Frances H. Arnold, California Institute of Technology, USA

Chemistry encoded in DNA and optimized by evolution promises efficient, clean, sustainable routes to fuels, chemicals, materials, foods, medicines, and more. Evolution not only optimizes and tunes features such as activity or stereoselectivity—it also innovates. We are using evolution to create entirely new biocatalysts that catalyze reactions unknown in biology and sometimes unprecedented in human-invented chemistry. New-to-nature ‘carbene transferase’ and ‘nitrene transferase’ enzymes increase the scope of molecules and materials that can be made using biology’s remarkable chemical machinery. Such enzymes unlock chemicial transformations that were inaccessible to small-molecule catalysts. And, with modern machine learning and AI tools to aid discovery and optimization, we are closer than ever to encoding a vast array of chemical transformations in DNA.

Marc Fontecave, Collège de France, FR

The energy transition requires new sources of carbon, as alternatives to fossil sources, for the production of organic compounds for the chemical industry (fuels, polymers, etc.). Carbon dioxide transrformation to C1 or C2 compounds can be achieved by electroreduction or photoreduction using low-carbon electricity or sunlight directly, respectively. The reactions at work imply multiple electrons and protons and thus require catalysts for minimizing energy barriers and controlling selectivity. Whereas synthetic coordination complexes and solid metal materials are extensively studied, it is interesting to note that living organisms have evolved complex metalloenzymes for the reduction of CO₂ into formic acid and for the interconversion between CO₂ and CO, providing a source of inspiration for the design of biomimetic homogeneous and heterogeneous catalysts. Another class of bioinspired catalysts, namely artificial enzymes, deserves more attention. Such a strategy has been very little explored so far for CO₂ electro- or photo-reduction. This the theme of this presentation which will focus on our recent studies of artificial CO₂ reductases.

Anthony Green, University of Manchester, UK

Protein cavities can offer highly versatile and engineerable environments for hosting new catalytic sites. However, only a narrow range of functional elements are available to enzyme designers when building new active sites, meaning that many important modes of reactivity are not accessible. Here I will discuss our efforts to overcome these limitations, by encoding new catalytic elements into proteins as non-canonical amino acid side chains. This approach has allowed us to build enzymes with new functions and reactivity modes that were previously inaccessible with protein catalysts. Significantly as our catalyts are genetically encoded, their activities and selectivities can be optimized using directed evoution workflows adpated to an expanded amino acid alphabet. We are optimistic that this integration of enzyme design, genetic code expansion and laboratory evolution can provide a versatile strategy for creating enzymes with catalytic functions not accessible to nature.

John F. Hartwig, University of California, Berkeley, USA

The introduction of functional groups at the positions of typically unreactive C-H bonds site-selectively and the stereo- and regio-selective functionalization of unconjugated C=C bonds have been longstanding challenges in catalysis. To this end, our group has been motivated by the limits of small-molecule catalysts for such reactions to begin to investigate artificial metalloenzymes. These artificial metalloenzymes contain synthetic cofactors possessing abiotic metal centers that catalyze unnatural reactions with control over selectivity resulting from the protein environment. In the best-case scenario such reactions could occur within the cells of E. coli or other microorganisms and in the long-term to occur as part of an unnatural biosynthetic pathway to produce unnatural products by fermentation.

This talk will include results on new transformations, new mechanisms, new reactive intermediates, and new methods for in vivo assembly of artificial metalloenzymes. This combination of results has enabled us to combine an unnatural carbene-transfer reaction catalyzed by natural and artificial metalloenzymes with the biosynthesis of diazo compounds and natural reactions of a heterologous biosynthetic pathway to create engineered microorganisms that produce unnatural products by artificial biosynthesis encompassing organometallic chemistry.

Xiongyi Huang, Johns Hopkins University, USA

Repurposing natural enzymes to catalyze synthetic transformations absent in nature has emerged as a significant research field bridging chemistry and biology. A key challenge in this pursuit is the introduction of synthetic reaction mechanisms into natural protein scaffolds. Over the past decades, substantial breakthroughs have been achieved in this field, with many enzymatic systems developed to catalyze critical chemical transformations not previously observed in biology. However, much of this progress has focused on proteins or enzymes containing heme or organic cofactors. In this context, our group has drawn inspiration from mechanistic connections between synthetic and biocatalytic systems to explore the vast, untapped potential of nonheme enzymes for new-to-nature biocatalysis. This talk will highlight several enzymatic systems developed by our group over the past five years, which utilize diverse reaction mechanisms in transition metal catalysis for the formation of C‒N, C‒S, C‒C, C‒O, and C‒halogen bonds. We hope these systems will further advance the integration of synthetic chemistry and biology to innovate chemical synthesis, as well as deepen our understanding of both biochemical and synthetic reaction mechanisms.

