Biocatalysis@TUDelft

Communications Chemistry, Published online: 12 December 2025; doi:10.1038/s42004-025-01783-w

The sustainable and scalable synthesis of chiral amines remains a significant challenge in industrial chemistry, particularly due to the limitations of current biocatalytic processes in achieving high productivity and selectivity at scale. This study reports a dynamic kinetic resolution strategy that integrates flash thermal racemization with enzymatic resolution in a continuous flow system, achieving unprecedented productivity, excellent enantioselectivity, and industry-relevant green chemistry metrics at scales up to 100 grams.

Communications Chemistry, Published online: 16 December 2025; doi:10.1038/s42004-025-01781-y

Fluorochemicals pose significant environmental challenges due to their persistence and strong C–F bonds. Here, the authors engineer transaminases to catalyze a hydrodefluorination reaction of perfluorinated methylene groups, achieving efficient biodegradation of difluorinated ketones.

This study established an efficient system for the debenzylation of N-benzyl groups in the tert-Butyl (S-5-benzyl-5-azaspiro [2.4] heptan-7-yl) carbamate, catalyzed by laccase. Specifically, within a biphasic system of butyl acetate, the laccase (E.C.1.10.3.2)-TEMPO medium system effectively catalyzed the debenzylation reaction of N-benzyl groups in the tert-Butyl (S-5-benzyl-5-azaspiro [2.4] heptan-7-yl) carbamate, achieving a conversion rate of 42.3% and a yield of 40.6%.

Abstract

This study we present an environmentally benign biphasic laccase-TEMPO catalytic system (n-butyl acetate/water) for oxidative N-debenzylation as a safe and sustainable alternative to conventional Pd-catalyzed hydrogenolysis. The optimized system achieves 42.3% conversion of tert-butyl (S-5-benzyl-5-azaspiro[2.4]heptan-7-yl)carbamate, representing a 42-fold enhancement over monophasic laccase catalysis (≤1%). Strategic biphasic engineering overcomes substrate solubility constraints while pH optimization (5.0) and O₂-coupled TEMPO⁺/TEMPO redox cycling synergistically enhance catalytic efficiency. Mechanistic investigations elucidate a nonradical ionic pathway with substrate selectivity governed by electronic and steric factors. The compartmentalized system uniquely decouples enzymatic oxidation (aqueous phase) from TEMPO⁺ regeneration (organic phase), resolving intrinsic stability limitations of homogeneous catalysis. The oxidation proceeds efficiently under mild, metal-free co-oxidant conditions, offering a fundamentally greener approach to pharmaceutical amine synthesis by eliminating hazardous hydrogen gas and precious metal catalysts.

The power of catalytic promiscuity: The enantioselective conjugate hydrocyanation of enals remains a long-standing challenge for biocatalysis. Here, we report the redesign of 2-deoxy-D-ribose-5-phosphate aldolase for the asymmetric conjugate hydrocyanation of aromatic enals, expanding the reaction scope of iminium-based enzyme catalysis to include an additional new-to-nature reaction.

ABSTRACT

The enantioselective conjugate hydrocyanation of α,β-unsaturated aldehydes remains a long-standing challenge in synthetic chemistry. Here, we report the redesign of 2-deoxy-D-ribose-5-phosphate aldolase (DERA) into an efficient biocatalyst capable of promoting the asymmetric conjugate addition of hydrogen cyanide (generated in situ from trimethylsilyl cyanide) to aromatic enals via an iminium activation pathway. The evolved variant DERA-CN enables the efficient formation of various C4-nitriles with high conversions (up to 99%) and good enantioselectivity (up to 98% e.e.). Control experiments revealed a stepwise process involving enzyme-catalyzed conjugate hydrocyanation followed by spontaneous 1,2-addition of cyanide. Substrates with various electron-donating and electron-withdrawing groups are tolerated, providing access to various enantioenriched nitriles. This work expands the scope of DERA-promoted iminium catalysis and provides a rare enzymatic platform for asymmetric conjugate hydrocyanation under mild aqueous conditions.

The strobilurin iterative PKS assembles a remarkable EZE triene by varying its stereoselectivity during synthesis. Stereoselective 2R-methylation of the triketide intermediate by the C-terminal C-methyltransferase domain induces the KR and DH domains to invert their native stereoselectivities and produce a Z-configured intermediate.

