Search

Homochiral 2-hydroxy-4-butyrolactone derivatives are important structural motifs frequently found in naturally occurring biologically active products, synthetic drugs, biodegradable polymers as well as building blocks and chiral auxiliaries in asymmetric organic synthesis (1). Among the methods described in the literature, the stereoselective reduction of 4-hydroxy-2-oxoacid derivatives to produce 2-hydroxy-4-butyrolactone derivatives remains unexplored. In this communication we report a tandem biocatalytic stereodivergent route for the preparation of these compounds using two stereocomplementary aldolases and oxidoreductases, using simple and achiral starting materials. The strategy comprises (Scheme 1) (i) aldol addition reaction of 2-oxoacids to formaldehyde using both 3-methyl-2-oxobutanoate hydroxymethyltransferase and 2-keto-3-deoxy-L-rhamnonate aldolase and variants thereof, and (ii) subsequent 2-oxogroup reduction of the aldol adduct by both ketopantoate reductase and a delta-1-piperidine-2-carboxylate/delta-1-pyrroline-2-carboxylate reductase with uncovered promiscuous ketoreductase activity. Finally, the formation of the corresponding lactone takes place during the work up and purification steps. Yields, enantiomeric excesses and diasteromeric ratios for a total of 23 structurally diverse 2-hydroxy-3-substituted-4-butyrolactones will be discussed. Moreover, one-pot one-step cascade reactions with the aldolases and reductases operating in tandem will be presented.

NAD(P)H-dependent ¿1-piperideine-2-carboxylate/¿1-pyrroline-2-carboxylate reductase from Pseudomonas syringae pv. tomato (DpkAPsyrin, EC 1.5.1.21) catalyzes the stereoselective reduction of ¿1-piperideine-2-carboxylate and ¿1-pyrroline-2-carboxylate to L-pipecolic acid and L-proline respectively. Furthermore, DpkAPsyrin is also able to catalyze the enantioselective synthesis of N-methyl-L-amino acids from methylamine and various 2-oxo acids [1]. In our research group, we observed that in addition to the imine reductase activity, DpkAPsyrin have a promiscuous ketoreductase activity and, consequently, we consider interesting to exploit its underdeveloped synthetic capabilities for the synthesis of 2-hydroxy-4-arylbut-3-enoic acid (4). These 2-hydroxyacids are intermediates in the synthesis of 2-hydroxy-4-arylbutanoic acid, important building blocks for the production of a variety of angiotensin converting enzyme (ACE) inhibitors (e.g. enalapril, lisinopril, cilapril, and benazepril) [2]. In this communication we reported the preparation of structurally diverse (S)-2-hydroxy-4-arylbut-3-enoic acid (4) (isolated yield: 57%-78% and ee: 87%->99%) from aromatic aldehydes (1) and pyruvate (2) by enzymatic cascade reaction using trans-o-hydroxybenzylidene pyruvate hydratase-aldolase from Pseudomonas putida HBPAPputida) [3] and DpkAPsyrin (Scheme 1). The cascade process was possible because it was found that the ketoreductase preferentially reduced the aldol condensation adducts (3) rather than the pyruvate (2). This allows for high levels of 2-hydroxy-4-arylbut-3-enoic acids (4) conversions ( >95% after 24 h).

Stevia rebaudiana leaf extracts contain stevioside and rebaudioside A, two steviol glycosides (SGs) used as natural sweeteners because of their non-toxic, thermally stable and non-caloric properties. Indeed, leaf extracts can be up to 300 times sweeter than sucrose. Stevioside and rebaudioside A have organoleptic differences, the first one having an undesirable bitterness and the second one a higher sweetener capacity. Selection of the S. rebaudiana varieties and the best environmental conditions that elicit higher SGs content and the appropriate composition is an important goal. In this study we quantified and compared the amount of stevioside and rebaudioside A in two of the most used S. rebaudiana cultivars, Morita II and Criolla. Our results show a strong differential ratio of stevioside and rebaudioside A accumulated in the leaf between these cultivars. The Criolla cultivar showed about 3 times more stevioside per mg of dry weight than Morita II, whereas the Morita II accumulated almost 10 times more rebaudioside A than that produced in Criolla. We observed an enhanced expression in Morita II of three genes (SrKA13H, SrUGT74G1 and SrUGT76G1) known to encode three enzymes that participate in SGs biosynthesis, likely contributing to the differences in the stevioside and rebaudioside A accumulation. Not only genetic variation can affect SGs composition, but also environmental factors and crop management. Numerous studies have shown that the light regime in which S. rebaudiana cultivars grow can affect SGs accumulation. However, the optimal light regime to increase total SGs content is currently controversial. By applying various light intensities, we detected an increase of expression of these three biosynthetic genes at higher light intensity, accompanied by higher levels of stevioside and rebaudioside A, demonstrating that light intensity influences the synthesis of SGs.

