New Methods for the High-Throughput Screening of Enantioselective Catalysts and Biocatalysts

Biomacromolecule-Assisted Screening for Reaction Discovery and Catalyst Optimization

Reaction discovery and catalyst screening lie at the heart of synthetic organic chemistry. While there are efforts at de novo catalyst design using computation/artificial intelligence, at its core, synthetic chemistry is an experimental science. This review overviews biomacromolecule-assisted screening methods and the follow-on elaboration of chemistry so discovered. All three types of biomacromolecules discussed─enzymes, antibodies, and nucleic acids─have been used as "sensors" to provide a readout on product chirality exploiting their native chirality. Enzymatic sensing methods yield both UV-spectrophotometric and visible, colorimetric readouts. Antibody sensors provide direct fluorescent readout upon analyte binding in some cases or provide for cat-ELISA (Enzyme-Linked ImmunoSorbent Assay)-type readouts. DNA biomacromolecule-assisted screening allows for templation to facilitate reaction discovery, driving bimolecular reactions into a pseudo-unimolecular format. In addition, the ability to use DNA-encoded libraries permits the barcoding of reactants. All three types of biomacromolecule-based screens afford high sensitivity and selectivity. Among the chemical transformations discovered by enzymatic screening methods are the first Ni(0)-mediated asymmetric allylic amination and a new thiocyanopalladation/carbocyclization transformation in which both C-SCN and C-C bonds are fashioned sequentially. Cat-ELISA screening has identified new classes of sydnone-alkyne cycloadditions, and DNA-encoded screening has been exploited to uncover interesting oxidative Pd-mediated amido-alkyne/alkene coupling reactions.

Chiral stationary phases and applications in gas chromatography

Chiral compounds are ubiquitous in nature and play a pivotal role in biochemical processes, in chiroptical materials and applications, and as chiral drugs. The analysis and determination of the enantiomeric ratio (er) of chiral compounds is of enormous scientific, industrial, and economic importance. Chiral separation techniques and methods have become indispensable tools to separate chiral compounds into their enantiomers on an analytical as well on a preparative level to obtain enantiopure compounds. Chiral gas chromatography and high-performance liquid chromatography have paved the way and fostered several research areas, that is, asymmetric synthesis and catalysis in organic, medicinal, pharmaceutical, and supramolecular chemistry. The development of highly enantioselective chiral stationary phases was essential. In particular, the elucidation and understanding of the underlying enantioselective supramolecular separation mechanisms led to the design of new chiral stationary phases. This review article focuses on the development of chiral stationary phases for gas chromatography. The fundamental mechanisms of the recognition and separation of enantiomers and the selectors and chiral stationary phases used in chiral gas chromatography are presented. An overview over syntheses and applications of these chiral stationary phases is presented as a practical guidance for enantioselective separation of chiral compound classes and substances by gas chromatography.

Learning the structure-activity relationship (SAR) of the Wittig reaction from genetically-encoded substrates

The Wittig reaction can be used for late stage functionalization of proteins and peptides to ligate glycans, pharmacophores, and many other functionalities. In this manuscript, we modified 160 000 N-terminal glyoxaldehyde peptides displayed on phage with the Wittig reaction by using a biotin labeled ylide under conditions that functionalize only 1% of the library population. Deep-sequencing of the biotinylated and input populations estimated the rate of conversion for each sequence. This “deep conversion” (DC) from deep sequencing correlates with rate constants measured by HPLC. Peptide sequences with fast and slow reactivity highlighted the critical role of primary backbone amides (N–H) in accelerating the rate of the aqueous Wittig reaction. Experimental measurement of reaction rates and density functional theory (DFT) computation of the transition state geometries corroborated this relationship. We also collected deep-sequencing data to build structure–activity relationship (SAR) models that can predict the DC value of the Wittig reaction. By using these data, we trained two classifier models based on gradient boosted trees. These classifiers achieved area under the ROC (receiver operating characteristic) curve (ROC AUC) of 81.2 ± 0.4 and 73.7 ± 0.8 (90–92% accuracy) in determining whether a sequence belonged to the top 5% or the bottom 5% in terms of its reactivity. This model can suggest new peptides never observed experimentally with ‘HIGH’ or ‘LOW’ reactivity. Experimental measurement of reaction rates for 11 new sequences corroborated the predictions for 8 of them. We anticipate that phage-displayed peptides and related mRNA or DNA-displayed substrates can be employed in a similar fashion to study the substrate scope and mechanisms of many other chemical reactions.