Todd Hyster, Princeton University, USA

Enzymes are exquisite catalysts for chemical synthesis, capable of providing unparalleled levels of chemo-, regio-, diastereo- and enantioselectivity. Unfortunately, biocatalysts are often limited to the reactivity patterns found in nature. In this talk, I will share my groups efforts to use light to expand the reactivity profile of enzymes. In our studies, we have developed novel photoexcitation mechanisms involving common biological cofactors, such as nicotinamide, flavin, and pyridoxal, to facilitate electron transfer to substrates bound within enzyme active sites. Alternatively, proteins can be used to electronically activate substrates for reduction by exogenous photoredox catalysts enabling radical formation to be localized to the protein active site. The resulting radicals can engage in a variety of inter- and intramolecular reactions with high levels of enantioselectivity. These approaches enable biocatalysts to solve long-standing selectivity challenges in chemical synthesis.

Yi Lu, University of Texas at Austin, USA

Metalloenzymes play important roles in many biological processes, yet the structural features underlying their remarkable reactivity and selectivity remain incompletely understood. To address this issue, we have designed artificial metalloenzymes (ArMs) using small, stable proteins as scaffolds. These scaffolds are designed to incorporate key residues essential for functions in native enzymes, including O₂ and N₂ reduction as well as nitrosylation and polysaccharide oxidative cleavage reactions.

Our findings reveal that replicating the primary coordination sphere may suffice for creating structural models of metalloenzymes. However, achieving functional ArMs with high activity and turnover rates—on par with native enzymes—requires precise engineering of non-covalent secondary coordination sphere interactions, such as hydrophobic effects and hydrogen bonding, including those mediated by water molecules.

This presentation will highlight recent advances in ArM design, the insights gained from these studies, and the expansion of ArM activities beyond the capabilities of natural enzymes.

Paolo Melchiorre, University of Bologna, IT

The combination of photocatalysis, biocatalysis, and organocatalysis provides a powerful yet underexplored strategy for addressing major challenges in asymmetric synthesis. By combining these distinct catalytic disciplines, we aim to develop novel enantioselective radical processes that are otherwise difficult to achieve using conventional approaches. Central to this concept is the ability to harness enzyme-bound organocatalytic intermediates as photoactive intermediates, unlocking new mechanistic pathways for radical generation and control.

Recently, we developed a new approach to light-driven biocatalysis, where engineered enzymes utilize iminium ion intermediates—formed transiently within their active sites—as single-electron oxidants upon visible-light excitation. This strategy enables the activation of chiral carboxylic acids, triggering radical decarboxylation and subsequent stereospecific cross-coupling to construct complex chiral architectures with multiple stereocenters and complete enantiocontrol. Notably, the enzyme’s active site prevents racemization of chiral radicals via a rare "memory of chirality" mechanism, ensuring high stereochemical fidelity. By leveraging the unique synergy between light, biocatalysts, and organocatalytic intermediates, this work expands the scope of radical chemistry and sets the stage for new, sustainable methods in asymmetric synthesis.

Hannah S. Shafaat, University of California, Los Angeles, USA

Metalloenzymes catalyze the challenging chemical reactions that lie at the core of vital life processes, from carbon and nitrogen fixation to photosynthesis and respiration. Nickel-containing enzymes, specifically, are essential for global hydrogen and carbon cycling and the metabolisms of diverse microbes, with implications in human health, clean energy conversion, and sustainable fuel generation. In this presentation, I will discuss our recent efforts to recapitulate key structural and functional elements of microbial nickel enzymes such as hydrogenase, carbon monoxide dehydrogenase (CODH), and acetyl coenzyme A synthase using protein-based scaffolds. By designing model metalloenzymes from the “inside out”, each contribution can be clearly delineated. Functional studies of our model proteins are combined with diverse spectroscopic techniques and computational investigations, allowing us to obtain a comprehensive understanding of how the entire protein matrix contributes to reactivity. These fundamental structure-function-dynamics relationships will be discussed in the context of understanding native metalloenzymes and providing design guidelines for new biological and anthropogenic catalyst development.

