174.
46. M. T. Reetz and J. D. Carballeira, Nat. Protoc., 2007, 2, 891–903.
47. G. Qu, A. Li, C. G. Acevedo-Rocha, Z. Sun and M. T. Reetz, Angew. Chem., Int. Ed., 2020, 59, 13204–13231.
48. W. Jiang and Y. Wang, J. Microbiol. Biotechnol., 2020, 30, 146–154.
49. L. J. Ye, H. H. Toh, Y. Yang, J. P. Adams, R. Snajdrova and Z. Li, ACS Catal., 2015, 5, 1119–1122.
50. A. Pushpanath, E. Siirola, A. Bornadel, D. Woodlock and U. Schell, ACS Catal., 2017, 7, 3204–3209.
51. V. Tseliou, T. Knaus, M. F. Masman, M. L. Corrado and F. G. Mutti, Nat. Commun., 2019, 10, 3717.
52. V. Tseliou, T. Knaus, J. Vilim, M. F. Masman and F. G. Mutti, ChemCatChem, 2020, 12, 2184–2188.
53. V. Tseliou, D. Schilder, M. F. Masman, T. Knaus and F. G. Mutti, Chem. – Eur. J., 2021, 27, 3315–3325.
54. N. Itoh, C. Yachi and T. Kudome, J. Mol. Catal. B: Enzym., 2000, 10, 281–290.
55. O. Mayol, S. David, E. Darii, A. Debard, A. Mariage, V. Pellouin, J.-L. Petit, M. Salanoubat, V. de Berardinis, A. Zaparucha and C. Vergne-Vaxelaire, Catal. Sci. Technol., 2016, 6, 7421–7428.
56. K. Bastard, A. A. T. Smith, C. Vergne-Vaxelaire, A. Perret, A. Zaparucha, R. De Melo-Minardi, A. Mariage, M. Boutard, A. Debard, C. Lechaplais, C. Pelle, V. Pellouin, N. Perchat, J.-L. Petit, A. Kreimeyer, C. Medigue, J. Weissenbach, F. Artiguenave, V. De Berardinis, D. Vallenet and M. Salanoubat, Nat. Chem. Biol., 2014, 10, 42–49.
57. A. A. Caparco, E. Pelletier, J. L. Petit, A. Jouenne, B. R. Bommarius, V. de Berardinis, A. Zaparucha, J. A. Champion, A. S. Bommarius and C. Vergne-Vaxelaire, Adv. Synth. Catal., 2020, 362, 2427–2436.
58. B. R. Bommarius, M. Schurmann and A. S. Bommarius, Chem. Commun., 2014, 50, 14953–14955.
59. S. K. Au, B. R. Bommarius and A. S. Bommarius, ACS Catal., 2014, 4, 4021–4026.
60. J. Loewe, A. A. Ingram and H. Groeger, Bioorg. Med. Chem., 2018, 26, 1387–1392.
61. J. Li, X. Mu, T. Wu and Y. Xu, Bioresour. Bioprocess., 2022, 9, 33.
62. F. G. Mutti, T. Knaus, N. S. Scrutton, M. Breuer and N. J. Turner, Science, 2015, 349, 1525–1529.
63. F.-F. Chen, Y.-Y. Liu, G.-W. Zheng and J.-H. Xu, ChemCatChem, 2015, 7, 3838–3841.
64. J. H. Sattler, M. Fuchs, K. Tauber, F. G. Mutti, K. Faber, J. Pfeffer, T. Haas and W. Kroutil, Angew. Chem., Int. Ed., 2012, 51, 9156–9159.
65. S. L. Montgomery, J. Mangas-Sanchez, M. P. Thompson, G. A. Aleku, B. Dominguez and N. J. Turner, Angew. Chem., Int. Ed., 2017, 56, 10491–10494.
66. T. Knaus, F. G. Mutti, L. D. Humphreys, N. J. Turner and N. S. Scrutton, Org. Biomol. Chem., 2015, 13, 223.
67. E. P. J. Jongkind, A. Fossey-Jouenne, O. Mayol, A. Zaparucha, C. Vergne-Vaxelaire and C. E. Paul, ChemCatChem, 2021, 14, e202101576.
