Search

The success of AlphaFold in protein structure prediction highlights the power of data-driven approaches in scientific research. However, developing machine learning models to design and engineer proteins with desirable functions is hampered by limited access to high-quality data sets and experimental feedback. The Critical Assessment of Protein Engineering (CAPE) challenge addresses these issues through a student-focused competition, utilizing cloud computing and biofoundries to lower barriers to entry. CAPE serves as an open platform for community learning, where mutant data sets and design algorithms from past contestants help improve overall performance in subsequent rounds. Through two competition rounds, student participants collectively designed >1500 new mutant sequences, with the best-performing variants exhibiting catalytic activity up to 5-fold higher than the wild-type parent. We envision CAPE as a collaborative platform to engage young researchers and promote computational protein engineering.

Inspired by the properties of natural protein-based biomaterials, protein nanomaterials are increasingly designed with natural or engineered peptides or with protein building blocks. Few examples describe the design of functional protein-based materials for biotechnological applications that can be readily manufactured, are amenable to functionalization, and exhibit robust assembly properties for macroscale material formation. Here, we designed a protein-scaffolding system that self-assembles into robust, macroscale materials suitable for in vitro cell-free applications. By controlling the coexpression in Escherichia coli of self-assembling scaffold building blocks with and without modifications for covalent attachment of cross-linking cargo proteins, hybrid scaffolds with spatially organized conjugation sites are overproduced that can be readily isolated. Cargo proteins, including enzymes, are rapidly cross-linked onto scaffolds for the formation of functional materials. We show that these materials can be used for the in vitro operation of a coimmobilized two-enzyme reaction and that the protein material can be recovered and reused. We believe that this work will provide a versatile platform for the design and scalable production of functional materials with customizable properties and the robustness required for biotechnological applications.

Salicylate plays a pivotal role as a pharmaceutical intermediate in drugs, such as aspirin and lamivudine. The low catalytic efficiency of key enzymes and the inherent toxicity of salicylates to cells pose significant challenges to large-scale microbial production. In this study, we introduced the salicylate synthase Irp9 into an l-phenylalanine-producing Escherichia coli , constructing the shortest salicylate biosynthetic pathway. Subsequent protein engineering increased the catalytic efficiency of Irp9 by 33.5%. Furthermore, by integrating adaptive evolution with transcriptome analysis, we elucidated the crucial mechanism of efflux proteins in salicylate tolerance. The elucidation of this mechanism guided us in the targeted modification of these transport proteins, achieving a reported maximum level of 3.72 g/L of salicylate in a shake flask. This study highlights the importance of efflux proteins for enhancing the productivity of microbial cell factories in salicylate production, which also holds potential for application in the green synthesis of other phenolic acids.

Herein, we report macroscopic materials formed by the assembly of engineered bacterial spores. Spores were engineered by using a T7-driven expression system to display a high density of pH-responsive self-associating proteins on their surface. The engineered surface protein on the spore surface enabled pH-dependent binding at the protein level and enabled the assembly of granular materials. Mechanical properties remained largely constant with changing pH, but erosion stability was pH-dependent in a manner consistent with the pH-dependent interaction between the engineered surface proteins. Our finding utilizes synthetic biology for the design of macroscopic materials and illuminates the impact of coiled-coil interaction across various length scales.

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.

Activity-regulated cytoskeleton-associated protein (Arc) forms virus-like capsids for mRNA transport between neurons. Unlike HIV-1 Group-specific Antigen (Gag), which uses its Nucleocapsid (NC) domain to bind HIV-1 genomic mRNA, mammalian Arc lacks the NC domain, and their direct mRNA binding interactions remain underexplored. This study examined rat Arc's binding to rat Arc 5' UTR (A5U), HIV-1 5' UTR (H5U), and GFP mRNAs, revealing weak binding with no significant preference. Adding the HIV-1 NC domain to rArc's C-terminus significantly improved binding to H5U, while also showing substantial binding to A5U at about 60% of its H5U level and exhibiting twice the affinity for A5U over GFP mRNA. Importantly, rArc-NC binds 3.4 times more A5U and H5U than GST-NC, indicating that rArc NTD-CA aids mRNA binding by HIV-1 NC. These findings suggest a conserved Gag protein-mRNA interaction mechanism, highlighting the potential for developing mRNA delivery systems that combine endogenous Gag NTD-CA with retroviral NC and UTRs.

The design and optimization of metabolic pathways, genetic systems, and engineered proteins rely on high-throughput assays to streamline design-build-test-learn cycles. However, assay development is a time-consuming and laborious process. Here, we create a generalizable approach for the tailored optimization of automated cell-free gene expression (CFE)-based workflows, which offers distinct advantages over in vivo assays in reaction flexibility, control, and time to data. Centered around designing highly accurate and precise transfers on the Echo Acoustic Liquid Handler, we introduce pilot assays and validation strategies for each stage of protocol development. We then demonstrate the efficacy of our platform by engineering transcription factor-based biosensors. As a model, we rapidly generate and assay libraries of 127 MerR and 134 CadR transcription factor variants in 3682 unique CFE reactions in less than 48 h to improve limit of detection, selectivity, and dynamic range for mercury and cadmium detection. This was achieved by assessing a panel of ligand conditions for sensitivity (to 0.1, 1, 10 μM Hg and 0, 1, 10, 100 μM Cd for MerR and CadR, respectively) and selectivity (against Ag, As, Cd, Co, Cu, Hg, Ni, Pb, and Zn). We anticipate that our Echo-based, cell-free approach can be used to accelerate multiple design workflows in synthetic biology.

Cancer therapy can be limited by potential side effects, and bacteria-based living cancer therapeutics have gained scientific interest in recent years. However, the full potential of bacteria as therapeutics has yet to be explored due to engineering challenges. In this study, we present a bacterial device designed to specifically target and eliminate breast cancer cells. We have engineered Escherichia coli ( E. coli ) to bind to HER2 receptors on breast cancer cells while also secreting a toxin, HlyE, which is a pore-forming protein. The binding of E. coli to HER2 is facilitated by a nanobody expressed on the bacteria's surface via the Ag43 autotransporter protein system. Our findings demonstrate that the nanobody efficiently binds to HER2+ cells in vitro , and we have utilized the YebF secretion tag to secrete HlyE and kill the target cancer cells. Overall, our results highlight the potential of our engineered bacteria as an innovative strategy for breast cancer treatment.

