Consensus protein design without phylogenetic bias

Thermostability enhancement and insight of L-asparaginase from Mycobacterium sp. via consensus-guided engineering

Acrylamide alleviation in food has represented as a critical issue due to its neurotoxic effect on human health. L-Asparaginase (ASNase, EC 3.5.1.1) is considered a potential additive for acrylamide alleviation in food. However, low thermal stability hinders the application of ASNase in thermal food processing. To obtain highly thermal stable ASNase for its industrial application, a consensus-guided approach combined with site-directed saturation mutation (SSM) was firstly reported to engineer the thermostability of Mycobacterium gordonae L-asparaginase (GmASNase). The key residues Gly97, Asn159, and Glu249 were identified for improving thermostability. The combinatorial triple mutant G97T/N159Y/E249Q (TYQ) displayed significantly superior thermostability with half-life values of 61.65 ± 8.69 min at 50 °C and 5.12 ± 1.66 min at 55 °C, whereas the wild-type was completely inactive at these conditions. Moreover, its T_m value increased by 8.59 °C from parent wild-type. Interestingly, TYQ still maintained excellent catalytic efficiency and specific activity. Further molecular dynamics and structure analysis revealed that the additional hydrogen bonds, increased hydrophobic interactions, and favorable electrostatic potential were essential for TYQ being in a more rigid state for thermostability enhancement. These results suggested that our strategy was an efficient engineering approach for improving fundamental properties of GmASNase and offering GmASNase as a potential agent for efficient acrylamide mitigation in food industry. KEY POINTS: • The thermostability of GmASNase was firstly improved by consensus-guided engineering. • The half-life and T_m value of triple mutant TYQ were significantly increased. • Insight on improved thermostability of TYQ was revealed by MD and structure analysis.

Therapeutic enzyme engineering using a generative neural network

Enhancing the potency of mRNA therapeutics is an important objective for treating rare diseases, since it may enable lower and less-frequent dosing. Enzyme engineering can increase potency of mRNA therapeutics by improving the expression, half-life, and catalytic efficiency of the mRNA-encoded enzymes. However, sequence space is incomprehensibly vast, and methods to map sequence to function (computationally or experimentally) are inaccurate or time-/labor-intensive. Here, we present a novel, broadly applicable engineering method that combines deep latent variable modelling of sequence co-evolution with automated protein library design and construction to rapidly identify metabolic enzyme variants that are both more thermally stable and more catalytically active. We apply this approach to improve the potency of ornithine transcarbamylase (OTC), a urea cycle enzyme for which loss of catalytic activity causes a rare but serious metabolic disease.

An adapted consensus protein design strategy for identifying globally optimal biotherapeutics

Biotherapeutic optimization, whether to improve general properties or to engineer specific attributes, is a time-consuming process with uncertain outcomes. Conversely, Consensus Protein Design has been shown to be a viable approach to enhance protein stability while retaining function. In adapting this method for a more limited number of protein sequences, we studied 21 consensus single-point variants from eight publicly available CD3 binding sequences with high similarity but diverse biophysical and pharmacological properties. All single-point consensus variants retained CD3 binding and performed similarly in cell-based functional assays. Using Ridge regression analysis, we identified the variants and sequence positions with overall beneficial effects on developability attributes of the CD3 binders. A second round of sequence generation that combined these substitutions into a single molecule yielded a unique CD3 binder with globally optimized developability attributes. In this first application to therapeutic antibodies, adapted Consensus Protein Design was found to be highly beneficial within lead optimization, conserving resources and minimizing iterations. Future implementations of this general strategy may help accelerate drug discovery and improve success rates in bringing novel biotherapeutics to market.

Comprehensive mutagenesis to identify amino acid residues contributing to the difference in thermostability between two originally thermostable ancestral proteins

Further improvement of the thermostability of inherently thermostable proteins is an attractive challenge because more thermostable proteins are industrially more useful and serve as better scaffolds for protein engineering. To establish guidelines that can be applied for the rational design of hyperthermostable proteins, we compared the amino acid sequences of two ancestral nucleoside diphosphate kinases, Arc1 and Bac1, reconstructed in our previous study. Although Bac1 is a thermostable protein whose unfolding temperature is around 100°C, Arc1 is much more thermostable with an unfolding temperature of 114°C. However, only 12 out of 139 amino acids are different between the two sequences. In this study, one or a combination of amino acid(s) in Bac1 was/were substituted by a residue(s) found in Arc1 at the same position(s). The best mutant, which contained three amino acid substitutions (S108D, G116A and L120P substitutions), showed an unfolding temperature more than 10°C higher than that of Bac1. Furthermore, a combination of the other nine amino acid substitutions also led to improved thermostability of Bac1, although the effects of individual substitutions were small. Therefore, not only the sum of the contributions of individual amino acids, but also the synergistic effects of multiple amino acids are deeply involved in the stability of a hyperthermostable protein. Such insights will be helpful for future rational design of hyperthermostable proteins.

Catalytic mechanism of ancestral L-lysine oxidase assigned by sequence data mining

A large number of protein sequences are registered in public databases such as PubMed. Functionally uncharacterized enzymes are included in these databases, some of which likely have potential for industrial applications. However, assignment of the enzymes remained difficult tasks for now. In this study, we assigned a total of 28 original sequences to uncharacterized enzymes in the FAD-dependent oxidase family expressed in some species of bacteria including Chryseobacterium, Flavobacterium, and Pedobactor. Progenitor sequence of the assigned 28 sequences was generated by ancestral sequence reconstruction, and the generated sequence exhibited L-lysine oxidase activity; thus, we named the enzyme AncLLysO. Crystal structures of ligand-free and ligand-bound forms of AncLLysO were determined, indicating that the enzyme recognizes L-Lys by hydrogen bond formation with R76 and E383. The binding of L-Lys to AncLLysO induced dynamic structural change at a plug loop formed by residues 251 to 254. Biochemical assays of AncLLysO variants revealed the functional importance of these substrate recognition residues and the plug loop. R76A and E383D variants were also observed to lose their activity, and the kcat/Km value of G251P and Y253A mutations were approximately 800- to 1800-fold lower than that of AncLLysO, despite the indirect interaction of the substrates with the mutated residues. Taken together, our data demonstrate that combinational approaches to sequence classification from database and ancestral sequence reconstruction may be effective not only to find new enzymes using databases of unknown sequences but also to elucidate their functions.

