Biocatalysis@TUDelft

Although significant progress has been made in creating de novo metalloenzymes that hydrolyze activated esters1,2, the energetically demanding cleavage of amide bonds has remained a major challenge for enzyme design: amide bonds are significantly more stable than ester bonds, the amine leaving groups in proteins are not activated, and peptide substrates are flexible making them difficult to bind precisely. Here, we report the de novo design of zinc proteases from minimal catalytic motifs using a fine-tuned version of RoseTTAFold Diffusion 2, called RoseTTAFold Diffusion 2 for Molecular Interfaces3, optimized for both enzyme and protein-protein interaction design. In a single one-shot design round of 135 designs, 36% of the designs had activity and cleaved precisely at the intended site. The most active design accelerated peptide bond hydrolysis more than 108-fold over the uncatalyzed reaction4. These results demonstrated that de novo enzyme design has advanced well beyond model reactions with activated substrates, and open the door to design of proficient metallohydrolases for medicine and bioremediation.

Despite advances in de novo enzyme design, success has been largely limited to low energy barrier model reactions. Amide bonds such as those linking amino acids along the peptide backbone are stable for hundreds of years in neutral aqueous solution because of the high energy barrier to hydrolysis1. Here we describe the use of a new deep learning method, RFD2-MI2, to de novo design enzymes which utilize an activated cysteine nucleophile to hydrolyze the polypeptide backbone in a sequence-dependent manner, achieving rate enhancements over the background reaction (kcat/kuncat) of up to 3 x 107. The generated designs have folds very different from the proteases in nature (TM score < 0.50), and crystal structures are very close to the design models (C RMSDs < 1.2 [A]), highlighting the accuracy of the design methodology. Our approach has broad utility for advancing the design of novel proteases for both biotechnical and medical applications.

Nature Catalysis, Published online: 21 November 2025; doi:10.1038/s41929-025-01443-1

Engineering protein catalysts represents an attractive approach for enantioselective energy-transfer photochemistry. By combining a genetically encoded photosensitizer in the protein catalyst and a judiciously selected triplet quencher to suppress the racemic background reaction in the solution, photobiocatalytic [2+2] cycloaddition offers improved enantiocontrol in a triplet sensitization catalysis.

Noncanonical amino acids (ncAAs) are valuable building blocks for novel therapeutics. Prolines are particularly useful because their cyclic cores can limit the conformational flexibility of a larger molecular framework, allowing such scaffolds to be further tuned to enhance engagement with their biological targets. Herein, we describe a convergent approach to synthesizing noncanonical prolines from readily available starting materials mediated by a tryptophan synthase β-subunit (TrpB)-imine reductase (IRED) cascade. In the first step, the TrpB catalyzes ketone substrate activation and C–C bond formation with an L-serine- or L-threonine (L-Thr)-derived amino acrylate intermediate. In the second step, the IRED reduces the cyclic imine intermediate, in some cases simultaneously setting two stereocenters via a dynamic kinetic resolution. With this one-pot, sequential cascade, we demonstrate the synthesis of mono-, bi-, and tricyclic prolines bearing as many as four chiral centers, including three examples possessing a remote desymmetrized stereocenter and one example possessing an L-Thr-derived, fully functionalized pyrrolidine ring. Furthermore, we show that D-prolines are accessible from L-amino acid starting materials, which has not been observed in tryptophan synthase catalysis previously. This cascade is a scalable, operation-ally simple method to synthesize new prolines, and is poised to expand the development of novel therapeutics featuring structurally complex ncAAs.

The Michaelis constant (K _m) is often reported with misleading confidence when only standard errors are considered. The accuracy confidence interval (ACI) framework is adapted to K _m by recasting velocity–substrate data as a binding isotherm, propagating systematic uncertainties in enzyme and substrate concentrations. A free web tool (https://aci.sci.yorku.ca) implements ACI-Km for more reliable K _m determination.

The Michaelis constant (K _m) is central to enzyme kinetics, guiding variant selection, inhibitor screening, and metabolic modeling. However, K _m obtained by nonlinear regression can be substantially inaccurate even when the reported standard error (SE) appears small. Common software reports SE but provides no accuracy metric. This gap is addressed by extending the accuracy confidence interval (ACI) framework to K _m (ACI-K _m) through a binding-isotherm formulation of the velocity–substrate fit. Given confidence intervals for concentration accuracy, the method quantifies how residual systematic uncertainties in enzyme and substrate concentrations ( E ₀ and S ₀) propagate into the determined K _m values and provides a probabilistic interval expected to enclose the accurate value. The approach requires no additional kinetic experiments and is directly applicable to existing datasets. Concentration-accuracy intervals can be estimated from calibration data, reagent specifications, or quality-control records. ACI-K _m is valid across a wide range of E ₀/K _m conditions, including relatively high E ₀. A free web application (https://aci.sci.yorku.ca) implements ACI-K _m. Tests on synthetic and experimental datasets show that K _m ± SE can severely underestimate uncertainty, whereas ACI provides more reliable accuracy bounds for decision-making, complementing rather than replacing traditional precision metrics by providing quantitative diagnostic bounds for concentration-related uncertainties in K _m determination.

