Microbial cell factory, which converts biomass feedstock to value-added compounds such as fuels, chemicals, materials and pharmaceuticals, has been proposed as a sustainable and renewable alternative to the traditional petrochemical industry.Saccharomyces cerevisiae is one of the most widely used microbial cell factories, to produce ethanol as the first generation of biofuel.To enable this yeast as a producer for 1-butanol, which is a next-generation gasoline substitute, I discovered, characterized and engineered an endogenous 1-butanol pathway in S. cerevisiae.Upon introduction of a single gene deletion adh1Δ, S. cerevisiae was able to accumulate more than 120 mg/L 1-butanol from glucose in the rich medium.Precursor feeding, 13C-isotope labeling and gene deletion experiments demonstrated that the endogenous 1-butanol production was dependent on catabolism of threonine in a manner similar to fusel alcohol production by the Ehrlich pathway.Overexpression of the pathway enzymes and elimination of competing pathways achieved the highest reported 1-butanol titer in S. cerevisiae (243 mg/L).Though 1-butanol titer was improved through pathway-based engineering, such rational design often meets with great challenges in cellular reprogramming due to our limited knowledge of complex biological systems.Directed evolution, on the other hand, has been proved as a better strategy by performing iterative cycles of mutagenesis and selection.Current practice of directed evolution is mostly confined to individual proteins, due to the lack of efficient tools to introduce mutations globally and iteratively in a genome.In the rest of this dissertation, I sought to develop a new method, RNA-interference assisted genome evolution (RAGE), to apply directed evolution strategy in genome scale engineering of S. cerevisiae. A functional RNA interference (RNAi) pathway was reconstituted in S. cerevisiae by introducing the Dicer and Argonaute proteins from Saccharomyces castellii as previously reported.We then performed the first RNAi screening in S. cerevisiae.The RNAi plasmid library was constructed with random genomic fragments and a convergent promoter expression cassette, which drives the in vivo synthesis of double-stranded RNAs (dsRNAs) to mediate knockdown of homologous genes.The library was confirmed with a complete coverage of the yeast genome, and employed to perform a suppressor analysis of a telomere-defect mutation yku70Δ.Two known and three novel knockdown modifications were identified to alleviate the growth arrest of the Δyku70 strain at higher temperature, confirming the effectiveness of our RNAi library for genotyping.After establishing RNAi screening in S. cerevisiae, we combined it with directed evolution to rapidly engineer yeast cells for improved acetic acid (HAc) tolerance.Three rounds of iterative RNAi screening resulted in accumulation of three gene knockdown modifications that acted synergistically to confer substantially improved HAc tolerance.Together, these results demonstrated the RAGE method as an efficient, genome-scale and generally applicable strategy for directed genome evolution in S. cerevisiae. I then expanded the application of RAGE to create a comprehensive genetic library (RAGE2.0).By directional cloning of a full-length, normalized cDNA library, one-step construction of the genome-wide ORF-overexpression and anti-sense RNA libraries was achieved.In the presence of the RNAi pathway, the RAGE2.0 library resulted in genome-wide overexpression and knockdown modifications simultaneously.A wide range of phenotypes, including protein secretion, substrate utilization, and fuel molecule production, were screened with the RAGE2.0 library in a high-throughput manner.Both overexpression and knockdown targets were successfully identified to improve these phenotypes.I further developed the RAGE3.0 method for automated genome engineering in yeast.Upon introduction of specific double-stranded breaks (DSBs) in the repetitive sequences by CRISPR nucleases, the genome-wide overexpression and knockdown cassettes in the RAGE2.0 library can be integrated into the genomic loci of repetitive sequences at high efficiency.This process can be iteratively performed to accumulate multiple genetic modifications in a single cell of an evolving yeast population.RAGE3.0 only involves simple liquid handling steps, hence it is readily automated with an integrated robotic platform.We envision this automated genome engineering method can enable generation of vast genetic diversity from which new or improved properties may emerge, and therefore greatly accelerate basic and applied biological research in S. cerevisiae.
PDF
download
878987797
028-85220240