Woon Ju Song, Seoul National University, KR

Protein scaffolds offer an expansive platform to construct novel structures and functions of metallocofactors. While numerous methods have been developed to optimize protein environments near or distant from active sites, the design of the first coordination sphere still relies largely on structure-based approaches derived from coordination chemistry. Here, we present two strategies to build first coordination spheres for new functions: (i) pinpointing specific positions for mutations to create divalent transition metal-binding sites by developing an in silico program, Metal-Installer, and (ii) genetically incorporating noncanonical amino acids followed by symmetry-based ligand mulitplication. By integrating geometric parameters derived from both natural metalloproteins and synthetic inorganic complexes, we created tailor-made metal-binding sites that mimic the structure and chemical properties of mono- and dinuclear metalloproteins. These designs closely matched our structural predictions and fulfilled the minimal requirements for metal-dependent catalysis or photochemical properties. This work significantly broadens the accessible chemical space of metalloproteins, enabling the repurposing of natural protein/enzyme scaffolds for various applications such as metalloenzyme mimics, biocatalysts, and protein-based photochemical materials.

Yuzhou Wu, Huazhong University of Science and Technology, CN

Biomanufacturing is a transformative technology with the potential to revolutionize diverse industrial sectors. An underlying limitation in this field is the lack of enzymes capable of driving unnatural transformations necessary for the synthesis of a broad range of chemical products. However, enzyme design is inherently restricted by the availability of structural and functional units provided by nature. This presentation will detail our ongoing efforts to develop new enzymes that incorporate unnatural entities into their scaffolds, with a particular emphasis on photoenzymes that utilize synthetic photosensitizers. I will discuss strategies for incorporating these entities, including genetic encoding approaches and chemical modification techniques, and studying their compatibility with high-throughput directed evolution in living cells. Additionally, I will highlight the unique catalytic properties observed in these artificial enzymes and their potential to be integrated with natural cellular metabolic pathways. We envision that artificial enzymes, enriched with unnatural entities, will serve as versatile tools for constructing "cell factories" adaptable to a variety of synthetic applications.

Nico Bruns, Technical University of Darmstadt, DE

Atom transfer radical reactions are chemical transformations that do not occur naturally in biological systems. Yet, metalloenzymes can catalyze such new-to-nature reactions, thereby paving the way to biocatalytic routes to atom transfer radical polymerizations (bioATRP) and atom transfer radical cyclizations (bioATRC). However, native heme proteins such as myoglobin and horseradish peroxidase display a limited degree of control or activity in these reactions. By rational design, myoglobin mutants that show enhanced performance over their wild-type counterparts in radical polymerizations and greatly enhanced catalytic turnover in radical cyclization reactions were created. Moreover, bioATRP is not only an enzymatic route to polymers but allows the synthesis of polymers in situ in biological systems. This opens up the possibility to engineer living cells on their surface and within their cytosol by biocatalytic radical polymerizations and to create artificial cells that can express their own proteins. Thus, repurposed metalloenzymes play a crucial role in developing novel communicating life-like biomaterials, semi-synthetic engineered living materials, and synthetic biology systems.

Rudi Fasan, University of Texas at Dallas, USA

Expanding the reaction scope of biological catalysts beyond the realm of enzymatic transformations occurring in nature can create new opportunities for the exploitation of biocatalysis for asymmetric synthesis and other applications. In this seminar, I will present recent progress made by our group toward the design and application of engineered cofactor-dependent enzymes for catalyzing ‘new-to-nature’ transformations, with a focus on selective C(sp³)–H functionalization reactions achieved through abiological reactive intermediates and the synergistic integration of enzyme catalysis with chemical hydrogen atom transfer. These systems make available new approaches for the asymmetric construction of carbon-carbon and carbon-heteroatom bonds beyond the scope of currently available biocatalytic or chemocatalytic methods.

Jared C. Lewis, Indiana University, USA

Metalloenzymes perform some of the most remarkable transformations in nature under ambient conditions in complex cellular milieu. The possibility of leveraging molecular recognition and evolution for non-biological metal catalysts has driven efforts to engineer artificial metalloenzymes (ArMs), hybrid catalysts comprised of synthetic metal cofactors linked to protein scaffolds. In this talk, I will discuss recent efforts from my group to design and evolve ArMs containing dirhodium and metal polypyridine cofactors for selective catalysis.

Angela Lombardi, University of Naples Federico II, IT

Metalloenzymes are capable of catalyzing a variety of reactions, and a given metal ion can be used in a number of oxidative, reductive, and hydrolytic transformations in different enzymes. This functional diversity arises from a strong partnership between the metal cofactor and protein matrix: the metal ion provides the protein with an array of chemical properties, while the protein stabilizes it in solution and directs its reactivity toward a unique and distinct path.

Bioinorganic chemists tackled the challenge to unravel the mechanisms that allow the protein matrix to modulate the catalytic activity of metal-containing cofactors, through the development of artificial systems. In this respect, de novo protein design, involving the construction of proteins “from scratch”, has contributed to tremendous advances in manufacturing metalloenzymes with unique structures and functionalities.