Drought is a natural, erratic phenomenon that has a widespread and significant impact on socioeconomic and environmental development. The early monitoring and evaluation of drought through forecasting models would allow the articulation of early control and mitigation strategies, thus achieving an optimal development in planning and preparation for climate change. Therefore, this research developed a methodology for spatiotemporal analysis of drought patterns using automatic learning tools in the dry corridor of Central America. To this end, some specific milestones were defined. These include: (i) To assess temporal and spatial meteorological and agricultural droughts events, (ii) Identifying and validate results of the spatiotemporal events using computer vision techniques and finally (iii) Implementing machine learning drought forecasting models. ERA 5 monthly land average dataset was used as input for index estimation, spatiotemporal analysis and forecasting models. The frequency of drought events was calculated using standardized SPI and SPEI indices for accumulation periods of 1,3,6,9. However, 3,6 allowed a more realistic analysis of the seasonal change conditions in the hydrological regime of the area and the identification of the existing teleconnection between drought events and scale propagation. Regarding the spatiotemporal dynamics, 97 drought events of greater extension were identified, which are generally originated in countries such as Guatemala, Nicaragua, and El Salvador between seasonal periods not longer than 7 months. Additionally, the suitability of automatic learning models such as SVR, ANN and deep learning such as LSTM for index forecasting (r2=0.80) and drought dynamics in a temporal window of 1 to 6 months ahead was verified with considerable performance. The presented methodology provides an important basis for drought characterization and forecasting through the integration of spatiotemporal tracking models and machine learning technique

Drought is a natural, erratic phenomenon that has a widespread and significant impact on socioeconomic and environmental development. The early monitoring and evaluation of drought through forecasting models would allow the articulation of early control and mitigation strategies, thus achieving an optimal development in planning and preparation for climate change. Therefore, this research developed a methodology for spatiotemporal analysis of drought patterns using automatic learning tools in the dry corridor of Central America. To this end, some specific milestones were defined. These include: (i) To assess temporal and spatial meteorological and agricultural droughts events, (ii) Identifying and validate results of the spatiotemporal events using computer vision techniques and finally (iii) Implementing machine learning drought forecasting models. ERA 5 monthly land average dataset was used as input for index estimation, spatiotemporal analysis and forecasting models. The frequency of drought events was calculated using standardized SPI and SPEI indices for accumulation periods of 1,3,6,9. However, 3,6 allowed a more realistic analysis of the seasonal change conditions in the hydrological regime of the area and the identification of the existing teleconnection between drought events and scale propagation. Regarding the spatiotemporal dynamics, 97 drought events of greater extension were identified, which are generally originated in countries such as Guatemala, Nicaragua, and El Salvador between seasonal periods not longer than 7 months. Additionally, the suitability of automatic learning models such as SVR, ANN and deep learning such as LSTM for index forecasting (r2=0.80) and drought dynamics in a temporal window of 1 to 6 months ahead was verified with considerable performance. The presented methodology provides an important basis for drought characterization and forecasting through the integration of spatiotemporal tracking models and machine learning technique

