The road to fully programmable protein catalysis

Light-Driven Deracemization by a Designed Photoenzyme

The creation of enzymes with abiological abilities offers exciting opportunities to access new-to-nature biocatalysis beyond that found in nature. Here, we repurpose a novel protein scaffold, CTB10, as an artificial photoenzyme through genetic code expansion. It enables catalytic deracemization of cyclopropane, a process that remains inaccessible to traditional biocatalysis due to its thermodynamically unfavorable nature. Following structural optimization through directed evolution, a broad substrate scope with high enantioselectivities is achieved. Furthermore, the crystal structure of the CTB10-based photoenzyme-substrate complex well demonstrates how the catalytic chiral cavity is sculpted to promote efficient and selective light-enabled deracemization. Therefore, this study unlocks the potential for achieving challenging deracemization through biocatalysis.

Computational design of serine hydrolases

The design of enzymes with complex active sites that mediate multistep reactions remains an outstanding challenge. With serine hydrolases as a model system, we combined the generative capabilities of RFdiffusion with an ensemble generation method for assessing active site preorganization at each step in the reaction to design enzymes starting from minimal active site descriptions. Experimental characterization revealed catalytic efficiencies (kcat/Km) up to 2.2 × 105 M-1 s-1 and crystal structures that closely match the design models (Cα root mean square deviations <1 angstrom). Selection for structural compatibility across the reaction coordinate enabled identification of new catalysts remove with five different folds distinct from those of natural serine hydrolases. Our de novo approach provides insight into the geometric basis of catalysis and a roadmap for designing enzymes that catalyze multistep transformations.

Methods for Theoretical Treatment of Local Fields in Proteins and Enzymes

Electric fields generated by protein scaffolds are crucial in enzymatic catalysis. This review surveys theoretical approaches for detecting, analyzing, and comparing electric fields, electrostatic potentials, and their effects on the charge density within enzyme active sites. Pioneering methods like the empirical valence bond approach rely on evaluating ionic and covalent resonance forms influenced by the field. Strategies employing polarizable force fields also facilitate field detection. The vibrational Stark effect connects computational simulations to experimental Stark spectroscopy, enabling direct comparisons. We highlight how protein dynamics induce fluctuations in local fields, influencing enzyme activity. Recent techniques assess electric fields throughout the active site volume rather than only at specific bonds, and machine learning helps relate these global fields to reactivity. Quantum theory of atoms in molecules captures the entire electron density landscape, providing a chemically intuitive perspective on field-driven catalysis. Overall, these methodologies show protein-generated fields are highly dynamic and heterogeneous, and understanding both aspects is critical for elucidating enzyme mechanisms. This holistic view empowers rational enzyme engineering by tuning electric fields, promising new avenues in drug design, biocatalysis, and industrial applications. Future directions include incorporating electric fields as explicit design targets to enhance catalytic performance and biochemical functionalities.

Enzymatic Synthesis of Saturated Bioisosteres of Ortho-Substituted Benzenes by Artificial Photoenzyme

Saturated bioisosteres of ortho-substituted benzenes are of significant interest due to their enhanced pharmacokinetic properties, such as improved metabolic stability and reduced toxicity, making them valuable in drug design and development. However, efficient synthesis of them remains a challenge in organic chemistry. Herein, we report the biocatalytic synthesis of saturated bioisosteres of ortho-substituted benzenes using engineered artificial photoenzymes. The artificial photoenzyme, incorporating genetically encoded unnatural amino acids with benzophenone photosensitizer residue, facilitate the formation of chiral saturated bioisosteres with moderate enantiomeric excess via the energy transfer process. Our results demonstrate the versatility of artificial photoenzymes in mediating new-to-nature reactions that are difficult to achieve with conventional chemical or enzymatic methods.

Toward deep learning sequence-structure co-generation for protein design

Deep generative models that learn from the distribution of natural protein sequences and structures may enable the design of new proteins with valuable functions. While the majority of today's models focus on generating either sequences or structures, emerging co-generation methods promise more accurate and controllable protein design, ideally achieved by modeling both modalities simultaneously. Here we review recent advances in deep generative models for protein design, with a particular focus on sequence-structure co-generation methods. We describe the key methodological and evaluation principles underlying these methods, highlight recent advances from the literature, and discuss opportunities for continued development of sequence-structure co-generation approaches.

Recent advances and future challenges in predictive modeling of metalloproteins by artificial intelligence

Metal coordination is essential for structural/catalytic functions of metalloproteins that mediate a wide range of biological processes in living organisms. Advances in bioinformatics have significantly enhanced our understanding of metal-binding sites and their functional roles in metalloproteins. State-of-the-art computational models developed for metal-binding sites seamlessly integrate protein sequence and structural data to unravel the complexities of metal coordination environments. Our goal in this mini-review is to give an overview of these tools and highlight the current challenges (predicting dynamic metal-binding sites, determining functional metalation states, and designing intricate coordination networks) remaining in the predictive models of metal-binding sites. Addressing these challenges will not only deepen our knowledge of natural metalloproteins but also accelerate the development of artificial metalloproteins with novel and precisely engineered functionalities.

Enzyme Enhancement Through Computational Stability Design Targeting NMR-Determined Catalytic Hotspots

Enzymes are the quintessential green catalysts, but realizing their full potential for biotechnology typically requires improvement of their biomolecular properties. Catalysis enhancement, however, is often accompanied by impaired stability. Here, we show how the interplay between activity and stability in enzyme optimization can be efficiently addressed by coupling two recently proposed methodologies for guiding directed evolution. We first identify catalytic hotspots from chemical shift perturbations induced by transition-state-analogue binding and then use computational/phylogenetic design (FuncLib) to predict stabilizing combinations of mutations at sets of such hotspots. We test this approach on a previously designed de novo Kemp eliminase, which is already highly optimized in terms of both activity and stability. Most tested variants displayed substantially increased denaturation temperatures and purification yields. Notably, our most efficient engineered variant shows a ∼3-fold enhancement in activity (kcat ∼ 1700 s-1, kcat/KM ∼ 4.3 × 105 M-1 s-1) from an already heavily optimized starting variant, resulting in the most proficient proton-abstraction Kemp eliminase designed to date, with a catalytic efficiency on a par with naturally occurring enzymes. Molecular simulations pinpoint the origin of this catalytic enhancement as being due to the progressive elimination of a catalytically inefficient substrate conformation that is present in the original design. Remarkably, interaction network analysis identifies a significant fraction of catalytic hotspots, thus providing a computational tool which we show to be useful even for natural-enzyme engineering. Overall, our work showcases the power of dynamically guided enzyme engineering as a design principle for obtaining novel biocatalysts with tailored physicochemical properties, toward even anthropogenic reactions.