Al-Hilfi et al. present a biocatalytic strategy for synthesizing 5-methyl-5,6-dihydrothymidine (5-MDHT), a sensitive MRI contrast agent. The study demonstrates that recombinant enzyme catalysis offers an efficient, sustainable, and eco-friendly alternative to traditional chemical synthesis for producing clinically relevant imaging probes.

Magnetic resonance imaging (MRI) is a cornerstone of modern clinical diagnostics, often enhanced by contrast agents. Traditionally, these agents are chemically synthesized, which can involve complex, costly, and environmentally unfriendly processes. Here, we report a novel biocatalytic approach for the efficient, safe, and eco-friendly synthesis of 5-methyl-5,6-dihydrothymidine (5-MDHT), a potent chemical exchange saturation transfer (CEST)-MRI probe for imaging in vivo expression of the Herpes Simplex Virus Type-1 Thymidine Kinase (HSV1-TK) reporter gene. We demonstrate that 5-MDHT can be biosynthesized via one- or two-step enzymatic reactions using human purine nucleoside phosphorylase (hPNPase) and the SgvM^VAV SAM-dependent methyltransferase. hPNPase catalyzed the base-exchange reaction with catalytic efficiencies (k _cat /K _M) between 138 and 316 s⁻¹·M⁻¹, while SgvM^VAV methylation of 5,6-dihydrothymidine yielded 5-MDHT with a catalytic efficiency of 26  s⁻¹·M⁻¹. Molecular dynamics simulations supported the enzymatic binding and selectivity observed experimentally. The resulting 5-MDHT was validated using CEST-MRI, showing a distinct exchangeable imino proton signal at 5.3 ppm. These findings highlight the chemo- and regioselectivity of the biocatalysts and establish biocatalysis as a viable platform for producing clinically relevant MRI contrast agents.

Polyketides constitute a large class of natural products with important biological activities and applications such as antibiotics, antitumor agents, pesticides, and pigments. Their biosynthesis is catalyzed by polyketide synthases (PKSs) which are multi-domain enzymes evolutionarily related to fatty acid synthases (FASs). Despite their close homology in structure and the chemistry they perform, FASs and PKSs differ fundamentally in their catalytic programming: FASs run fully reducing elongation reactions to yield saturated fatty acids, while iterative PKSs execute reductions just in selected cycles, generating complex oxidized compounds. In this study, we aimed at engineering the metazoan FAS in its KR domain to switch from fully reducing to a non-reducing mode during chain elongation. Guided by recent insights into KR programming, we incorporated a helix into metazoan FAS, which is found in KRs from iterative PKSs and type II FASs with chain length programming. These FAS variants initially catalyze complete fatty acid cycles but lose the ability of {beta}-keto reduction in later elongation rounds, producing intermediates that spontaneously cyclize to pyrone products. Finally, our study provides valuable insight into the mechanism of KR catalysis identifying another amino acid next to the active tyrosine which is capable for intermediate protonation.

Succinimide (SNN), an intermediate formed during asparaginyl deamidation or aspartyl dehydration in proteins, is generally hydrolysis-prone, leading to isomerization to L/D /{beta}-aspartyl residue, with the latter being considered deleterious to protein structure and function. An unusually stable SNN-mediated conformational rigidity through restriction of the backbone dihedral angle, {psi}, enhances the thermostability of glutamine amidotransferase (GATase) from Methanocaldococcus jannaschii (Mj). Although several structural features involved in maintaining a stable SNN and imparting SNN-mediated thermostability have been identified in MjGATase, the residues in the protein that catalyse the rapid and complete conversion of Asn109 to SNN remain unknown. Here, we investigated several site-directed mutants of MjGATase for their ability to retain Asn109 side chain in the unmodified form. Mass spectrometric analysis of 10 single mutants enabled the identification of residues that impacted the proportion of SNN and Asn population in the protein sample. This led to the generation of two double mutants that retained intact Asn109 side chain as observed in the mass spectra and crystal structures. These mutants with intact Asn residue at position 109, displayed lower thermal stability than the protein with the SNN modification. Further understanding of the deprotonation mechanism was addressed using QM/MM MD metadynamics simulations. HighlightsO_LIStable succinimide (SNN) arising from deamidation of Asn109 residue imparts hyperthermostability to MjGATase. C_LIO_LIExamination of the structure of MjGATase suggests neighbouring residues playing possible roles in deamidation and cyclization. C_LIO_LIExamination by LC-MS of single site-directed mutants of residues contacting SNN revealed varied levels of intact Asn109 enabling generation of double mutants with complete absence of deamidation. C_LIO_LIPresence of intact Asn109 confirmed by X-ray crystallography highlights the role of Y158, D110, and K151 in mediating SNN formation. C_LIO_LIQM/MM MD metadynamics simulations support experimental findings. C_LI

