Enzymes, as biological catalysts, underpin countless biochemical reactions and are indispensable across industries ranging from food production to pharmaceuticals. However, naturally occurring enzymes often fall short of the robustness, specificity, and catalytic efficiency demanded by commercial processes. Traditional approaches to improve these biomolecules rely heavily on directed evolution, a strategy that mimics natural selection by generating vast libraries of enzyme variants through gene mutagenesis and then screening for desirable traits. Although powerful, the scale of mutant libraries—sometimes reaching up to 100 trillion variants—renders screening laborious and expensive.

Addressing these challenges head-on, the SMART system represents a paradigm shift in enzyme screening. Unlike conventional methods that struggle to correlate phenotypic function with genotype efficiently, SMART leverages the power of mRNA display technology. By chemically coupling an enzyme directly to its messenger RNA (mRNA) through a puromycin linker, SMART maintains the physical connection between a variant’s catalytic output and its corresponding genetic information. This unique molecular tether enables precise tracking of enzymes and their encoding sequences simultaneously, facilitating high-throughput functional assessments at an unprecedented single-molecule resolution.

A key innovation within SMART lies in its auxiliary detection unit, which amplifies the readout of enzymatic activity. The research team employed an engineered enzyme, ascorbate peroxidase 2 (APEX2), fused with the scCro DNA-binding domain and tethered to the system via ORC hairpin DNA structures. When the target enzyme catalyzes reactions producing hydrogen peroxide, APEX2 reacts by biotinylating nearby proteins. This biotin tag serves as a handle for magnetic bead-based isolation of active enzyme-mRNA complexes, enabling selective capture and enrichment for downstream analysis.

To evaluate SMART’s capabilities, scientists focused on a yeast-derived oxidase known as SpDAAO. Oxidases are particularly challenging due to the subtle and transient nature of their catalytic outputs. The choice of SpDAAO also aligns with growing interest in D-amino acids for pharmaceutical applications, offering a model system with both industrial and therapeutic relevance. By generating an extensive mutant library targeting the critical 232nd amino acid residue, the team employed iterative cycles of in vitro transcription, translation, enzymatic activity labeling, and next-generation sequencing to map functional landscapes rapidly.

Their experimental data demonstrated remarkable enrichment of active variants after just a single selection round, showcasing SMART’s sensitivity and selectivity. Significantly, the wild-type residue at position 232 was consistently prioritized in enriched pools, validating the platform’s accuracy. Subsequent statistical rigor revealed that minor variants initially indicated by raw sequencing data were likely experimental noise, underscoring the importance of comprehensive bioinformatic scrutiny alongside biochemical assays.

SMART’s integration of mRNA display techniques, advanced sequencing, and enzymatic biotinylation provides a robust framework that potentially transcends oxidase targets. By manipulating the auxiliary enzyme component to suit different enzymatic classes, this platform could broaden its scope to encompass hydrolases, lyases, and beyond. Additionally, its streamlined workflow reduces screening times from months or years to mere days or weeks, dramatically accelerating the pace of enzyme optimization.

The implications of this technology are profound. Industries dependent on enzyme catalysts may soon harness SMART to tailor biocatalysts with enhanced stability, substrate range, or reaction rates, enabling greener chemical processes and novel drug production routes. Moreover, the single-molecule resolution offers insights into enzyme heterogeneity and mechanistic nuances that bulk assays obscure, paving the way for fundamental advances in protein science.

Despite its transformative promise, the team acknowledges that rigorous experimental design and data analysis remain crucial to harness SMART’s full potential. Ensuring reproducibility and minimizing noise will be vital, especially as the platform is adapted to more complex enzymatic systems and real-world substrates. The researchers envision creating standardized protocols and computational tools to streamline SMART’s adoption across laboratories and industries worldwide.

Envisioning a future where enzyme engineering no longer wrestles with the constraints of scale and efficiency, the Nagoya team’s SMART technology stands as a beacon of innovation. Beyond expanding the enzyme repertoire available for biotechnology, this method embodies a new era where molecular precision and rapid iteration converge, unlocking biological functions with unprecedented agility.