Effects of binding partners on thermal reversion rates of photoswitchable molecules

The binding of photoswitchable molecules to partners forms the basis of many naturally occurring light-dependent signaling pathways and various photopharmacological and optogenetic tools. A critical parameter affecting the function of these molecules is the thermal half-life of the light state. Reports in the literature indicate that, in some cases, a binding partner can significantly influence the thermal half-life, while in other cases it has no effect. Here, we present a unifying framework for quantitatively analyzing the effects of binding partners on thermal reversion rates. We focus on photoswitchable protein/binder interactions involving LOV domains, photoactive yellow protein, and CBCR GAF domains with partners that bind either the light or the dark state of the photoswitchable domain. We show that the effect of a binding partner depends on the extent to which the transition state for reversion resembles the dark state or the light state. We quantify this resemblance with a ϕswitching value, where ϕswitching = 1 if the conformation of the part of the photoswitchable molecule that interacts with the binding partner closely resembles its dark state conformation and ϕswitching = 0 if it resembles its light state. In addition to providing information on the transition state for switching, this analysis can guide the design of photoswitchable systems that retain useful thermal half-lives in practice. The analysis also provides a basis for the use of simple kinetic measurements to determine effective changes in affinity even in complex milieu.

Engineered enzymes for enantioselective nucleophilic aromatic substitutions

Nucleophilic aromatic substitutions (SNAr) are among the most widely used processes in the pharmaceutical and agrochemical industries1-4, allowing convergent assembly of complex molecules through C-C and C-X (X = O, N, S) bond formation. SNAr reactions are typically carried out using forcing conditions, involving polar aprotic solvents, stoichiometric bases and elevated temperatures, which do not allow for control over reaction selectivity. Despite the importance of SNAr chemistry, there are only a handful of selective catalytic methods reported that rely on small organic hydrogen-bonding or phase-transfer catalysts5-11. Here we establish a biocatalytic approach to stereoselective SNAr chemistry by uncovering promiscuous SNAr activity in a designed enzyme featuring an activated arginine12. This activity was optimized over successive rounds of directed evolution to afford an engineered biocatalyst, SNAr1.3, that is 160-fold more efficient than the parent and promotes the coupling of electron-deficient arenes with carbon nucleophiles with near-perfect stereocontrol (>99% enantiomeric excess (e.e.)). SNAr1.3 can operate at a rate of 0.15 s-1, perform more than 4,000 turnovers and can accept a broad range of electrophilic and nucleophilic coupling partners, including those that allow construction of challenging 1,1-diaryl quaternary stereocentres. Biochemical, structural and computational studies provide insights into the catalytic mechanism of SNAr1.3, including the emergence of a halide binding pocket shaped by key catalytic residues Arg124 and Asp125. This study brings a landmark synthetic reaction into the realm of biocatalysis to provide an efficient and versatile platform for catalytic SNAr chemistry.

Beyond symmetric self-assembly and effective molarity: unlocking functional enzyme mimics with robust organic cages

The bespoke environments in enzyme active sites can selectively accelerate chemical reactions by as much as 1019. Macromolecular and supramolecular chemists have been inspired to understand and mimic these accelerations and selectivities for applications in catalysis for sustainable synthesis. Over the past 60+ years, mimicry strategies have evolved with changing interests, understanding, and synthetic advances but, ubiquitously, research has focused on use of a molecular "cavity". The activities of different cavities vary with the subset of features available to a particular cavity type. Unsurprisingly, without synthetic access to mimics able to encompass more/all of the functional features of enzyme active sites, examples of cavity-catalyzed processes demonstrating enzyme-like rate accelerations remain rare. This perspective will briefly highlight some of the key advances in traditional cavity catalysis, by cavity type, in order to contextualize the recent development of robust organic cage catalysts, which can exploit stability, functionality, and reduced symmetry to enable promising catalytic modes.

AI-enabled alkaline-resistant evolution of protein to apply in mass production

Artificial intelligence (AI) models have been used to study the compositional regularities of proteins in nature, enabling it to assist in protein design to improve the efficiency of protein engineering and reduce manufacturing cost. However, in industrial settings, proteins are often required to work in extreme environments where they are relatively scarce or even non-existent in nature. Since such proteins are almost absent in the training datasets, it is uncertain whether AI model possesses the capability of evolving the protein to adapt extreme conditions. Antibodies are crucial components of affinity chromatography, and they are hoped to remain active at the extreme environments where most proteins cannot tolerate. In this study, we applied an advanced large language model (LLM), the Pro-PRIME model, to improve the alkali resistance of a representative antibody, a VHH antibody capable of binding to growth hormone. Through two rounds of design, we ensured that the selected mutant has enhanced functionality, including higher thermal stability, extreme pH resistance, and stronger affinity, thereby validating the generalized capability of the LLM in meeting specific demands. To the best of our knowledge, this is the first LLM-designed protein product, which is successfully applied in mass production.

Directed Evolution's Selective Use of Quantum Tunneling in Designed Enzymes─A Combined Theoretical and Experimental Study

Natural enzymes are powerful catalysts, reducing the apparent activation energy for reactions and enabling chemistry to proceed as much as 1015 times faster than the corresponding solution reaction. It has been suggested for some time that, in some cases, quantum tunneling can contribute to this rate enhancement by offering pathways through a barrier inaccessible to activated events. A central question of interest to both physical chemists and biochemists is the extent to which evolution introduces mechanisms below the barrier, or tunneling mechanisms. In view of the rapidly expanding chemistries for which artificial enzymes have been created, it is of interest to see how quantum tunneling has been used in these reactions. In this paper, we study the evolution of possible proton tunneling during C-H bond cleavage in enzymes that catalyze the Morita-Baylis-Hillman (MBH) reaction. The enzymes were generated by theoretical design, followed by laboratory evolution. We employ classical and centroid molecular dynamics approaches in path sampling computations to determine whether there is a quantum contribution to lowering the free energy of the proton transfer for various experimentally generated protein and substrate combinations. These data are compared to experiments reporting on the observed kinetic isotope effect (KIE) for the relevant reactions. Our results indicate the modest involvement of tunneling when laboratory evolution has resulted in a system with a higher classical free energy barrier to chemistry (that is, when optimization of processes other than chemistry results in a higher chemical barrier).