Cystobactamids are non-ribosomal peptide natural products that function as DNA gyrase inhibitors, exhibiting significant antibacterial activity. They are isolated from Cystobacter sp. Cbv34 and contain various alkoxy groups on para-aminobenzoic acid moieties, which are believed to play a crucial role in antibacterial functions. The alkoxy groups are generated by iterative methylations on a methoxy group by the cobalamin (Cbl)-dependent radical S-adenosylmethionine (SAM) enzyme CysS. CysS catalyzes up to three methylations to give ethoxy, isopropoxy, sec-butoxy, and tert-butoxy groups. For each methylation, CysS uses a ping-pong mechanism in which two molecules of SAM are consumed. One SAM is used to methylate cob(I)alamin, while another generates a 5'-deoxyadenosyl 5'-radical to initiate substrate methylation. However, little is known about how the enzyme promotes both Cbl methylation and iterative substrate methylation, which occur by polar SN2 and radical processes, respectively. Here, we report three X-ray crystal structures of a homolog of CysS from Corallococcus sp. CA054B. Two were determined in the presence of methoxy- and ethoxy-containing substrates, showing how CysS accommodates substrates and products during iterative methylation. The third structure, determined in the absence of a substrate, exhibits structural changes that reorient the SAMs conformation to allow for the methylation of cob(I)alamin.

Significant effort has been directed toward characterization of nonheme iron enzymes owing to their breadth of unique reactivity. Through genome mining, we identified a conserved biosynthetic gene cluster within Pseudomonadota encoding one such family, the multinuclear nonheme iron-dependent oxidative enzymes (MNIO, formerly DUF692). Using a representative gene cluster from Fontimonas thermophila, we heterologously produced the post-translationally modified peptide fontiphorin, and detailed spectral analysis revealed MNIO-catalyzed installation of seven 5-thiooxazole (5TO) moieties. During our work, additional MNIO products were reported with conflicting structural assignments, so we investigated the related biosynthetic gene clusters from Haemophilus influenzae and Neisseria gonorrhoeae. Using alkylation-assisted HMBC correlations, we demonstrated that these products also contain 5TO resulting in a revision of the structure of oxazolin. We further provide evidence supporting a role for 5TO-containing peptides in copper detoxification and recommended this emerging class of Cu-associated peptidic thiooxazole metallophores be referred to as captophorins. To further explore the captophorins, we reconstituted fontiphorin biosynthesis in vitro and investigated its enzymatic requirements. Using cell-free production of single-site, double-site, and naturally occurring variants, we examined enzyme-substrate interactions to determine key sites governing catalysis by 5TO-forming MNIOs. Through our detailed spectroscopic approach for 5TO assignment and investigation of enzyme-substrate interactions, our work unifies tens of thousands of MNIOs in the biosynthesis of captophorins.

Isoxazolines are an important class of heterocycles with a broad range of biological activities. One of the most prevalent synthetic strategies to access isoxazolines is through the oxidative [3+2]-cycloaddition between nitrile oxides and alkenes initiated by halogenation of a starting aldoxime, but current methods rely on toxic oxidants that produce significant quantities of waste and are largely bioincompatible. We have recently discovered that the vanadium-dependent haloperoxidase (VHPO) class of enzymes are an efficient catalyst system for the in situ generation of nitrile oxides. Herein, we have developed a chemoenzymatic protocol for the conversion of aldehydes to nitrile oxides that features a sequential condensation of hydroxylamine with a starting aldehyde followed by oxidative [3+2]-cycloaddition with alkenes enabled by VHPO-catalyzed halogenation of aldoximes on a broad range of structurally diverse substrates in high yield and excellent chemoselectivity. The protocol is conducted on gram-scale, demonstrated using whole cells and cell lysate, and extended to isoxazole synthesis. Finally, this process is coupled to lipase-mediated conversion of amines to oximes to generate isoxazolines.

The semipinacol rearrangement comprises a valuable class of synthetic organic transformations that provides access to useful molecular scaffolds via C–C bond cleavage and formation. Biocatalytic semipinacol rearrangements are rare, however, and the only naturally occurring semipinacolase catalyzes a reaction that proceeds via chirality transfer on a complex natural product substrate. Herein, we report that flavin-dependent halogenases (FDHs) can catalyze enantioselective halogenative semipinacol rearrangement of pro-chiral allylic alcohols. This biocatalytic platform exhibits a broad substrate scope, affording chiral ketones bearing all-carbon quaternary stereocenters with high enantioselectivity. The reaction system displays a kcat value of 16.86 ± 0.97 min-1, representing the highest reported to date for FDH catalysis, and the T52G mutation was found to play a key role in reshaping the enzyme active site to enable this non-native transformation. This first example of asymmetric C−C bond construction using an FDH highlights the catalytic flexibility of these enzymes and provides access to a diverse range of enantioenriched carbocycles and heterocycles.