Abstract

Type I Iterative polyketide synthases (PKS) use a limited set of catalytic extension and β-processing domains to create complex polyketides. A remarkable case is that of the strobilurin PKS where a single dehydratase (DH) domain creates an EZE triene over three dehydration cycles. Here we dissect the strobilurin PKS and assay catalytic domains individually and in combination with stereo-defined synthetic substrates in vitro, to reveal the complex and varying selectivities of methylation, ketoreduction and dehydration that lead to this remarkable result. At the diketide stage all stereoselectivities are consistent with those known for other related systems, giving an E product. But at the triketide stage, 2R-methylation is selectively achieved, that is followed by rapid keto-reduction to give an unusual 3-L-alcohol. In-turn, this is eliminated to give the unusual Z-alkene. These selectivities are flexible and change in response to the structure of the substrate at every stage. This uncovers the complete and intricate regio- and stereo-selectivities of a highly reducing iterative Type I PKS for the first time, and highlights the important differences to the well-studied cis-AT modular PKS β-processing enzymes that appear to have inflexible selectivities.

Enzyme catalysis: The di-nickel enzyme GdmH catalyzes guanidinium hydrolysis through a mechanism involving nucleophilic attack, proton transfer, and C-N bond cleavage. The positive charge of guanidinium is essential for initiating the reaction. In contrast, GdmH demonstrates significantly reduced efficiency in catalyzing urea hydrolysis.

Abstract

The nitrogen-rich compound guanidine is widely distributed in nature, but its utilization is hindered by strong resonance stabilization. GdmH, a binuclear nickel enzyme from Synechocystis sp. PCC 6803, is capable of directly converting guanidinium cation into urea and ammonium. In this study, we employed density functional calculations to investigate the reaction mechanism of GdmH using a chemical model derived from the enzyme's X-ray crystal structure. The calculations revealed that the GdmH-catalyzed guanidinium hydrolysis proceeds through a nucleophilic attack by the di-nickel bridging hydroxide on the guanidinium carbon forming a tetrahedral intermediate, two proton transfer steps from the hydroxyl to an amino facilitated by Asp203, C─N bond cleavage yielding urea and ammonia, and regeneration of the bridging hydroxide accompanied by ammonium release. The rate-limiting step is the first proton transfer from hydroxyl to Asp203, with an energy barrier of 11.8 kcal mol⁻¹. Comparative analyses demonstrated that neutral guanidine cannot be hydrolyzed by GdmH due to the absence of a positive charge, which is essential for effective catalysis. Further investigations showed that GdmH is inefficient in catalyzing urea hydrolysis. These findings enhance our understanding of the catalytic specificity of GdmH and the role of nickel cofactors in biological enzymatic processes.

Use of a H₂-driven flavin recycling system based on the O₂-tolerant [NiFe] hydrogenase 1 from E. coli supports efficient halogenation catalyzed by the flavin-dependent halogenase, PyrH. Operation under atmospheric H₂ with controlled delivery of safe, low-level O₂ enables intensification of biocatalytic halogenation.

Abstract

We report a simplified, H₂-driven method for operating biocatalytic halogenation by a flavin-dependent halogenase, PyrH, which enables intensification of biocatalytic conversion of L-tryptophan to its 5-halo product. Flavin-dependent halogenases are gaining traction in biotechnology as their substrate scope is expanded by enzyme discovery and engineering, but their application remains impeded by a particularly complex electron transfer chain and the fact that full conversion is generally only achieved at sub-millimolar substrate concentrations. Here, we apply nickel-iron hydrogenase and ambient pressure H₂ in place of the NAD(P)⁺, glucose, glucose dehydrogenase and reductase which are normally used to supply reduced FAD to halogenases. Together with controlled delivery of O₂, which is needed for generating a hypohalous acid intermediate, we achieve full conversion of 5.5 mM tryptophan, with a PyrH total turnover number of 275, and PyrH turnover frequency of 0.76 min⁻¹ over a 6 h reaction, comparable with rates sustained only for short reaction times using the conventional glucose-driven system. This should help to facilitate application of flavin-dependent halogenases in fine chemical synthesis.