The fusion of cutting-edge molecular biology techniques, protein engineering, and bioinformatics underscores the holistic approach required for next-generation biocatalyst development. As researchers continue to refine SMART and explore its myriad applications, the boundaries of what enzymes can achieve—in medicine, sustainability, and manufacturing—are poised to expand dramatically.

This pioneering work, published in ACS Synthetic Biology, not only showcases technological ingenuity but also exemplifies the collaborative spirit bridging academia and interdisciplinary research institutes. Supported by multiple grants and initiatives, the SMART platform exemplifies how sustained investment in scientific innovation translates into tools that could reshape biotechnology’s future landscape.

Subject of Research: Not applicable

Article Title: Harnessing the Power of SMART Single-Molecule Display for Enzyme Evolution: A Focus on Oxidase

News Publication Date: 23-Feb-2026

Web References:
10.1021/acssynbio.5c00968

Image Credits: Hideo Nakano and Jasmina Damnjanović

Keywords: enzyme engineering, directed evolution, mRNA display, single-molecule assay, oxidase, biocatalysis, protein evolution, SMART system, next-generation sequencing, puromycin linker, ascorbate peroxidase 2, biotinylation

At the heart of the study lies the ambitious goal of crafting enzymes capable of catalyzing Kemp elimination reactions—a model reaction of great interest due to its relevance in studying enzyme mechanisms and design strategies. Unlike traditional directed evolution approaches, this work harnesses modular assembly and sophisticated design algorithms to generate entirely new enzymatic backbones, finely tuned active sites, and enhanced catalytic capabilities without reliance on natural enzyme templates.

The initial backbone generation involved an ingenious modular strategy leveraging multiple homologous protein structures. Through precise alignment of five distinct imidazole glycerol-phosphate synthase (IGPS) protein backbones, segments were dissected and recombined at structurally conserved junctions. Computational sequence design refined these chimeric constructs, guided by position-specific scoring matrices to ensure stability and compatibility across fragments, thus generating thousands of candidate backbones tailored for subsequent catalytic modification.

To complement this, the research expanded its scope by mining sequence databases extensively, identifying thousands of IGPS homologues. These sequences underwent clustering and rigorous structural prediction via the latest AlphaFold2 implementations. The high-confidence models then entered a cycle of stability design focused on preserving structural integrity while allowing active site flexibility—a crucial balance for efficient catalysis.

Engineering the catalytic site itself was meticulously executed through theozyme modeling. By incorporating precise geometric constraints derived from quantum chemical calculations, the team employed Rosetta’s Matcher algorithm to embed catalytic residues within the designed scaffolds, optimizing their positioning to mimic transition states effectively. This level of geometrical rigor ensured that the catalytic apparatus would be primed for the Kemp elimination’s mechanistic demands.

Once the initial active sites were integrated, the designed enzymes underwent an exhaustive round of sequence optimization focused on the microenvironment surrounding the ligand and catalytic residues. Utilizing Rosetta’s sequence design within spatial proximity to the active site and leveraging mutational scanning data, the design process strategically narrowed the sequence landscape. A ‘fuzzy’-logic objective function balanced energy considerations, van der Waals interactions, solvation effects, and geometric fidelity, filtering millions of designs to identify promising candidates.

Active-site and core stabilization efforts pushed the designs closer to practical viability. Enumerating low-energetic mutations in the active site and scanning the broader protein structure for positions amenable to beneficial substitutions permitted iterative rounds of refinement. Employing FuncLib calculations allowed for the judicious combination of stabilizing mutations while avoiding deleterious effects, ultimately enhancing the robustness of the enzyme across diverse conditions.