The use of enzymes in industrial processes is often limited by the unavailability of biocatalysts with prolonged stability. Thermostable enzymes allow increased process temperature and thus higher substrate and product solubility, reuse of expensive biocatalysts, resistance against organic solvents, and better "evolvability"of enzymes. In this work, we have used an activity-independent method for the selection of thermostable variants of any protein in Thermus thermophilus through folding interference at high temperature of a thermostable antibiotic reporter protein at the C-terminus of a fusion protein. To generate a monomeric folding reporter, we have increased the thermostability of the moderately thermostable Hph5 variant of the hygromycin B phosphotransferase from Escherichia coli to meet the method requirements. The final Hph17 variant showed 1.5 °C higher melting temperature (Tm) and 3-fold longer half-life at 65 °C compared to parental Hph5, with no changes in the steady-state kinetic parameters. Additionally, we demonstrate the validity of the reporter by stabilizing the 2-keto-3-deoxy-l-rhamnonate aldolase from E. coli (YfaU). The most thermostable multiple-mutated variants thus obtained, YfaU99 and YfaU103, showed increases of 2 and 2.9 °C in Tm compared to the wild-type enzyme but severely lower retro-aldol activities (150- and 120-fold, respectively). After segregation of the mutations, the most thermostable single variant, Q107R, showed a Tm 8.9 °C higher, a 16-fold improvement in half-life at 60 °C and higher operational stability than the wild-type, without substantial modification of the kinetic parameters.

A multi-enzyme route for the production of an atorvastatin side-chain was proposed. The approach includes the consecutive double aldol addition of acetaldehyde to the phenylacetamide amino-protected propanal by 2-deoxyribose-5-phosphate aldolase (DERA), the oxidation by ketoreductase, the lipase-catalysed acylation and amino group deprotection by penicillin G-acylase. To explore the feasibility of the process, the DERA reaction was studied into detail. Based on the kinetic and stability studies, a mathematical model was developed and validated in a batch and fed-batch reactor. The highest productivity was obtained in repetitive batch reactor (229.1 g/(L day)). The highest final product concentration of 124 g/L was obtained in fed-batch reactor. The mathematical model-based optimisation provided insight into the possibilities for process metrics improvement. Further, the second step was explored. The results showed that the DERA reaction and the oxidation reaction can be carried out as the multi-enzyme one pot process in the sequential manner.

Aldolase and transaminase coexpressed in Escherichia coli cells and lyophilized (i.e., lyophilized whole-cell biocatalyst (LWCB)) were used as biocatalysts for the one-pot cascade synthesis of l-homoserine with substrate cycling. The kinetic analysis of enzymes within lyophilized cells was performed to evaluate the behavior of the system. The best result among the performed fed-batch reactor experiments achieved was 640.8 mM (76.3 g L–1) of l-homoserine with a volume productivity of 2.6 g L–1 h–1. This is comparable with the results of the same cascade synthesis using cell-free extracts (CFEs) and significantly better than the reports in the literature applying fermentation technology. The approach applied here can serve as guidance for the design of microbial cells with an optimal ratio of expressed enzymes that act as biocatalysts in the cascade, resulting in lower biocatalyst cost, no need for the addition of expensive coenzymes, and enhanced enzyme stability as compared with cell-free extracts.

L-Homoserine is an important compound and a building block in chemical and pharmaceutical industry. The one-pot cascade synthesis of L-homoserine with substrate recycling catalysed by an aldolase and a transaminase was investigated in detail, with special interest in the reaction’s kinetics and process development. Complex relationships between the process variables were discovered and the kinetic and reactor model were developed. In addition, new chemical insights were developed through this model-driven process development, and for the first time a kinetic model was proposed to describe the YfaU-catalysed reaction. The modelling was crucial for a better understanding of the reaction system, as well as for process optimisation. Finding a proper strategy for process improvement by manipulating the reaction conditions, as well as developing process design, was imperative to achieve the process metrics necessary for scale-up. Using the optimized reaction conditions in the fed-batch reactor with a continuous formaldehyde supply, and with subsequent addition of pyruvate and L-alanine, 80.1 g L–1 of L-homoserine was obtained with a volume productivity of 3.2 g L–1 h–1 after 25 h. These results present a 100% increase in product concentration and approximately 18% increase in volume productivity in comparison with previously published work. This work shows the benefit of the synergistic approach of bio(chemical) knowledge and process model development working hand-in-hand which is crucial for stronger breakthrough of biocatalysis to chemical industry.