160 000 peptides displayed on phage were subjected to the Wittig reaction with a biotinylated ylide. Deep-sequencing estimated the conversion rate for each sequence and unveiled the relationship between sequences and the rate of the Wittig reaction.

Direct monitoring of biocatalytic deacetylation of amino acid substrates by 1H NMR reveals fine details of substrate specificity

Amino acids are key synthetic building blocks that can be prepared in an enantiopure form by biocatalytic methods. We show that the l-selective ornithine deacetylase ArgE catalyses hydrolysis of a wide-range of N-acyl-amino acid substrates. This activity was revealed by 1H NMR spectroscopy that monitored the appearance of the well resolved signal of the acetate product. Furthermore, the assay was used to probe the subtle structural selectivity of the biocatalyst using a substrate that could adopt different rotameric conformations.

High-throughput experimentation meets artificial intelligence: a new pathway to catalyst discovery

High throughput experimentation in heterogeneous catalysis provides an efficient solution to the generation of large datasets under reproducible conditions. Knowledge extraction from these datasets has mostly been performed using statistical methods, targeting the optimization of catalyst formulations. The combination of advanced machine learning methodologies with high-throughput experimentation has enormous potential to accelerate the predictive discovery of novel catalyst formulations that do not exist with current statistical design of experiments. This perspective describes selective examples ranging from statistical design of experiments for catalyst synthesis to genetic algorithms applied to catalyst optimization, and finally random forest machine learning using experimental data for the discovery of novel catalysts. Lastly, this perspective also provides an outlook on advanced machine learning methodologies as applied to experimental data for materials discovery.

High-Throughput Screening of a 2-Keto-L-Gulonic Acid-Producing Gluconobacter oxydans Strain Based on Related Dehydrogenases

High-throughput screening is a powerful tool for discovering strains in the natural environment that may be suitable for target production. Herein, a novel enzyme-based high-throughput screening method was developed for rapid screening of strains overproducing 2-keto-L-gulonic acid (2-KLG). The screening method detects changes in the fluorescence of reduced nicotinamide adenine dinucleotide (NADH) at 340 nm using a microplate reader when 2-KLG is degraded by 2-KLG reductase. In this research, three different 2-KLG reductases were expressed, purified, and studied. The 2-KLG reductase from Aspergillus niger were selected as the best appropriate reductase to establishment the method for its high activity below pH 7. Using the established method, and coupled with fluorescence-activated cell sorting, we achieved a high 2-KLG-producing strain of Gluconobacter oxydans WSH-004 from soil. When cultured with D-sorbitol as the substrate, the 2-KLG yield was 2.5 g/L from 50 g/L D-sorbitol without any side products. Compared with other reported screening methods, our enzyme-based method is more efficient and accurate for obtaining high-producing 2-KLG strains, and it is also convenient and cost-effective. The method is broadly applicable for screening keto acids and other products that can be oxidized via nicotinamide adenine dinucleotide (NAD⁺) or nicotinamide adenine dinucleotide phosphate (NADP⁺).

Machine learning and molecular descriptors enable rational solvent selection in asymmetric catalysis

Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO)2(acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.

An indicator displacement assay recognizes enantiomers of chiral carboxylates

Analyte chirality induces changes in fluorescence.