A Brief History of De Novo Protein Design: Minimal, Rational, and Computational

Protein design has come of age, but how will it mature? In the 1980s and the 1990s, the primary motivation for de novo protein design was to test our understanding of the informational aspect of the protein-folding problem; i.e., how does protein sequence determine protein structure and function? This necessitated minimal and rational design approaches whereby the placement of each residue in a design was reasoned using chemical principles and/or biochemical knowledge. At that time, though with some notable exceptions, the use of computers to aid design was not widespread. Over the past two decades, the tables have turned and computational protein design is firmly established. Here, I illustrate this progress through a timeline of de novo protein structures that have been solved to atomic resolution and deposited in the Protein Data Bank. From this, it is clear that the impact of rational and computational design has been considerable: More-complex and more-sophisticated designs are being targeted with many being resolved to atomic resolution. Furthermore, our ability to generate and manipulate synthetic proteins has advanced to a point where they are providing realistic alternatives to natural protein functions for applications both in vitro and in cells. Also, and increasingly, computational protein design is becoming accessible to non-specialists. This all begs the questions: Is there still a place for minimal and rational design approaches? And, what challenges lie ahead for the burgeoning field of de novo protein design as a whole?

Partial Consensus Design and Enhancement of Protein Function by Secondary-Structure-Guided Consensus Mutations

Consensus design (CD) is a representative sequence-based protein design method that enables the design of highly functional proteins by analyzing vast amounts of protein sequence data. This study proposes a partial consensus design (PCD) of a protein as a derivative approach of CD. The method replaces the target protein sequence with a consensus sequence in a secondary-structure-dependent manner (i.e., regionally dependent and divided into α-helix, β-sheet, and loop regions). In this study, we generated several artificial partial consensus l-threonine 3-dehydrogenases (PcTDHs) by PCD using the TDH from Cupriavidus necator (CnTDH) as a target protein. Structural and functional analysis of PcTDHs suggested that thermostability would be independently improved when consensus mutations are introduced into the loop region of TDHs. On the other hand, enzyme kinetic parameters (k_cat/K_m) and average productivity would be synergistically enhanced by changing the combination of the mutations-replacement of one region of CnTDH with a consensus sequence provided only negative effects, but the negative effects were nullified when the two regions were replaced simultaneously. Taken together, we propose the hypothesis that there are protein regions that encode individual protein properties, such as thermostability and activity, and that the introduction of consensus mutations into these regions could additively or synergistically modify their functions.

Complete and cooperative in vitro assembly of computationally designed self-assembling protein nanomaterials

Recent advances in computational methods have enabled the predictive design of self-assembling protein nanomaterials with atomic-level accuracy. These design strategies focus exclusively on a single target structure, without consideration of the mechanism or dynamics of assembly. However, understanding the assembly process, and in particular its robustness to perturbation, will be critical for translating this class of materials into useful technologies. Here we investigate the assembly of two computationally designed, 120-subunit icosahedral complexes in detail using several complementary biochemical methods. We found that assembly of each material from its two constituent protein building blocks was highly cooperative and yielded exclusively complete, 120-subunit complexes except in one non-stoichiometric regime for one of the materials. Our results suggest that in vitro assembly provides a robust and controllable route for the manufacture of designed protein nanomaterials and confirm that cooperative assembly can be an intrinsic, rather than evolved, feature of hierarchically structured protein complexes.

Recent advances in computational methods have enabled the predictive design of self-assembling protein nanomaterials with atomic-level accuracy. Here authors investigate the assembly of two computationally designed, 120-subunit icosahedral complexes and find that assembly of each material from its two constituent protein building blocks was highly cooperative.

Protein Engineering Approaches to Enhance Fungal Laccase Production in S. cerevisiae

Laccases secreted by saprotrophic basidiomycete fungi are versatile biocatalysts able to oxidize a wide range of aromatic compounds using oxygen as the sole requirement. Saccharomyces cerevisiae is a preferred host for engineering fungal laccases. To assist the difficult secretion of active enzymes by yeast, the native signal peptide is usually replaced by the preproleader of S. cerevisiae alfa mating factor (MFα1). However, in most cases, only basal enzyme levels are obtained. During directed evolution in S. cerevisiae of laccases fused to the α-factor preproleader, we demonstrated that mutations accumulated in the signal peptide notably raised enzyme secretion. Here we describe different protein engineering approaches carried out to enhance the laccase activity detected in the liquid extracts of S. cerevisiae cultures. We demonstrate the improved secretion of native and engineered laccases by using the fittest mutated α-factor preproleader obtained through successive laccase evolution campaigns in our lab. Special attention is also paid to the role of protein N-glycosylation in laccase production and properties, and to the introduction of conserved amino acids through consensus design enabling the expression of certain laccases otherwise not produced by the yeast. Finally, we revise the contribution of mutations accumulated in laccase coding sequence (CDS) during previous directed evolution campaigns that facilitate enzyme production.

Improved thermostability of creatinase from Alcaligenes Faecalis through non-biased phylogenetic consensus-guided mutagenesis

Background

Enzymatic quantification of creatinine has become an essential method for clinical evaluation of renal function. Although creatinase (CR) is frequently used for this purpose, its poor thermostability severely limits industrial applications. Herein, we report a novel creatinase from Alcaligenes faecalis (afCR) with higher catalytic activity and lower K_M value, than currently used creatinases. Furthermore, we developed a non-biased phylogenetic consensus method to improve the thermostability of afCR.

Results

We applied a non-biased phylogenetic consensus method to identify 59 candidate consensus residues from 24 creatinase family homologs for screening afCR mutants with improved thermostability. Twenty-one amino acids of afCR were selected to mutagenesis and 11 of them exhibited improved thermostability compared to the parent enzyme (afCR-M0). Combination of single-site mutations in sequential screens resulted in a quadruple mutant D17V/T199S/L6P/T251C (M4-2) which showed ~ 1700-fold enhanced half-life at 57 °C and a 4.2 °C higher T₅₀¹⁵ than that of afCR-M0. The mutant retained catalytic activity equivalent to afCR-M0, and thus showed strong promise for application in creatinine detection. Structural homology modeling revealed a wide range of potential molecular interactions associated with individual mutations that contributed to improving afCR thermostability.

Conclusions

Results of this study clearly demonstrated that the non-biased-phylogenetic consensus design for improvement of thermostability in afCR is effective and promising in improving the thermostability of more enzymes.