Nitrogen heterocycles are structural motifs found in many bioactive natural products and of utmost importance in pharmaceutical drug development. In this work, a stereoselective synthesis of functionalized N-heterocycles was accomplished in two steps, comprising the biocatalytic aldol addition of ethanal and simple aliphatic ketones such as propanone, butanone, 3-pentanone, cyclobutanone, and cyclopentanone to N -Cbz-protected aminoaldehydes using engineered variants of -fructose-6-phosphate aldolase from Escherichia coli (FSA) or 2-deoxy--ribose-5-phosphate aldolase from Thermotoga maritima (DERA) as catalysts. FSA catalyzed most of the additions of ketones while DERA was restricted to ethanal and propanone. Subsequent treatment with hydrogen in the presence of palladium over charcoal, yielded low-level oxygenated N-heterocyclic derivatives of piperidine, pyrrolidine and N-bicyclic structures bearing fused cyclobutane and cyclopentane rings, with stereoselectivities of 96-98 ee and 97:3 dr in isolated yields ranging from 35 to 79%.

3-Hydroxyisobutyric acid is an important intermediate in the biosynthesis of methacrylic acid. Its biocatalytic synthesis can be performed by aldolase-catalyzed aldol addition of propanal to formaldehyde followed by an enzymatic oxidation of the resulting 3-hydroxy-2-methylpropanal to 3-hydroxyisobutyric acid. In this work, D-fructose-6-phosphate aldolase D6Q variant was investigated as a key step for the biocatalytic preparation of 3-hydroxy-2-methylpropanal, a commercially unavailable precursor of 3-hydroxyisobutyric acid. The kinetic model of this step was developed for the purpose of reactor selection and process optimization. It was found that enzyme operational stability decay is co-dependent on the initial formaldehyde concentration. Thus, the choice of the initial conditions is crucial for a successful process set-up. It was concluded that fed-batch was the best reactor choice for this reaction due to enzyme inhibition by formaldehyde and propanal, and its operational stability decay. At the optimal process conditions, the product concentration, product yield, and volume productivity after 5.5 h were 72 g L−1, 88.5% and 313.7 g L−1 d−1, respectively. Enzymatic oxidation of 3-hydroxy-2-methylpropanal to the corresponding acid was performed as a proof of concept using an aldehyde dehydrogenase in the presence of NAD+, regenerated by water-forming NADH oxidase, and 2.5 g L−1 (24 mM) of 3-hydroxyisobutyric acid was obtained. © 2019 Institution of Chemical Engineers

Supports based on poly-styrene-divinylbenzene (PSD) are commercially available since a long time ago. However, they are not commonly used as enzyme immobilization matrices. The main reason for this lies in the negative effect of the very hydrophobic surface on enzyme stability that produces the instantaneous enzyme inactivation in many instances. However, they have recently regained some impact in enzyme immobilization. They are easy to modify, and have been prepared with different modifiers. We will pay special attention to the coating of these supports with ionic liquids, which permits to have the ionic liquid phase anchored to the solid and modulate the enzyme properties without risk of losing these expensive and potentially toxic compounds. Thus, this review will present the covalent or physical immobilization of enzymes on PSD supports, submitted to different modifications. Moreover, lipases immobilized via interfacial activation on some naked PSD supports have shown some unexpected improvement in their catalytic properties, with uses in reactions like hydrolysis, esterification or transesterification.

Supports based on poly-styrene-divinylbenzene (PSD) are commercially available since a long time ago. However, they are not commonly used as enzyme immobilization matrices. The main reason for this lies in the negative effect of the very hydrophobic surface on enzyme stability that produces the instantaneous enzyme inactivation in many instances. However, they have recently regained some impact in enzyme immobilization. They are easy to modify, and have been prepared with different modifiers. We will pay special attention to the coating of these supports with ionic liquids, which permits to have the ionic liquid phase anchored to the solid and modulate the enzyme properties without risk of losing these expensive and potentially toxic compounds. Thus, this review will present the covalent or physical immobilization of enzymes on PSD supports, submitted to different modifications. Moreover, lipases immobilized via interfacial activation on some naked PSD supports have shown some unexpected improvement in their catalytic properties, with uses in reactions like hydrolysis, esterification or transesterification.