Machine learning meets volcano plots: computational discovery of cross-coupling catalysts

The application of modern machine learning to challenges in atomistic simulation is gaining attraction. We present new machine learning models that can predict the energy of the oxidative addition process between a transition metal complex and a substrate for C-C cross-coupling reactions. In turn, this quantity can be used as a descriptor to estimate the activity of homogeneous catalysts using molecular volcano plots. The versatility of this approach is illustrated for vast libraries of organometallic catalysts based on Pt, Pd, Ni, Cu, Ag, and Au combined with 91 ligands. Out-of-sample machine learning predictions were made on a total of 18 062 compounds leading to 557 catalyst candidates falling into the ideal thermodynamic window. This number was further refined by searching for candidates with an estimated price lower than 10 US$ per mmol. The 37 catalyst finalists are dominated by palladium phosphine ligand combinations but also include the earth abundant transition metal (Cu) with less common ligands. Our results indicate that modern statistical learning techniques can be applied to the computational discovery of readily available and promising catalyst candidates.

Computational design of high-performance ligand for enantioselective Markovnikov hydroboration of aliphatic terminal alkenes

Finding optimal chiral ligands for transition-metal-catalyzed asymmetric reactions using trial-and-error methods is often time-consuming and costly, even if the details of the reaction mechanism are already known. Although modern computational analyses allow the prediction of the stereoselectivity, there are only very few examples for the attempted design of chiral ligands using a computational approach for the improvement of the stereoselectivity. Herein, we report a systematic method for the design of chiral ligands for the enantioselective Markovnikov hydroboration of aliphatic terminal alkenes based on a computational and experimental evaluation sequence. We developed a three-hindered-quadrant P-chirogenic bisphosphine ligand that was designed in accordance with the design guidelines derived from this method, which allowed the Markovnikov hydroboration to proceed with high enantioselectivity (up to 99% ee).

Predictive and mechanistic multivariate linear regression models for reaction development

Multivariate Linear Regression (MLR) models utilizing computationally-derived and empirically-derived physical organic molecular descriptors are described in this review. Several reports demonstrating the effectiveness of this methodological approach towards reaction optimization and mechanistic interrogation are discussed. A detailed protocol to access quantitative and predictive MLR models is provided as a guide for model development and parameter analysis.

The In Situ Enzymatic Screening (ISES) Approach to Reaction Discovery and Catalyst Identification

The importance of discovering new chemical transformations and/or optimizing catalytic combinations has led to a flurry of activity in reaction screening. The in situ enzymatic screening (ISES) approach described here utilizes biological tools (enzymes/cofactors) to advance chemistry. The protocol interfaces an organic reaction layer with an adjacent aqueous layer containing reporting enzymes that act upon the organic reaction product, giving rise to a spectroscopic signal. ISES allows the experimentalist to rapidly glean information on the relative rates of a set of parallel organic/organometallic reactions under investigation, without the need to quench the reactions or draw aliquots. In certain cases, the real-time enzymatic readout also provides information on sense and magnitude of enantioselectivity and substrate specificity. This article contains protocols for single-well (relative rate) and double-well (relative rate/enantiomeric excess) ISES, in addition to a colorimetric ISES protocol and a miniaturized double-well procedure. © 2017 by John Wiley & Sons, Inc.

Monoamine oxidase and transaminase screening: biotransformation of 2-methyl-6-alkylpiperidines by Neopestalotiopsis sp. CBMAI 2030

High-throughput screening detected transaminases (TAs) and monoamine oxidases (MAOs) in fungi by applying a fluorogenic probe. Strains F026, F037, F041, F053, and F057 showed the highest enzymatic conversions (31, 60, 30, 40, and 32%, respectively) and where evaluated for their ability to transform piperidines. Strain F053 (Neopestalotiopsis sp. CBMAI 2030) revealed unusual enzymatic activity to deracemize 2-methyl-6-alkylpiperidines. Neopestalotiopsis sp. CBMAI 2030 was capable to convert 2-methyl-6-propylpiperidine, 2-methyl-6-butylpiperidine, and 2-methyl-6-pentylpiperidine in piperideine with 11, 14, and 24% conversion, respectively. The activity was enhanced by cultivating the fungus with 2-methyl-6-pentylpiperidine (38% conversion and 73% ee).