Confinement and Catalysis within De Novo Designed Peptide Barrels

De novo protein design has advanced such that many peptide assemblies and protein structures can be generated predictably and quickly. The drive now is to bring functions to these structures, for example, small-molecule binding and catalysis. The formidable challenge of binding and orienting multiple small molecules to direct chemistry is particularly important for paving the way to new functionalities. To address this, here we describe the design, characterization, and application of small-molecule:peptide ternary complexes in aqueous solution. This uses α-helical barrel (αHB) peptide assemblies, which comprise 5 or more α helices arranged around central channels. These channels are solvent accessible, and their internal dimensions and chemistries can be altered predictably. Thus, αHBs are analogous to "molecular flasks" made in supramolecular, polymer, and materials chemistry. Using Förster resonance energy transfer as a readout, we demonstrate that specific αHBs can accept two different organic dyes, 1,6-diphenyl-1,3,5-hexatriene and Nile red, in close proximity. In addition, two anthracene molecules can be accommodated within an αHB to promote anthracene photodimerization. However, not all ternary complexes are productive, either in energy transfer or photodimerization, illustrating the control that can be exerted by judicious choice and design of the αHB.

Identification of Scaffold Specific Energy Transfer Networks in the Enthalpic Activation of Orotidine 5'-Monophosphate Decarboxylase

Orotidine 5'-monophosphate decarboxylase (OMPDC) is one of the most efficient enzyme systems studied, enhancing the decarboxylation of OMP to uridine 5'-monophosphate (UMP) by ca. 17 orders of magnitude, primarily by reducing the enthalpy of activation by ca. 28 kcal/mol. Despite a substantial reduction in activation enthalpy, OMPDC requires 15 kcal/mol of activation energy post-ES complex formation. This study investigates the physical basis of how thermal energy from solvent collisions is directed into the active site of enzyme to enable efficient thermal activation of the reaction. Comparative study of temperature-dependent hydrogen-deuterium exchange mass spectrometry (TDHDX) for WT and mutant forms of enzymes has recently been shown to uncover site specific protein networks for thermal energy transfer from solvent to enzyme active sites. In this study, we interrogate region-specific changes in the enthalpic barrier for local protein flexibility using a native OMPDC from Methanothermobacter thermautotrophicus (Mt-OMPDC) and a single site variant (Leu123Ala) that alters the activation enthalpy for catalytic turnover. The data obtained implicate four spatially resolved, thermally sensitive networks that originate at different protein/solvent interfaces and terminate at sites surrounding the substrate near the substrate phosphate-binding region (R203), the substrate- ribose binding region (K42), and a reaction enhancing loop5 (S127). These are proposed to act synergistically, transiently optimizing the position and electrostatics of the reactive carboxylate of the substrate to facilitate activated complex formation. The uncovered complexity of thermal activation networks in Mt-OMPDC distinguishes this enzyme from other members of the TIM barrel family previously investigated by TDHDX. The new findings extend the essential role of protein scaffold dynamics in orchestrating enzyme activity, with broad implications for the design of highly efficient biocatalysts.

Genetic Incorporation of a Thioxanthone-Containing Amino Acid for the Design of Artificial Photoenzymes

Genetically encodable photosensitizers allow the design of artificial photoenzymes to expand the scope of abiological reactions. Herein, we report the genetic incorporation of a thioxanthone-containing amino acid into a protein scaffold via an engineered pyrrolysyl-tRNA/pyrrolysyl-tRNA synthetase pair. The designer enzyme was engineered to catalyze a dearomative [2+2] cycloaddition reaction in high yields (up to>99 % yield) with excellent enantioselectivity (up to 98 : 2 e.r.). This work provides a robust and facile method for photoenzyme design and lays the foundation for the development of further photoenzymatic reactions.

Genetic Code Expansion: Recent Developments and Emerging Applications

The concept of genetic code expansion (GCE) has revolutionized the field of chemical and synthetic biology, enabling the site-specific incorporation of noncanonical amino acids (ncAAs) into proteins, thus opening new avenues in research and applications across biology and medicine. In this review, we cover the principles of GCE, including the optimization of the aminoacyl-tRNA synthetase (aaRS)/tRNA system and the advancements in translation system engineering. Notable developments include the refinement of aaRS/tRNA pairs, enhancements in screening methods, and the biosynthesis of noncanonical amino acids. The applications of GCE technology span from synthetic biology, where it facilitates gene expression regulation and protein engineering, to medicine, with promising approaches in drug development, vaccine production, and gene editing. The review concludes with a perspective on the future of GCE, underscoring its potential to further expand the toolkit of biology and medicine. Through this comprehensive review, we aim to provide a detailed overview of the current state of GCE technology, its challenges, opportunities, and the frontier it represents in the expansion of the genetic code for novel biological research and therapeutic applications.

Protein Engineering of Substrate Specificity toward Nitrilases: Strategies and Challenges

Nitrilase is extensively applied across diverse sectors owing to its unique catalytic properties. Nevertheless, in industrial production, nitrilases often face issues such as low catalytic efficiency, limited substrate range, suboptimal selectivity, and side reaction products, which have garnered heightened attention. With the widespread recognition that the structure of enzymes has a direct impact on their catalytic properties, an increasing number of researchers are beginning to optimize the functional characteristics of nitrilases by modifying their structures, in order to meet specific industrial or biotechnology application needs. Particularly in the artificial intelligence era, the innovative application of computer-aided design in enzyme engineering offers remarkable opportunities to tailor nitrilases for the widespread production of high-value products. In this discussion, we will briefly examine the structural mechanism of nitrilase. An overview of the protein engineering strategies of substrate preference, regioselectivity and stereoselectivity are explored combined with some representative examples recently in terms of the substrate specificity of enzyme. The future research trends in this field are also prospected.

Biocatalytic synthesis of ribonucleoside analogues using nucleoside transglycosylase-2

Ribonucleosides are essential building blocks used extensively in antiviral and oligonucleotide therapeutics. A major challenge in the further development of nucleoside analogues for therapeutic applications is access to scalable and environmentally sustainable synthetic strategies. This study uses the type II nucleoside 2'-deoxyribosyltransferase from Lactobacillus leichmannii (LlNDT-2) to prepare a suite of ribonucleoside analogues using naturally-occurring uridine and cytidine sugar donors. Crystal structure and mutational analyses are used to define the substrate tolerance of the nucleobase exchange and the 2'-substituent of the nucleoside sugar donor. Nucleobase profiling identified acceptance of both purine and pyrimidine nucleobases. Finally, the scalability of the approach is showcased, enabling the preparation of ribonucleosides on millimolar scales. This biocatalytic strategy opens up opportunities to establish chemoenzymatic routes to prepare nucleoside analogues incorporating 2' modifications that are of therapeutic importance.

