Microbes can be engineered to synthesize a wide array of bioproducts, yet production phenotype evaluation remains a frequent bottleneck in the design-build-test cycle where strain development requires iterative rounds of library construction and testing. Here, we present Syntrophic Co-culture Amplification of Production phenotype (SnoCAP). Through a metabolic cross-feeding circuit, the production level of a target molecule is translated into highly distinguishable co-culture growth characteristics, which amplifies differences in production into highly distinguishable growth phenotypes. We demonstrate SnoCAP with the screening of Escherichia coli strains for production of two target molecules: 2-ketoisovalerate, a precursor of the drop-in biofuel isobutanol, and L-tryptophan. The dynamic range of the screening can be tuned by employing an inhibitory analog of the target molecule. Screening based on this framework requires compartmentalization of individual producers with the sensor strain. We explore three formats of implementation with increasing throughput capability: confinement in microtiter plates (10 2 -10 4 assays/experiment), spatial separation on agar plates (10 4 -10 5 assays/experiment), and encapsulation in microdroplets (10 5 -10 7 assays/experiment). Using SnoCAP, we identified an efficient isobutanol production strain from a random mutagenesis library, reaching a final titer that is 5-fold higher than that of the parent strain. The framework can also be extended to screening for secondary metabolite production using a push-pull strategy.We expect that SnoCAP can be readily adapted to the screening of various microbial species, to improve production of a wide range of target molecules. We expect the cross-feeding system to have two major advantages for high-throughput screening over traditional auxotrophic biosensor implementation: i) amplification of production improvement through the co-culture's exponential growth, and ii) a wider and more tunable dynamic range. The former lies in the co-culture's amplification of differences between library members (Section 3, Calculation). Since strain development generally proceeds through incremental improvements, it is critical to be able to discern modest increases in production.Amplification of differences through the cross-feeding metabolic circuit makes relatively small improvements conspicuous. The second advantage is achieved because, when neither of the cross-fed molecules is exogenously supplied, the production rates of the cross-fed molecules generally limit total co-culture growth, keeping the target molecule within the dynamic range of the sensor strain. This is in contrast to direct application of the sensor to the supernatant of a secretor strain or in co-culture with a non-auxotrophic secretor strain when the level of target molecule will often exceed sensor dynamic range. 6
Materials and MethodsKey elements are described here. Additional details are provided in the Supplementary Information.
Strains and plasmidsStrains and plasmids used are lis...