This lecture will give an overview of our results on the design of artificial helical bundles, housing different metal cofactors, which catalyze a variety of oxidative reactions. Starting from the Due Ferri (DF) family of artificial diiron-oxo-proteins, our design has integrated rational and computational strategies to engineer mononuclear and dinuclear copper sites, mimicking natural lytic polysaccharide monooxygenases (LPMOs) and polyphenol oxidases.

Alison Narayan, University of Michigan, USA

Chemical methods that facilitate a desired transformation with precise chemo-, site- or stereoselectivity can allow for more efficient synthetic routes, thus expanding the practical-to-target molecules. Biocatalytic methods present the opportunity to develop exquisite catalyst-controlled selectivity, enabling highly streamlined synthetic routes. This is exemplified by nature’s ability to make intricate secondary metabolites with potent biological activity such as taxol and vancomycin. Although biocatalysis has been embraced by industrial chemists for the commercial production of pharmaceutical agents, more work is needed to promote the implementation of biocatalysis in target-oriented synthesis more broadly, including limitations in the breadth of well-developed reactions, the unknown substrate scope of functionally characterized enzymes, and the perceived incompatibility with multistep, preparative-scale sequences. In this talk, the use of iron-dependent enzymes for C–C bond formation, C–C bond cleavage, and C–C bond rotation will be discussed from mechanism to application in the synthesis of complex molecule synthesis.

Osami Shoji, Nagoya University, JP

Cytochrome P450BM3 is a highly efficient heme enzyme, but its native activity is largely restricted to long-chain fatty acids. To overcome this limitation without relying on mutagenesis, we developed a decoy molecule strategy that employs inert compounds mimicking native substrates to trigger oxygen activation. This approach enabled wild-type P450BM3 to hydroxylate abiological substrates such as benzene, propane, and methane. Optimized decoy molecules, particularly those derived from N-acyl amino acids and amino acid dimers, enhanced substrate turnover and, in some cases, promoted crystallization of the enzyme, which facilitated structural analysis. This strategy has also been successfully applied to other P450 enzymes, demonstrating its broader applicability. In addition, by combining the decoy approach with directed evolution, we developed variants capable of utilizing microbial signaling molecules such as N-acyl homoserine lactones as functional decoys. Our decoy molecule strategy offers new opportunities for sustainable and programmable oxidation chemistry.

Cathleen Zeymer, Technical University of Munich, DE

Cerium photoredox catalysis is a powerful method to activate organic molecules under mild conditions. However, it remains a major challenge to achieve stereocontrol in these light-driven radical reactions. We thus developed a cerium-dependent photoenzyme enabling this chemistry in the chiral environment of a de novo protein. Our work is based on a de novo TIM barrel scaffold designed previously in a physics-based approach. We equipped the protein with a high-affinity metal binding site for lanthanide ions and demonstrated its photocatalytic potential. Upon visible-light irradiation, the cerium-bound enzyme enables the radical C–C bond cleavage of 1,2-diols in aqueous solution. To optimize the initially low activity and enantioselectivity, we redesigned the scaffold computationally. We decreased the cavity size between the two independently folded domains and used ProteinMPNN to redesign the sequences. Selected redesigns were characterized experimentally and showed significantly improved k_cat/K_M and enantiomeric excess in kinetic resolutions of diols.

Huimin Zhao, University of Illinois Urbana-Champaign, USA

Enzymes have been increasingly used for practical synthesis of chemicals, fuels, and materials thanks to recent advances in enzyme engineering, synthetic biology, artificial intelligence (AI)/machine learning (ML), and laboratory automation. In this talk, I will discuss our recent effort in designing repurposed enzymes with new-to-nature reactivity for asymmetric synthesis by exploring the synergy between enzymatic catalysis and photocatalysis. The representative new-to-nature photoenzymatic reactions that we have demonstrated so far include but are not limited to intermolecular radical hydroalkylation, intermolecular radical conjugate addition, and intermolecular radical hydroamination. In addition, I will introduce a new strategy to address the scalability issue of these new-to-nature photoenzymatic reactions by directly integrating them into microbial metabolism. Finally, I will highlight the development of machine learning and laboratory automation tools for enzyme discovery and engineering. Taken together, these strategies and tools should greatly accelerate the development of biocatalysts for applications related to human health, energy, and sustainability.