Hyalucidin A, the most complex polycyclic bacterial RiPP identified to date, features a unique benzofuranoindoline motif and was discovered by targeting dual-P450 gene clusters. The synergistic P450 catalysis that drives its formation was elucidated, and peptide engineering led to the creation of 22 derivatives, emonstrating a scalable platform for macrocyclic RiPP diversification.

Abstract

Despite the emergence of P450-modified ribosomally synthesized and post-translationally modified peptides (RiPPs) as a distinct and rapidly expanding class of natural products, their structural diversity has been limited to scaffolds catalyzed solely by single P450 enzymes. To access greater structural complexity, we targeted unexplored biosynthetic gene clusters encoding two P450s. This effort led to the discovery of hyalucidin A, which features the most intricate polycyclic cross-linked architecture among reported P450-RiPPs and represents the first bacterial RiPP containing a benzofuranoindoline motif. Through stepwise enzymatic analysis, combinatorial biosynthesis, and precursor peptide engineering, we deciphered the cooperative function and substrate tolerance of both P450s. Our work expands the biosynthetic logic of P450-driven cyclization and provides a platform for engineering complex macrocyclic peptides.

Isoxazolines are an important class of heterocycles with a broad range of biological activities. One of the most prevalent synthetic strategies to access isoxazolines is through the oxidative [3+2]-cycloaddition between nitrile oxides and alkenes initiated by halogenation of a starting aldoxime, but current methods rely on toxic oxidants that produce significant quantities of waste and are largely bioincompatible. We have recently discovered that the vanadium-dependent haloperoxidase (VHPO) class of enzymes are an efficient catalyst system for the in situ generation of nitrile oxides. Herein, we have developed a chemoenzymatic protocol for the conversion of aldehydes to nitrile oxides that features a sequential condensation of hydroxylamine with a starting aldehyde followed by oxidative [3+2]-cycloaddition with alkenes enabled by VHPO-catalyzed halogenation of aldoximes on a broad range of structurally diverse substrates in high yield and excellent chemoselectivity. The protocol is conducted on gram-scale, demonstrated using whole cells and cell lysate, and extended to isoxazole synthesis. Finally, this process is coupled to lipase-mediated conversion of amines to oximes to generate isoxazolines.

Genetic code expansion with a unique hyper-fluorinated phosphotyrosine analog. In this work, we successfully incorporated the unnatural amino acid pentafluorophosphato-difluoromethyl-phenylalanine, carrying seven fluorine atoms and a permanent negative charge into three different proteins via the use of mutated orthogonal aminoacyl-tRNA synthetases. Biological testing revealed the great potential of this approach for furnishing functional phosphoprotein mimetics.

Abstract

Protein phosphorylation is one of the most important posttranslational modifications altering the structure, stability, and activity of more than 13 000 human proteins. In this work, the phosphotyrosine mimetic pentafluorophosphato-difluoromethyl-phenylalanine (PF₅CF₂Phe) was genetically encoded and incorporated into three different proteins. Screening two libraries of orthogonal aminoacyl-tRNA synthetases identified enzymes enabling the efficient and specific incorporation of PF₅CF₂Phe into red fluorescent protein (RFP) via amber stop codon suppression. Two model proteins, human ubiquitin (Ubq) and the B1 immunoglobulin-binding domain of streptococcal protein G (GB1), were prepared with PF₅CF₂Phe mutations and investigated for potential interaction partners. While native GB1 showed no binding to protein tyrosine phosphatases (PTP), PF₅-GB1, with PF₅CF₂Phe at position 17, was a strong inhibitor of the phosphatases PTP1B and SHP2. PF₅-Ubq was produced and converted into the first example of a protein carrying the most prominent phosphotyrosine mimetic, phosphono-difluoromethyl phenylalanine (PO₃CF₂Phe). With increasing need in the biosciences to delineate the functions of complex phosphorylation patterns, genetic encoding of PF₅CF₂Phe yielding phosphoprotein mimetics opens unique opportunities for precise functional studies where site-specific and homogeneous protein modifications are required.

Nature Chemical Biology, Published online: 14 November 2025; doi:10.1038/s41589-025-02070-4

Nitrogen fixation by nitrogenase requires a metallocofactor built by a dedicated multiprotein machinery. Here, the authors captured structural snapshots of a precursor entering a key maturase, revealing a dynamic process involving extensive structural rearrangements and partial protein unfolding.