The Development of Multidimensional Analysis Tools for Asymmetric Catalysis and Beyond

In most modern organic chemistry reports, including many of ours, reaction optimization schemes are typically presented to showcase how reaction conditions have been tailored to augment the reaction's yield and selectivity. In asymmetric catalysis, this often involves evaluation of catalyst, solvent, reagent, and, sometimes, substrate features. Such an article will then detail the process's scope, which mainly focuses on its successes and briefly outlines the "limitations". These limitations or poorer-performing substrates are occasionally the result of obvious, significant changes to structure (e.g., a Lewis basic group binds to a catalyst), but frequently, a satisfying explanation for inferior performance is not clear. This is one of several reasons such results are not often reported. These apparent outliers are also commonplace in the evaluation of catalyst structure, although most of this information is placed in the Supporting Information. These practices are unfortunate because results that appear at first glance to be peculiar or poor are considerably more interesting than ones that follow obvious or intuitive trends. In other words, all of the data from an optimization campaign contain relevant information about the reaction under study, and the "outliers" may be the most revealing. Realizing the power of outliers as an entry point to entirely new reaction development is not unusual. Nevertheless, the concept that no data should be wasted when considering the underlying phenomena controlling the observations of a given reaction is at the heart of the strategy we describe in this Account. The idea that one can concurrently optimize a reaction to expose the structural features that control its outcomes would represent a transformative addition to the arsenal of catalyst development and, ultimately, de novo design. Herein we outline the development of a recently initiated program in our lab that unites optimization with mechanistic interrogation by correlating reaction outputs (e.g., electrochemical potential or enantio-, site, or chemoselectivity) with structural descriptors of the molecules involved. The ever-evolving inspiration for this program is rooted in outliers of classical linear free energy relationships. These outliers encouraged us to ask questions about the parameters themselves, suggest potential interactions at the source of the observed effects, and, of particular applicability, identify more sophisticated physical organic descriptors. Throughout this program, we have integrated techniques from disparate fields, including synthetic methodology development, mechanistic investigations, statistics, computational chemistry, and data science. The implementation of many of these strategies is described, and the resulting tools are illustrated in a wide range of case studies, which include data sets with simultaneous and multifaceted changes to the reagent, substrate, and catalyst structures. This tactic constitutes a modern approach to physical organic chemistry wherein no data are wasted and mechanistic hypotheses regarding sophisticated processes can be developed and probed.

Organic chemistry. A data-intensive approach to mechanistic elucidation applied to chiral anion catalysis

Knowledge of chemical reaction mechanisms can facilitate catalyst optimization, but extracting that knowledge from a complex system is often challenging. Here, we present a data-intensive method for deriving and then predictively applying a mechanistic model of an enantioselective organic reaction. As a validating case study, we selected an intramolecular dehydrogenative C-N coupling reaction, catalyzed by chiral phosphoric acid derivatives, in which catalyst-substrate association involves weak, noncovalent interactions. Little was previously understood regarding the structural origin of enantioselectivity in this system. Catalyst and substrate substituent effects were probed by means of systematic physical organic trend analysis. Plausible interactions between the substrate and catalyst that govern enantioselectivity were identified and supported experimentally, indicating that such an approach can afford an efficient means of leveraging mechanistic insight so as to optimize catalyst design.

Enzyme immunoassays as screening tools for catalysts and reaction discovery

This feature article summarizes the development and use of immunoassay techniques (ELISA) as screening tools for fast identification of efficient catalysts in libraries and for the discovery of new chemical reactions.