Mary C. Andorfer, Michigan State University, USA

Glycyl radical enzyme (GRE) hydroalkylases use amino acid residue-based radicals to directly convert C(sp³)–H bonds into stereodefined C–C bonds. Despite their potential for biocatalysis, GRE hydroalkylases are underutilized due to the challenge of installing these amino acid-based radicals in vitro. We recently developed a platform that overcomes this limitation, enabling radical generation within purified enzyme. Using this approach, we demonstrated that a wild-type GRE efficiently catalyzes its native reaction—stereoselective toluene addition to fumarate—with high total turnover (>17,000 TTN) and robust catalysis (kcat = 18 s^-1). The glycyl radical species persists in an anaerobic environment without losing activity for up to 11 days. Moreover, we have found that this enzyme accommodates a range of simple hydrocarbons and heterocycles as substrates and that scope and selectivity are tunable through mutagenesis. Beyond native reactions, we have also discovered non-native transformations that are only catalyzed by enzyme variants. This work lays the foundation for developing GRE hydroalkylases as versatile biocatalysts for selective coupling reactions.

Andrew R. Buller, University of Wisconsin-Madison, USA

We report the in vivo synthesis of a new cofactor where cobalt instead of iron is inserted into protoporphyrin. The resulting cofactor can access a meta-stable metal-hydride that can be directed to perform metallo-hydrogen atom transfer chemistry. We report the molecular mechanisms of how this new cofactor is made in E. coli. When this cofactor is inserted into P450 scaffolds, reaction with phenylsilanes access a unique cobalt-hydride intermediate. Synthetic limitations of using phenylsilane as a hydride source are discussed, as well as potential solutions. Evolution for reductive deallylation led to the serendipitous discovery of a novel mode of reductive dearomatization, which was characterized through detailed enzymological study.

Adrian Bunzel, ETH Zurich, CH

The global energy crisis challenges us to develop more sustainable strategies to produce electricity. Given the excellent efficiency of natural photosynthetic complexes, biohybrid photovoltaic devices present an attractive solution for solar energy conversion. However, the low stability and high complexity of natural photoactive systems limit their application in photovoltaics.

Previously, our group has combined computational design and directed evolution to engineer de novo enzymes. Here, we adopted this strategy to create photoenzymes that overcome the limitations of natural photosystems. Photobiocatalysts were designed by introducing photosensitizer binding sites into heme-containing helical bundle proteins. The designed binding sites were highly specific for the target photosensitizer and reached nanomolar ligand affinity. Photosensitizer binding to the protein scaffold improved photostability by at least an order of magnitude, substantially extending the lifetime in a model biohybrid solar cell. By screening directly for photocurrents, photovoltaic activity could be improved 4-fold within just two initial rounds of directed evolution. This promising result suggests that enzyme engineering can yield photocatalysts with activities approaching those of state-of-the-art solar cells.

Our work provides a robust methodological framework for creating photoenzymes addressing critical sustainability challenges, such as solar energy conversion, nitrogen fixation, carbon capture, and hydrogen production.

Dongping Chen, University of Basel, CH

Metal-hydride species are pivotal in organometallic chemistry, enabling a wide range of transformations of unsaturated C=X (X = CR₂, O, NR, etc.) bonds. While metal-catalyzed hydrogen atom transfer (HAT) offers an appealing alternative through an outer-sphere, radical- based pathway, achieving stereoselectivity in homogeneous catalysis remains a formidable challenge. Enzymes provide a blueprint for stereocontrol—thanks to their chiral microenvironments—and have been shown to facilitate radical reactions with remarkable selectivity. However, natural metal-hydride containing enzymes are rare and often exhibit limited catalytic activities compared to synthetic catalysts. To bridge this gap, we explored the integration of abiotic metal-hydrides into biological contexts to achieve enantioselective radical transformations using RepArtZymes (Repurposed and Artificial metalloenzymes).

Artificial metalloenzymes (ArMs) present a fascinating means of introducing non-natural metallic complexes into an evolvable protein scaffold, providing a unique approach to tackle some of great challenges in homogeneous catalysis. A Schiff-base cobalt complex was anchored into a chimeric streptavidin scaffold relying on the biotin-streptavidin supramolecular interaction as well as an axial histidine coordination to the cobalt-ion. The resulting Artificial Radical Cyclase was engineered to catalyze the formation of enantioenriched bicyclic terpenoid scaffolds via a Co–H mediated hydrogen atom transfer radical cyclization.

Repurposed metalloenzymes offer an alternative strategy to achieve enantioselective radical reactions. Building on our finding with Artificial Radical Cyclase and relying on enzyme engineering, we repurposed hemoproteins to catalyze asymmetric abiological radical reactions, proposing a transient Fe–H species as key intermediate. Tailored variants of cytochrome P450s catalyzed enantioselective radical cyclizations through hydrofunctionalization of unactivated alkenes, proceeding via a homolytic metal-hydride HAT mechanism.

