Facilitating Unconventional Yeast Engineering for Biorenewables Production

Many of the 1800 known yeast species have highly unusual metabolic, biosynthetic, physiological and fermentative capacities that are not possessed by model yeasts such as Saccharomyces cerevisiae.  To date, strides have primarily made on transferring genetic elements attributed to the desired traits from native sources to model hosts.  However, as the outcomes of long-term natural evolution in particular environments, many high-performing functions are granted by a network of genes through a complex hierarchy of regulations, many of which have not been clearly elucidated.  When the predicted relevant genes were incorporated into model hosts, many functions were impaired due to the alteration of physiological backgrounds, general missing of desired regulations, lack of mechanisms for cofactor recycling, or simply expression imbalance.  On the other hand, nowadays, we are no longer retarded by genome sequencing, and if we could establish generic design rules, hence shortening tool development cycle for unconventional microbes, we can directly leverage, engineer and continuously improve the treasures granted by nature.

 Here we report our recent success in establishing the first stable expression platform for engineering Scheffersomyces stipitis, one of the species with the highest native capability for xylose fermentation.  The current expression plasmid is extremely unstable, being lost within 2 days of cultivation.  With the perfect integration of in silico prediction and high throughput screening, we were able to isolate all eight centromere sequences from S. stipitis genome in a monthly manner.  The identified centromeres significantly enhanced the stability of the episomal vector, resulting in 3-fold increase in lactic acid production.  Isolating a functional centromere in an efficient manner by far is still a missing piece from literature.  Immunoprecipitation with specific kinetochore protein antibodies is the predominant method, but suffers from the extensive experiment procedures and the prerequisite of antibody that needs to be developed for each species.  We are currently investigating the generalizability of our hybrid approach in rapid centromere identification in a broad spectrum of unconventional yeasts.  In contrast to the previous developments in this field that were fairly scattered and short of systematicness, we are ambitious to elucidate design principles and establish platform technologies to enable rapid functional modifications of a series of high-potential yeast species.