High-throughput enrichment of temperature-sensitive argininosuccinate synthetase for two-stage citrulline production in E. coli.

Controlling metabolism of engineered microbes is important to modulate cell growth and production during a bioprocess. For example, external parameters such as light, chemical inducers, or temperature can act on metabolism of production strains by changing the abundance or activity of enzymes. Here, we created temperature-sensitive variants of an essential enzyme in arginine biosynthesis of Escherichia coli (argininosuccinate synthetase, ArgG) and used them to dynamically control citrulline overproduction and growth of E. coli. We show a method for high-throughput enrichment of temperature-sensitive ArgG variants with a fluorescent TIMER protein and flow cytometry. With 90 of the thus derived ArgG variants, we complemented an ArgG deletion strain showing that 90% of the strains exhibit temperature-sensitive growth and 69% of the strains are auxotrophic for arginine at 42 °C and prototrophic at 30 °C. The best temperature-sensitive ArgG variant enabled precise and tunable control of cell growth by temperature changes. Expressing this variant in a feedback-dysregulated E. coli strain allowed us to realize a two-stage bioprocess: a 33 °C growth-phase for biomass accumulation and a 39 °C stationary-phase for citrulline production. With this two-stage strategy, we produced 3 g/L citrulline during 45 h cultivation in a 1-L bioreactor. These results show that temperature-sensitive enzymes can be created en masse and that they may function as metabolic valves in engineered bacteria.

Multi-omics Analysis of CRISPRi-Knockdowns Identifies Mechanisms that Buffer Decreases of Enzymes in E. coli Metabolism.

Enzymes maintain metabolism, and their concentration affects cellular fitness: high enzyme levels are costly, and low enzyme levels can limit metabolic flux. Here, we used CRISPR interference (CRISPRi) to study the consequences of decreasing E. coli enzymes below wild-type levels. A pooled CRISPRi screen with 7,177 strains demonstrates that metabolism buffers fitness defects for hours after the induction of CRISPRi. We characterized the metabolome and proteome responses in 30 CRISPRi strains and elucidated three gene-specific buffering mechanisms: ornithine buffered the knockdown of carbamoyl phosphate synthetase (CarAB) by increasing CarAB activity, S-adenosylmethionine buffered the knockdown of homocysteine transmethylase (MetE) by de-repressing expression of the methionine pathway, and 6-phosphogluconate buffered the knockdown of 6-phosphogluconate dehydrogenase (Gnd) by activating a bypass. In total, this work demonstrates that CRISPRi screens can reveal global sources of metabolic robustness and identify local regulatory mechanisms that buffer decreases of specific enzymes. A record of this paper's transparent peer review process is included in the Supplemental Information.

Selective Enrichment of Slow-Growing Bacteria in a Metabolism-Wide CRISPRi Library with a TIMER Protein.

Construction of pooled genetic variant libraries has become very fast and versatile. The current limitation of this technique is to select cells with a desired phenotype from very large libraries. Especially cells with poor fitness and slow growth are difficult to select because they are rapidly outcompeted by fitter cells. Here, we demonstrate selective and high-throughput enrichment of slow-growing strains using a fluorescent TIMER protein and flow cytometry. As a proof of principle, we created a metabolism-wide CRISPR interference library for Escherichia coli and enriched targets that interfere with amino acid metabolism. After enrichment of slow-growing cells, the CRISPRi library consisted almost entirely of targets that block amino acid biosynthesis. These results provide general guidelines for how to enrich slow-growing strains from a large pool of genetic variants, with applications in genetic screens, metabolic engineering, and synthetic biology.