Well Plate Circular Dichroism Reader for the Rapid Determination of Enantiomeric Excess

Circular dichroism (CD) spectropolarimeters typically employ one photoelastic modulator. However, spectropolarimeters employing two or even four modulators are more versatile and can be used to subvert common measurement errors arising from imperfectly isotropic samples or sample holders. Small linear anisotropies that can cause large errors in CD measurement can be associated with multi-well sample holders. Thus, high-throughput CD analyses in multi-well plates have not yet been demonstrated. One such application is the determination of enantiomeric excess of a library of reaction products. Herein, a spectropolarimeter employing four photoelastic modulators and a translation stage was used to determine the enantiomeric excess of a family of chiral amine complexes much more rapidly than could be achieved with a robotic fluid injection system. These experiments are proof of concept for high-throughput CD analysis. In practice, commercially available glass bottomed well plates are sufficiently strain free such that a simple instrument with just one photoelastic modulator and a vertical optical train should be able to deliver the CD without special considerations given herein. On the other hand, polystyrene well plates cannot be used in this way.

Combinatorial approach to chiral tris-ligated carbophilic platinum complexes: application to asymmetric catalysis

A straightforward methodology for the synthesis of libraries of chiral tris-ligated cationic platinum complexes and their in situ evaluation as asymmetric carbophilic catalysts in a model domino hydroarylation/cyclization reaction of a 1,6-enyne was developed. A catalyst-generation process based on a combination of a monodentate and a bidentate phosphorus ligand allowed the formation of 108 chiral complexes. One-pot screening of the stereoinduction obtained with this library in a test domino addition/cyclization reaction validated this approach and stressed the key role played by the monodentate ligand partner in obtaining high enantioselectivities. In the case of two challenging substrate/nucleophile combinations, the combinatorial approach resulted in a significant gain in enantioselectivity.

High throughput technology: approaches of research in homogeneous and heterogeneous catalysis

High throughput experimentation (HTE) approaches and the choice of the design of experiment (DoE) tools are discussed with regard to their convenience and applicability in homogeneous and heterogeneous catalysis as a concerted workflow. Much attention is given to diverse methodologies and strategies, which are fundamental for the experimental planning. For two target reactions in two case studies presented in this chapter, HTE methods were applied to create and evaluate catalyst libraries. A homogeneous catalyst case study is illustrated first, which deals with parallel synthesis and screening of organometallic catalysts in the polymerisation of ethylene. The second case study (heterogeneous catalysis) focuses on coherent synthesis and testing of dopant effects on the performance of oxidation catalysts in a reaction of transformation of n-butane to maleic anhydride. Supporting examples from the literature described here show that careful planning of libraries and test conditions is vital in high throughput experimentation in order to deliver meaningful results leading to performance improvements or disruptive new findings.

Efficient asymmetric synthesis of lamivudine via enzymatic dynamic kinetic resolution

The anti-HIV nucleoside lamivudine was asymmetrically synthesized in only three steps via a novel surfactant-treated subtilisin Carlsberg-catalyzed dynamic kinetic resolution protocol.

In situ evaluation of kinetic resolution catalysts for nitroaldol by rationally designed fluorescence probe

Development of effective chemical catalysts is a key concern in organic chemistry. Therefore, convenient screening systems for chemical catalysts are required, and although some fluorescence-based HTS systems have been developed, little attempt has been made to apply them to asymmetric catalysts. Therefore, we tried to develop a chiral fluorescence probe which can evaluate the reactivity and enantioselectivity of asymmetric catalysts. We focused on kinetic resolution catalysts as a target of our novel fluorescence probe, employing β-elimination following acylation of nitroaldol. Once the hydroxyl group of nitroaldol is acylated, β-elimination occurs immediately, affording nitro olefin. Therefore, we designed and synthesized a fluorescence probe with an asymmetric nitroaldol moiety. Its fluorescence intensity decreases dramatically upon β-elimination, so the fluorescence decrease is an indicator of the reaction yield. Thus, the enantioselectivity of kinetic resolution catalysts can be assessed simply by measuring the fluorescence intensities of the reaction mixtures of the two enantiomers; it is not necessary to purify the product. This fluorescence probe revealed that benzotetramisole is a superior catalyst for kinetic resolution of nitroaldol. Furthermore, we established an HTS system for asymmetric catalysts, using a fluorescence probe and benzotetramisole. To our knowledge, this is the first fluorescence-based HTS system for asymmetric catalysts.