Single-Electron Oxidation Triggered by Visible-Light-Excited Enzymes for Asymmetric Biocatalysis

By integrating enzymatic catalysis with photocatalysis, photoenzymatic catalysis emerges as a powerful strategy to enhance enzyme catalytic capabilities and provide superior stereocontrol in reactions involving reactive intermediates. Repurposing naturally occurring enzymes using visible light is among the most active directions of photoenzymatic catalysis. This Minireview focuses on a cutting-edge strategy in this direction, namely single-electron-oxidation-triggered non-natural biotransformations catalyzed by photoexcited enzymes. These straightforward transformations feature a unique radical mechanism initiated by single-electron oxidation, achieving redox-neutral non-natural C-C, C-O, and C-S bond formation, and expanding the chemical toolbox of enzymes. By highlighting recent advances in this field and emphasizing their catalytic mechanisms and synthetic potential, innovative approaches for photobiomanufacturing are anticipated.

Seven critical challenges in synthetic one-carbon assimilation and their potential solutions

Synthetic C1 assimilation holds the promise of facilitating carbon capture while mitigating greenhouse gas emissions, yet practical implementation in microbial hosts remains relatively limited. Despite substantial progress in pathway design and prototyping, most efforts stay at the proof-of-concept stage, with frequent failures observed even under in vitro conditions. This review identifies seven major barriers constraining the deployment of synthetic C1 metabolism in microorganisms and proposes targeted strategies for overcoming these issues. A primary limitation is the low catalytic activity of carbon-fixing enzymes, particularly carboxylases, which restricts the overall pathway performance. In parallel, challenges in expressing multiple heterologous genes-especially those encoding metal-dependent or oxygen-sensitive enzymes-further hinder pathway functionality. At the systems level, synthetic C1 pathways often exhibit poor flux distribution, limited integration with the host metabolism, accumulation of toxic intermediates, and disruptions in redox and energy balance. These factors collectively reduce biomass formation and compromise product yields in biotechnological setups. Overcoming these interconnected challenges is essential for moving synthetic C1 assimilation beyond conceptual stages and enabling its application in scalable, efficient bioprocesses towards a circular bioeconomy.

Structure Prediction and Computational Protein Design for Efficient Biocatalysts and Bioactive Proteins

The ability to predict and design protein structures has led to numerous applications in medicine, diagnostics and sustainable chemical manufacture. In addition, the wealth of predicted protein structures has advanced our understanding of how life's molecules function and interact. Honouring the work that has fundamentally changed the way scientists research and engineer proteins, the Nobel Prize in Chemistry in 2024 was awarded to David Baker for computational protein design and jointly to Demis Hassabis and John Jumper, who developed AlphaFold for machine-learning-based protein structure prediction. Here, we highlight notable contributions to the development of these computational tools and their importance for the design of functional proteins that are applied in organic synthesis. Notably, both technologies have the potential to impact drug discovery as any therapeutic protein target can now be modelled, allowing the de novo design of peptide binders and the identification of small molecule ligands through in silico docking of large compound libraries. Looking ahead, we highlight future research directions in protein engineering, medicinal chemistry and material design that are enabled by this transformative shift in protein science.

Conformational ensembles reveal the origins of serine protease catalysis

Enzymes exist in ensembles of states that encode the energetics underlying their catalysis. Conformational ensembles built from 1231 structures of 17 serine proteases revealed atomic-level changes across their reaction states. By comparing the enzymatic and solution reaction, we identified molecular features that provide catalysis and quantified their energetic contributions to catalysis. Serine proteases precisely position their reactants in destabilized conformers, creating a downhill energetic gradient that selectively favors the motions required for reaction while limiting off-pathway conformational states. The same catalytic features have repeatedly evolved in proteases and additional enzymes across multiple distinct structural folds. Our ensemble-function analyses revealed previously unknown catalytic features, provided quantitative models based on simple physical and chemical principles, and identified motifs recurrent in nature that may inspire enzyme design.

Synergistic photobiocatalysis for enantioselective triple-radical sorting

Multicomponent reactions-those where three or more substrates combine into a product-have been highly useful in rapidly building chemical building blocks of increased complexity1, but achieving this enzymatically has remained rare2-5. This limitation primarily arises because an enzyme's active site is not typically set up to address multiple substrates, especially in cases involving multiple radical intermediates6. Recently, chemical catalytic radical sorting has emerged as an enabling strategy for a variety of useful reactions7,8. However, making such processes enantioselective is highly challenging owing to the inherent difficulty in the stereochemical control of radicals9. Here we repurpose a thiamine-dependent enzyme10,11 through directed evolution and combine it with photoredox catalysis to achieve a photobiocatalytic enantioselective three-component radical cross-coupling. This approach combines three readily available starting materials-aldehydes, α-bromo-carbonyls and alkenes-to give access to enantioenriched ketone products. Mechanistic investigations provide insights into how this dual photocatalyst-enzyme system precisely directs the three distinct radicals involved in the transformation, unlocking enzyme reactivity. Our approach has achieved exceptional stereoselectivity, with 24 out of 33 examples achieving ≥97% enantiomeric excess.

EvoAI enables extreme compression and reconstruction of the protein sequence space

Designing proteins with improved functions requires a deep understanding of how sequence and function are related, a vast space that is hard to explore. The ability to efficiently compress this space by identifying functionally important features is extremely valuable. Here we establish a method called EvoScan to comprehensively segment and scan the high-fitness sequence space to obtain anchor points that capture its essential features, especially in high dimensions. Our approach is compatible with any biomolecular function that can be coupled to a transcriptional output. We then develop deep learning and large language models to accurately reconstruct the space from these anchors, allowing computational prediction of novel, highly fit sequences without prior homology-derived or structural information. We apply this hybrid experimental-computational method, which we call EvoAI, to a repressor protein and find that only 82 anchors are sufficient to compress the high-fitness sequence space with a compression ratio of 1048. The extreme compressibility of the space informs both applied biomolecular design and understanding of natural evolution.