Protein Sequence Selection Method That Enables Full Consensus Design of Artificial l-Threonine 3-Dehydrogenases with Unique Enzymatic Properties

Exponentially increasing protein sequence data enables artificial enzyme design using sequence-based protein design methods, including full-consensus protein design (FCD). The success of artificial enzyme design is strongly dependent on the nature of the sequences used. Hence, sequences must be selected from databases and curated libraries prepared to enable a successful design by FCD. In this study, we proposed a selection approach regarding several key residues as sequence motifs. We used l-threonine 3-dehydrogenase (TDH) as a model to test the validity of this approach. In the classification, four residues (143, 174, 188, and 214) were used as key residues. We classified thousands of TDH homologous sequences into five groups containing hundreds of sequences. Utilizing sequences in the libraries, we designed five artificial TDHs by FCD. Among the five, we successfully expressed four in soluble form. Biochemical analysis of artificial TDHs indicated that their enzymatic properties vary; half of the maximum measured enzyme activity (t_1/2) and activation energies were distributed from 53 to 65 °C and from 38 to 125 kJ/mol, respectively. The artificial TDHs had unique kinetic parameters, distinct from one another. Structural analysis indicates that consensus mutations are mainly introduced in the secondary or outer shell. The functional diversity of the artificial TDHs is due to the accumulation of mutations that affect their physicochemical properties. Taken together, our findings indicate that our proposed approach can help generate artificial enzymes with unique enzymatic properties.

Sortase mutants with improved protein thermostability and enzymatic activity obtained by consensus design

Staphylococcus aureus sortase A (SaSrtA) is an enzyme that anchors proteins to the cell surface of Gram-positive bacteria. During the transpeptidation reaction performed by SaSrtA, proteins containing an N-terminal glycine can be covalently linked to another protein with a C-terminal LPXTG motif (X being any amino acid). Since the sortase reaction can be performed in vitro as well, it has found many applications in biotechnology. Although sortase-mediated ligation has many advantages, SaSrtA is limited by its low enzymatic activity and dependence on Ca2+. In our study, we evaluated the thermodynamic stability of the SaSrtA wild type and found the enzyme to be stable. We applied consensus analysis to further improve the enzyme's stability while at the same time enhancing the enzyme's activity. As a result, we found thermodynamically improved, more active and Ca2+-independent mutants. We envision that these new variants can be applied in conjugation reactions in low Ca2+ environments.

The use of consensus sequence information to engineer stability and activity in proteins

The goal of protein design is to create proteins that are stable, soluble, and active. Here we focus on one approach to protein design in which sequence information is used to create a "consensus" sequence. Such consensus sequences comprise the most common residue at each position in a multiple sequence alignment (MSA). After describing some general ideas that relate MSA and consensus sequences and presenting a statistical thermodynamic framework that relates consensus and non-consensus sequences to stability, we detail the process of designing a consensus sequence and survey reports of consensus design and characterization from the literature. Many of these consensus proteins retain native biological activities including ligand binding and enzyme activity. Remarkably, in most cases the consensus protein shows significantly higher stability than extant versions of the protein, as measured by thermal or chemical denaturation, consistent with the statistical thermodynamic model. To understand this stability increase, we compare various features of consensus sequences with the extant MSA sequences from which they were derived. Consensus sequences show enrichment in charged residues (most notably glutamate and lysine) and depletion of uncharged polar residues (glutamine, serine, and asparagine). Surprisingly, a survey of stability changes resulting from point substitutions show little correlation with residue frequencies at the corresponding positions within the MSA, suggesting that the high stability of consensus proteins may result from interactions among residue pairs or higher-order clusters. Whatever the source, the large number of reported successes demonstrates that consensus design is a viable route to generating active and in many cases highly stabilized proteins.

Common mechanism of thermostability in small α- and β-proteins studied by molecular dynamics

Protein thermostability is important to evolution, diseases, and industrial applications. Proteins use diverse molecular strategies to achieve stability at high temperature, yet reducing the entropy of unfolding seems required. We investigated five small α-proteins and five β-proteins with known, distinct structures and thermostability (T_m ) using multi-seed molecular dynamics simulations at 300, 350, and 400 K. The proteins displayed diverse changes in hydrogen bonding, solvent exposure, and secondary structure with no simple relationship to T_m . Our dynamics were in good agreement with experimental B-factors at 300 K and insensitive to force-field choice. Despite the very distinct structures, the native-state (300 + 350 K) free-energy landscapes (FELs) were significantly broader for the two most thermostable proteins and smallest for the three least stable proteins in both the α- and β-group and with both force fields studied independently (tailed t-test, 95% confidence level). Our results suggest that entropic ensembles stabilize proteins at high temperature due to reduced entropy of unfolding, viz., ΔG = ΔH - TΔS. Supporting this mechanism, the most thermostable proteins were also the least kinetically stable, consistent with broader FELs, typified by villin headpiece and confirmed by specific comparison to a mesophilic ortholog of Thermus thermophilus apo-pyrophosphate phosphohydrolase. We propose that molecular strategies of protein thermostabilization, although diverse, tend to converge toward highest possible entropy in the native state consistent with the functional requirements. We speculate that this tendency may explain why many proteins are not optimally structured and why molten-globule states resemble native proteins so much.

Consensus Design of an Evolved High-Redox Potential Laccase

Among the broad repertory of protein engineering methods that set out to improve stability, consensus design has proved to be a powerful strategy to stabilize enzymes without compromising their catalytic activity. Here, we have applied an in-house consensus method to stabilize a laboratory evolved high-redox potential laccase. Multiple sequence alignments were carried out and computationally refined by applying relative entropy and mutual information thresholds. Through this approach, an ensemble of 20 consensus mutations were identified, 18 of which were consensus/ancestral mutations. The set of consensus variants was produced in Saccharomyces cerevisiae and analyzed individually, while site directed recombination of the best mutations did not produce positive epistasis. The best single variant carried the consensus-ancestral A240G mutation in the neighborhood of the T2/T3 copper cluster, which dramatically improved thermostability, kinetic parameters and secretion.

Phylogenetic spread of sequence data affects fitness of consensus enzymes: Insights from triosephosphate isomerase

The concept of consensus in multiple sequence alignments (MSAs) has been used to design and engineer proteins previously with some success. However, consensus design implicitly assumes that all amino acid positions function independently, whereas in reality, the amino acids in a protein interact with each other and work cooperatively to produce the optimum structure required for its function. Correlation analysis is a tool that can capture the effect of such interactions. In a previously published study, we made consensus variants of the triosephosphate isomerase (TIM) protein using MSAs that included sequences form both prokaryotic and eukaryotic organisms. These variants were not completely native-like and were also surprisingly different from each other in terms of oligomeric state, structural dynamics, and activity. Extensive correlation analysis of the TIM database has revealed some clues about factors leading to the unusual behavior of the previously constructed consensus proteins. Among other things, we have found that the more ill-behaved consensus mutant had more broken correlations than the better-behaved consensus variant. Moreover, we report three correlation and phylogeny-based consensus variants of TIM. These variants were more native-like than the previous consensus mutants and considerably more stable than a wild-type TIM from a mesophilic organism. This study highlights the importance of choosing the appropriate diversity of MSA for consensus analysis and provides information that can be used to engineer stable enzymes.