68. F.-F. Chen, G.-W. Zheng, L. Liu, H. Li, Q. Chen, F.-L. Li, C.-X. Li and J.-H. Xu, ACS Catal., 2018, 8, 2622–2628.
69. F.-F. Chen, S. C. Cosgrove, W. R. Birmingham, J. Mangas-Sanchez, J. Citoler, M. P. Thompson, G.-W. Zheng, J.-H. Xu and N. J. Turner, ACS Catal., 2019, 9, 11813–11818.
70. X. Mu, T. Wu, Y. Mao, Y. Zhao, Y. Xu and Y. Nie, ChemCatChem, 2021, 13, 5243–5253.
71. R. D. Franklin, C. J. Mount, B. R. Bommarius and A. S. Bommarius, ChemCatChem, 2020, 12, 2436–2439.
72. D.-H. Wang, Q. Chen, S.-N. Yin, X.-W. Ding, Y.-C. Zheng, Z. Zhang, Y.-H. Zhang, F.-F. Chen, J.-H. Xu and G.-W. Zheng, ACS Catal., 2021, 11, 14274–14283.
73. T. Knaus, W. Bohmer and F. G. Mutti, Green Chem., 2017, 19, 453–463.
74. W. Kong, Y. Liu, C. Huang, L. Zhou, J. Gao, N. J. Turner and Y. Jiang, Angew. Chem., Int. Ed., 2022, 61, e202202264.
75. R.-F. Cai, L. Liu, F.-F. Chen, A. Li, J.-H. Xu and G.-W. Zheng, ACS Sustainable Chem. Eng., 2020, 8, 17054–17061.
76. T. Lv, J. Feng, X. Chen, Y. Luo, Q. Wu, D. Zhu and Y. Ma, ACS Catal., 2023, 13, 5053–5061.
77. H. Wang, G. Qu, J.-K. Li, J.-A. Ma, J. Guo, Y. Miao and Z. Sun, Catal. Sci. Technol., 2020, 10, 5945–5952.
78. F. Tong, Z. Qin, H. Wang, Y. Jiang, J. Li, H. Ming, G. Qu, Y. Xiao and Z. Sun, Front. Bioeng. Biotechnol., 2021, 9, 778584.
79. Z. Mei, K. Zhang, G. Qu, J.-K. Li, B. Liu, J.-A. Ma, R. Tu and Z. Sun, ACS Omega, 2020, 5, 13588–13594.
80. H. Ming, B. Yuan, G. Qu and Z. Sun, Catal. Sci. Technol., 2022, 12, 5952–5960.
81. M. P. Thompson, I. Penafiel, S. C. Cosgrove and N. J. Turner, Org. Process Res. Dev., 2019, 23, 9–18.
82. J. Coloma, Y. Guiavarc’h, P.-L. Hagedoorn and U. Hanefeld, Chem. Commun., 2021, 57, 11416–11428.
83. J. Liu, B. Q. W. Pang, J. P. Adams, R. Snajdrova and Z. Li, ChemCatChem, 2017, 9, 425–431.
84. H. Ren, Y. Zhang, J. Su, P. Lin, B. Wang, B. Fang and S. Wang, J. Biotechnol., 2017, 241, 33–41.
85. A. A. Caparco, A. S. Bommarius and J. A. Champion, AIChE J., 2018, 64, 2934–2946.
86. A. A. Caparco, B. R. Bommarius, A. S. Bommarius and J. A. Champion, Biotechnol. Bioeng., 2020, 117, 1979–1989.
87. R. D. Franklin, J. A. Whitley, A. A. Caparco, B. R. Bommarius, J. A. Champion and A. S. Bommarius, Chem. Eng. J., 2021, 407, 127065.
88. H. L. Yu, T. Li, F. F. Chen, X. J. Luo, A. T. Li, C. Yang, G.-W. Zheng and J. H. Xu, Metab. Eng., 2018, 47, 184–189.
89. J. A. Houwman, T. Knaus, M. Costa and F. G. Mutti, Green Chem., 2019, 21, 3846–3857.
90. J. Liu and Z. Li, Biotechnol. Bioeng., 2019, 116, 536–542.
91. S. C. Cosgrove, M. P. Thompson, S. T. Ahmed, F. Parmeggiani and N. J. Turner, Angew. Chem., Int. Ed., 2020, 59, 18156–18160.