Microcin J25 (MccJ25), a lasso peptide antibiotic with a unique structure that resembles the lariat knot, has been a topic of intense interest since its discovery in 1992. The precursor (McjA) contains a leader and a core segment. McjB is a protease activated upon binding to the leader, and McjC converts the core segment into the mature MccJ25. Previous studies suggested that these biosynthetic steps likely proceed in a (nearly) concerted fashion; however, there is only limited information regarding the structural and molecular intricacies of MccJ25 biosynthesis. To close this knowledge gap, we used AlphaFold2 to predict the structure of the precursor (McjA) in complex with its biosynthetic enzymes (McjB and McjC) and queried the critical predicted features by protein engineering. Based on the predicted structure, we designed protein variants to show that McjB can still be functional and form a proficient biosynthetic complex with McjC when its recognition and protease domains were circularly permutated or split into separate proteins. Specific residues important for McjA recognition were also identified, which permitted us to pinpoint a compensatory mutation (McjB M108T ) to restore McjA/McjB interaction that rescued an otherwise nearly nonproductive precursor variant (McjA T-2M ). Studies of McjA, McjB, and McjC have long been mired by them being extremely difficult to handle experimentally, and our results suggest that the AF2 predicted ternary complex structure may serve as a reasonable starting point for understanding MccJ25 biosynthesis. The prediction-validation workflow presented herein combined artificial intelligence and laboratory experiments constructively to gain new insights.

Engineered type I polyketide synthases (type I PKSs) can enable access to diverse polyketide pharmacophores and generate non-natural natural products. However, the promise of type I PKS engineering remains modestly realized at best. Here, we report that ketosynthase (KS) domains, the key carbon-carbon bond-forming catalysts, control which intermediates are allowed to progress along the PKS assembly lines and which intermediates are excluded. Using bimodular PKSs, we demonstrate that KSs can be exquisitely selective for the upstream polyketide substrate while retaining promiscuity for the extender unit that they incorporate. It is then the downstream KS that acts as a gatekeeper to ensure the fidelity of the extender unit incorporation by the upstream KS. We also demonstrate that these findings are not universally applicable; substrate-tolerant KSs do allow engineered polyketide intermediates to be extended. Our results demonstrate the utility for evaluating the KS-induced bottlenecks to gauge the feasibility of engineering PKS assembly lines.

Relaxin-2 is a peptide hormone with important roles in human cardiovascular and reproductive biology. Its ability to activate cellular responses such as vasodilation, angiogenesis, and anti-inflammatory and antifibrotic effects has led to significant interest in using relaxin-2 as a therapeutic for heart failure and several fibrotic conditions. However, recombinant relaxin-2 has a very short serum half-life, limiting its clinical applications. Here, we present protein engineering efforts targeting the relaxin-2 hormone in order to increase its serum half-life while maintaining its ability to activate the G protein-coupled receptor RXFP1. To achieve this, we optimized a fusion between relaxin-2 and an antibody Fc fragment, generating a version of the hormone with a circulating half-life of around 3 to 5 days in mice while retaining potent agonist activity at the RXFP1 receptor both in vitro and in vivo.

Protein engineering through directed evolution and (semi)rational approaches is routinely applied to optimize protein properties for a broad range of applications in industry and academia. The multitude of possible variants, combined with limited screening throughput, hampers efficient protein engineering. Data-driven strategies have emerged as a powerful tool to model the protein fitness landscape that can be explored in silico , significantly accelerating protein engineering campaigns. However, such methods require a certain amount of data, which often cannot be provided, to generate a reliable model of the fitness landscape. Here, we introduce MERGE, a method that combines direct coupling analysis (DCA) and machine learning (ML). MERGE enables data-driven protein engineering when only limited data are available for training, typically ranging from 50 to 500 labeled sequences. Our method demonstrates remarkable performance in predicting a protein's fitness value and rank based on its sequence across diverse proteins and properties. Notably, MERGE outperforms state-of-the-art methods when only small data sets are available for modeling, requiring fewer computational resources, and proving particularly promising for protein engineers who have access to limited amounts of data.

Protein synthesis methods have been adapted to incorporate an ever-growing level of non-natural components. Meanwhile, design of de novo protein structure and function has rapidly emerged as a viable capability. Yet, these two exciting trends have yet to intersect in a meaningful way. The ability to perform de novo design with non-proteinogenic components requires that synthesis and computation align on common targets and applications. This perspective examines the state of the art in these areas and identifies specific, consequential applications to advance the field toward generalized macromolecule design.

Enzymatic DNA writing technologies based on the template-independent DNA polymerase terminal deoxynucleotidyl transferase (TdT) have the potential to advance DNA information storage. TdT is unique in its ability to synthesize single-stranded DNA de novo but has limitations, including catalytic inhibition by ribonucleotide presence and slower incorporation rates compared to replicative polymerases. We anticipate that protein engineering can improve, modulate, and tailor the enzyme's properties, but there is limited information on TdT sequence-structure-function relationships to facilitate rational approaches. Therefore, we developed an easily modifiable screening assay that can measure the TdT activity in high-throughput to evaluate large TdT mutant libraries. We demonstrated the assay's capabilities by engineering TdT mutants that exhibit both improved catalytic efficiency and improved activity in the presence of an inhibitor. We screened for and identified TdT variants with greater catalytic efficiency in both selectively incorporating deoxyribonucleotides and in the presence of deoxyribonucleotide/ribonucleotide mixes. Using this information from the screening assay, we rationally engineered other TdT homologues with the same properties. The emulsion-based assay we developed is, to the best of our knowledge, the first high-throughput screening assay that can measure TdT activity quantitatively and without the need for protein purification.

