Functional annotation of novel proteins lags behind the number of sequences discovered by the next-generation sequencing. The throughput of conventional testing methods is far too low compared to sequencing; thus, experimental alternatives are needed. Microfluidics offer high throughput and reduced sample consumption as a tool to keep up with a sequence-based exploration of protein diversity. The most promising droplet-based systems have a significant limitation: leakage of hydrophobic compounds from water compartments to the carrier prevents their use with hydrophilic reagents. Here, we present a novel approach of substrate delivery into microfluidic droplets and apply it to high-throughput functional characterization of enzymes that convert hydrophobic substrates. Substrate delivery is based on the partitioning of hydrophobic chemicals between the oil and water phases. We applied a controlled distribution of 27 hydrophobic haloalkanes from oil to reaction water droplets to perform substrate specificity screening of eight model enzymes from the haloalkane dehalogenase family. This droplet-on-demand microfluidic system reduces the reaction volume 65 000-times and increases the analysis speed almost 100-fold compared to the classical test tube assay. Additionally, the microfluidic setup enables a convenient analysis of dependences of activity on the temperature in a range of 5 to 90 °C for a set of mesophilic and hyperstable enzyme variants. A high correlation between the microfluidic and test tube data supports the approach robustness. The precision is coupled to a considerable throughput of >20 000 reactions per day and will be especially useful for extending the scope of microfluidic applications for high-throughput analysis of reactions including compounds with limited water solubility.

Department of Biochemistry University of Cambridge 80 Tennis Court Road Cambridge CB2 1GA United Kingdom

Institute of Scientific Instruments Czech Academy of Sciences Kralovopolska 147 Brno 612 64 Czech Republic

International Clinical Research Center St Anne's University Hospital Pekarska 53 Brno 656 91 Czech Republic

Living Systems Institute University of Exeter Exeter EX4 4QD United Kingdom

Loschmidt Laboratories Department of Experimental Biology and RECETOX Faculty of Science Masaryk University Kamenice 5 Brno 625 00 Czech Republic

Citace poskytuje Crossref.org

Engineering Dehalogenase Enzymes Using Variational Autoencoder-Generated Latent Spaces and Microfluidics

Advancing Enzyme's Stability and Catalytic Efficiency through Synergy of Force-Field Calculations, Evolutionary Analysis, and Machine Learning

Engineering the protein dynamics of an ancestral luciferase