Phenotypic Phase Plane Mapping of Batch Culture Trajectories for Xylose Fermenting Yeast Scheffersomyces Stipitis Using Genome-Scale Metabolic Modelling

Scheffersomyces stipitis is well known for its ability to produce ethanol from xylose, it does not present the so called Crabtree effect or overflow phenomenon as Saccharomyces cerevisiae does and it shows a strong dependency between production and oxygen availability. Determining aeration conditions to reach high ethanol yield in S. stipitis cultures is difficult. Therefore the identification of metabolic characteristics that determine ethanol production capacity may help improving S. stipitis fermentation performance. Fermentative metabolism is affected by NAD(H/+) cofactors, and evidence shows that high NADH availability promotes fermentation, this is supported by a number of reports indicating that cofactor manipulation is an useful tool for improving fermentation.

The sensitivity analysis of the linear optimization problem involved in the flux balance analysis allows characterizing the optimal metabolic response of a given flux with respect to a change in the availability of a cofactor. The metabolic response to changes in the reducing power supplied by the redox couple NAD(H/+) has been studied in terms of sensitivity to available reducing capacity (ARC), which corresponds to the difference between the sensitivities to changes in NADH and NAD.

The aim of this study is to characterize the trajectory of S. stipitis batch culture in the context of the metabolic phenotypes defined by the optimal flux distribution and the sensitivity of ethanol production to variations in ARC. Using a genome-scale metabolic model for S. stipitisand flux balance analysis, the sensitivity of ethanol flux to ARC changes was calculated throughout their phenotypic phase planes determining whether or not these changes favour ethanol production. Then the trajectories of batch cultures in terms of oxygen and carbon source uptakes were mapped over the planes.

Regarding the sensitivity of the phenotypic phases it was found that the metabolic network presents three different responses to an increase in ARC: i) negative effect, ii) no effect and iii) a positive effect on ethanol production, also it was found that ARC may not control the ethanol production in phases next to the line of optimality for ethanol yield. Thereby the increment of ethanol production is not always characterized by an increase in ARC.  On the other hand, by mapping the trajectories it was observed that cultures transit in a subset of the phases which comprised growth limited by carbon source, optimal aerobic growth and growth limited by oxygen uptake, where the last corresponds to ethanol producing phenotypes. Furthermore, according to the sensitivities associated with the trajectories, ARC limits ethanol production during the culture. This is in line with previous reports showing that S. stipitis favours aerobic oxidation of NADH instead of inducing ethanol synthesis. Hence, the sensitivity analysis, along with the trajectory over the phenotypic phases allowed characterizing the ethanol production capacity in connection with available reducing capacity.