The engineering of enzymatic activity generally involves alteration of the protein primary sequences, which introduce structural changes that give rise to functional improvements. Mechanical forces have been used to interrogate protein biophysics, leading to deep mechanistic insights in single-molecule studies. Here, we use simple DNA springs to apply small pulling forces to perturb the active site of a thermostable alcohol dehydrogenase. Methods were developed to enable the study of different spring lengths and spring orientations under bulk catalysis conditions. Tension applied across the active site expanded the binding pocket volume and shifted the preference of the enzyme for longer chain-length substrates, which could be tuned by altering the spring length and the resultant applied force. The substrate specificity changes did not occur when the DNA spring was either severed or rotated by ∼90°. These findings demonstrate an alternative approach in protein engineering, where active site architectures can be dynamically and reversibly remodeled using applied mechanical forces.

Enzyme engineering involves the customization of enzymes by introducing mutations to expand the application scope of natural enzymes. One limitation of that is the complex interaction between two key properties, activity and stability, where the enhancement of one often leads to the reduction of the other, also called the trade-off mechanism. Although dozens of methods that predict the change of protein stability upon mutations have been developed, the prediction of the effect on activity is still in its early stage. Therefore, developing a fast and accurate method to predict the impact of the mutations on enzyme activity is helpful for enzyme design and understanding of the trade-off mechanism. Here, we introduce a novel approach, EnzyACT, a deep learning method that fuses graph technique and protein embedding to predict activity changes upon single or multiple mutations. Our model combines graph-based techniques and language models to predict the activity changes. Moreover, EnzyACT is trained on a new curated data set including both single- and multiple-point mutations. When benchmarked on multiple independent data sets, it shows uniform performance on problems affected by mutations. This work also provides insights into the impact of distant mutations within activity design, which could also be useful for predicting catalytic residues and developing improved enzyme-engineering strategies.

Proteins implement many useful functions, including binding ligands with unparalleled affinity and specificity, catalyzing stereospecific chemical reactions, and directing cell behavior. Incorporating proteins into materials has the potential to imbue devices with these desirable traits. This review highlights recent advances in creating active materials by genetically fusing a self-assembling protein to a functional protein. These fusion proteins form materials while retaining the function of interest. Key advantages of this approach include elimination of a separate functionalization step during materials synthesis, uniform and dense coverage of the material by the functional protein, and stabilization of the functional protein. This review focuses on macroscale materials and discusses (i) multiple strategies for successful protein fusion design, (ii) successes and limitations of the protein fusion approach, (iii) engineering solutions to bypass any limitations, (iv) applications of protein fusion materials, including tissue engineering, drug delivery, enzyme immobilization, electronics, and biosensing, and (v) opportunities to further develop this useful technique.

Transcription factor (TF)-based biosensors are useful synthetic biology tools for applications in a variety of areas of biotechnology. A major challenge of biosensor circuits is the limited repertoire of identified and well-characterized TFs for applications of interest, in addition to the challenge of optimizing selected biosensors. In this work, we implement the IclR family repressor TF TtgV from Pseudomonas putida DOT-T1E as an indole-derivative biosensor in Escherichia coli . We optimize the genetic circuit utilizing different components, providing insights into biosensor design and expanding on previous studies investigating this TF. We discover novel physiologically relevant ligands of TtgV, such as skatole. The broad specificity of TtgV makes it a useful target for directed evolution and protein engineering toward desired specificity. TtgV, as an indole-derivative biosensor, is a promising genetic component for the detection of compounds with biological activities relevant to health and the gut microbiome.

Interactions among biomacromolecules, predominantly noncovalent, underpin biological processes. However, recent advancements in biospecific chemistry have enabled the creation of specific covalent bonds between biomolecules, both in vitro and in vivo. This Review traces the evolution of biospecific chemistry in proteins, emphasizing the role of genetically encoded latent bioreactive amino acids. These amino acids react selectively with adjacent natural groups through proximity-enabled bioreactivity, enabling targeted covalent linkages. We explore various latent bioreactive amino acids designed to target different protein residues, ribonucleic acids, and carbohydrates. We then discuss how these novel covalent linkages can drive challenging protein properties and capture transient protein-protein and protein-RNA interactions in vivo. Additionally, we examine the application of covalent peptides as potential therapeutic agents and site-specific conjugates for native antibodies, highlighting their capacity to form stable linkages with target molecules. A significant focus is placed on proximity-enabled reactive therapeutics (PERx), a pioneering technology in covalent protein therapeutics. We detail its wide-ranging applications in immunotherapy, viral neutralization, and targeted radionuclide therapy. Finally, we present a perspective on the existing challenges within biospecific chemistry and discuss the potential avenues for future exploration and advancement in this rapidly evolving field.

Phosphate plays a vital role in spider silk spinning and has been utilized in numerous artificial silk spinning attempts to replicate the remarkable mechanical properties of natural silk fiber. Its application in artificial processes has, however, yielded varying outcomes. It is thus necessary to investigate the origins and mechanisms behind these differences. By using recombinant silk protein SC-ADF3 derived from the garden spider Araneus diadematus , here, we describe its conformational changes under various conditions, elucidating the effect of phosphate on SC-ADF3 silk protein properties and interactions. Our results demonstrate that elevated phosphate levels induce the irreversible conformational conversion of SC-ADF3 from random coils to β-sheet structures, leading to decreased protein solubility over time. Furthermore, exposure of SC-ADF3 to phosphate stiffens already formed structures and reduces the ability to form new interactions. Our findings offer insights into the underlying mechanism through which phosphate-induced β-sheet structures in ADF3-related silk proteins impede fiber formation in the subsequent phases. From a broader perspective, our studies emphasize the significance of silk protein conformation for functional material formation, highlighting that the formation of β-sheet structures at the initial stages of protein assembly will affect the outcome of material forming processes.

The mono(2-hydroxyethyl) terephthalate hydrolase (MHETase) from Ideonella sakaiensis carries out the second step in the enzymatic depolymerization of poly(ethylene terephthalate) (PET) plastic into the monomers terephthalic acid (TPA) and ethylene glycol (EG). Despite its potential industrial and environmental applications, poor recombinant expression of MHETase has been an obstacle to its industrial application. To overcome this barrier, we developed an assay allowing for the medium-throughput quantification of MHETase activity in cell lysates and whole-cell suspensions, which allowed us to screen a library of engineered variants. Using consensus design, we generated several improved variants that exhibit over 10-fold greater whole-cell activity than wild-type (WT) MHETase. This is revealed to be largely due to increased soluble expression, which biochemical and structural analysis indicates is due to improved protein folding.