Validation extended into computational dynamical analyses, where the preorganization of the active site was tested through rigid-body minimization simulations absent of ligands. Designs failing to maintain catalytic residue alignment beyond acceptable root-mean-square deviations were discarded, ensuring that functionally viable geometries were preserved intrinsically. Additionally, the congruence between model predictions from different computational approaches was verified to further guarantee reliability.

The iterative optimization of active-site constellations was key to achieving high catalytic efficiency. Targeted mutagenesis simulated via FuncLib—eschewing homologous sequence constraints due to the de novo nature of designs—facilitated exploration of sequence diversity while maintaining stability. The best-performing designs were advanced for experimental validation, narrowing the gap between in silico predictions and laboratory realization.

Protein expression protocols were developed with precision to ensure that the computationally designed enzymes could be produced reliably and at scale. Harnessing bacterial expression systems and affinity purification techniques, the team prepared high-purity samples essential for detailed biochemical characterization and crystallographic analysis, confirming the fidelity of designs from sequence to structure.

Activity assays monitored enzymatic function using spectrophotometric methods detecting product formation, enabling kinetic parameter determination under varying substrate concentrations. The data fitting to Michaelis-Menten kinetics provided insights into catalytic turnover rates and substrate affinity, highlighting the practical effectiveness of the computationally designed enzymes compared to natural counterparts.

Thermal stability assessments using nano differential scanning fluorimetry revealed the robustness of the new enzymes, a critical factor for potential industrial or therapeutic applications. The temperature ramping experiments showcased the engineered proteins’ ability to maintain structural integrity under stress, consistent with the stability enhancements incorporated during design.

Crystallographic studies offered definitive structural validation, with multiple enzyme variants crystallized and their structures solved to resolutions near or below 2.1 Å. These analyses verified the accuracy of the computational models and provided atomic-level insights into active-site architecture, substrate positioning, and dynamic features instrumental for catalysis.

To complement static structural data, extensive molecular dynamics simulations spanning multiple microseconds illuminated the enzymes’ dynamic behaviors in bound and unbound states. Employing enhanced sampling techniques and state-of-the-art force fields, these simulations elucidated substrate binding modes, active-site flexibility, and solvent interactions, painting a comprehensive picture of the catalytic process in motion.

Electrostatic Valence Bond (EVB) simulations further probed the reaction mechanism at a quantum-mechanical/molecular-mechanical interface, distinguishing between reactive substrate conformers and capturing transient states of the Kemp elimination process. These simulations offered quantitative free-energy profiles that correlated closely with experimental activity, underpinning the rationale behind the designed enzymes’ functionality.

Collectively, this multidisciplinary approach—spanning computational modeling, structural biology, biophysical characterization, and dynamic simulations—presents a paradigm shift in enzyme design. The researchers’ ability to computationally generate highly efficient Kemp eliminases from scratch portends transformative impacts on enzyme engineering, enabling custom biocatalysts for diverse chemical transformations without exhaustive laboratory evolution.

By harnessing sophisticated algorithms, thorough validation pipelines, and rigorous biochemical assays, the study sets a new standard for the scope and precision of computer-aided enzyme design. This breakthrough holds promise not only for academic exploration but also for practical applications in sustainable manufacturing, drug development, and synthetic biology, where tailored catalysts can accelerate innovation and reduce environmental impact.

The confluence of modular backbone assembly, advanced design algorithms, and comprehensive dynamic simulations represents a masterclass in modern enzymology, painting a hopeful future where enzyme engineering is limited only by imagination and computational power. This work invites further refinement and expansion, including exploration of other challenging reactions and incorporation of allosteric regulation, ushering in an era of bespoke enzymes crafted entirely by computation.

Subject of Research: Computational design and engineering of high-efficiency Kemp elimination enzymes.

Article Title: Complete computational design of high-efficiency Kemp elimination enzymes.

Article References:
Listov, D., Vos, E., Hoffka, G. et al. Complete computational design of high-efficiency Kemp elimination enzymes. Nature (2025). https://doi.org/10.1038/s41586-025-09136-2

Image Credits: AI Generated