Rhamnulose-1-phosphate aldolase from Thermotoga maritima (Rhu1PATm) has been recently cloned and characterized. This hyperthermophilic enzyme offers intriguing possibilities for practical catalysis. This is due to its high stability under extreme reaction conditions, such as high temperature or the presence of organic co-solvents. The Rhu1PATm potentiality in organic synthesis has been explored focusing on i) the reaction stereocontrol, ii) the possibility of combining it with other mesophilic enzymes in multienzyme systems and iii) its application to the synthesis of imino- and nitrocyclitols. In our study, Rhu1PATm diastereoselectivity was similar to the one reported by Rhu1PA from Escherichia coli (Rhu1PAEc). However, we observed significant differences for some aldehyde acceptors. Indeed, the diastereoselectivity control was not complete, since mixtures of L-threo (2R,3S; natural stereopreference) and D-erythro (2R,3R) diastereoisomers were obtained as described for Rhu1PAEc. Rhu1PATm was able to catalyse aldol reactions using nitroaldehydes and N-Cbz-aminoaldehydes. Conversion of the selected nitroaldehydes was complete, leading to mixtures of L-threo and D-erythro diastereoisomers, in 95:5 and 75:25 ratios. When N-Cbz-aminoaldehydes were used, aldol reaction yields were lower and the stereoselectivity L-threo:D-erythro ranged between 32:68 to >95:5.

Hydrogen peroxide is a substrate or side-product in many enzyme-catalyzed reactions. For example, it is a side-product of oxidases, resulting from the re-oxidation of FAD with molecular oxygen, and it is a substrate for peroxidases and other enzymes. However, hydrogen peroxide is able to chemically modify the peptide core of the enzymes it interacts with, and also to produce the oxidation of some cofactors and prostetic groups (e.g., the hemo group). Thus, the development of strategies that may permit to increase the stability of enzymes in the presence of this deleterious reagent is an interesting target. This enhancement in enzyme stability has been attempted following almost all available strategies: site-directed mutagenesis (eliminating the most reactive moieties), medium engineering (using stabilizers), immobilization and chemical modification (trying to generate hydrophobic environments surrounding the enzyme, to confer higher rigidity to the protein or to generate oxidation-resistant groups), or the use of systems capable of decomposing hydrogen peroxide under very mild conditions. If hydrogen peroxide is just a side-product, its immediate removal has been reported to be the best solution. In some cases, when hydrogen peroxide is the substrate and its decomposition is not a sensible solution, researchers coupled one enzyme generating hydrogen peroxide “in situ” to the target enzyme resulting in a continuous supply of this reagent at low concentrations thus preventing enzyme inactivation. This review will focus on the general role of hydrogen peroxide in biocatalysis, the main mechanisms of enzyme inactivation produced by this reactive and the different strategies used to prevent enzyme inactivation caused by this “dangerous liaison”.

Hydrogen peroxide is a substrate or side-product in many enzyme-catalyzed reactions. For example, it is a side-product of oxidases, resulting from the re-oxidation of FAD with molecular oxygen, and it is a substrate for peroxidases and other enzymes. However, hydrogen peroxide is able to chemically modify the peptide core of the enzymes it interacts with, and also to produce the oxidation of some cofactors and prostetic groups (e.g., the hemo group). Thus, the development of strategies that may permit to increase the stability of enzymes in the presence of this deleterious reagent is an interesting target. This enhancement in enzyme stability has been attempted following almost all available strategies: site-directed mutagenesis (eliminating the most reactive moieties), medium engineering (using stabilizers), immobilization and chemical modification (trying to generate hydrophobic environments surrounding the enzyme, to confer higher rigidity to the protein or to generate oxidation-resistant groups), or the use of systems capable of decomposing hydrogen peroxide under very mild conditions. If hydrogen peroxide is just a side-product, its immediate removal has been reported to be the best solution. In some cases, when hydrogen peroxide is the substrate and its decomposition is not a sensible solution, researchers coupled one enzyme generating hydrogen peroxide “in situ” to the target enzyme resulting in a continuous supply of this reagent at low concentrations thus preventing enzyme inactivation. This review will focus on the general role of hydrogen peroxide in biocatalysis, the main mechanisms of enzyme inactivation produced by this reactive and the different strategies used to prevent enzyme inactivation caused by this “dangerous liaison”.