Cas12a is a widely used programmable nuclease for genome editing across a variety of organisms, but its application is limited by its PAM recognition restriction. To alleviate these PAM constraints, protein engineering efforts have been applied to expand the PAM recognition range. In this study, we designed and constructed 990 synthetic hybrid Cas12a chimeras through domain shuffling and screened an efficient hybrid Cas12a (ehCas12a) that could recognize a broad range PAM of 5'-TYYN-3' (Y is T or C and N is A, T, C, or G). Furthermore, we constructed an ehCas12a variant, ehCas12a RRVR (T167R/N572R/K578V/N582R), with expanded PAM preference to 5'-TNYN, TWRV-3' (W is A or T, R is A or G, and V is A, C, or G), which can efficiently recognize -2* A/G PAMs that are barely recognized by Cas12a-type proteins and their mutants. Finally, we demonstrated that the DNase-inactivated ehCas12a RRVR base editor (dehCas12a RRVR-BE) was capable of targeting noncanonical PAMs in vivo and disease-related loci for potential therapeutic applications. Overall, our findings highlight the modular design and reconfiguration of Cas proteins for enhanced functionality.

Post-translational modifications (PTMs) such as phosphorylation and dephosphorylation can rapidly alter protein surface chemistry and structural conformation, which can switch protein-protein interactions (PPIs) within signaling networks. Recently, de novo -designed phosphorylation-responsive protein switches have been created that harness kinase- and phosphatase-mediated phosphorylation to modulate PPIs. PTM-driven protein switches are promising tools for investigating PTM dynamics in living cells, developing biocompatible nanodevices, and engineering signaling pathways to program cell behavior. However, little is known about the physical and kinetic constraints of PTM-driven protein switches, which limits their practical application. In this study, we present a framework to evaluate two-component PTM-driven protein switches based on four performance metrics: effective concentration, dynamic range, response time, and reversibility. Our computational models reveal an intricate relationship between the binding kinetics, phosphorylation kinetics, and switch concentration that governs the sensitivity and reversibility of PTM-driven protein switches. Building upon the insights of the interaction modeling, we built and evaluated novel phosphorylation-driven protein switches consisting of phosphorylation-sensitive coiled coils as sensor domains fused to fluorescent proteins as actuator domains. By modulating the phosphorylation state of the switches with a specific protein kinase and phosphatase, we demonstrate fast, reversible transitions between "on" and "off" states. The response of the switches linearly correlated to the kinase concentration, demonstrating its potential as a biosensor for kinase measurements in real time. As intended, the switches responded to specific kinase activity with an increase in the fluorescence signal and our model could be used to distinguish between two mechanisms of switch activation: dimerization or a structural rearrangement. The protein switch kinetics model developed here should enable PTM-driven switches to be designed with ideal performance for specific applications.

Despite their diverse functions, proteins are inherently constructed from a limited set of building blocks. These compositional constraints pose significant challenges to protein research and its practical applications. Strategically manipulating the cellular protein synthesis system to incorporate novel building blocks has emerged as a critical approach for overcoming these constraints in protein research and application. In the past two decades, the field of genetic code expansion (GCE) has achieved significant advancements, enabling the integration of numerous novel functionalities into proteins across a variety of organisms. This technological evolution has paved the way for the extensive application of genetic code expansion across multiple domains, including protein imaging, the introduction of probes for protein research, analysis of protein-protein interactions, spatiotemporal control of protein function, exploration of proteome changes induced by external stimuli, and the synthesis of proteins endowed with novel functions. In this comprehensive Review, we aim to provide an overview of cellular and biophysical applications that have employed GCE technology over the past two decades.

Chalcone synthase (CHS) catalyzes the rate-limiting step of (2 S )-naringenin (the essential flavonoid skeleton) biosynthesis. Improving the activity of the CHS by protein engineering enhances (2 S )-naringenin production by microbial fermentation and can facilitate the production of valuable flavonoids. A (2 S )-naringenin biosensor based on the TtgR operon was constructed in Escherichia coli and its detection range was expanded by promoter optimization to 0-300 mg/L, the widest range for (2 S )-naringenin reported. The high-throughput screening scheme for CHS was established based on this biosensor. A mutant, Sj CHS1 S208N with a 2.34-fold increase in catalytic activity, was discovered by directed evolution and saturation mutagenesis. A pathway for de novo biosynthesis of (2 S )-naringenin by Sj CHS1 S208N was constructed in Saccharomyces cerevisiae , combined with CHS precursor pathway optimization, increasing the (2 S )-naringenin titer by 65.34% compared with the original strain. Fed-batch fermentation increased the titer of (2 S )-naringenin to 2513 ± 105 mg/L, the highest reported so far. These findings will facilitate efficient flavonoid biosynthesis and further modification of the CHS in the future.

Immunogenicity is a major caveat of protein therapeutics. In particular, the long-term administration of protein therapeutic agents leads to the generation of antidrug antibodies (ADAs), which reduce drug efficacy while eliciting adverse events. One promising solution to this issue is the use of mirror-image proteins consisting of d-amino acids, which are resistant to proteolytic degradation in immune cells. We have recently reported the chemical synthesis of the enantiomeric form of the variable domain of the antibody heavy chain (d-VHH). However, identifying mirror-image antibodies capable of binding to natural ligands remains challenging. In this study, we developed a novel screening platform to identify a d-VHH specific for vascular endothelial growth factor A (VEGF-A). We performed mirror-image screening of two newly constructed synthetic VHH libraries displayed on T7 phage and identified VHH sequences that effectively bound to the mirror-image VEGF-A target (d-VEGF-A). We subsequently synthesized a d-VHH candidate that preferentially bound the native VEGF-A (l-VEGF-A) with submicromolar affinity. Furthermore, immunization studies in mice demonstrated that this d-VHH elicited no ADAs, unlike its corresponding l-VHH. Our findings highlight the utility of this novel d-VHH screening platform in the development of protein therapeutics exhibiting both reduced immunogenicity and improved efficacy.

