| dc.description |
Microbial biosynthesis is a sustainable and high-specificity means of producing various bioproducts, including pharmaceuticals, specialty and commodity chemicals, and biofuels. Due to the complexity of microorganisms, it is frequently difficult to rationally engineer them, which necessitates iterative rounds of design, construction, and testing to generate high-producing strains. Furthermore, it can be difficult to optimize multiple functions in the same microorganism. Microbial consortia are abundant in natural environments and can offer unique properties that are not attainable by monocultures. Design principles have begun to be developed for synthetic consortia and further maturation of this field will lead to many exciting new opportunities in microbial bioprocessing. In this dissertation, we describe two cases of utilizing microbial consortia, one as a tool for screening microbial libraries, and the other as a division-of-labor approach for accomplishing the complex task of lignocellulosic biofuel production.
First, we demonstrate that a cross-feeding metabolic circuit can convert production phenotypes into growth phenotypes, which are highly screenable. This technology, which we term Syntrophic Co-culture Amplification of Production phenotype (SnoCAP), has two valuable properties that are not present in monocultures: (1) it has a highly tunable dynamic range, and (2) it amplifies small differences between strains. We implemented three different compartmentalization schemes of increasing throughput capability: microplates (10^2-10^4 strains evaluated/experiment), agar plates (10^4-10^5 strains evaluated/experiment), and microdroplets (10^5-10^7 strains evaluated/experiment). We demonstrated SnoCAP’s ability to differentiate between Escherichia coli strains of differing production levels for 2-ketoisovalerate (2-KIV), a precursor of the drop-in biofuel isobutanol, and L-tryptophan, a precursor for several pharmaceutically active compounds. We then used SnoCAP to screen a chemically mutagenized library and identify an efficient isobutanol production strain that reaches a 5-fold higher titer than the parent strain. We expect SnoCAP can be applied to the screening of a wide variety of target molecules for which high-throughput screening assays do not currently exist.
Second, we examine a previously developed co-culture of the cellulolytic fungus Trichoderma reesei and isobutanol-producing E. coli for consolidated bioprocessing of lignocellulosic biomass to biofuel. This approach provides division-of-labor, distributing the metabolic burden and allowing optimization of hydrolysis and fermentation separately. We work toward improving this co-culture by engineering the E. coli strains for improved performance under co-culture conditions. Due to observed issues with plasmid loss, we developed strains with the isobutanol pathway integrated into the genome. We used the chemically inducible chromosomal evolution (CIChE) method to achieve high copy number of the genes responsible for the conversion of 2-KIV to isobutanol. We then explored the use of position-dependent expression variation, in conjunction with SnoCAP screening, to optimize expression of another gene crucial for the synthesis of 2-KIV. Additionally, we developed a framework for adaptive evolution of the T. reesei/E. coli co-culture. We expect that this method may be used on a strain with the isobutanol pathway integrated into the genome to select for variants that are well-suited to production under co-culture conditions.
In summary, this work contributes to the development of synthetic microbial consortia for biochemical production. We have demonstrated that the properties of cross-feeding metabolic circuits can be exploited as a useful high-throughput screening tool. We have also explored a synthetic fungal-bacterial consortium that divides the labor of lignocellulosic biomass conversion between two specialist strains and developed new approaches to optimize the fermentation specialist for the unusual conditions it encounters in the co-culture. |
|