92. F.-F. Chen, Y.-H. Zhang, Z.-J. Zhang, L. Liu, J.-P. Wu, J.-H. Xu and G.-W. Zheng, J. Org. Chem., 2019, 84, 14987–14993.
93. K. Mitsukura, M. Suzuki, K. Tada, T. Yoshida and T. Nagasawa, Org. Biomol. Chem., 2010, 8, 4533–4535.
94. A. K. Gilio, T. W. Thorpe, N. Turner and G. Grogan, Chem. Sci., 2022, 13, 4697–4713.
95. M. Lenz, N. Borlinghaus, L. Weinmann and B. M. Nestl, World J. Microbiol. Biotechnol., 2017, 33, 199.
96. J. H. Schrittwieser, S. Velikogne and W. Kroutil, Adv. Synth. Catal., 2015, 357, 1655–1685.
97. S. C. Cosgrove, A. Brzezniak, S. P. France, J. I. Ramsden, J. Mangas-Sanchez, S. L. Montgomery, R. S. Heath and N. J. Turner, Methods Enzymol., ed. N. Scrutton, 2018, 608, 131–149.
98. G. Grogan and N. J. Turner, Chem. – Eur. J., 2016, 22, 1900–1907.
99. L. Ducrot, M. Bennett, G. Grogan and C. Vergne-Vaxelaire, Adv. Synth. Catal., 2020, 363, 328–351.
100. K. Wu, J. Huang and L. Shao, ChemCatChem, 2022, 14, e202200921.
101. M. Rodriguez-Mata, A. Frank, E. Wells, F. Leipold, N. J. Turner, S. Hart, J. P. Turkenburg and G. Grogan, ChemBioChem, 2013, 14, 1372–1379.
102. G. A. Aleku, H. Man, S. P. France, F. Leipold, S. Hussain, L. Toca-Gonzalez, R. Marchington, S. Hart, J. P. Turkenburg, G. Grogan and N. J. Turner, ACS Catal., 2016, 6, 3880–3889.
103. T. Huber, L. Schneider, A. Präg, S. Gerhardt, O. Einsle and M. Müller, ChemCatChem, 2014, 6, 2248–2252.
104. H. Man, E. Wells, S. Hussain, F. Leipold, S. Hart, J. P. Turkenburg, N. J. Turner and G. Grogan, ChemBioChem, 2015, 16, 1052–1059.
105. T. Meyer, N. Zumbraegel, C. Geerds, H. Groeger and H. H. Niemann, Biomolecules, 2020, 10(8), 1130.
106. S. P. France, G. A. Aleku, M. Sharma, J. Mangas-Sanchez, R. M. Howard, J. Steflik, R. Kumar, R. W. Adams, I. Slabu, R. Crook, G. Grogan, T. W. Wallace and N. J. Turner, Angew. Chem., Int. Ed., 2017, 56, 15589–15593.
107. M. Lenz, S. Fademrecht, M. Sharma, J. Pleiss, G. Grogan and B. M. Nestl, Protein Eng., Des. Sel., 2018, 31, 109–120.
108. P. Stockinger, N. Borlinghaus, M. Sharma, B. Aberle, G. Grogan, J. Pleiss and B. M. Nestl, ChemCatChem, 2021, 13, 5210–5215.
109. J. Zhang, D. Liao, R. Chen, F. Zhu, Y. Ma, L. Gao, G. Qu, C. Cui, Z. Sun, X. Lei and S. S. Gao, Angew. Chem., Int. Ed., 2022, 61, e202201908.
110. T. W. Thorpe, J. R. Marshall, V. Harawa, R. E. Ruscoe, A. Cuetos, J. D. Finnigan, A. Angelastro, R. S. Heath, F. Parmeggiani, S. J. Charnock, R. M. Howard, R. Kumar, D. S. B. Daniels, G. Grogan and N. J. Turner, Nature, 2022, 604, 86–91.