The SpyCatcher/SpyTag system is a protein pair that forms a covalent isopeptide bond without an additional energy supply. The ability to connect fused proteins makes this system an attractive tool for several protein engineering applications. Conditional activation of the SpyCatcher/SpyTag complex formation further expands the use of this system. Here, we evaluated the pH activation of SpyTag using alkoxyaspartic acids in the isopeptide-forming residue. We found that a peptide with an ethoxy group can be activated by hydrolysis under high pH conditions. However, the hydrolysis induces isoaspartate (isoAsp) formation, which is confirmed by an isoAsp-inserted short peptide. We overcame this problem by changing the C-terminal side of the aspartic acid position to Pro, which does not form isoAsp under high pH conditions. The findings of this study provide fundamental knowledge of the synthetic construction of the modified SpyTag peptide.

Polyhydroxyalkanoate (PHA) synthases (PhaCs) are useful and versatile tools for the production of aliphatic polyesters. Here, the chimeric PHA synthase PhaC AR was engineered to increase its capacity to incorporate unusual 6-hydroxyhexanoate (6HHx) units. Mutations at positions 149 and 314 in PhaC AR were previously found to increase the incorporation of an analogous natural monomer, 3-hydroxyhexanoate (3HHx). We attempted to repurpose the mutations to produce 6HHx-containing polymers. Site-directed saturation mutants at these positions were applied for P(3HB- co -6HHx) synthesis in Escherichia coli . As a result, the N149D and F314Y mutants effectively increased the 6HHx fraction. Moreover, the pairwise NDFY mutation further increased the 6HHx fraction, which reached 22 mol %. This increase was presumably caused by altered enzyme activity rather than altered expression levels, as assessed based on immunoblot analysis. The glass transition temperature and crystallinity of P(3HB- co -6HHx) decreased as the 6HHx fraction increased.

Protein engineering faces challenges in finding optimal mutants from a massive pool of candidate mutants. In this study, we introduce a deep-learning-based data-efficient fitness prediction tool to steer protein engineering. Our methodology establishes a lightweight graph neural network scheme for protein structures, which efficiently analyzes the microenvironment of amino acids in wild-type proteins and reconstructs the distribution of the amino acid sequences that are more likely to pass natural selection. This distribution serves as a general guidance for scoring proteins toward arbitrary properties on any order of mutations. Our proposed solution undergoes extensive wet-lab experimental validation spanning diverse physicochemical properties of various proteins, including fluorescence intensity, antigen-antibody affinity, thermostability, and DNA cleavage activity. More than 40% of ProtLGN-designed single-site mutants outperform their wild-type counterparts across all studied proteins and targeted properties. More importantly, our model can bypass the negative epistatic effect to combine single mutation sites and form deep mutants with up to seven mutation sites in a single round, whose physicochemical properties are significantly improved. This observation provides compelling evidence of the structure-based model's potential to guide deep mutations in protein engineering. Overall, our approach emerges as a versatile tool for protein engineering, benefiting both the computational and bioengineering communities.

Bispecific antibodies (BsAbs) represent an emerging class of biologics that can recognize two different antigens or epitopes. T-cell engagers (TcEs) bind two targets in trans on the cell surface of the effector and target cell to induce proximal immune effects, opening exciting windows for immunotherapies. To date, the engineering of BsAbs has been mainly focused on tuning the molecular weight and valency. However, the effects of spatial factors on the biological functions of BsAbs have been less explored due to the lack of biochemical methods to precisely manipulate protein geometry. Here, we studied the geometric effects of the TcEs. First, by genetically inserting rigidly designed ankyrin repeat proteins into TcEs, we revealed that the efficacy progressively decreased as the spacer distance of the two binding domains increased. Then, we constructed 26 pairs of TcEs with the same size but varying orientations using click chemistry-mediated conjugation at different mutation sites. We found that linear ligation sites play a minor role in modulating cell-killing efficacy. Next, we rendered the TcEs' advanced topology by cyclization chemistry using the SpyTag/SpyCatcher pair or sortase ligation approaches. Cyclized TcEs were generally more potent than their linear counterparts. Particularly, sortase A cyclized TcEs, bearing a minimal tagging motif, exhibited better cell-killing efficacy in vitro and improved stability both in vitro and in vivo compared to the linear TcE. This work combines modern bioconjugation chemistry and protein engineering tools for antibody engineering, shedding light on the elusive spatial factors of BsAbs functionality.

The acquisition of a thermostable enzyme is an indispensable prerequisite for its successful implementation in industrial applications and the development of novel functionalities. Various protein engineering approaches, including rational design, semirational design, and directed evolution, have been employed to enhance thermostability. However, all of these approaches require sensitive and reliable high-throughput screening (HTS) technologies to efficiently and rapidly identify variants with improved properties. While numerous reviews focus on modification strategies for enhancing enzyme thermostability, there is a dearth of literature reviewing HTS methods specifically aimed at this objective. Herein, we present a comprehensive overview of various HTS methods utilized for modifying enzyme thermostability across different screening platforms. Additionally, we highlight significant recent examples that demonstrate the successful application of these methods. Furthermore, we address the technical challenges associated with HTS technologies used for screening thermostable enzyme variants and discuss valuable perspectives to promote further advancements in this field. This review serves as an authoritative reference source offering theoretical support for selecting appropriate screening strategies tailored to specific enzymes with the aim of improving their thermostability.

The tolerance of the translation apparatus toward noncanonical amino acids (ncAAs) has enabled the creation of diverse natural-product-like peptide libraries using mRNA display for use in drug discovery. Typical experiments testing for ribosomal ncAA incorporation involve radioactive end point assays to measure yield alongside mass spectrometry experiments to validate incorporation. These end point assays require significant postexperimental manipulation for analysis and prevent higher throughput analysis and optimization experiments. Continuous assays for in vitro translation involve the synthesis of fluorescent proteins which require the full complement of canonical AAs for function and are therefore of limited utility for testing of ncAAs. Here, we describe a new, continuous fluorescence assay for in vitro translation based on detection of a short peptide tag using an affinity clamp protein, which exhibits changes in its fluorescent properties upon binding. Using this assay in a 384-well format, we were able to validate the incorporation of a variety of ncAAs and also quickly test for the codon reading specificities of a variety of Escherichia coli tRNAs. This assay enables rapid assessment of ncAAs and optimization of translation components and is therefore expected to advance the engineering of the translation apparatus for drug discovery and synthetic biology.