These findings highlight the fusion of abiological metal-hydride catalysis with enzymatic stereocontrol, offering new avenues for enantioselective synthesis in biocatalysis and expanding the catalytic repertoire of enzymes.

Nobutaka Fujieda, Osaka Metropolitan University, JP

In recent years, it has become clear that the Diels-Alder reaction occurs in biological system, where the Diels-Alderases catalyze this reaction as a part of metabolic pathway in living organisms. As the enzymes could be applied to the synthesis of desired cycloadducts, they would be very promising as an environmentally-friendly catalyst because of their industrial usefulness. However, such application has not been reached yet due to its high substrate specificity. On the other hand, a variety of approach toward developing artificial metalloenzymes (ArMs) have recently emerged all over the world. ArMs are defined as highly regio- and/or enantioselective catalysts consisting of a protein matrix and a synthetic metal complex. Therefore, ArMs can harness excellent reactivity derived from the metal complexes as well as enzymatic ability such as exquisite chemical environment to accelerate even difficult and desirable chemical reactions.

Our group developed artificial metalloenzymes with a cupin-type protein (TM1459) obtained from hyperthermophile, Thermotoga maritima, where well-defined amino acid residues are disposed around the metal center. This metal binding motif consists of 4- histidine tetrad in a almost identical geometry to that of the tris(2-pyridylmethyl)amine (TPA) ligand system. Using this protein as metal-ligands, we have recently developed the artificial non-heme metalloenzymes with high stereoselectivity by mutating 4 histidines at the metal binding site.

In this study, we screened thus obtained mini-library of mutants for the inverse electron- demand hetero-Diels-Alder reaction (Scheme). As a result, H52A mutants which has 3-his triad, showed high endo-selectively, but low enantioselectivity and yield as well. Therefore, the pose of substrate docked into cavity by in silico simulation suggested that there are some steric repulsion between the substrate and surrounded amino acids. Based on this hypothesis, we constructed the H52G/I108A mutant which showed excellent selectivity and yield. Finally based on this notion, we constructed the Cu-H52G/I108D mutant which showed enhanced selectivity (94 % ee) and yield (92 %). The X-ray crystallographic analysis of this mutant exhibited that copper center would migrate during the catalytic cycle. In addition, further substrate scope was investigated and the series of substrate also showed good stereoselectivity.

Marc Garcia-Borràs, University of Girona, ES

Enzymes catalyze complex chemical reactions with high specificity, efficiency, and selectivity. Many enzymes require cofactors that participate in biocatalytic cycles involving the formation of organometallic, ionic, or radical reactive intermediates. These reactive intermediates can follow divergent reaction pathways, leading to enzymatic promiscuity and providing opportunities for designing new challenging enzyme-catalyzed transformations. Experimental characterization of these “fleeting” reactive intermediates is challenging due to limitations in structural and spectroscopic techniques. However, computational methods offer a powerful alternative for describing these intermediates with atomic-level and real-time precision.

Our research program aims to develop and apply new multiscale computational protocols to study, characterize, and rationally improve the formation and stabilization of highly reactive intermediates in enzyme active sites, directing them towards desired reaction pathways. This can be achieved through designing precision confinement and specific polar environment within the protein’s active site. The ultimate goal is to implement new synthetically useful biocatalytic reactions. Herein, we present recent successful cases in computationally guided enzyme engineering, showcasing how precision confinement and tailored active-site electrostatics can be leveraged to direct enzymatic transformations toward synthetically valuable outcomes. These strategies have led to the successful implementation of metalloenzyme-catalyzed non-natural oxidative processes, carbene and nitrene transfers, and radical-mediated C–H functionalization. Our findings provide atomistic descriptions and fundamental insights into enzymatic catalysis and offer a blueprint for designing enzymes capable of performing novel transformations with potential applications in green chemistry, pharmaceutical synthesis, and sustainable catalysis.

Stephan Hammer, Bielefeld University, DE

Enzyme catalysis is frequently highly efficient and selective, as enzymes precisely control the entire environment in which a reaction takes place. Beyond primary catalytic interactions, enzymes provide a multitude of secondary interactions that can pre-organize substrates (reagents), control conformations of highly reactive intermediates (e.g., radicals and carbocations) in each step of a catalytic cycle and distinguish between competing reaction pathways and transitions states. To me, one of the most exciting questions in catalysis is: What reactions can proteins catalyze that are not readily achieved with other types of catalysts?