111. V. Harawa, T. W. Thorpe, J. R. Marshall, J. J. Sangster, A. K. Gilio, L. Pirvu, R. S. Heath, A. Angelastro, J. D. Finnigan, S. J. Charnock, J. W. Nafie, G. Grogan, R. C. Whitehead and N. J. Turner, J. Am. Chem. Soc., 2022, 144, 21088–21095.
112. A. K. Gilio, T. W. Thorpe, A. Heyam, M. R. Petchey, B. Pogranyi, S. P. France, R. M. Howard, M. J. Karmilowicz, R. Lewis, N. Turner and G. Grogan, ACS Catal., 2023, 13, 1669–1677.
113. T. Nagasawa, T. Yoshida, K. Ishida, H. Yamamoto and N. Kimoto, US Pat., WO2010024445A1, 2009.
114. K. Mitsukura, T. Kuramoto, T. Yoshida, N. Kimoto, H. Yamamoto and T. Nagasawa, Appl. Microbiol. Biotechnol., 2013, 97, 8079–8086.
115. K. Mitsukura, M. Suzuki, S. Shinoda, T. Kuramoto, T. Yoshida and T. Nagasawa, Biosci., Biotechnol., Biochem., 2011, 75, 1778–1782.
116. F. Leipold, S. Hussain, D. Ghislieri and N. J. Turner, ChemCatChem, 2013, 5, 3505–3508.
117. P. N. Scheller, S. Fademrecht, S. Hofelzer, J. Pleiss, F. Leipold, N. J. Turner, B. M. Nestl and B. Hauer, ChemBioChem, 2014, 15, 2201–2204.
118. S. Hussain, F. Leipold, H. Man, E. Wells, S. P. France, K. R. Mulholland, G. Grogan and N. J. Turner, ChemCatChem, 2015, 7, 579–583.
119. D. Wetzl, M. Berrera, N. Sandon, D. Fishlock, M. Ebeling, M. Muller, S. Hanlon, B. Wirz and H. Iding, ChemBioChem, 2015, 16, 1749–1756.
120. D. Wetzl, M. Gand, A. Ross, H. Müller, P. Matzel, S. P. Hanlon, M. Müller, B. Wirz, M. Höhne and H. Iding, ChemCatChem, 2016, 8, 2023–2026.
121. M. Prejano, X. Sheng and F. Himo, ChemistryOpen, 2022, 11, e202100250.
122. N. Zumbraegel, C. Merten, S. M. Huber and H. Groeger, Nat. Commun., 2018, 9, 1949.
123. N. Zumbraegel and H. Groeger, J. Biotechnol., 2019, 291, 35–40.
124. N. Zumbraegel, K. Wagner, N. Weissing and H. Groeger, J. Heterocycl. Chem., 2019, 56, 788–794.
125. N. Zumbraegel, P. Machui, J. Nonnhoff and H. Groeger, J. Org. Chem., 2019, 84, 1440–1447.
126. P. N. Scheller, M. Lenz, S. C. Hammer, B. Hauer and B. M. Nestl, ChemCatChem, 2015, 7, 3239–3242.
127. D. González-Martínez, A. Cuetos, M. Sharma, M. García-Ramos, I. Lavandera, V. Gotor-Fernández and G. Grogan, ChemCatChem, 2020, 12, 2421–2425.
128. P. Matzel, M. Gand and M. Höhne, Green Chem., 2017, 19, 385–389.
129. G.-D. Roiban, M. Kern, Z. Liu, J. Hyslop, P. L. Tey, M. S. Levine, L. S. Jordan, K. K. Brown, T. Hadi, L. A. F. Ihnken and M. J. B. Brown, ChemCatChem, 2017, 9, 4475–4479.
130. N. Borlinghaus, S. Gergel and B. M. Nestl, ACS Catal., 2018, 8, 3727–3732.
131. S. P. France, R. M. Howard, J. Steflik, N. J. Weise, J. Mangas-Sanchez, S. L. Montgomery, R. Crook, R. Kumar and N. J. Turner, ChemCatChem, 2018, 10, 510–514.