The development of solid biomaterials has rapidly progressed in recent years in applications in bionanotechnology. The immobilization of proteins, such as enzymes, within protein crystals is being used to develop solid catalysts and functionalized materials. However, an efficient method for encapsulating protein assemblies has not yet been established. This work presents a novel approach to displaying protein cages onto a crystalline protein scaffold using in-cell protein crystal engineering. The polyhedra crystal (PhC) scaffold, which displays a ferritin cage, was produced by coexpression of polyhedrin monomer (PhM) and H1-ferritin (H1-Fr) monomer in Escherichia coli . The H1-tag is derived from the H1-helix of PhM. Our technique represents a unique strategy for immobilizing protein assemblies onto in-cell protein crystals and is expected to contribute to various applications in bionanotechnology.

Although recent advances in deep learning approaches for protein engineering have enabled quick prediction of hot spot residues improving protein solubility, the predictions do not always correspond to an actual increase in solubility under experimental conditions. Therefore, developing methods that rapidly confirm the linkage between computational predictions and empirical results is essential to the success of improving protein solubility of target proteins. Here, we present a simple hybrid approach to computationally predict hot spots possibly improving protein solubility by sequence-based analysis and empirically explore valuable mutants using split GFP as a reporter system. Our approach, Consen sus design S oluble Mutant Screen ing ( ConsenSing ), utilizes consensus sequence prediction to find hot spots for improvement of protein solubility and constructs a mutant library using Darwin assembly to cover all possible mutations in one pot but still keeps the library as compact as possible. This approach allowed us to identify multiple mutants of Escherichia coli lysine decarboxylase, LdcC, with substantial increases in soluble expression. Further investigation led us to pinpoint a single critical residue for the soluble expression of LdcC and unveiled its mechanism for such improvement. Our approach demonstrated that following a protein's natural evolutionary path provides insights to improve protein solubility and/or increase protein expression by a single residue mutation, which can significantly change the profile of protein solubility.

l-Histidine is an essential proteinogenic amino acid in food with extensive applications in the pharmaceutical field. Herein, we constructed a Corynebacterium glutamicum recombinant strain for efficient biosynthesis of l-histidine. First, to alleviate the l-histidine feedback inhibition, the ATP phosphoribosyltransferase mutant HisG T235P-Y56M was constructed based on molecular docking and high-throughput screening, resulting in the accumulation of 0.83 g/L of l-histidine. Next, we overexpressed rate-limiting enzymes including HisG T235P-Y56M and PRPP synthetase and knocked out the pgi gene in the competing pathway, which increased the l-histidine production to 1.21 g/L. Furthermore, the energy status was optimized by decreasing the reactive oxygen species level and enhancing the supply of adenosine triphosphate, reaching a titer of 3.10 g/L in a shake flask. The final recombinant strain produced 5.07 g/L of l-histidine in a 3 L bioreactor, without the addition of antibiotics and chemical inducers. Overall, this study developed an efficient cell factory for l-histidine biosynthesis by combinatorial protein engineering and metabolic engineering.

Here we develop a mechanism of protein optimization using a computational approach known as "genetic programming". We developed an algorithm called Protein Optimization Engineering Tool (POET). Starting from a small library of literature values, the use of this tool allowed us to develop proteins that produce four times more MRI contrast than what was previously state-of-the-art. Interestingly, many of the peptides produced using POET were dramatically different with respect to their sequence and chemical environment than existing CEST producing peptides, and challenge prior understandings of how those peptides function. While existing algorithms for protein engineering rely on divergent evolution, POET relies on convergent evolution and consequently allows discovery of peptides with completely different sequences that perform the same function with as good or even better efficiency. Thus, this novel approach can be expanded beyond developing imaging agents and can be used widely in protein engineering.

Many successful stories in enzyme engineering are based on the creation of randomized diversity in large mutant libraries, containing millions to billions of enzyme variants. Methods that enabled their evaluation with high throughput are dominated by spectroscopic techniques due to their high speed and sensitivity. A large proportion of studies relies on fluorogenic substrates that mimic the chemical properties of the target or coupled enzymatic assays with an optical read-out that assesses the desired catalytic efficiency indirectly. The most reliable hits, however, are achieved by screening for conversions of the starting material to the desired product. For this purpose, functional group assays offer a general approach to achieve a fast, optical read-out. They use the chemoselectivity, differences in electronic and steric properties of various functional groups, to reduce the number of false-positive results and the analytical noise stemming from enzymatic background activities. This review summarizes the developments and use of functional group probes for chemoselective derivatizations, with a clear focus on screening for enzymatic activity in protein engineering.

Yeast surface display is a valuable tool for protein engineering and directed evolution; however, significant variability in the copy number (i.e., avidity) of displayed variants on the yeast cell wall complicates screening and selection campaigns. Here, we report an engineered titratable display platform that modulates the avidity of Aga2-fusion proteins on the yeast cell wall dependent on the concentration of the anhydrotetracycline (aTc) inducer. Our design is based on a genomic Aga1 gene copy and an episomal Aga2-fusion construct both under the control of an aTc-dependent transcriptional regulator that enables stoichiometric and titratable expression, secretion, and display of Aga2-fusion proteins. We demonstrate tunable display levels over 2-3 orders of magnitude for various model proteins, including glucose oxidase enzyme variants, mechanostable dockerin-binding domains, and anti-PDL1 affibody domains. By regulating the copy number of displayed proteins, we demonstrate the effects of titratable avidity levels on several specific phenotypic activities, including enzyme activity and cell adhesion to surfaces under shear flow. Finally, we show that titrating down the display level allows yeast-based binding affinity measurements to be performed in a regime that avoids ligand depletion effects while maintaining small sample volumes, avoiding a well-known artifact in yeast-based binding assays. The ability to titrate the multivalency of proteins on the yeast cell wall through simple inducer control will benefit protein engineering and directed evolution methodology relying on yeast display for broad classes of therapeutic and diagnostic proteins of interest.

