Rational engineering of Luminiphilus syltensis (R)-selective amine transaminase for the acceptance of bulky substrates

Advances in the Molecular Modification of Microbial ω-Transaminases for Asymmetric Synthesis of Bulky Chiral Amines

ω-Transaminases are biocatalysts capable of asymmetrically synthesizing high-value chiral amines through the reductive amination of carbonyl compounds, and they are ubiquitously distributed across diverse microorganisms. Despite their broad natural occurrence, the industrial utility of naturally occurring ω-transaminases remains constrained by their limited catalytic efficiency toward sterically bulky substrates. Over recent decades, the use of structure-guided molecular modifications, leveraging three-dimensional structures, catalytic mechanisms, and machine learning-driven predictions, has emerged as a transformative strategy to address this limitation. Notably, these advancements have unlocked unprecedented progress in the asymmetric synthesis of bulky chiral amines, which is exemplified by the industrial-scale production of sitagliptin using engineered ω-transaminases. This review systematically explores the structural and mechanistic foundations of ω-transaminase engineering. We first delineate the substrate binding regions of these enzymes, focusing on their defining features such as substrate tunnels and dual pockets. These structural elements serve as critical targets for rational design to enhance substrate promiscuity. Next, we dissect the catalytic and substrate recognition mechanisms of (S)- and (R)-ω-transaminases. Drawing on these insights, we consolidate recent advances in engineering ω-transaminases to highlight their performance in synthesizing bulky chiral amines and aim to guide future research and the industrial implementation of tailored ω-transaminases.

Challenges and good practices on transaminase-catalysed synthesis of optically pure amines

Transaminase-catalysed synthesis of chiral amines has been highlighted as an important strategy to get access to optically pure primary amines with high selectivity. However, their application is still hindered from several factors. The co-factor instability leads to instability of the protein itself, while the thermodynamics typically do not favor the desired amination reaction. Thus, several strategies have been suggested to tackle the thermodynamic issue, while parameters that initially may seem trivial, such as selection of buffer salt, pH and temperature, recently were studied more thoroughly. In this chapter we provide a review on the suggested strategies with specific commentaries on their application, as well as protocol for the synthesis of optically pure amines with the two most commonly used amine donors, namely alanine and isopropylamine.

Structure-guided engineering an (R)-transaminase from Mycobacterium neoaurum for efficient synthesis of chiral N-heterocyclic amines

(R)-selective amine transaminases (R-ATAs) show considerable potential for the asymmetric synthesis of chiral drug intermediates. However, the low catalytic efficiency of natural R-ATAs toward bulky ketone substrates, such as N-heterocyclic compounds, severely limits its industrial application. In this study, five putative (R)-ATAs were mined from NCBI database, among which MnTA showed the highest activity for N-Boc-3-pyrrolidinone (1a) and N-Boc-3-piperidone (2a), and its crystal structure was performed. Furthermore, a structure-guided engineering strategy combined with directed evolution and in silico design was executed. Four key sites for substrate binding were identified based on alanine scanning. Then, a saturated mutation library was constructed, and residues G66 and F127 were found to be the key sites affecting substrate binding. By further combining mutation and iterative saturation mutation, variants with markedly improved activity were obtained. The optimal mutant MnTA-M1 (F127M) and MnTA-M5 (G66L/H67N/F127M/L160I) also displayed significantly enhanced activity toward various cyclic ketones or bulky N-heterocyclic ketone analogs. Finally, the gram-scale synthesis of (R)-3-amino-N-Boc-pyrrolidin (1b) and (R)-3-amino-N-Boc-piperidine (2b) was performed by the best mutants, achieving the space-time yields (STY) of 108 and 214 g/L·d, respectively. This research provides efficient biocatalysts for the synthesis of various chiral N-heterocyclic amines, along with a structural insight into the molecular mechanism for enhanced catalytic performance.

Generalized biomolecular modeling and design with RoseTTAFold All-Atom

Deep-learning methods have revolutionized protein structure prediction and design but are presently limited to protein-only systems. We describe RoseTTAFold All-Atom (RFAA), which combines a residue-based representation of amino acids and DNA bases with an atomic representation of all other groups to model assemblies that contain proteins, nucleic acids, small molecules, metals, and covalent modifications, given their sequences and chemical structures. By fine-tuning on denoising tasks, we developed RFdiffusion All-Atom (RFdiffusionAA), which builds protein structures around small molecules. Starting from random distributions of amino acid residues surrounding target small molecules, we designed and experimentally validated, through crystallography and binding measurements, proteins that bind the cardiac disease therapeutic digoxigenin, the enzymatic cofactor heme, and the light-harvesting molecule bilin.