Retargeting CD19 Chimeric Antigen Receptor T Cells via Engineered CD19-Fusion Proteins

CD19-targeted chimeric antigen receptor (CAR) T-cells (CAR19s) show remarkable efficacy in the treatment of relapsed/refractory acute lymphocytic leukemia and Non-Hodgkin's lymphoma. However, the use of CAR T-cell therapy against CD19-negative hematological cancers and solid tumors has been challenging. We propose CD19-fusion proteins (CD19-FPs) to leverage the benefits of CAR19s while retargeting this validated cellular therapy to alternative tumor antigens. We demonstrate the ability of a fusion of CD19 extracellular domain (ECD) and a human epidermal growth factor receptor 2 (HER2) single-chain antibody fragment to retarget CAR19s to kill HER2⁺ CD19^- tumor cells. To enhance the modularity of this technology, we engineered a more robust CD19 ECD via deep mutational scanning with yeast display and flow cytometric selections for improved protease resistance and anti-CD19 antibody binding. These enhanced CD19 ECDs significantly increase, and in some cases recover, fusion protein expression while maintaining target antigen affinity. Importantly, CD19-FPs retarget CAR19s to kill tumor cells expressing multiple distinct antigens, including HER2, CD20, EGFR, BCMA, and Clec12A as N- or C-terminal fusions and linked to both antibody fragments and fibronectin ligands. This study provides fundamental insights into CD19 sequence-function relationships and defines a flexible and modular platform to retarget CAR19s to any tumor antigen.

Following the Evolutionary Track of a Highly Specific l-Arginine Oxidase by Reconstruction and Biochemical Analysis of Ancestral and Native Enzymes

Following the evolutionary track of enzymes can help elucidate how enzymes attain their characteristic functions, such as thermal adaptation and substrate selectivity, during the evolutionary process. Ancestral sequence reconstruction (ASR) is effective for following evolutionary processes if sufficient sequence data are available. Selecting sequences from the data to generate a curated sequence library is necessary for the successful design of artificial proteins by ASR. In this study, we tried to follow the evolutionary track of l-arginine oxidase (AROD), a flavin adenine dinucleotide (FAD)-dependent amino acid oxidase (LAAO) that exhibits high specificity for l-arginine. The library was generated by selecting sequences in which the 15th, 50th, 332nd, and 580th residues are Gly, Ser, Trp, and Thr, respectively. We excluded sequences that are either extremely short or long and those with a low degree of sequence identity. Three ancestral ARODs (AncARODn0, AncARODn1, and AncARODn2) were designed using the library. Subsequently, we expressed the ancestral ARODs as well as native Oceanobacter kriegii AROD (OkAROD) in bacteria. AncARODn0 is phylogenetically most remote from OkAROD, whereas AncARODn2 is most similar to OkAROD. Thermal stability was gradually increased by extending AROD sequences back to the progenitor, while the temperature at which the residual activity is half of the maximum measured activity (T _1/2) of AncARODn0 was >20°C higher than that of OkAROD. Remarkably, only AncARODn0 exhibited broad substrate selectivity similar to that of conventional promiscuous LAAO. Taken together, our findings led us to infer that AROD may have evolved from a highly thermostable and promiscuous LAAO.IMPORTANCE In this study, we attempted to infer the molecular evolution of a recently isolated FAD-dependent l-arginine oxidase (AROD) that oxidizes l-arginine to 2-ketoarginine. Utilizing 10 candidate AROD sequences, we obtained a total of three ancestral ARODs. In addition, one native AROD was obtained by cloning one of the candidate ARODs. The candidate sequences were selected utilizing a curation method defined in this study. All the ARODs were successfully expressed in Escherichia coli for analysis of their biochemical functions. The catalytic activity of our bacterially expressed ancestral ARODs suggests that our ASR was successful. The ancestral AROD that is phylogenetically most remote from a native AROD has the highest thermal stability and substrate promiscuity. Our findings led us to infer that AROD evolved from a highly thermostable and promiscuous LAAO. As an application, we can design artificial ARODs with improved functions compared with those of native ones.

Validation and Stabilization of a Prophage Lysin of Clostridium perfringens by Using Yeast Surface Display and Coevolutionary Models

Bacteriophage lysins are compelling antimicrobial proteins whose biotechnological utility and evolvability would be aided by elevated stability. Lysin catalytic domains, which evolved as modular entities distinct from cell wall binding domains, can be classified into one of several families with highly conserved structure and function, many of which contain thousands of annotated homologous sequences. Motivated by the quality of these evolutionary data, the performance of generative protein models incorporating coevolutionary information was analyzed to predict the stability of variants in a collection of 9,749 multimutants across 10 libraries diversified at different regions of a putative lysin from a prophage region of a Clostridium perfringens genome. Protein stability was assessed via a yeast surface display assay with accompanying high-throughput sequencing. Statistical fitness of mutant sequences, derived from second-order Potts models inferred with different levels of sequence homolog information, was predictive of experimental stability with areas under the curve (AUCs) ranging from 0.78 to 0.85. To extract an experimentally derived model of stability, a logistic model with site-wise score contributions was regressed on the collection of multimutants. This achieved a cross-validated classification performance of 0.95. Using this experimentally derived model, 5 designs incorporating 5 or 6 mutations from multiple libraries were constructed. All designs retained enzymatic activity, with 4 of 5 increasing the melting temperature and with the highest-performing design achieving an improvement of +4°C.IMPORTANCE Bacteriophage lysins exhibit high specificity and activity toward host bacteria with which the phage coevolved. These properties of lysins make them attractive for use as antimicrobials. Although there has been significant effort to develop platforms for rapid lysin engineering, there have been numerous shortcomings when pursuing the ultrahigh throughput necessary for the discovery of rare combinations of mutations to improve performance. In addition to validation of a putative lysin and stabilization thereof, the experimental and computational methods presented here offer a new avenue for improving protein stability and are easily scalable to analysis of tens of millions of mutations in single experiments.

Monomerization of far-red fluorescent proteins

Anthozoa-class red fluorescent proteins (RFPs) are frequently used as biological markers, with far-red (λem ∼ 600-700 nm) emitting variants sought for whole-animal imaging because biological tissues are more permeable to light in this range. A barrier to the use of naturally occurring RFP variants as molecular markers is that all are tetrameric, which is not ideal for cell biological applications. Efforts to engineer monomeric RFPs have typically produced dimmer and blue-shifted variants because the chromophore is sensitive to small structural perturbations. In fact, despite much effort, only four native RFPs have been successfully monomerized, leaving the majority of RFP biodiversity untapped in biomarker development. Here we report the generation of monomeric variants of HcRed and mCardinal, both far-red dimers, and describe a comprehensive methodology for the monomerization of red-shifted oligomeric RFPs. Among the resultant variants is mKelly1 (emission maximum, λem = 656 nm), which, along with the recently reported mGarnet2 [Matela G, et al. (2017) Chem Commun (Camb) 53:979-982], forms a class of bright, monomeric, far-red FPs.