A Dual-Purpose Non-Canonical Amino Acid for the Expanded Genetic Code: Combining Metal-Binding and Click Chemistry

A rationally designed dual-purpose non-canonical amino acid (Trz) has been synthesised and successfully incorporated into a protein scaffold by genetic code expansion. Trz contains a 5-pyridyl-1,2,4-triazine system, which allows for inverse-electron-demand Diels-Alder (IEDDA) reactions to occur on the triazine ring and for metal ions to be chelated both before and after the click reaction. Trz was successfully incorporated into a protein scaffold and the IEDDA utility of Trz demonstrated through the site-specific labelling of the purified protein with a bicyclononyne. Additionally, Trz was shown to successfully coordinate a cyclometallated iridium(III) centre, providing access to a bioorthogonal luminogenic probe. The luminescent properties of the Ir(III)-bound protein blue-shift upon IEDDA click reaction with bicyclononyne, providing a unique method for monitoring the extent and location of the labelling reaction. In summary, Trz is a new dual-purpose non-canonical amino acid with great potential for myriad bioapplications where metal-based functionality is required, for example in imaging, catalysis, and photo-dynamic therapy, in conjunction with a bioorthogonal reactive handle to impart additional functionalities, such as dual-modality imaging or therapeutic payloads.

Boron Designer Enzyme with a Hybrid Catalytic Dyad

Genetically encoded noncanonical amino acids can introduce new-to-nature activation modes into enzymes. While these amino acids can act as catalysts on their own due to their inherent chemical properties, interactions with adjacent residues in an enzyme, such as those present in natural catalytic dyads or triads, unlock a higher potential for designer enzymes. We incorporated a boron-containing amino acid into the protein scaffold RamR to create an active enzyme for the kinetic resolution of α-hydroxythioesters. We found that a closely positioned lysine residue is crucial for the catalytic activity of the designer enzyme by forming a hybrid catalytic dyad with the boronic acid residue. The enzyme is capable of resolving differently substituted α-hydroxythioesters with good selectivities. High-resolution mass spectrometry, 11B NMR spectroscopy, and crystal structure analysis of the designer enzyme gave insight into the three steps of the mechanism (substrate binding, hydroxide transfer, product release). Mutations of a residue around the catalytic dyad led to a variant of the enzyme with 2-fold improvement of catalytic activity and selectivity.

Organophosphate Hydrolysis by a Designed Metalloenzyme: Impact of Mutations Explained

The efficient design of novel enzymes has been attainable only by a combination of theoretical approaches and experimental refinement, suggesting inadequate performance of de novo design protocols. Based on the analysis of the evolutionary trajectory of a designed organophosphate hydrolase, this work aimed at developing and validating the improved theoretical models describing the catalytic activity of five enzyme variants (including wild-type as well as theoretically derived and experimentally refined enzymes) performing the hydrolysis of diethyl 7-hydroxycoumarinyl phosphate. The following aspects possibly important for enzyme design were addressed: the level of theory sufficient for a reliable description of enzyme-reactant interactions, the issue of ground state (GS) destabilization versus transition state (TS) stabilization, and the derivation of the proper side chain rotamers of amino acid residues. For enzyme variants analyzed herein, differential transition state stabilization (DTSS, i.e., preferential TS binding by an enzyme over the GS binding) calculated with a non-empirical model of the interaction energy (i.e., multipole electrostatic plus approximate dispersion terms, MED) displayed a superior performance in ranking the enzyme catalytic activity. The MED DTSS-based systematic rotamer refinement performed with an efficient scanning procedure and accounting for long-range interaction energy terms is an important step capable of unlocking the full potential impact of the given residue that could be otherwise overlooked with a conventional static approach featuring optimization to the nearby minimum. While TS stabilization is the main factor contributing to the increased catalytic activity of the de novo-designed variant studied in this work, directed evolution refinement appears to impact the catalytic activity of another enzyme variant analyzed herein via GS destabilization.

Understanding the Directed Evolution of a Natural-like Efficient Artificial Metalloenzyme

The artificial metalloenzyme containing iridium in place of iron along with four directed evolution mutations C317G, T213G, L69V, and V254L in a natural cytochrome P450 presents an important milestone in merging the extraordinary efficiency of biocatalysts with the versatility of small molecule chemical catalysts in catalyzing a new-to-nature carbene insertion reaction. This is a show-stopper enzyme, as it exhibits a catalytic efficiency similar to that of natural enzymes. Despite this remarkable discovery, there is no mechanistic and structural understanding as to why it displays extraordinary efficiency after the incorporation of the four active site mutations by directed evolution methods, which so far has been intractable to any experimental methods. In this study, we have deciphered how directed evolution mutations gradually alter the protein conformational ensemble to populate a catalytically active conformation to boost a multistep catalysis in a natural-like artificial metalloenzyme using large-scale molecular dynamics simulations, rigorous quantum chemical (QM), and multiscale quantum chemical/molecular mechanics (QM/MM) calculations. It reveals how evolution precisely positions the cofactor-substrate in an unusual but effective orientation within a reshaped active site in the catalytically active conformation stabilized by C-H···π interactions from more ordered mutated L69V and V254L residues to achieve preferential transition state stabilization compared to the ground state. This work essentially tracks down in atomistic detail the shift in the conformational ensemble of the highly active conformation from the less efficient single mutant to the most efficient quadruple mutant and offers valuable insights for designing better enzymes. The active conformation correctly reproduces the experimental barrier height and also accounts for the catalytic effect, which is in good agreement with experimental observations. Moreover, this conformation features an unusual bonding interaction in a metal-carbene species that preferentially stabilizes the rate-determining formation of an iridium porphyrin carbene intermediate to render the observed high catalytic rate acceleration. Our study provides crucial insights into the underlying rationale for directed evolution, reports the major catalytic role of nonelectrostatic interactions in enzyme catalysis different from the electrostatic model, and suggests a crucial principle toward designing enzymes with natural efficiency.

A general temperature-guided language model to design proteins of enhanced stability and activity

Designing protein mutants with both high stability and activity is a critical yet challenging task in protein engineering. Here, we introduce PRIME, a deep learning model, which can suggest protein mutants with improved stability and activity without any prior experimental mutagenesis data for the specified protein. Leveraging temperature-aware language modeling, PRIME demonstrated superior predictive ability compared to current state-of-the-art models on the public mutagenesis dataset across 283 protein assays. Furthermore, we validated PRIME's predictions on five proteins, examining the impact of the top 30 to 45 single-site mutations on various protein properties, including thermal stability, antigen-antibody binding affinity, and the ability to polymerize nonnatural nucleic acid or resilience to extreme alkaline conditions. More than 30% of PRIME-recommended mutants exhibited superior performance compared to their premutation counterparts across all proteins and desired properties. We developed an efficient and effective method based on PRIME to rapidly obtain multisite mutants with enhanced activity and stability. Hence, PRIME demonstrates broad applicability in protein engineering.

