Smart Microbial Cells Couple Catalysis and Sensing to Provide High-Throughput Selection of an Organophosphate Hydrolase

Targeted mutagenesis and high-throughput screening of diversified gene and promoter libraries for isolating gain-of-function mutations

Targeted mutagenesis of a promoter or gene is essential for attaining new functions in microbial and protein engineering efforts. In the burgeoning field of synthetic biology, heterologous genes are expressed in new host organisms. Similarly, natural or designed proteins are mutagenized at targeted positions and screened for gain-of-function mutations. Here, we describe methods to attain complete randomization or controlled mutations in promoters or genes. Combinatorial libraries of one hundred thousands to tens of millions of variants can be created using commercially synthesized oligonucleotides, simply by performing two rounds of polymerase chain reactions. With a suitably engineered reporter in a whole cell, these libraries can be screened rapidly by performing fluorescence-activated cell sorting (FACS). Within a few rounds of positive and negative sorting based on the response from the reporter, the library can rapidly converge to a few optimal or extremely rare variants with desired phenotypes. Library construction, transformation and sequence verification takes 6-9 days and requires only basic molecular biology lab experience. Screening the library by FACS takes 3-5 days and requires training for the specific cytometer used. Further steps after sorting, including colony picking, sequencing, verification, and characterization of individual clones may take longer, depending on number of clones and required experiments.

Tackling the Catch-22 Situation of Optimizing a Sensor and a Transporter System in a Whole-Cell Microbial Biosensor Design for an Anthropogenic Small Molecule

Whole-cell biosensors provide a convenient detection tool for the high-throughput screening of genetically engineered biocatalytic activity. But establishing a biosensor for an anthropogenic molecule requires both a custom transporter and a transcription factor. This results in an unavoidable "Catch-22" situation in which transporter activity cannot be easily confirmed without a biosensor and a biosensor cannot be established without a functional transporter in a host organism. We overcame this type of circular problem while developing an adipic acid (ADA) sensor. First, leveraging an established cis,cis-muconic acid (ccMA) sensor, an annotated ccMA transporter MucK, which is expected to be broadly responsive to dicarboxylates, was stably expressed in the genome of Pseudomonas putida to function as a transporter for ADA, and then a PcaR transcription factor (endogenous to the strain and naturally induced by β-ketoadipic acid, BKA) was diversified and selected to detect the ADA molecule. While MucK expression is otherwise very unstable in P. putida under strong promoter expression, our optimized mucK expression was functional for over 70 generations without loss of function, and we selected an ADA sensor that showed a specificity switch of over 35-fold from BKA at low concentrations (typically <0.1 mM of inducers). Our ADA and BKA biosensors show high sensitivity (low detection concentration <10 μM) and dynamic range (∼50-fold) in an industrially relevant organism and will open new avenues for high throughput discovery and optimization of enzymes and metabolic pathways for the biomanufacture of these molecules. In particular, the novel ADA sensor will aid the discovery and evolution of efficient biocatalysts for the biological recycling of ADA from the degradation of nylon-6,6 waste.

Carbohydrate-Active enZyme (CAZyme) enabled glycoengineering for a sweeter future

One of the stumbling blocks to advance the field of glycobiology has been the difficulty in synthesis of bespoke carbohydrate-based molecules like glycopolymers (e.g. human milk oligosaccharides) and glycoconjugates (e.g. glycosylated monoclonal antibodies). Recent strides towards using engineered Carbohydrate-Active enZymes (CAZymes) like glycosyl transferases, transglycosidases, and glycosynthases for glycans synthesis has allowed production of diverse glycans. Here, we discuss enzymatic routes for glycans biosynthesis and recent advances in protein engineering strategies that enable improvement of CAZyme specificity and catalytic turnover. We focus on rational and directed evolution methods that have been developed to engineer CAZymes. Finally, we discuss how improved CAZymes have been used in recent years to remodel and synthesize glycans for biotherapeutics and biotechnology related applications.

Automated classification of bacterial cell sub-populations with convolutional neural networks

Quantification of phenotypic heterogeneity present amongst bacterial cells can be a challenging task. Conventionally, classification and counting of bacteria sub-populations is achieved with manual microscopy, due to the lack of alternative, high-throughput, autonomous approaches. In this work, we apply classification-type convolutional neural networks (cCNN) to classify and enumerate bacterial cell sub-populations (B. subtilis clusters). Here, we demonstrate that the accuracy of the cCNN developed in this study can be as high as 86% when trained on a relatively small dataset (81 images). We also developed a new image preprocessing algorithm, specific to fluorescent microscope images, which increases the amount of training data available for the neural network by 72 times. By summing the classified cells together, the algorithm provides a total cell count which is on parity with manual counting, but is 10.2 times more consistent and 3.8 times faster. Finally, this work presents a complete solution framework for those wishing to learn and implement cCNN in their synthetic biology work.