The CaspBase: a curated database for evolutionary biochemical studies of caspase functional divergence and ancestral sequence inference

Sequence databases are powerful tools for the contemporary scientists' toolkit. However, most functional annotations in public databases are determined computationally and are not verified by a human expert. While hypotheses generated from computational studies are now amenable to experimentation, the quality of the results relies on the quality of input data. We developed the CaspBase to expedite high-quality dataset compilation of annotated caspase sequences, to maximize phylogenetic signal, and to reduce the noise contributed from public databanks. We describe our methods of curation for the CaspBase and how researchers can acquire sequences from CaspBase.org. Our immediate goal for developing the CaspBase was to optimize the ancestral protein reconstruction (APR) of caspases, and we demonstrate the utility of the CaspBase in APR studies. We also developed the Common Position (CP) system for comparing human caspase family paralogs and suggest the CP system as an update to current reporting methods of caspase amino acid positions. We present a standardized multiple sequence alignment (MSA) for the CP system and show the advantage of using large databases such as the CaspBase in defining structural positions in proteins. Although the results described here pertain to caspase evolution and structure-function studies, the methods can be adapted to any gene family.

Consensus design for improved thermostability of lipoxygenase from Anabaena sp. PCC 7120

Background

Lipoxygenase (LOX) from Anabaena sp. PCC 7120 (Ana-rLOX) offers important applications in the food industry, especially for improving aroma and dough rheological properties. However, industrial applications of LOXs have been limited by their poor thermostability. Herein, we report a bioinformatics-based consensus concept approach for the engineering of thermostable Ana-rLOX.

Results

A series of mutations (N130D, G260A, S437T, N130D/G260Q, N130D/S437Y) showed higher thermostability and activity than the wild-type enzyme. Thus, N130D/G260Q exhibited a 6.6-fold increase in half-life and 2.45 °C increase in unfolding temperature; N130D/S437Y showed a 10 °C increase in optimal temperature. The secondary structure did not change much that contributed to improved thermostability were investigated in detail using circular dichroism. Homology modeling suggested that enhanced thermostability and specific activity may result from favorable hydrophobic interactions.

Conclusions

A series of mutations were achieved, showing higher thermostability and activity than the wild-type enzyme by semi-rational mutagenesis with limited structure information. Our findings provide important new insights into molecular modifications aimed at improving Ana-rLOX thermostability and activity.

Electronic supplementary material

The online version of this article (10.1186/s12896-018-0468-4) contains supplementary material, which is available to authorized users.

Bacterially expressed HIV-1 gp120 outer-domain fragment immunogens with improved stability and affinity for CD4-binding site neutralizing antibodies

Protein minimization is an attractive approach for designing vaccines against rapidly evolving pathogens such as human immunodeficiency virus, type 1 (HIV-1), because it can help in focusing the immune response toward conserved conformational epitopes present on complex targets. The outer domain (OD) of HIV-1 gp120 contains epitopes for a large number of neutralizing antibodies and therefore is a primary target for structure-based vaccine design. We have previously designed a bacterially expressed outer-domain immunogen (OD_EC) that bound CD4-binding site (CD4bs) ligands with 3-12 μm affinity and elicited a modest neutralizing antibody response in rabbits. In this study, we have optimized OD_EC using consensus sequence design, cyclic permutation, and structure-guided mutations to generate a number of variants with improved yields, biophysical properties, stabilities, and affinities (K_D of 10-50 nm) for various CD4bs targeting broadly neutralizing antibodies, including the germline-reverted version of the broadly neutralizing antibody VRC01. In contrast to OD_EC, the optimized immunogens elicited high anti-gp120 titers in rabbits as early as 6 weeks post-immunization, before any gp120 boost was given. Following two gp120 boosts, sera collected at week 22 showed cross-clade neutralization of tier 1 HIV-1 viruses. Using a number of different prime/boost combinations, we have identified a cyclically permuted OD fragment as the best priming immunogen, and a trimeric, cyclically permuted gp120 as the most suitable boosting molecule among the tested immunogens. This study also provides insights into some of the biophysical correlates of improved immunogenicity.

Benchmark Analysis of Native and Artificial NAD+-Dependent Enzymes Generated by a Sequence-Based Design Method with or without Phylogenetic Data

The expansion of protein sequence databases has enabled us to design artificial proteins by sequence-based design methods, such as full-consensus design (FCD) and ancestral-sequence reconstruction (ASR). Artificial proteins with enhanced activity levels compared with native ones can potentially be generated by such methods, but successful design is rare because preparing a sequence library by curating the database and selecting a method is difficult. Utilizing a curated library prepared by reducing conservation energies, we successfully designed two artificial l-threonine 3-dehydrogenases (SDR-TDH) with higher activity levels than native SDR-TDH, FcTDH-N1, and AncTDH, using FCD and ASR, respectively. The artificial SDR-TDHs had excellent thermal stability and NAD⁺ recognition compared to native SDR-TDH from Cupriavidus necator (CnTDH); the melting temperatures of FcTDH-N1 and AncTDH were about 10 and 5 °C higher than that of CnTDH, respectively, and the dissociation constants toward NAD⁺ of FcTDH-N1 and AncTDH were 2- and 7-fold lower than that of CnTDH, respectively. Enzymatic efficiency of the artificial SDR-TDHs were comparable to that of CnTDH. Crystal structures of FcTDH-N1 and AncTDH were determined at 2.8 and 2.1 Å resolution, respectively. Structural and MD simulation analysis of the SDR-TDHs indicated that only the flexibility at specific regions was changed, suggesting that multiple mutations introduced in the artificial SDR-TDHs altered their flexibility and thereby affected their enzymatic properties. Benchmark analysis of the SDR-TDHs indicated that both FCD and ASR can generate highly functional proteins if a curated library is prepared appropriately.