Using the Theozyme Model to Study the Dynamical Mechanism of the Post-Transition State Bifurcation Reaction by NgnD Enzyme

Post-transition state bifurcation (PTSB) is a fundamental process in which a single transition state leads to multiple products. This phenomenon is important in both biological and chemical contexts and offers valuable insights into reaction mechanisms and their applications. The theozyme model, which focuses on key residues within enzymes, offers a computationally efficient method for studying these processes while preserving the enzyme's catalytic properties. This approach enhances our understanding of how enzymes stabilize and direct the transition state, thereby influencing product distribution and selectivity. In this study, we investigate the dynamics and regulatory mechanisms of the PTSB reaction catalyzed by the enzyme NgnD. The enzyme NgnD facilitates a cycloaddition reaction that produces both [6 + 4] and [4 + 2] adducts, with a preference for the [6 + 4] adduct. By analyzing the potential energy surface, bond length distribution, and interactions between the theozyme and the ambimodal transition state, we elucidate the role of the enzyme's active site residues in determining product selectivity. We illustrate how these key residues contribute to the formation of different adducts, providing insights from various perspectives. Using theozyme models, we propose how the four most influential active residues collectively might control the direction of adduct formation through their cumulative effects.

Installation of an organocatalyst into a protein scaffold creates an artificial Stetterase

Using a protein scaffold covalently functionalised with a thiamine-inspired N-heterocyclic carbene (NHC), we created an artificial Stetterase (ArtiSt) which catalyses a stereoselective, intramolecular Stetter reaction. We demonstrate that ArtiSt functions under ambient conditions with low catalyst loading. Furthermore, activity can be increased >20 fold by altering the protein scaffold.

Reaching New Heights in Genetic Code Manipulation with High Throughput Screening

The chemical and physical properties of proteins are limited by the 20 canonical amino acids. Genetic code manipulation allows for the incorporation of noncanonical amino acids (ncAAs) that enhance or alter protein functionality. This review explores advances in the three main strategies for introducing ncAAs into biosynthesized proteins, focusing on the role of high throughput screening in these advancements. The first section discusses engineering aminoacyl-tRNA synthetases (aaRSs) and tRNAs, emphasizing how novel selection methods improve characteristics including ncAA incorporation efficiency and selectivity. The second section examines high-throughput techniques for improving protein translation machinery, enabling accommodation of alternative genetic codes. This includes opportunities to enhance ncAA incorporation through engineering cellular components unrelated to translation. The final section highlights various discovery platforms for high-throughput screening of ncAA-containing proteins, showcasing innovative binding ligands and enzymes that are challenging to create with only canonical amino acids. These advances have led to promising drug leads and biocatalysts. Overall, the ability to discover unexpected functionalities through high-throughput methods significantly influences ncAA incorporation and its applications. Future innovations in experimental techniques, along with advancements in computational protein design and machine learning, are poised to further elevate this field.

A resurrected ancestor of Cas12a expands target access and substrate recognition for nucleic acid editing and detection

The properties of Cas12a nucleases constrict the range of accessible targets and their applications. In this study, we applied ancestral sequence reconstruction (ASR) to a set of Cas12a orthologs from hydrobacteria to reconstruct a common ancestor, ReChb, characterized by near-PAMless targeting and the recognition of diverse nucleic acid activators and collateral substrates. ReChb shares 53% sequence identity with the closest Cas12a ortholog but no longer requires a T-rich PAM and can achieve genome editing in human cells at sites inaccessible to the natural FnCas12a or the engineered and PAM-flexible enAsCas12a. Furthermore, ReChb can be triggered not only by double-stranded DNA but also by single-stranded RNA and DNA targets, leading to non-specific collateral cleavage of all three nucleic acid substrates with similar efficiencies. Finally, tertiary and quaternary structures of ReChb obtained by cryogenic electron microscopy reveal the molecular details underlying its expanded biophysical activities. Overall, ReChb expands the application space of Cas12a nucleases and underscores the potential of ASR for enhancing CRISPR technologies.

Machine learning-assisted amidase-catalytic enantioselectivity prediction and rational design of variants for improving enantioselectivity

Biocatalysis is an attractive approach for the synthesis of chiral pharmaceuticals and fine chemicals, but assessing and/or improving the enantioselectivity of biocatalyst towards target substrates is often time and resource intensive. Although machine learning has been used to reveal the underlying relationship between protein sequences and biocatalytic enantioselectivity, the establishment of substrate fitness space is usually disregarded by chemists and is still a challenge. Using 240 datasets collected in our previous works, we adopt chemistry and geometry descriptors and build random forest classification models for predicting the enantioselectivity of amidase towards new substrates. We further propose a heuristic strategy based on these models, by which the rational protein engineering can be efficiently performed to synthesize chiral compounds with higher ee values, and the optimized variant results in a 53-fold higher E-value comparing to the wild-type amidase. This data-driven methodology is expected to broaden the application of machine learning in biocatalysis research.

Noncanonical Amino Acids: Bringing New-to-Nature Functionalities to Biocatalysis

Biocatalysis has become an important component of modern organic chemistry, presenting an efficient and environmentally friendly approach to synthetic transformations. Advances in molecular biology, computational modeling, and protein engineering have unlocked the full potential of enzymes in various industrial applications. However, the inherent limitations of the natural building blocks have sparked a revolutionary shift. In vivo genetic incorporation of noncanonical amino acids exceeds the conventional 20 amino acids, opening new avenues for innovation. This review provides a comprehensive overview of applications of noncanonical amino acids in biocatalysis. We aim to examine the field from multiple perspectives, ranging from their impact on enzymatic reactions to the creation of novel active sites, and subsequent catalysis of new-to-nature reactions. Finally, we discuss the challenges, limitations, and promising opportunities within this dynamic research domain.

Rational Design of an Artificial Metalloenzyme by Constructing a Metal-Binding Site Close to the Heme Cofactor in Myoglobin

In this study, we constructed a metal-binding site close to the heme cofactor in myoglobin (Mb) by covalently attaching a nonnative metal-binding ligand of bipyridine to Cys46 through the F46C mutation in the heme distal site. The X-ray structure of the designed enzyme, termed F46C-mBpy Mb, was solved in the Cu(II)-bound form, which revealed the formation of a heterodinuclear center of Cu-His-H2O-heme. Cu(II)-F46C-mBpy Mb exhibits not only nitrite reductase reactivity but also cascade reaction activity involving both hydrolysis and oxidation. Furthermore, F46C-mBpy Mb displays Mn-peroxidase activity by the oxidation of Mn2+ to Mn3+ using H2O2 as an oxidant. This study shows that the construction of a nonnative metal-binding site close to the heme cofactor is a convenient approach to creating an artificial metalloenzyme with a heterodinuclear center that confers multiple functions.