Directed evolution, a revolutionary biotechnology in protein engineering, optimizes protein fitness by searching an astronomical mutational space via expensive experiments. The cluster learning-assisted directed evolution (CLADE) efficiently explores the mutational space via a combination of unsupervised hierarchical clustering and supervised learning. However, the initial-stage sampling in CLADE treats all clusters equally despite many clusters containing a large portion of non-functional mutations. Recent statistical and deep learning tools enable evolutionary density modeling to access protein fitness in an unsupervised manner. In this work, we construct an ensemble of multiple evolutionary scores to guide the initial sampling in CLADE. The resulting evolutionary score-enhanced CLADE, called CLADE 2.0, efficiently selects a training set within a small informative space using the evolution-driven clustering sampling. CLADE 2.0 is validated by using two benchmark libraries both having 160,000 sequences from four-site mutational combinations. Extensive computational experiments and comparisons with existing cutting-edge methods indicate that CLADE 2.0 is a new state-of-art tool for machine learning-assisted directed evolution.

Two fundamentally different approaches are routinely used for protein engineering: user-defined mutagenesis and random mutagenesis, each with its own strengths and weaknesses. Here, we invent a unique mutagenesis protocol, which combines the advantages of user-defined mutagenesis and random mutagenesis. The new method, termed the reverse Kunkel method, allows the user to create random mutations at multiple specified regions in a one-pot reaction. We demonstrated the reverse Kunkel method by mimicking the somatic hypermutation in antibodies that introduces random mutations concentrated in complementarity-determining regions. Coupling with the phage display and yeast display selections, we successfully generated dramatically improved antibodies against a model protein and a neurotransmitter peptide in terms of affinity and immunostaining performance. The reverse Kunkel method is especially suitable for engineering proteins whose activities are determined by multiple variable regions, such as antibodies and adeno-associated virus capsids, or whose functional domains are composed of several discontinuous sequences, such as Cas9 and Cas12a.

In addition to its biological function, the stability of a protein is a major determinant for its applicability. Unfortunately, engineering proteins for improved functionality usually results in destabilization of the protein. This so-called stability-function trade-off can be explained by the simple fact that the generation of a novel protein function─or the improvement of an existing one─necessitates the insertion of mutations, i.e. , deviations from the evolutionarily optimized wild-type sequence. In fact, it was demonstrated that gain-of-function mutations are not more destabilizing than other random mutations. The stability-function trade-off is a universal phenomenon during protein evolution that has been observed with completely different types of proteins, including enzymes, antibodies, and engineered binding scaffolds. In this review, we discuss three types of strategies that have been successfully deployed to overcome this omnipresent obstacle in protein engineering approaches: (i) using highly stable parental proteins, (ii) minimizing the extent of destabilization during functional engineering (by library optimization and/or coselection for stability and function), and (iii) repairing damaged mutants through stability engineering. The implementation of these strategies in protein engineering campaigns will facilitate the efficient generation of protein variants that are not only functional but also stable and therefore better-suited for subsequent applications.

Engineering G-protein-coupled receptors (GPCRs) for improved stability or altered function is of great interest, as GPCRs consist of the largest protein family, are involved in many important signaling pathways, and thus, are one of the major drug targets. Here, we report the development of a high-throughput screening method for GPCRs using a reconstituted in vitro transcription-translation (IVTT) system. Human endothelin receptor type-B (ETBR), a class A GPCR that binds endothelin-1 (ET-1), a 21-residue peptide hormone, was synthesized in the presence of nanodisc (ND) composed of a phospholipid, 1-palmitoyl-2-oleoyl- sn -glycero-3-phospho-(1'-rac-glycerol) (POPG). The ET-1 binding of ETBR was significantly reduced or was undetectable when other phospholipids were used for ND preparation. However, when functional ETBR purified from Sf9 cells was reconstituted into NDs, ET-1 binding was observed with two different phospholipids tested, including POPG. These results suggest that POPG likely supports the folding of ETBR into its functional form in the IVTT system. Using the same conditions as ETBR, whose three-dimensional structure has been solved, human endothelin receptor type-A (ETAR), whose three-dimensional structure remains unsolved, was also synthesized in its functional form. By adding POPG-ND to the IVTT system, both ETAR and ETBR were successfully subjected to ribosome display, a method of in vitro directed evolution that facilitates the screening of up to 10 12 mutants. Finally, using a mock library, we showed that ribosome display can be applied for gene screening of ETBR, suggesting that high-throughput screening and directed evolution of GPCRs is possible in vitro.

Escherichia coli BL21 (DE3) is the most widely used host for recombinant protein expression. However, not every protein can be highly expressed in BL21 (DE3), so individual optimization strategies are often required for different proteins, which is time-consuming and difficult to apply rapidly for industrial production. Constructing more hosts is a good choice to enrich protein expression selection. The expression level of T7 RNAP is the core control node of the pET expression system, so regulating its expression level is an effective way of improving the production of difficult-to-express proteins. Various BL21 (DE3)-derived variant hosts with different translation levels of T7 RNAP could be obtained by changing the ribosomal binding site (RBS) sequences of T7 RNAP in a genome. Here, a BL21 (DE3)-derived variant strain library with different RBS sequences of T7 RNAP was constructed using a base editor and CRISPR-Cas9. Notably, the CRISPR-Cas9 system combined with degenerate primers enabled the construction of an RBS library with 87.5% of the theoretical coverage in single editing, which is more convenient and efficient than the use of a base editor. The expression level of a target gene in the variant strain library ranged from 28 to 220% of the parental strain. Furthermore, a high-throughput host-screening platform for recombinant protein production was constructed, which enabled us to obtain the best expression host for certain target proteins in only 3 days. As a proof of concept, the production of all eight difficult-to-express proteins was greatly improved, including autolytic protein, membrane proteins, antimicrobial peptides, and hardly soluble proteins. Among them, the expression of glucose dehydrogenase in the best host exhibited a 298-fold increase compared to the parental strain. This strategy is simple and effective, requires no advanced equipment, and can be carried out in any laboratory.