132. S. L. Montgomery, A. Pushpanath, R. S. Heath, J. R. Marshall, U. Klemstein, J. L. Galman, D. Woodlock, S. Bisagni, C. J. Taylor, J. Mangas-Sanchez, J. I. Ramsden, B. Dominguez and N. J. Turner, Sci. Adv., 2020, 6, eaay9320.
133. J. R. Marshall, P. Yao, S. L. Montgomery, J. D. Finnigan, T. W. Thorpe, R. B. Palmer, J. Mangas-Sanchez, R. A. M. Duncan, R. S. Heath, K. M. Graham, D. J. Cook, S. J. Charnock and N. J. Turner, Nat. Chem., 2021, 13, 140–148.
134. J. Citoler, V. Harawa, J. R. Marshall, H. Bevinakatti, J. D. Finnigan, S. J. Charnock and N. J. Turner, Angew. Chem., Int. Ed., 2021, 60, 24456–24460.
135. Y. Li, N. Hu, Z. Xu, Y. Cui, J. Feng, P. Yao, Q. Wu, D. Zhu and Y. Ma, Angew. Chem., Int. Ed., 2022, 61, e202116344.
136. Z. Xu, P. Yao, X. Sheng, J. Li, J. Li, S. Yu, J. Feng, Q. Wu and D. Zhu, ACS Catal., 2020, 10, 8780–8787.
137. Y.-H. Zhang, F.-F. Chen, B.-B. Li, X.-Y. Zhou, Q. Chen, J.-H. Xu and G.-W. Zheng, Org. Lett., 2020, 22, 3367–3372.
138. Q. Chen, B.-B. Li, L. Zhang, X.-R. Chen, X.-X. Zhu, F.-F. Chen, M. Shi, C.-C. Chen, Y. Yang, R.-T. Guo, W. Liu, J.-H. Xu and G.-W. Zheng, ACS Catal., 2022, 12, 14795–14803.
139. J. Zhang, X. Li, R. Chen, X. Tan, X. Liu, Y. Ma, F. Zhu, C. An, G. Wei, Y. Yao, L. Yang, P. Zhang, Q. Wu, Z. Sun, B. G. Wang, S. S. Gao and C. Cui, Commun. Chem., 2022, 5, 123.
140. Z. Zhan, Z. Xu, S. Yu, J. Feng, F. Liu, P. Yao, Q. Wu and D. Zhu, Adv. Synth. Catal., 2022, 364, 2380–2386.
141. L. Yang, J. Li, Z. Xu, P. Yao, Q. Wu, D. Zhu and Y. Ma, Org. Lett., 2022, 24, 6531–6536.
142. F. F. Chen, X. F. He, X. X. Zhu, Z. Zhang, X. Y. Shen, Q. Chen, J. H. Xu, N. J. Turner and G.-W. Zheng, J. Am. Chem. Soc., 2023, 145, 4015–4025.
143. E. J. Ma, E. Siirola, C. Moore, A. Kummer, M. Stoeckli, M. Faller, C. Bouquet, F. Eggimann, M. Ligibel, D. Huynh, G. Cutler, L. Siegrist, R. A. Lewis, A.-C. Acker, E. Freund, E. Koch, M. Vogel, H. Schlingensiepen, E. J. Oakeley and R. Snajdrova, ACS Catal., 2021, 11, 12433–12445.
144. Y. Yu, A. Rué Casamajo, W. Finnigan, C. Schnepel, R. Barker, C. Morrill, R. S. Heath, L. De Maria, N. J. Turner and N. S. Scrutton, ACS Catal., 2023, 13, 12310–12321.
145. M. Schober, C. MacDermaid, A. A. Ollis, S. Chang, D. Khan, J. Hosford, J. Latham, L. A. F. Ihnken, M. J. B. Brown, D. Fuerst, M. J. Sanganee and G.-D. Roiban, Nat. Catal., 2019, 2, 909–915.
146. W. Zawodny, S. L. Montgomery, J. R. Marshall, J. D. Finnigan, N. J. Turner and J. Clayden, J. Am. Chem. Soc., 2018, 140, 17872–17877.