Our research group combines approaches from organic chemistry, enzymology and directed evolution to design, evolve, understand, and apply new enzyme function. We develop biocatalysts for sought-after chemical transformations that currently lack efficient catalytic solutions. This includes enzymes for Wacker-type alkene oxidation, regioselective N-alkylation of azoles with “off the shelf” reagents, as well as asymmetric hydration to synthesize chiral alcohols simply from alkenes and water. During this talk, I will discuss unpublished examples of directed evolution, mechanistic studies and applications of new biocatalysts developed in our research group.

Xiaoqiang Huang, Nanjing University, CN

Thiamine diphosphate(ThDP)-dependent enzymes are versatile biocatalysts that naturally facilitate asymmetric C−C forming/breaking through the enzymatic Breslow intermediate via a 2-electron umpolung polar mechanism. Recently, our group has repurposed ThDP-dependent enzymes for non-natural asymmetric radical transformations triggered by single-electron radical pathways. Mechanistically, an active site ketyl radical is generated by the oxidation of the resting enzymatic Breslow intermediate through photoredox catalysis. Concurrently, a prochiral benzylic radical is formed via single-electron trasnfers from an appropriate precursor. These radicals can undergo enantioselective cross-coupling within the active site, leading to the synthesis of chiral ketones. Specifically, we successfully employed N-(acyloxy)phthalimides, which are one-step synthesized from carboxylic acids, as radical precursors. More recently, we designed in situ generation of prochiral radicals through the addition of electron-deficient carbon-centred radicals to olefins, followed by enzymatic acyl radical transfer, achieving a three-component photobiocatalytic transformation. In this talk, I will also present our latest progress, including a triple activation strategy to utilize C-H bonds as radical precursors, as well as a radical repositioning strategy.

Zhen Liu, National Institute of Biological Sciences, CN

Efficient methods for achieving desaturation of α,β-unsaturated carbonyl compounds are highly sought after in organic chemistry. In contrast to traditional synthetic approaches, enzymatic desaturation systems offer the potential for enhanced sustainability and selectivity but have remained elusive. In this talk, I will introduce a versatile and general enzymatic desaturation system based on flavin-dependent ene-reductases for desymmetrizing cyclohexanones. This innovative platform facilitates the synthesis of a wide array of chiral cyclohexenones bearing quaternary stereocenters—structural motifs commonly present in bioactive molecules—with remarkable yields and enantioselectivities. Mechanistic insights into this novel enzymatic desaturation process are provided through a combination of experimental investigations and computational studies. Furthermore, leveraging these insights gained, we have devised an additional biocatalytic strategy for the synthesis of α,β-unsaturated carbonyl compounds by reductively desymmetrizing cyclohexadienones. This method yields the opposite enantiomer compared to our desaturation system, underscoring the complementary nature and broad applicability of our flavin-based desymmetrization approaches.

Jeffrey D. Martell, University of Wisconsin–Madison, USA

We have developed new platforms for directed evolution that merge cellular protein expression systems with chemical synthesis, enabling ultrahigh-throughput selection of enzyme mutants (from libraries of millions) with high activity toward abiotic substrates. In one project area, we are evolving enzymes for enhanced activity in plastic recycling. Existing methods to evolve plastic-degrading enzymes require low-throughput testing, creating a bottleneck in discovery of high-activity mutants. To overcome this limitation, we developed an ultrahigh-throughput platform to evolve polymer-degrading enzymes, combining yeast display with the synthesis of a probe resembling the target polymer chain. We discovered mutants of polyethylene-terephthalate (PET)-degrading enzymes with enhanced activity in degrading bulk plastics. We are applying the platform to diverse enzymes, synthetic polymer chains, and reaction conditions. In another project area, we are using high-throughput evolution with rational protein engineering and mechanistic analysis to discover catalysts with improved activity for generating reactive intermediates, which can be employed for “proximity labeling” inside living cells to map the locations of specific biomolecules. The biocatalysts evolved using our approach also have applications for green synthetic methodology.

Hui-Jie Pan, Nanjing University, CN

Enzymes are nature’s most efficient and selective catalysts, yet their utility is often constrained by the limited scope of reactions they catalyze. To unlock new possibilities, artificial enzymes offer a compelling solution. However, current methods for constructing artificial enzymes typically rely on the irreversible incorporation of non-natural active sites into protein scaffolds, complicating design and severely limiting scaffold diversity. Here, inspired by natural cofactors like NAD(P)⁺, we present a reversible binding strategy for artificial enzyme development. We introduce BpAD, a photoactive general cofactor that can seamlessly integrate into a wide range of NAD⁺-dependent protein scaffolds. This cofactor enables the efficient catalysis of both intermolecular and intramolecular [2+2] cycloaddition reactions with exceptional enantioselectivity, demonstrating a broad substrate scope and remarkable enantiodivergence. Computational studies confirm the precise, dynamic binding of BpAD within these scaffolds and unveil a key exo-attack pathway in the stepwise C–C bond formation mechanism. Furthermore, BpAD displays strong orthogonality with NAD⁺, allowing both cofactors to operate simultaneously without interference. This reversible cofactor binding strategy not only simplifies artificial enzyme design but also opens the door to leveraging a diverse array of protein scaffolds for tailored catalytic applications.