Design and Evolution of an Artificial Friedel-Crafts Alkylation Enzyme Featuring an Organoboronic Acid Residue

Creating artificial enzymes by the genetic incorporation of noncanonical amino acids with catalytic side chains would expand the enzyme chemistries that have not been discovered in nature. Here, we report the design of an artificial enzyme that uses p-boronophenylalanine as the catalytic residue. The artificial enzyme catalyzes Michael-type Friedel-Crafts alkylation through covalent activation. The designer enzyme was further engineered to afford high yields with excellent enantioselectivities. We next developed a practical method for preparative-scale reactions by whole-cell catalysis. This enzymatic C-C bond formation reaction was combined with palladium-catalyzed dearomative arylation to achieve the efficient synthesis of spiroindolenine compounds.

Enzymatic production of xylooligosaccharide from lignocellulosic and marine biomass: A review of current progress, challenges, and its applications in food sectors

Over the last decade, xylooligosaccharides (XOS) have attracted great attentions because of their unique chemical properties and excellent prebiotic effects. Among the current strategies for XOS production, enzymatic hydrolysis is preferred due to its green and safe process, simplicity in equipment, and high control of the degrees of polymerization. This paper comprehensively summarizes various lignocellulosic biomass and marine biomass employed in enzymatic production of XOS. The importance and advantages of enzyme immobilization in XOS production are also discussed. Many novel immobilization techniques for xylanase are presented. In addition, bioinformatics techniques for the mining and designing of new xylanase are also described. Moreover, XOS has exhibited great potential applications in the food industry as diverse roles, such as a sugar replacer, a fat replacer, and cryoprotectant. This review systematically summarizes the current research progress on the applications of XOS in food sectors, including beverages, bakery products, dairy products, meat products, aquatic products, food packaging film, wall materials, and others. It is anticipated that this paper will act as a reference for the further development and application of XOS in food sectors and other fields.

Cellular Site-Specific Incorporation of Noncanonical Amino Acids in Synthetic Biology

Over the past two decades, genetic code expansion (GCE)-enabled methods for incorporating noncanonical amino acids (ncAAs) into proteins have significantly advanced the field of synthetic biology while also reaping substantial benefits from it. On one hand, they provide synthetic biologists with a powerful toolkit to enhance and diversify biological designs beyond natural constraints. Conversely, synthetic biology has not only propelled the development of ncAA incorporation through sophisticated tools and innovative strategies but also broadened its potential applications across various fields. This Review delves into the methodological advancements and primary applications of site-specific cellular incorporation of ncAAs in synthetic biology. The topics encompass expanding the genetic code through noncanonical codon addition, creating semiautonomous and autonomous organisms, designing regulatory elements, and manipulating and extending peptide natural product biosynthetic pathways. The Review concludes by examining the ongoing challenges and future prospects of GCE-enabled ncAA incorporation in synthetic biology and highlighting opportunities for further advancements in this rapidly evolving field.

Characterization and rational engineering of an alkaline-tolerant azoreductase derived from Roseibium sp. H3510 for enhanced decolorization of azo dyes

rAzoR2326, an azoreductase derived from Roseibium sp. H3510, functions as an FMN-dependent homodimer utilizing NADH as cofactor. It demonstrated maximum activity at 45 °C and retained moderate activity above 50 °C, exhibiting stability from pH 7-10. Evolution and structure guided rational design of wild-type rAzoR2326 (WT) efficiently yielded 6 single-point mutants with improved thermostability and activity from a 22-variant library. Further combinatorial mutation led to mutant M20 with substantially enhanced thermostability (15-fold longer half-life at 50 °C) and activity (3.24-fold higher kcat/Km). M20 exhibited superior catalytic properties for decolorizing Allura Red compared to WT. Specifically, its decolorization capacity at pH 10.0 was 4.26-fold higher than WT. Additionally, M20 demonstrated remarkable thermostability, retaining 76.83 % decolorization activity for Allura Red after 120 min at 50 °C, whereas WT nearly lost all catalytic activity under the same conditions. Molecular dynamics simulations revealed the structural changes in M20, such as improved hydrogen bonding and a new C-H···π interaction, led to a more compact and rigid enzyme structure. This resulted in a more stable FMN-binding pocket and substrate tunnel, thereby improving the catalytic stability and activity of M20. Given its enhanced dye decolorization ability and alkaline tolerance, M20 shows promise as a biocatalyst for treating azo dye effluents.

BhrPETase catalyzed polyethylene terephthalate depolymerization: A quantum mechanics/molecular mechanics approach

Polyethylene terephthalate (PET) is a widely used material in our daily life, particularly in areas such as packaging, fibers, and engineering plastics. However, PET waste can accumulate in the environment and pose a great threat to our ecosystem. Recently enzymatic conversion has emerged as an efficient and green strategy to address the PET crisis. Here, using a theoretical approach combining molecular dynamics simulation and quantum mechanics/molecular mechanics calculations, the depolymerization mechanism of the thermophilic cutinase BhrPETase was fully deciphered. Surprisingly, unlike the previously studied cutinase LCCICCG, our results indicate that the first step, catalytic triad assisted nucleophilic attack, is the rate-determining step. The corresponding Boltzmann weighted average energy barrier is 18.2 kcal/mol. Through extensive comparison between BhrPETase and LCCICCG, we evidence that key features like charge CHis@N1 and angle APET@C1-Ser@O1-His@H1 significantly impact the depolymerization efficiency of BhrPETase. Non-covalent bond interaction and distortion/interaction analysis inform new insights on enzyme engineer and may aid the recycling of enzymatic PET waste. This study will aid the advancement of the plastic bio-recycling economy and promote resource conservation and reuse.

Computation-guided engineering of distal mutations in an artificial enzyme

Artificial enzymes are valuable biocatalysts able to perform new-to-nature transformations with the precision and (enantio-)selectivity of natural enzymes. Although they are highly engineered biocatalysts, they often cannot reach catalytic rates akin those of their natural counterparts, slowing down their application in real-world industrial processes. Typically, their designs only optimise the chemistry inside the active site, while overlooking the role of protein dynamics on catalysis. In this work, we show how the catalytic performance of an already engineered artificial enzyme can be further improved by distal mutations that affect the conformational equilibrium of the protein. To this end, we subjected a specialised artificial enzyme based on the lactococcal multidrug resistance regulator (LmrR) to an innovative algorithm that quickly inspects the whole protein sequence space for hotpots which affect the protein dynamics. From an initial predicted selection of 73 variants, two variants with mutations distant by more than 11 Å from the catalytic pAF residue showed increased catalytic activity towards the new-to-nature hydrazone formation reaction. Their recombination displayed a 66% higher turnover number and 14 °C higher thermostability. Microsecond time scale molecular dynamics simulations evidenced a shift in the distribution of productive enzyme conformations, which are the result of a cascade of interactions initiated by the introduced mutations.