Phylogenetic spread of sequence data affects fitness of SOD1 consensus enzymes: Insights from sequence statistics and structural analyses

Non-natural protein sequences with native-like structures and functions can be constructed successfully using consensus design. This design strategy is relatively well understood in repeat proteins with simple binding function, however detailed studies are lacking in globular enzymes. The SOD1 family is a good model for such studies due to the availability of large amount of sequence and structure data motivated by involvement of human SOD1 in the fatal motor neuron disease amyotrophic lateral sclerosis (ALS). We constructed two consensus SOD1 enzymes from multiple sequence alignments from all organisms and eukaryotic organisms. A significant difference in their catalytic activities shows that the phylogenetic spread of the sequences used affects the fitness of the construct obtained. A mutation in an electrostatic loop and overall design incompatibilities between bacterial and eukaryotic sequences were implicated in this disparity. Based on this analysis, a bioinformatics approach was used to classify mutations thought to cause familial ALS providing a unique high level view of the physical basis of disease-causing aggregation of human SOD1.

Evaluation of affibody charge modification identified by synthetic consensus design in molecular PET imaging of epidermal growth factor receptor

Tumor overexpression of epidermal growth factor receptor (EGFR) correlates to therapeutic response in select patient populations. Thus, molecular positron emission tomography (PET) imaging of EGFR could stratify responders versus non-responders. We previously demonstrated effectiveness of a "synthetic consensus" design principle to identify six neutralizing mutations within a 58-amino acid EGFR-targeted affibody domain. Herein, we extend the approach to identify additional neutralized variants that vary net charge from -2 to either -4 or +4 while retaining high affinity (1.6 ± 1.2 nM and 2.5 ± 0.7 nM), specific binding to EGFR, secondary structure, and stability (T_m = 68 °C and 59 °C). We radiolabeled the resultant collection of five charge variants with ⁶⁴Cu and evaluated PET imaging performance in murine models with subcutaneously xenografted EGFR^high and EGFR^low tumors. All variants exhibited good EGFR^high tumor imaging as early as 1 h, with EA35S (+3/-5) achieving 7.7 ± 1.4 %ID/g tumor at 4 h with 1.5 ± 0.3%ID/g EGFR^low tumor, 34 ± 5 tumor:muscle and 12 ± 3 tumor:blood ratios. The positively charged EA62S mutant (+6/-2) exhibited 2.2-3.3-fold higher liver signal than the other variants (p<0.01). The EA68 variant with higher charge density was more stable to human and mouse serum than neutralized variants. In a comparison of radiometal chelators, 1,4,7-triazacyclononane,1-glutaric acid-4,7-acetic acid (NODAGA) exhibited superior physiological specificity to 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA). In total, these studies comparatively evaluated a set of EGFR-targeted affibodies varying in net charge and charge density, which revealed functional variations that are useful in engineering an ideal probe for translational studies.

Stabilization of a nucleotide-binding domain of the cystic fibrosis transmembrane conductance regulator yields insight into disease-causing mutations

Characterization of the second nucleotide-binding domain (NBD2) of the cystic fibrosis transmembrane conductance regulator (CFTR) has lagged behind research into the NBD1 domain, in part because NBD1 contains the F508del mutation, which is the dominant cause of cystic fibrosis. Research on NBD2 has also been hampered by the overall instability of the domain and the difficulty of producing reagents. Nonetheless, multiple disease-causing mutations reside in NBD2, and the domain is critical for CFTR function, because channel gating involves NBD1/NBD2 dimerization, and NBD2 contains the catalytically active ATPase site in CFTR. Recognizing the paucity of structural and biophysical data on NBD2, here we have defined a bioinformatics-based method for manually identifying stabilizing substitutions in NBD2, and we used an iterative process of screening single substitutions against thermal melting points to both produce minimally mutated stable constructs and individually characterize mutations. We present a range of stable constructs with minimal mutations to help inform further research on NBD2. We have used this stabilized background to study the effects of NBD2 mutations identified in cystic fibrosis (CF) patients, demonstrating that mutants such as N1303K and G1349D are characterized by lower stability, as shown previously for some NBD1 mutations, suggesting a potential role for NBD2 instability in the pathology of CF.

Computational tools for enzyme improvement: why everyone can - and should - use them

This review presents computational methods that experimentalists can readily use to create smart libraries for enzyme engineering and to obtain insights into protein-substrate complexes. Computational tools have the reputation of being hard to use and inaccurate compared to experimental methods in enzyme engineering, yet they are essential to probe datasets of ever-increasing size and complexity. In recent years, bioinformatics groups have made a huge leap forward in providing user-friendly interfaces and accurate algorithms for experimentalists. These methods guide efficient experimental planning and allow the enzyme engineer to rationalize time and resources. Computational tools nevertheless face challenges in the realm of transient modern technology.

Protein stability: computation, sequence statistics, and new experimental methods

Calculating protein stability and predicting stabilizing mutations remain exceedingly difficult tasks, largely due to the inadequacy of potential functions, the difficulty of modeling entropy and the unfolded state, and challenges of sampling, particularly of backbone conformations. Yet, computational design has produced some remarkably stable proteins in recent years, apparently owing to near ideality in structure and sequence features. With caveats, computational prediction of stability can be used to guide mutation, and mutations derived from consensus sequence analysis, especially improved by recent co-variation filters, are very likely to stabilize without sacrificing function. The combination of computational and statistical approaches with library approaches, including new technologies such as deep sequencing and high throughput stability measurements, point to a very exciting near term future for stability engineering, even with difficult computational issues remaining.

Accurate design of megadalton-scale two-component icosahedral protein complexes

Nature provides many examples of self- and co-assembling protein-based molecular machines, including icosahedral protein cages that serve as scaffolds, enzymes, and compartments for essential biochemical reactions and icosahedral virus capsids, which encapsidate and protect viral genomes and mediate entry into host cells. Inspired by these natural materials, we report the computational design and experimental characterization of co-assembling, two-component, 120-subunit icosahedral protein nanostructures with molecular weights (1.8 to 2.8 megadaltons) and dimensions (24 to 40 nanometers in diameter) comparable to those of small viral capsids. Electron microscopy, small-angle x-ray scattering, and x-ray crystallography show that 10 designs spanning three distinct icosahedral architectures form materials closely matching the design models. In vitro assembly of icosahedral complexes from independently purified components occurs rapidly, at rates comparable to those of viral capsids, and enables controlled packaging of molecular cargo through charge complementarity. The ability to design megadalton-scale materials with atomic-level accuracy and controllable assembly opens the door to a new generation of genetically programmable protein-based molecular machines.