147. P. Yao, Z. Xu, S. Yu, Q. Wu and D. Zhu, Adv. Synth. Catal., 2019, 361, 556–561.
148. S. Velikogne, V. Resch, C. Dertnig, J. H. Schrittwieser and W. Kroutil, ChemCatChem, 2018, 10, 3236–3246.
149. N. Alvarenga, S. E. Payer, P. Petermeier, C. Kohlfuerst, A. L. Meleiro Porto, J. H. Schrittwieser and W. Kroutil, ACS Catal., 2020, 10, 1607–1620.
150. R. L. Booth, G. Grogan, K. S. Wilson and A.-K. Duhme-Klair, RSC Chem. Biol., 2020, 1, 369–378.
151. R. Kumar, M. J. Karmilowicz, D. Burke, M. Burns, L. A. Clark, C. G. Connor, E. Cordi, N. M. Do, K. M. Doyle, S. Hoagland, C. A. Lewis, D. Mangan, C. A. Martinez, E. L. McInturff, K. Meldrum, R. Pearson, J. Steflik, A. Rane and J. Weaver, Nat. Catal., 2021, 4, 775–782.
152. L. Yang, J. Zhu, C. Sun, Z. Deng and X. Qu, Chem. Sci., 2020, 11, 364–371.
153. P. Yao, J. R. Marshall, Z. Xu, J. Lim, S. J. Charnock, D. Zhu and N. J. Turner, Angew. Chem., Int. Ed., 2021, 60, 8717–8721.
154. J. Zhu, L. Yang, J. Wu, Z. Deng and X. Qu, ACS Catal., 2022, 12, 9823–9830.
155. J. Nonnhoff, H. G. Stammler and H. Groger, J. Org. Chem., 2022, 87, 11369–11378.
156. M. Li, Y. Cui, Z. Xu, X. Chen, J. Feng, M. Wang, P. Yao, Q. Wu and D. Zhu, Adv. Synth. Catal., 2022, 364, 372–379.
157. M. Cárdenas-Fernández, R. Roddan, E. M. Carter, H. C. Hailes and J. M. Ward, ChemCatChem, 2023, 15, e202201126.
158. Y. Li, X. Yue, Z. Li, Z. Huang and F. Chen, Org. Lett., 2023, 25, 1285–1289.
159. J. J. Sangster, R. E. Ruscoe, S. C. Cosgrove, J. Mangas-Sanchez and N. J. Turner, J. Am. Chem. Soc., 2023, 145(8), 4431–4437.
160. L. J. Hepworth, S. P. France, S. Hussain, P. Both, N. J. Turner and S. L. Flitsch, ACS Catal., 2017, 7, 2920–2925.
161. S. P. France, S. Hussain, A. M. Hill, L. J. Hepworth, R. M. Howard, K. R. Mulholland, S. L. Flitsch and N. J. Turner, ACS Catal., 2016, 6, 3753–3759.
162. M. Tavanti, J. Mangas-Sanchez, S. L. Montgomery, M. P. Thompson and N. J. Turner, Org. Biomol. Chem., 2017, 15, 9790–9793.
163. T. W. Thorpe, S. P. France, S. Hussain, J. R. Marshall, W. Zawodny, J. Mangas-Sanchez, S. L. Montgomery, R. M. Howard, D. S. B. Daniels, R. Kumar, F. Parmeggiani and N. J. Turner, J. Am. Chem. Soc., 2019, 141, 19208–19213.
164. F. Escalettes and N. J. Turner, ChemBioChem, 2008, 9, 857–860.
165. J. B. Rannes, A. Ioannou, S. C. Willies, G. Grogan, C. Behrens, S. L. Flitsch and N. J. Turner, J. Am. Chem. Soc., 2011, 133, 8436–8439.
166. G. J. Ford, N. Kress, A. P. Mattey, L. J. Hepworth, C. R. Baldwin, J. R. Marshall, L. S. Seibt, M. Huang, W. R. Birmingham, N. J. Turner and S. L. Flitsch, Chem. Commun., 2020, 56, 7949–7952.