Felix Raps, Princeton University, USA

Translation of small molecule reactivity into biocatalytic frameworks offers access to highly precise catalysts for selective synthesis. These systems can overcome limitations often innate to small molecule catalysts, frequently focusing on stereo- and regiocontrol of transformations. In addition, biocatalysts can offer unexpected results, harnessing emergent mechanisms that in some cases are unprecedented in organic synthesis.

This talk will feature the discovery of an emergent mechanism for a Markovnikov-selective, photoenzymatic hydroamination in a Baeyer-Villiger Monooxygenase that was amplified by directed evolution. The development allowed for the preparation of highly-congested α-tertiary amines difficult to access with traditional methodologies. Mechanistic investigations using small molecule probes, DFT, and QM/MM revealed the behavior within the enzyme active site. Furthermore, the more recent development of a hydroarylase will be presented, focusing on the requirements of the enzyme to facilitate the reaction, as well as studies of the changes to the active site by crystallography and computational tools.

Gerard Roelfes, University of Groningen, NL

Genetic code expansion, in particular stop codon suppression, has proven to be a valuable approach towards artificial enzymes displaying new-to-nature catalytic activity. We have used SCS to introduce non-canonical amino acid, e.g. bipyridyl alanine, that bind a catalytic metal ion, or that function directly as catalytic residue, e.g. para-aminophenylalanine and p-boronophenylalanine.

Now we introduce two new classes of artificial enzymes that have been created by the incorporation of non-canonical amino acids containing a thiophenol-derivative as side chain, such as 4-mercaptophenylalanine (pSHF) and 3-mercaptotyrosine (SHY), at various positions into the protein scaffolds LmrR and RamR.

These thiophenolates were exploited as soft ligands for noble metals, in particular Au(I). The new artificial gold enzymes proved to be active in intramolecular hydramination reactions and heterocyclization reactions. Active enzymes have been characterized structurally as well as kinetically. Currently we are evolving these enzymes towards higher activity and (enantio-)selectivity

Thiophenolates also are attractive for application as nucleophilic catalysts. We found especially some of the RamR-based enzymes to be good biocatalysts for enantioselective intramolecular Morita-Baylis-Hillman reactions. Depending on the position of the pSHF or SHY residue within RamR, up to 50 % ee has been obtained thus far. Further rounds of directed evolution are currently underway.

Zachary Wu, Google DeepMind, UK

AlphaFold 3 has expanded the frontiers of biomolecular modeling, including in the prediction of protein-ligand interactions. This presentation will cover the core principles of AlphaFold 3, and then focus on the methods developed for accurate ligand prediction. Finally, we will present key results showcasing the model's performance, accuracy, and practical utility.

Yuxuan Ye, Westlake University, CN

Flavin-dependent old yellow enzymes (OYEs) are privileged biocatalysts with extensive utilization in both academic and industrial settings. Natural OYEs catalyze the asymmetric reduction of activated C=C bonds using the reduced form of the flavin cofactor. We have successfully harnessed the promiscuity of OYEs to catalyze two underexplored new-to-nature reactions, desaturation and Michael addition, with complementary reactivity and selectivity to existing chemical methods.

Desaturation: Guided by the principle of microscopic reversibility, OYEs have been utilized to facilitate carbonyl dehydrogenation, the reverse process of their native reduction. This biocatalytic desaturation platform has achieved desymmetrizing desaturation of cyclohexanones and site-divergent late-stage functionalization of cyclic ketones. Furthermore, with the merger of photoenzymatic catalysis, selective β-C–H alkylation of carbonyl compounds has been realized.

Michael Addition: OYEs are well-known to catalyze reactions via hydride transfer mechanism (H^–). Recently, OYEs are employed in photoenzymatic reactions via hydrogen atom transfer pathway (H·). We have repurposed OYEs to catalyze asymmetric Michael addition via proton transfer mechanism (H⁺), an underdeveloped activation mode with great potentials. A variety of chiral allenoates and cabonyl compounds with a quaternary stereogenic center have been prepared efficiently.