Synthetic and natural consensus design for engineering charge within an affibody targeting epidermal growth factor receptor

Protein ligand charge can impact physiological delivery with charge reduction often benefiting performance. Yet neutralizing mutations can be detrimental to protein function. Herein, three approaches are evaluated to introduce charged-to-neutral mutations of three cations and three anions within an affibody engineered to bind epidermal growth factor receptor. These approaches-combinatorial library sorting or consensus design, based on natural homologs or library-sorted mutants-are used to identify mutations with favorable affinity, stability, and recombinant yield. Consensus design, based on 942 affibody homologs, yielded a mutant of modest function (Kd  = 11 ±4 nM, Tm  = 62°C, and yield = 4.0 ± 0.8 mg/L as compared to 5.3 ± 1.7 nM, 71°C, and 3.5 ± 0.3 mg/L for the parental affibody). Extension of consensus design to 10 additional mutants exhibited varied performance including a substantially improved mutant (Kd  = 6.9 ± 1.4 nM, Tm  = 71°C, and 12.7 ± 0.9 mg/L yield). Sorting a homolog-based combinatorial library of 7 × 10(5) mutants generated a distribution of mutants with lower stability and yield, but did identify one strongly binding variant (Kd  = 1.2 ± 0.3 nM, Tm  = 69°C, and 6.0 ± 0.4 mg/L yield). Synthetic consensus design, based on the amino acid distribution in functional library mutants, yielded higher affinities (P = 0.05) with comparable stabilities and yields. The best of four analyzed clones had Kd  = 1.7 ± 0.5 nM, Tm  = 68°C, and 7.0 ± 0.5 mg/L yield. While all three approaches were effective in creating targeted affibodies with six charged-to-neutral mutations, synthetic consensus design proved to be the most robust. Synthetic consensus design provides a valuable tool for ligand engineering, particularly in the context of charge manipulation. Biotechnol. Bioeng. 2016;113: 1628-1638. © 2016 Wiley Periodicals, Inc.

Consensus protein design

A popular and successful strategy in semi-rational design of protein stability is the use of evolutionary information encapsulated in homologous protein sequences. Consensus design is based on the hypothesis that at a given position, the respective consensus amino acid contributes more than average to the stability of the protein than non-conserved amino acids. Here, we review the consensus design approach, its theoretical underpinnings, successes, limitations and challenges, as well as providing a detailed guide to its application in protein engineering.

Epistasis effects of multiple ancestral-consensus amino acid substitutions on the thermal stability of glycerol kinase from Cellulomonas sp. NT3060

Thermostable variants of the Cellulomonas sp. NT3060 glycerol kinase have been constructed by through the introduction of ancestral-consensus mutations. We produced seven mutants, each having an ancestral-consensus amino acid residue that might be present in the common ancestors of both bacteria and of archaea, and that appeared most frequently at the position of 17 glycerol kinase sequences in the multiple sequence alignment. The thermal stabilities of the resulting mutants were assessed by determining their melting temperatures (Tm), which was defined as the temperature at which 50% of the initial catalytic activity is lost after 15 min of incubation, as well as when the half-life of the catalytic activity occurs at a temperature of 60°C (t1/2). Three mutants showed increased stabilities compared to the wild-type protein. We then produced five more mutants with multiple amino acid substitutions. Some of the resulting mutants showed thermal stabilities much greater than those expected given the stabilities of the respective mutants with single mutations. Therefore, the effects of mutations are not always simply additive and some amino acid substitutions, which do not affect or only slightly improve stability when individually introduced into the protein, show substantial stabilizing effects in combination with other mutations.

Protein evolution analysis of S-hydroxynitrile lyase by complete sequence design utilizing the INTMSAlign software

Development of software and methods for design of complete sequences of functional proteins could contribute to studies of protein engineering and protein evolution. To this end, we developed the INTMSAlign software, and used it to design functional proteins and evaluate their usefulness. The software could assign both consensus and correlation residues of target proteins. We generated three protein sequences with S-selective hydroxynitrile lyase (S-HNL) activity, which we call designed S-HNLs; these proteins folded as efficiently as the native S-HNL. Sequence and biochemical analysis of the designed S-HNLs suggested that accumulation of neutral mutations occurs during the process of S-HNLs evolution from a low-activity form to a high-activity (native) form. Taken together, our results demonstrate that our software and the associated methods could be applied not only to design of complete sequences, but also to predictions of protein evolution, especially within families such as esterases and S-HNLs.

Thermostable artificial enzyme isolated by in vitro selection

Artificial enzymes hold the potential to catalyze valuable reactions not observed in nature. One approach to build artificial enzymes introduces mutations into an existing protein scaffold to enable a new catalytic activity. This process commonly results in a simultaneous reduction of protein stability as an undesired side effect. While protein stability can be increased through techniques like directed evolution, care needs to be taken that added stability, conversely, does not sacrifice the desired activity of the enzyme. Ideally, enzymatic activity and protein stability are engineered simultaneously to ensure that stable enzymes with the desired catalytic properties are isolated. Here, we present the use of the in vitro selection technique mRNA display to isolate enzymes with improved stability and activity in a single step. Starting with a library of artificial RNA ligase enzymes that were previously isolated at ambient temperature and were therefore mostly mesophilic, we selected for thermostable active enzyme variants by performing the selection step at 65°C. The most efficient enzyme, ligase 10C, was not only active at 65°C, but was also an order of magnitude more active at room temperature compared to related enzymes previously isolated at ambient temperature. Concurrently, the melting temperature of ligase 10C increased by 35 degrees compared to these related enzymes. While low stability and solubility of the previously selected enzymes prevented a structural characterization, the improved properties of the heat-stable ligase 10C finally allowed us to solve the three-dimensional structure by NMR. This artificial enzyme adopted an entirely novel fold that has not been seen in nature, which was published elsewhere. These results highlight the versatility of the in vitro selection technique mRNA display as a powerful method for the isolation of thermostable novel enzymes.

Antibody variable domain interface and framework sequence requirements for stability and function by high-throughput experiments

Protein structural stability and biological functionality are dictated by the formation of intradomain cores and interdomain interfaces, but the intricate sequence-structure-function interrelationships in the packing of protein cores and interfaces remain difficult to elucidate due to the intractability of enumerating all packing possibilities and assessing the consequences of all the variations. In this work, groups of β strand residues of model antibody variable domains were randomized with saturated mutagenesis and the functional variants were selected for high-throughput sequencing and high-throughput thermal stability measurements. The results show that the sequence preferences of the intradomain hydrophobic core residues are strikingly flexible among hydrophobic residues, implying that these residues are coupled indirectly with antigen binding through energetic stabilization of the protein structures. By contrast, the interdomain interface residues are directly coupled with antigen binding. The interdomain interface should be treated as an integral part of the antigen-binding site.