A class III phosphoribosyl pyrophosphate synthetase (PfPRS) from the hyper thermophilic archaeon Pyrolobus fumarii 1A was identified and characterized. PfPRS exhibited exceptional thermal and pH stability, retaining high activity even at 100 °C. Its robust catalytic features, along with exploratory co-immobilization for nucleotide regeneration, highlight its promise for biocatalytic applications.

Phosphoribosyl pyrophosphate (PRPP) functions as a central metabolic intermediate, supplying ribose-5-phosphate moieties for the biosynthesis of nucleotides, certain amino acids, and a range of essential cofactors. In this study, a thermostable phosphoribosyl pyrophosphate synthetase (PfPRS) was identified from the hyper thermophilic archaeon Pyrolobus fumarii 1A, a hyper thermophilic archaeon that grows optimally at 90–113 °C. The prs gene was heterologously expressed in Escherichia coli, and the recombinant enzyme was purified and characterized. Peak catalytic activity of PfPRS was observed at approximately pH 7.5 and 55 °C and retained over 85% of its activity after 2 h of incubation across pH 4.0–10.5. PfPRS exhibited high thermal stability. The enzyme exhibited half-lives of 12 h at 90 °C, 5 h at 95 °C, and 3 h at 100 °C. Among the nucleotides tested as diphosphate donors, PfPRS showed a strong preference for ATP, whereas ADP served as an effective inhibitor. Kinetic analysis revealed K _m values of 35 µM for R5P and 46 µM for ATP, with turnover rates (k _cat) of 71 s⁻¹ and 56 s⁻¹. PfPRS was co-immobilized with polyphosphate kinase 2 (DrPPK2) from Deinococcus radiodurans using a cross-linked enzyme aggregate (CLEA) system to enable ATP regeneration and to explore the feasibility of using PfPRS for PRPP biosynthesis.

Applied and Environmental Microbiology, Volume 92, Issue 1, January 2026.

Computational models of macromolecules have many applications in biochemistry, but physical inaccuracies limit their utility. One class of models uses energy functions rooted in classical mechanics. The standard datasets used to train these models are limited in diversity, pointing to a need for new training data. Here, we sought to explore a new paradigm for training an energy function, where the Rosetta energy function was used to design de novo proteins. Experimental results on these designs were then used to identify failure modes of design, which were subsequently used as a "guiding principle" to retrain the energy function. Specifically, we examined a diverse set of de novo protein designs experimentally tested for their ability to stably fold, identifying unstable designs that were predicted to be stable by the Rosetta energy function. Using deep mutational scanning, we identified single amino-acid mutations that rescued the stability of these designs, providing insight into common failure modes of the energy function. We identified one key failure mode, involving steric clashing in protein cores. We identified similar overpacking when using Rosetta to refine high-resolution protein crystal structures, quantified the degree of overpacking, and refit a small set of energy-function parameters to better recapitulate native-like packing. Following fitting, we largely eliminated the failure mode in the refinement task, while retaining performance on other benchmarks, resulting in an updated version of the Rosetta energy function. This work shows how learning from protein designs can guide energy-function development. Author summaryComputational models of macromolecules have many applications, such as predicting structures, predicting mutational effects, or designing new proteins. One type of model uses an energy function to explicitly model the physical forces at play. These models are trained using experimental data. However, available training data are limited in number and diversity, prompting a need for new sources of training data. In this paper, we explore a new paradigm for training an energy function, which involves using the energy function to design de novo proteins and then learning from which designs succeed and which fail when experimentally tested in the lab. We used experimental data to learn common failure modes in design, identified examples of a failure mode in a high-quality benchmark, and then used this benchmark to retrain the energy function. Using this strategy, we identified and largely resolved a bias in the Rosetta energy function that involved energetically unfavorable steric clashes in protein cores. Overall, this work helps establish a framework for how learning from design can be used to guide the development of macromolecular models.