Identification, characterization and application of M16AT, a new organic solvent-tolerant (R)-enantio-selective type IV amine transaminase from Mycobacterium sp. ACS1612

Biocatalysis has emerged as a powerful alternative to traditional chemical methods, especially for asymmetric synthesis. As biocatalysts usually exhibit excellent chemical, regio- and enantioselectivity, they facilitate and simplify many chemical processes for the production of a broad range of products. Here, a new biocatalyst called, R-selective amine transaminases (R-ATAs), was obtained from Mycobacterium sp. ACS1612 (M16AT) using in-silico prediction combined with a genome and protein database. A two-step simple purification process could yield a high concentration of pure enzyme, suggesting that industrial application would be inexpensive. Additionally, the newly identified and characterized R-ATAs displayed a broad substrate spectrum and strong tolerance to organic solvents. Moreover, the synthetic applicability of M16AT has been demonstrated by the asymmetric synthesis of (R)-fendiline from of (R)-1-phenylethan-1-amine.

Identification of a novel thermostable transaminase and its application in L-phosphinothricin biosynthesis

Transaminase (TA) is a crucial biocatalyst for enantioselective production of the herbicide L-phosphinothricin (L-PPT). The use of enzymatic cascades has been shown to effectively overcome the unfavorable thermodynamic equilibrium of TA-catalyzed transamination reaction, also increasing demand for TA stability. In this work, a novel thermostable transaminase (PtTA) from Pseudomonas thermotolerans was mined and characterized. The PtTA showed a high specific activity (28.63 U/mg) towards 2-oxo-4-[(hydroxy)(methyl)phosphinoyl]butyric acid (PPO), with excellent thermostability and substrate tolerance. Two cascade systems driven by PtTA were developed for L-PPT biosynthesis, including asymmetric synthesis of L-PPT from PPO and deracemization of D, L-PPT. For the asymmetric synthesis of L-PPT from PPO, a three-enzyme cascade was constructed as a recombinant Escherichia coli (E. coli G), by co-expressing PtTA, glutamate dehydrogenase (GluDH) and D-glucose dehydrogenase (GDH). Complete conversion of 400 mM PPO was achieved using only 40 mM amino donor L-glutamate. Furthermore, by coupling D-amino acid aminotransferase (Ym DAAT) from Bacillus sp. YM-1 and PtTA, a two-transaminase cascade was developed for the one-pot deracemization of D, L-PPT. Under the highest reported substrate concentration (800 mM D, L-PPT), a 90.43% L-PPT yield was realized. The superior catalytic performance of the PtTA-driven cascade demonstrated that the thermodynamic limitation was overcome, highlighting its application prospect for L-PPT biosynthesis. KEY POINTS: • A novel thermostable transaminase was mined for L-phosphinothricin biosynthesis. • The asymmetric synthesis of L-phosphinothricin was achieved via a three-enzyme cascade. • Development of a two-transaminase cascade for D, L-phosphinothricin deracemization.

Comparison of Four Immobilization Methods for Different Transaminases

Biocatalytic syntheses often require unfavorable conditions, which can adversely affect enzyme stability. Consequently, improving the stability of biocatalysts is needed, and this is often achieved by immobilization. In this study, we aimed to compare the stability of soluble and immobilized transaminases from different species. A cysteine in a consensus sequence was converted to a single aldehyde by the formylglycine-generating enzyme for directed single-point attachment to amine beads. This immobilization was compared to cross-linked enzyme aggregates (CLEAs) and multipoint attachments to glutaraldehyde-functionalized amine- and epoxy-beads. Subsequently, the reactivity and stability (i.e., thermal, storage, and solvent stability) of all soluble and immobilized transaminases were analyzed and compared under different conditions. The effect of immobilization was highly dependent on the type of enzyme, the immobilization strategy, and the application itself, with no superior immobilization technique identified. Immobilization of HAGA-beads often resulted in the highest activities of up to 62 U/g beads, and amine beads were best for the hexameric transaminase from Luminiphilus syltensis. Furthermore, the immobilization of transaminases enabled its reusability for at least 10 cycles, while maintaining full or high activity. Upscaled kinetic resolutions (partially performed in a SpinChemTM reactor) resulted in a high conversion, maintained enantioselectivity, and high product yields, demonstrating their applicability.

Non-Canonical Amino Acid-Based Engineering of (R)-Amine Transaminase

Non-canonical amino acids (ncAAs) have been utilized as an invaluable tool for modulating the active site of the enzymes, probing the complex enzyme mechanisms, improving catalytic activity, and designing new to nature enzymes. Here, we report site-specific incorporation of p-benzoyl phenylalanine (pBpA) to engineer (R)-amine transaminase previously created from d-amino acid aminotransferase scaffold. Replacement of the single Phe88 residue at the active site with pBpA exhibits a significant 15-fold and 8-fold enhancement in activity for 1-phenylpropan-1-amine and benzaldehyde, respectively. Reshaping of the enzyme's active site afforded an another variant F86A/F88pBpA, with 30% higher thermostability at 55°C without affecting parent enzyme activity. Moreover, various racemic amines were successfully resolved by transaminase variants into (S)-amines with excellent conversions (∼50%) and enantiomeric excess (>99%) using pyruvate as an amino acceptor. Additionally, kinetic resolution of the 1-phenylpropan-1-amine was performed using benzaldehyde as an amino acceptor, which is cheaper than pyruvate. Our results highlight the utility of ncAAs for designing enzymes with enhanced functionality beyond the limit of 20 canonical amino acids.