In situ evaluation of lipase performances through dynamic asymmetric cyanohydrin resolution

A dynamic resolution process based on multiple reversible cyanohydrin formation coupled to lipase-mediated transesterification is demonstrated. The resulting process resulted in the efficient evaluation of complex lipase performances in asymmetric cyanohydrin acylate synthesis. Dynamic systems were generated and resolved in situ, and the effects of the reaction conditions could be directly monitored for the overall system. By this concept, the enzyme activity, chemo- and stereoselectivity for all involved substrates could be simultaneously evaluated.

Artificial metalloenzymes based on the biotin-avidin technology: enantioselective catalysis and beyond

Artificial metalloenzymes are created by incorporating an organometallic catalyst within a host protein. The resulting hybrid can thus provide access to the best features of two distinct, and often complementary, systems: homogeneous and enzymatic catalysts. The coenzyme may be positioned with covalent, dative, or supramolecular anchoring strategies. Although initial reports date to the late 1970s, artificial metalloenzymes for enantioselective catalysis have gained significant momentum only in the past decade, with the aim of complementing homogeneous, enzymatic, heterogeneous, and organic catalysts. Inspired by a visionary report by Wilson and Whitesides in 1978, we have exploited the potential of biotin-avidin technology in creating artificial metalloenzymes. Owing to the remarkable affinity of biotin for either avidin or streptavidin, covalent linking of a biotin anchor to a catalyst precursor ensures that, upon stoichiometric addition of (strept)avidin, the metal moiety is quantitatively incorporated within the host protein. In this Account, we review our progress in preparing and optimizing these artificial metalloenzymes, beginning with catalytic hydrogenation as a model and expanding from there. These artificial metalloenzymes can be optimized by both chemical (variation of the biotin-spacer-ligand moiety) and genetic (mutation of avidin or streptavidin) means. Such chemogenetic optimization schemes were applied to various enantioselective transformations. The reactions implemented thus far include the following: (i) The rhodium-diphosphine catalyzed hydrogenation of N-protected dehydroaminoacids (ee up to 95%); (ii) the palladium-diphosphine catalyzed allylic alkylation of 1,3-diphenylallylacetate (ee up to 95%); (iii) the ruthenium pianostool-catalyzed transfer hydrogenation of prochiral ketones (ee up to 97% for aryl-alkyl ketones and ee up to 90% for dialkyl ketones); (iv) the vanadyl-catalyzed oxidation of prochiral sulfides (ee up to 93%). A number of noteworthy features are reminiscent of homogeneous catalysis, including straightforward access to both enantiomers of the product, the broad substrate scope, organic solvent tolerance, and an accessible range of reactions that are typical of homogeneous catalysts. Enzyme-like features include access to genetic optimization, an aqueous medium as the preferred solvent, Michaelis-Menten behavior, and single-substrate derivatization. The X-ray characterization of artificial metalloenzymes provides fascinating insight into possible enantioselection mechanisms involving a well-defined second coordination sphere environment. Thus, such artificial metalloenzymes combine attractive features of both homogeneous and enzymatic kingdoms. In the spirit of surface borrowing, that is, modulating ligand affinity by harnessing existing protein surfaces, this strategy can be extended to selectively binding streptavidin-incorporated biotinylated ruthenium pianostool complexes to telomeric DNA. This application paves the way for chemical biology applications of artificial metalloenzymes.