Tryptophan serves as a versatile biosynthetic precursor across living organisms. While heme-binding proteins (HBPs) mediate key reactions in tryptophan transformation, the full diversity of HBPs remains largely unexplored. Here, we developed the novel Cofactor-Integrative Structural Inspector (CISSspector) to systematically identify HBPs in the extensive extant microbial genomic sequence database, which revealed several uncharacterized HBP families. We experimentally characterized one of the most prominent families, the cytochrome P422 (formerly DUF6875) family, distributed throughout the prokaryotes and eukaryotes. Strikingly, we discovered that this enzyme family orchestrates four chemically distinct and biochemically unprecedented transformations, with regioselectivity, including N1-, C6-, and C7-hydroxylations and intramolecular C-S bond formations. Notably, the discovery of enzymes capable of Trp N1- and C7-hydroxylation addresses a long-standing gap in the natural enzyme arsenal. Structural analysis of the representative cytochrome P422 enzyme Mc170 revealed a structurally unique HBP fold in which conserved residues form a substrate "clamp" that positions the tryptophan indole ring for selective modification. Our work unveils a hidden enzymatic repertoire of HBPs, expands the known landscape of tryptophan metabolism, and establishes an artificial intelligence-augmented framework for discovering cryptic enzymes with broad implications for synthetic biology and natural product discovery.

We show that Limosilactobacillus fermentum NKN51 isolated from fermented Himalayan yak milk converts cholesterol to coprostanol through a cholesterol 5β reductase (5βChR). The structure and mutation analysis of the enzyme confirms the residues involved in binding to NADPH and cholesterol. Phylogenetic analysis revealed that 5βChR classified as a new class of microbial short-chain dehydrogenases. Diabetic cohort metagenomic study highlights 5βChR abundance in healthy participants and its importance for human physiology.

Cholesterol serves as a fundamental molecule in various structural and biochemical pathways; however, high cholesterol levels are linked to cardiovascular diseases. Some selected strains of Lactobacilli are known for modulating cholesterol levels. However, the molecular mechanism underlying cholesterol transformation by lactobacilli has remained elusive. This study describes the discovery and function of a microbial 3β-OH-Δ^5-6-cholesterol-5β-reductase (5βChR) from Limosilactobacillus fermentum NKN51, which directly converts cholesterol to coprostanol, thereby unraveling this longstanding mystery. Protein engineering of the reductase enzyme identified the cholesterol and NADPH interacting amino acid residues, detailing the catalytic mechanism of 5βChR. Phylogenetic analyses highlight the prevalence of 5βChRs in gut commensal lactobacilli, which share a common evolutionary origin with plant 5β reductases. Meta-analysis of microbiomes from healthy individuals underscores the importance of 5βChR homologs, while a cohort study demonstrates an inverse association between 5βChR abundance and diabetes. The discovery of the 5βChR enzyme and its molecular mechanism in cholesterol metabolism paves the way for a better understanding of the gut-associated microbiome and the design of practical applications to ameliorate dyslipidemia.

Better enzymes are needed to develop sustainable methods to recycle plastics with C-X heterobonds such as polyurethane (PUR) and nylon, for which no industrial-scale solutions exist. Current methods rely largely on sequence mining based on a small number of known enzymes. Here we expand the pool of PURases and nylonases by bioprospecting legacy plastic waste with fluorophore plastic mimics combined with FACS. We identify 29 plastic-degrading bacteria, from which 12 enzymes are identified by mass spectrometry and homology searches. Compared to existing enzymes, these enzymes are superior in thermostability and the ability to hydrolyse different high-molecular weight PUR oligomers and nylon textiles. To our knowledge, this is the first reported example of enzymes capable of hydrolysing longer chains of PUR and nylon. This study significantly increases the number of known PURases and nylonases and provides starting points for optimization campaigns through protein engineering and for in silico discovery.

The conversion of CO₂ into value-added chemicals using sunlight is a major goal in sustainable chemistry. Here, we report the design and chemogenetic optimization of a Re(I)-based artificial metalloenzyme (ArM) for the photocatalytic reduction of CO₂ to CO in aqueous solution. By incorporating a biotinylated Re(I)-phenanthroline catalyst into a tetrameric streptavidin scaffold, we achieved a significant enhancement in catalytic turnover compared to free catalyst. Mutational screening at viable positions Ser112 and Lys121 revealed a range of catalytic activities, highlighting the tunability of the system. Through a combination of molecular dynamics simulations and experimental characterization, we demonstrate that the catalytic turnover number correlates strongly with the accessibility of CO₂ to the catalyst's active site, a factor not considered in free solution catalyst design, and further computation reveals the variable activation of a bound CO2 among mutants. This work establishes a clear design principle for enhancing Re(I)-based photocatalysis by leveraging the second coordination sphere of a protein scaffold to confer aqueous stability and control substrate access.