Here, we enhanced the popular yeast display method by multiple rounds of DNA and protein engineering. We introduced surface exposure-tailored reporters, eUnaG2 and DnbALFA, creating a new platform of C and N terminal fusion vectors. The optimization of eUnaG2 resulted in five times brighter fluorescence and 10 °C increased thermostability than UnaG. The optimized DnbALFA has 10-fold the level of expression of the starting protein. Following this, different plasmids were developed to create a complex platform allowing a broad range of protein expression organizations and labeling strategies. Our platform showed up to five times better separation between nonexpressing and expressing cells compared with traditional pCTcon2 and c-myc labeling, allowing for fewer rounds of selection and achieving higher binding affinities. Testing 16 different proteins, the enhanced system showed consistently stronger expression signals over c-myc labeling. In addition to gains in simplicity, speed, and cost-effectiveness, new applications were introduced to monitor protein surface exposure and protein retention in the secretion pathway that enabled successful protein engineering of hard-to-express proteins. As an example, we show how we optimized the WD40 domain of the ATG16L1 protein for yeast surface and soluble bacterial expression, starting from a nonexpressing protein. As a second example, we show how using the here-presented enhanced yeast display method we rapidly selected high-affinity binders toward two protein targets, demonstrating the simplicity of generating new protein-protein interactions. While the methodological changes are incremental, it results in a qualitative enhancement in the applicability of yeast display for many applications.

The formation of specific protein-protein interactions (PPIs) drive most biological processes. Malfunction of such interactions is the molecular driver of many diseases. Our ability to engineer existing PPIs or create new ones has become a vital research tool. In addition, engineered proteins with new or altered interactions are among the most critical drugs that have been developed in recent years. These include antibodies, cytokines, inhibitors, and others. Here, we provide a perspective on the current status of the methods used to engineer new or altered PPIs. The emergence of the COVID-19 pandemic, which resulted in a worldwide quest to develop specific PPI inhibitors as drugs, provided an up-to-date and state-of-the-art status report on the methodologies for engineering PPIs targeting the interaction of the viral spike protein with its cellular target, ACE2. Multiple, very high affinity binders were generated within a few months using in vitro evolution by itself, or in combination with computational design. The different experimental and computational methods used to block this interaction provide a road map for the future of PPI engineering.

Genetic code expansion is a powerful approach for advancing critical fields such as biological therapeutic discovery. However, the machinery for genetically encoding noncanonical amino acids (ncAAs) is only available in limited plasmid formats, constraining potential applications. In extreme cases, the introduction of two separate plasmids, one containing an orthogonal translation system (OTS) to facilitate ncAA incorporation and a second for expressing a ncAA-containing protein of interest, is not possible due to a lack of the available selection markers. One strategy to circumvent this challenge is to express the OTS and protein of interest from a single vector. For what we believe is the first time in yeast, we describe here several sets of single plasmid systems (SPSs) for performing genetic code manipulation and compare the ncAA incorporation capabilities of these plasmids against the capabilities of previously described dual plasmid systems (DPSs). For both dual fluorescent protein reporters and yeast display reporters tested with multiple OTSs and ncAAs, measured ncAA incorporation efficiencies with SPSs were determined to be equal to efficiencies determined with DPSs. Click chemistry on yeast cells displaying ncAA-containing proteins was also shown to be feasible in both formats, although differences in reactivity between formats suggest the need for caution when using such approaches. Additionally, we investigated whether these reporters would support the separation of yeast strains known to exhibit distinct ncAA incorporation efficiencies. Model sorts conducted with mixtures of two strains transformed with the same SPS or DPS both led to the enrichment of a strain known to support a higher efficiency ncAA incorporation, suggesting that these reporters will be suitable for conducting screens for strains exhibiting enhanced ncAA incorporation efficiencies. Overall, our results confirm that SPSs are well behaved in yeast and provide a convenient alternative to DPSs. SPSs are expected to be invaluable for conducting high-throughput investigations of the effects of genetic or genomic changes on ncAA incorporation efficiency and, more fundamentally, the eukaryotic translation apparatus.

Site-specific protein labeling is important in biomedical research and biotechnology. While many methods allow site-specific protein modification, a straightforward approach for efficient N-terminal protein labeling is not available. We introduce a novel sortase-mediated swapping approach for a one-step site-specific N-terminal labeling with a near-quantitative yield. We show that this method allows rapid and efficient cleavage and simultaneous labeling of the N or C termini of fusion proteins. The method does not require any prior modification beyond the genetic incorporation of the sortase recognition motif. This new approach provides flexibility for protein engineering and site-specific protein modifications.

Developing potent antimicrobials, and platforms for their study and engineering, is critical as antibiotic resistance grows. A high-throughput method to quantify antimicrobial peptide and protein (AMP) activity across a broad continuum would be powerful to elucidate sequence-activity landscapes and identify potent mutants. Yet the complexity of antimicrobial activity has largely constrained the scope and mechanistic bandwidth of AMP variant analysis. We developed a platform to efficiently perform sequence-activity mapping of AMPs via depletion (SAMP-Dep): a bacterial host culture is transformed with an AMP mutant library, induced to intracellularly express AMPs, grown under selective pressure, and deep sequenced to quantify mutant depletion. The slope of mutant growth rate versus induction level indicates potency. Using SAMP-Dep, we mapped the sequence-activity landscape of 170 000 mutants of oncocin, a proline-rich AMP, for intracellular activity against Escherichia coli . Clonal validation supported the platform's sensitivity and accuracy. The mapped landscape revealed an extended oncocin pharmacophore contrary to earlier structural studies, clarified the C-terminus role in internalization, identified functional epistasis, and guided focused, successful synthetic peptide library design, yielding a mutant with 2-fold enhancement in both intracellular and extracellular activity. The efficiency of SAMP-Dep poises the platform to transform AMP engineering, characterization, and discovery.