Used but not Sensed - The Paradox of D-xylose Metabolism in Saccharomyces cerevisiae

Abstract: The realization that the extraction and combustion of fossil fuels is having serious effects on the environment and the climate, together with the ever-growing need for fuels, has led to the development of the concept of the biorefinery. Biorefineries are refineries in which fossil resources, such as oil, are replaced by renewable biomaterials to produce biofuels and biochemicals. Non-edible biomass is used in a lignocellulose-based refinery, which avoids the conflict between fuel and food production, but a number of inherent technical challenges must be overcome. The robust and genetic engineering-friendly yeast Saccharomyces cerevisiae (baker’s yeast) is a promising platform organism for biomass fermentation, but it lacks functional assimilatory pathways to utilise D-xylose, the second most abundant sugar in a wide range of lignocellulosic materials. During the past two decades, recombinant forms of S. cerevisiae have been developed able to efficiently convert D-xylose to ethanol. However, the rate of conversion is slow, and D-xylose appears not to be recognised by S. cerevisiae as a fermentable sugar.This thesis is focused on investigating the role of the sugar sensing and signalling routes in the unusual behaviour of S. cerevisiae on D-xylose. A panel of in vivo biosensors coupled to D-glucose signalling routes was used under different physiological conditions and in the presence of different genetic modifications. The green fluorescent protein gene (yEGFP3) was coupled to different endogenous yeast promoters known to be regulated by at least one of the three main sugar pathways: Snf3p/Rgt2p, cAMP/PKA and SNF1/Mig1p.The signallome investigation revealed that a recombinant strain of S. cerevisiae able to assimilate D-xylose could sense high concentrations of D-xylose, but the signal was similar to that observed with low levels of D-glucose: inducing SUC2p (SNF1/Mig1p pathway) and HXT2p (Snf3p/Rgt2p pathway) but repressing HXT1p (Snf3p/Rgt2p and cAMP/PKA pathway). Strains unable to metabolise D-xylose provided no clear signal in the presence of D-xylose due to heterogeneity in the population of the biosensor strains. However, in strains that were able to assimilate D-xylose, the signalling induction pattern was completely opposite to the signal obtained when protein kinase A (PKA) was activated by high levels of D-glucose. It was therefore hypothesized that the signal triggered by a high D-xylose level similar to a low D-glucose signal was due to a low PKA activity.Further validation of the role of sugar signalling was obtained by using targeted deletants known to improve the D-xylose consumption rate without being directly associated with D-xylose catabolic routes. Notably, it was found that the signalling response on D-xylose changed from a low D-glucose signal in the background strain, to simultaneous signalling of high and low D-glucose in the best strain (ira2Δisu1Δ). Since IRA2 is a repressor of PKA activity, this finding supported the hypothesis of the malfunction of PKA activity on D-xylose due to poor sensing through this route.This study also focused on understanding whether the sensing signal observed in the presence of high concentrations of D-xylose could be linked to a metabolite acting as a pathway regulator. Using strains in which PGI1, which encodes an isomerase enabling the reversible conversion of glucose-6-phosphate and fructose-6-phosphate, had been deleted, it was possible to link changes in signalling to disturbances in the levels of glycolytic intermediates. The findings presented in this thesis support the hypothesis of a dysfunctional sugar signalling mechanism on D-xylose, and show that the phenotype is a result of the lack of membrane sensing in connection with alterations in intracellular signalling.