Rubisco adaptation is more limited by phylogenetic constraint than by catalytic trade-off

Rubisco assimilates CO2 to form the sugars that fuel life on earth. Correlations between rubisco kinetic traits across species have led to the proposition that rubisco adaptation is highly constrained by catalytic trade-offs. However, these analyses did not consider the phylogenetic context of the enzymes that were analysed. Thus, it is possible that the correlations observed were an artefact of the presence of phylogenetic signal in rubisco kinetics and the phylogenetic relationship between the species that were sampled. Here, we conducted a phylogenetically-resolved analysis of rubisco kinetics and show that there is a significant phylogenetic signal in rubisco kinetic traits. We re-evaluated the extent of catalytic trade-offs accounting for this phylogenetic signal and found that all were attenuated. Following phylogenetic correction, the largest catalytic trade-offs were observed between the Michaelis constant for CO2 and carboxylase turnover (∼21-37%), and between the Michaelis constants for CO2 and O2 (∼9-19%), respectively. All other catalytic trade-offs were substantially attenuated such that they were marginal (<9%) or non-significant. This phylogenetically resolved analysis of rubisco kinetic evolution also identified kinetic changes that occur concomitant with the evolution of C4 photosynthesis. Finally, we show that phylogenetic constraints have played a larger role than catalytic trade-offs in limiting the evolution of rubisco kinetics. Thus, although there is strong evidence for some catalytic trade-offs, rubisco adaptation has been more limited by phylogenetic constraint than by the combined action of all such trade-offs.

Characterization of a cryptic, pyrroloquinoline quinone-dependent dehydrogenase of Gluconobacter sp. strain CHM43

ABSTRACT We characterized the pyrroloquinoline quinone (PQQ)-dependent dehydrogenase 9 (PQQ-DH9) of Gluconobacter sp. strain CHM43, which is a homolog of PQQ-dependent glycerol dehydrogenase (GLDH). We used a plasmid construct to express PQQ-DH9. The expression host was a derivative strain of CHM43, which lacked the genes for GLDH and the membrane-bound alcohol dehydrogenase and consequently had minimal ability to oxidize primary and secondary alcohols. The membranes of the transformant exhibited considerable d-arabitol dehydrogenase activity, whereas the reference strain did not, even if it had PQQ-DH9-encoding genes in the chromosome and harbored the empty vector. This suggests that PQQ-DH9 is not expressed in the genome. The activities of the membranes containing PQQ-DH9 and GLDH suggested that similar to GLDH, PQQ-DH9 oxidized a wide variety of secondary alcohols but had higher Michaelis constants than GLDH with regard to linear substrates such as glycerol. Cyclic substrates such as cis-1,2-cyclohexanediol were readily oxidized by PQQ-DH9.

Effects of Imprinting and Water Activity on Transesterification and Thermostability with Lipases in Ionic Liquid

The effect of bio-imprinting and water activity on catalytic activities and the thermostability of lipases was investigated for transesterification using vinyl acetate and benzyl alcohol as substrates in ionic liquid, [Cnmim][PF6] (n=4,6,8), and benzene. The catalytic activities were enhanced by imprinting in benzene and [C4mim][PF6], and the relations between the transesterification activities and the water activity in both solvents were approximately bell shaped. The reactivity of the transesterification in benzene was higher than that in [C4<br /> mim][PF6]. The effects of water activity and imprinting on the kinetic parameters in [C4mim][PF6] were examined. Without controlling the water content, the values of Km,VA and Km,BA (Michaelis constants of vinyl acetate and benzyl alcohol, respectively) decreased, and the values of Vm (maximum rate) increased by imprinting. On the other hand, by controlling the water content in the organic media, the values of Vm, Km,VA, and Km,BA increased by imprinting. The activities of lipase in ionic liquid are more strongly affected by water activity and imprinting than those in benzene. We observed effects of water activity on thermostability but none from imprinting.

Microfluidics Featuring Multilayered Hydrogel Assemblies Enable Real-Time NMR-Monitoring of Enzyme Cascade Reactions

Compartmentalized chemical reactions at the microscale are interesting from many perspectives including (multi)functional surfaces and biotechnology. Monitoring the molecular content as a measure of functional performance at these small scales is challenging. As a means to address this challenge, we leverage microtechnology and biocompatible materials to integrate a compact, reconfigurable reaction cell featuring electrochemical functionality with high-resolution nuclear magnetic resonance spectroscopy (NMR). We demonstrate the operation of this system by monitoring the activity of enzymes immobilized in chemically distinct layers within a multi-layered chitosan hydrogel assembly. As a benchmark, we observed the parallel activities of urease (Urs), catalase (Cat), and glucose oxidase (GOx) by recording NMR spectra to extract reagent and product concentrations in real-time. As a result, simultaneous monitoring of a cooperative enzymatic process (GOx + Cat) together with an independent process (Urs) is achieved. Using Michaelis-Menten progress curve analysis of the NMR data, kinetic data is extracted: in the case of GOx, the Michaelis constants (K_M) are consistent with previous reports, while for Urs, deviations are observed, attributed to an inhibitory effect under our reaction conditions. The system therefore enables the construction of complex reaction cascades with spatial control, as would be interesting in, for example, metabolic engineering and multiplexed sensing applications.

Microfluidics Featuring Multilayered Hydrogel Assemblies Enable Real-Time NMR-Monitoring of Enzyme Cascade Reactions

Compartmentalized chemical reactions at the microscale are interesting from many perspectives including (multi)functional surfaces and biotechnology. Monitoring the molecular content as a measure of functional performance at these small scales is challenging. As a means to address this challenge, we leverage microtechnology and biocompatible materials to integrate a compact, reconfigurable reaction cell featuring electrochemical functionality with high-resolution nuclear magnetic resonance spectroscopy (NMR). We demonstrate the operation of this system by monitoring the activity of enzymes immobilized in chemically distinct layers within a multi-layered chitosan hydrogel assembly. As a benchmark, we observed the parallel activities of urease (Urs), catalase (Cat), and glucose oxidase (GOx) by recording NMR spectra to extract reagent and product concentrations in real-time. As a result, simultaneous monitoring of a cooperative enzymatic process (GOx + Cat) together with an independent process (Urs) is achieved. Using Michaelis-Menten progress curve analysis of the NMR data, kinetic data is extracted: in the case of GOx, the Michaelis constants (K_M) are consistent with previous reports, while for Urs, deviations are observed, attributed to an inhibitory effect under our reaction conditions. The system therefore enables the construction of complex reaction cascades with spatial control, as would be interesting in, for example, metabolic engineering and multiplexed sensing applications.

Prediction of Michaelis constants from structural features using deep learning

ABSTRACTThe Michaelis constant KM describes the affinity of an enzyme for a specific substrate, and is a central parameter in studies of enzyme kinetics and cellular physiology. As measurements of KM are often difficult and time-consuming, experimental estimates exist for only a minority of enzyme-substrate combinations even in model organisms. Here, we build and train an organism-independent model that successfully predicts KM values for natural enzyme-substrate combinations using machine and deep learning methods. Predictions are based on a task-specific molecular fingerprint of the substrate, generated using a graph neural network, and the domain structure of the enzyme. Model predictions can be used to estimate enzyme efficiencies, to relate metabolite concentrations to cellular physiology, and to fill gaps in the parameterization of kinetic models of cellular metabolism.