Dynamic combinatorial libraries of artificial repeat proteins

Repeat proteins are found in almost all cellular systems, where they are involved in diverse molecular recognition processes. Recent studies have suggested that de novo designed repeat proteins may serve as universal binders, and might potentially be used as practical alternative to antibodies. We describe here a novel chemical methodology for producing small libraries of repeat proteins, and screening in parallel the ligand binding of library members. The first stage of this research involved the total synthesis of a consensus-based three-repeat tetratricopeptide (TPR) protein (~14 kDa), via sequential attachment of the respective peptides. Despite the effectiveness of the synthesis and ligation steps, this method was found to be too demanding for the production of proteins containing variable number of repeats. Additionally, the analysis of binding of the individual proteins was time consuming. Therefore, we designed and prepared novel dynamic combinatorial libraries (DCLs), and show that their equilibration can facilitate the formation of TPR proteins containing up to eight repeating units. Interestingly, equilibration of the library building blocks in the presence of the biologically relevant ligands, Hsp90 and Hsp70, induced their oligomerization into forming more of the proteins with large recognition surfaces. We suggest that this work presents a novel simple and rapid tool for the simultaneous screening of protein mixtures with variable binding surfaces, and for identifying new binders for ligands of interest.

All repeats are not equal: a module-based approach to guide repeat protein design

Repeat proteins composed of tandem arrays of a short structural motif often mediate protein-protein interactions. Past efforts to design repeat protein-based molecular recognition tools have focused on the creation of templates from the consensus of individual repeats, regardless of their natural context. Such an approach assumes that all repeats are essentially equivalent. In this study, we present the results of a "module-based" approach in which modules composed of tandem repeats are aligned to identify repeat-specific features. Using this approach to analyze tetratricopeptide repeat modules that contain three tandem repeats (3TPRs), we identify two classes of 3TPR modules with distinct structural signatures that are correlated with different sets of functional residues. Our analyses also reveal a high degree of correlation between positions across the entire ligand-binding surface, indicative of a coordinated, coevolving binding surface. Extension of our analyses to different repeat protein modules reveals more examples of repeat-specific features, especially in armadillo repeat modules. In summary, the module-based analyses that we present effectively capture key repeat-specific features that will be important to include in future repeat protein design templates.

Recent trends in biocatalysis engineering

During the last 30 years the scope of biocatalysis has been expanding due to the advances in several technological fields. Diverse techniques as structural enzyme improvement (e.g. protein engineering, direct evolution), engineering approaches (e.g. ionic liquids, supercritical fluids) and physical stabilization (e.g. immobilization, CLEAS) have been developed, which in combination are powerful tools to improve biotransformation and to synthesize new products. In the present work, recent advances in biocatalysis are reviewed.

Stabilizing proteins from sequence statistics: the interplay of conservation and correlation in triosephosphate isomerase stability

Understanding the determinants of protein stability remains one of protein science's greatest challenges. There are still no computational solutions that calculate the stability effects of even point mutations with sufficient reliability for practical use. Amino acid substitutions rarely increase the stability of native proteins; hence, large libraries and high-throughput screens or selections are needed to stabilize proteins using directed evolution. Consensus mutations have proven effective for increasing stability, but these mutations are successful only about half the time. We set out to understand why some consensus mutations fail to stabilize, and what criteria might be useful to predict stabilization more accurately. Overall, consensus mutations at more conserved positions were more likely to be stabilizing in our model, triosephosphate isomerase (TIM) from Saccharomyces cerevisiae. However, positions coupled to other sites were more likely not to stabilize upon mutation. Destabilizing mutations could be removed both by removing sites with high statistical correlations to other positions and by removing nearly invariant positions at which "hidden correlations" can occur. Application of these rules resulted in identification of stabilizing mutations in 9 out of 10 positions, and amalgamation of all predicted stabilizing positions resulted in the most stable yeast TIM variant we produced (+8 °C). In contrast, a multimutant with 14 mutations each found to stabilize TIM independently was destabilized by 2 °C. Our results are a practical extension to the consensus concept of protein stabilization, and they further suggest the importance of positional independence in the mechanism of consensus stabilization.

Combinatorial reshaping of the Candida antarctica lipase A substrate pocket for enantioselectivity using an extremely condensed library

A highly combinatorial structure-based protein engineering method for obtaining enantioselectivity is reported that results in a thorough modification of the substrate binding pocket of Candida antarctica lipase A (CALA). Nine amino acid residues surrounding the entire pocket were simultaneously mutated, contributing to a reshaping of the substrate pocket to give increased enantioselectivity and activity for a sterically demanding substrate. This approach seems to be powerful for developing enantioselectivity when a complete reshaping of the active site is required. Screening toward ibuprofen ester 1, a substrate for which previously used methods had failed, gave variants with a significantly increased enantioselectivity and activity. Wild-type CALA has a moderate activity with an E value of only 3.4 toward this substrate. The best variant had an E value of 100 and it also displayed a high activity. The variation at each mutated position was highly reduced, comprising only the wild type and an alternative residue, preferably a smaller one with similar properties. These minimal binary variations allow for an extremely condensed protein library. With this highly combinatorial method synergistic effects are accounted for and the protein fitness landscape is explored efficiently.

Directed evolution to re-adapt a co-evolved network within an enzyme

We have previously used targeted active-site saturation mutagenesis to identify a number of transketolase single mutants that improved activity towards either glycolaldehyde (GA), or the non-natural substrate propionaldehyde (PA). Here, all attempts to recombine the singles into double mutants led to unexpected losses of specific activity towards both substrates. A typical trade-off occurred between soluble expression levels and specific activity for all single mutants, but many double mutants decreased both properties more severely suggesting a critical loss of protein stability or native folding. Statistical coupling analysis (SCA) of a large multiple sequence alignment revealed a network of nine co-evolved residues that affected all but one double mutant. Such networks maintain important functional properties such as activity, specificity, folding, stability, and solubility and may be rapidly disrupted by introducing one or more non-naturally occurring mutations. To identify variants of this network that would accept and improve upon our best D469 mutants for activity towards PA, we created a library of random single, double and triple mutants across seven of the co-evolved residues, combining our D469 variants with only naturally occurring mutations at the remaining sites. A triple mutant cluster at D469, E498 and R520 was found to behave synergistically for the specific activity towards PA. Protein expression was severely reduced by E498D and improved by R520Q, yet variants containing both mutations led to improved specific activity and enzyme expression, but with loss of solubility and the formation of inclusion bodies. D469S and R520Q combined synergistically to improve k_cat 20-fold for PA, more than for any previous transketolase mutant. R520Q also doubled the specific activity of the previously identified D469T to create our most active transketolase mutant to date. Our results show that recombining active-site mutants obtained by saturation mutagenesis can rapidly destabilise critical networks of co-evolved residues, whereas beneficial single mutants can be retained and improved upon by randomly recombining them with natural variants at other positions in the network.