Spiers Memorial Lecture: Engineering biocatalysts

Enzymes are being engineered to catalyze chemical reactions for many practical applications in chemistry and biotechnology. The approaches used are surveyed in this short review, emphasizing methods for accessing reactivities not expressed by native protein scaffolds. The successful generation of completely de novo enzymes that rival the rates and selectivities of their natural counterparts highlights the potential role that designer enzymes may play in the coming years in research, industry, and medicine. Some challenges that need to be addressed to realize this ambitious dream are considered together with possible solutions.

Concluding remarks: biocatalysis

Biocatalysis is a rapidly evolving field with increasing impact in organic synthesis, chemical manufacturing and medicine. The Faraday Discussion reflected the current state of biocatalysis, covering the design of de novo enzymatic activities, but especially methods for the improvement of enzymes targeting a broad range of applications (i.e., hydroxylations by P450 monooxygenases, enzymatic deprotection of organic compounds under mild conditions, synthesis of chiral intermediates, plastic degradation, silicone polymer synthesis, and peptide synthesis). Central themes have been how to improve an enzyme using methods of rational design and directed evolution, informed by computer modelling and machine learning, and the incorporation of new catalytic functionalities to create hybrid and artificial enzymes.

An efficient pyrrolysyl-tRNA synthetase for economical production of MeHis-containing enzymes

Genetic code expansion has emerged as a powerful tool in enzyme design and engineering, providing new insights into sophisticated catalytic mechanisms and enabling the development of enzymes with new catalytic functions. In this regard, the non-canonical histidine analogue Nδ-methylhistidine (MeHis) has proven especially versatile due to its ability to serve as a metal coordinating ligand or a catalytic nucleophile with a similar mode of reactivity to small molecule catalysts such as 4-dimethylaminopyridine (DMAP). Here we report the development of a highly efficient aminoacyl tRNA synthetase (G1PylRSMIFAF) for encoding MeHis into proteins, by transplanting five known active site mutations from Methanomethylophilus alvus (MaPylRS) into the single domain PylRS from Methanogenic archaeon ISO4-G1. In contrast to the high concentrations of MeHis (5-10 mM) needed with the Ma system, G1PylRSMIFAF can operate efficiently using MeHis concentrations of ∼0.1 mM, allowing more economical production of a range of MeHis-containing enzymes in high titres. Interestingly G1PylRSMIFAF is also a 'polyspecific' aminoacyl tRNA synthetase (aaRS), enabling incorporation of five different non-canonical amino acids (ncAAs) including 3-pyridylalanine and 2-fluorophenylalanine. This study provides an important step towards scalable production of engineered enzymes that contain non-canonical amino acids such as MeHis as key catalytic elements.

Learning from Protein Engineering by Deconvolution of Multi-Mutational Variants

This review analyzes a development in biochemistry, enzymology and biotechnology that originally came as a surprise. Following the establishment of directed evolution of stereoselective enzymes in organic chemistry, the concept of partial or complete deconvolution of selective multi-mutational variants was introduced. Early deconvolution experiments of stereoselective variants led to the finding that mutations can interact cooperatively or antagonistically with one another, not just additively. During the past decade, this phenomenon was shown to be general. In some studies, molecular dynamics (MD) and quantum mechanics/molecular mechanics (QM/MM) computations were performed in order to shed light on the origin of non-additivity at all stages of an evolutionary upward climb. Data of complete deconvolution can be used to construct unique multi-dimensional rugged fitness pathway landscapes, which provide mechanistic insights different from traditional fitness landscapes. Along a related line, biochemists have long tested the result of introducing two point mutations in an enzyme for mechanistic reasons, followed by a comparison of the respective double mutant in so-called double mutant cycles, which originally showed only additive effects, but more recently also uncovered cooperative and antagonistic non-additive effects. We conclude with suggestions for future work, and call for a unified overall picture of non-additivity and epistasis.

Advancing microbial production through artificial intelligence-aided biology

Microbial cell factories (MCFs) have been leveraged to construct sustainable platforms for value-added compound production. To optimize metabolism and reach optimal productivity, synthetic biology has developed various genetic devices to engineer microbial systems by gene editing, high-throughput protein engineering, and dynamic regulation. However, current synthetic biology methodologies still rely heavily on manual design, laborious testing, and exhaustive analysis. The emerging interdisciplinary field of artificial intelligence (AI) and biology has become pivotal in addressing the remaining challenges. AI-aided microbial production harnesses the power of processing, learning, and predicting vast amounts of biological data within seconds, providing outputs with high probability. With well-trained AI models, the conventional Design-Build-Test (DBT) cycle has been transformed into a multidimensional Design-Build-Test-Learn-Predict (DBTLP) workflow, leading to significantly improved operational efficiency and reduced labor consumption. Here, we comprehensively review the main components and recent advances in AI-aided microbial production, focusing on genome annotation, AI-aided protein engineering, artificial functional protein design, and AI-enabled pathway prediction. Finally, we discuss the challenges of integrating novel AI techniques into biology and propose the potential of large language models (LLMs) in advancing microbial production.

Hosts engineering and in vitro enzymatic synthesis for the discovery of novel natural products and their derivatives

Novel natural products (NPs) and their derivatives are important sources for drug discovery, which have been broadly applied in the fields of agriculture, livestock, and medicine, making the synthesis of NPs and their derivatives necessarily important. In recent years, biosynthesis technology has received increasing attention due to its high efficiency in the synthesis of high value-added novel products and its advantages of green, environmental protection, and controllability. In this review, the technological advances of biosynthesis strategies in the discovery of novel NPs and their derivatives are outlined, with an emphasis on two areas of host engineering and in vitro enzymatic synthesis. In terms of hosts engineering, multiple microorganisms, including Streptomyces, Aspergillus, and Penicillium, have been used as the biosynthetic gene clusters (BGCs) provider and host strain for the expression of BGCs to discover new compounds over the past years. In addition, the use of in vitro enzymatic synthesis strategy to generate novel compounds such as triterpenoid saponins and flavonoids is also hereby described.