167. G. A. Aleku, J. Mangas-Sanchez, J. Citoler, S. P. France, S. L. Montgomery, R. S. Heath, M. P. Thompson and N. J. Turner, ChemCatChem, 2018, 10, 515–519.
168. A. Al-Shameri, M.-C. Petrich, K. Junge Puring, U.-P. Apfel, B. M. Nestl and L. Lauterbach, Angew. Chem., Int. Ed., 2020, 59, 10929–10933.
169. P. Petermeier, C. Kohlfuerst, A. Torvisco, R. Fischer, A. Mata, D. Dallinger, C. O. Kappe, J. Schrittwieser and W. Kroutil, Adv. Synth. Catal., 2023, 365, 2057–2278.
170. M. Lenz, J. Meisner, L. Quertinmont, S. Lutz, J. Kastner and B. M. Nestl, ChemBioChem, 2017, 18, 253–256.
171. N. Borlinghaus, L. Weinmann, F. Krimpzer, P. N. Scheller, A. Al-Shameri, L. Lauterbach, A.-S. Coquel, C. Lattemann, B. Hauer and B. M. Nestl, ChemCatChem, 2019, 11, 5738–5742.
172. S. Roth, M. B. Kilgore, T. M. Kutchan and M. Mueller, ChemBioChem, 2018, 19, 1849–1852.
173. S. Roth, P. Stockinger, J. Steff, S. Steimle, V. Sautner, K. Tittmann, J. Pleiss and M. Mueller, ChemBioChem, 2020, 21, 2615–2619.
174. P. Stockinger, S. Roth, M. Mueller and J. Pleiss, ChemBioChem, 2020, 21, 2689–2695.
175. S. Duan, D. W. Widlicka, M. P. Burns, R. Kumar, I. Hotham, J.-N. Desrosiers, P. Bowles, K. N. Jones, L. D. Nicholson, M. T. Buetti-Weekly, L. Han, J. Steflik, E. Hansen, C. M. Hayward, H. Strohmeyer, S. Monfette, S. C. Sutton and C. Morris, Org. Process Res. Dev., 2021, 26, 879–890.
176. L.-E. Meyer, H. Brundiek and J. von Langermann, Biotechnol. Prog., 2020, 36, e3024.
177. A. P. Mattey, G. J. Ford, J. Citoler, C. Baldwin, J. R. Marshall, R. B. Palmer, M. Thompson, N. J. Turner, S. C. Cosgrove and S. L. Flitsch, Angew. Chem., Int. Ed., 2021, 60, 18660–18665.
178. N. Borlinghaus and B. M. Nestl, ChemCatChem, 2018, 10, 183–187.
179. J. Preissler, H. A. Reeve, T. Zhu, J. Nicholson, K. Urata, L. Lauterbach, L. L. Wong, K. A. Vincent and O. Lenz, ChemCatChem, 2020, 12, 4853–4861.
180. L. S. Vidal, J. W. Murray and J. T. Heap, Nat. Commun., 2021, 12, 6859.
181. H. C. Buechsenschuetz, V. Vidimce-Risteski, B. Eggbauer, S. Schmidt, C. K. Winkler, J. H. Schrittwieser, W. Kroutil and R. Kourist, ChemCatChem, 2020, 12, 726–730.
182. F. Schwizer, Y. Okamoto, T. Heinisch, Y. Gu, M. M. Pellizzoni, V. Lebrun, R. Reuter, V. Kohler, J. C. Lewis and T. R. Ward, Chem. Rev., 2018, 118, 142–231.
183. M. Hestericova, T. Heinisch, L. Alonso-Cotchico, J.-D. Marechal, P. Vidossich and T. R. Ward, Angew. Chem., Int. Ed., 2018, 57, 1863–1868.
184. M. Hestericova, T. Heinisch, M. Lenz and T. R. Ward, Dalton Trans., 2018, 47, 10837–10841.
185. T. Yu, A. G. Boob, M. J. Volk, X. Liu, H. Cui and H. Zhao, Nat. Catal., 2023, 6, 137–151.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3cs00391d

---