Yeast and cancer cells share the unusual characteristic of favoring fermentation of sugar over respiration. We now reveal an evolutionary conserved mechanism linking fermentation to activation of Ras, a major regulator of cell proliferation in yeast and mammalian cells, and prime proto-oncogene product. A yeast mutant (tps1∆) with overactive influx of glucose into glycolysis and hyperaccumulation of Fru1,6bisP, shows hyperactivation of Ras, which causes its glucose growth defect by triggering apoptosis. Fru1,6bisP is a potent activator of Ras in permeabilized yeast cells, likely acting through Cdc25. As in yeast, glucose triggers activation of Ras and its downstream targets MEK and ERK in mammalian cells. Biolayer interferometry measurements show that physiological concentrations of Fru1,6bisP stimulate dissociation of the pure Sos1/H-Ras complex. Thermal shift assay confirms direct binding to Sos1, the mammalian ortholog of Cdc25. Our results suggest that the Warburg effect creates a vicious cycle through Fru1,6bisP activation of Ras, by which enhanced fermentation stimulates oncogenic potency.

Whereas the yeast Saccharomyces cerevisiae shows great preference for glucose as a carbon source, a deletion mutant in trehalose-6-phosphate synthase, tps1Δ, is highly sensitive to even a few millimolar glucose, which triggers apoptosis and cell death. Glucose addition to tps1Δ cells causes deregulation of glycolysis with hyperaccumulation of metabolites upstream and depletion downstream of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The apparent metabolic barrier at the level of GAPDH has been difficult to explain. We show that GAPDH isozyme deletion, especially Tdh3, further aggravates glucose sensitivity and metabolic deregulation of tps1Δ cells, but overexpression does not rescue glucose sensitivity. GAPDH has an unusually high pH optimum of 8.0 to 8.5, which is not altered by tps1Δ. Whereas glucose causes short, transient intracellular acidification in wild-type cells, in tps1Δ cells, it causes permanent intracellular acidification. The hxk2Δ and snf1Δ suppressors of tps1Δ restore the transient acidification. These results suggest that GAPDH activity in the tps1Δ mutant may be compromised by the persistently low intracellular pH. Addition of NH4Cl together with glucose at high extracellular pH to tps1Δ cells abolishes the pH drop and reduces glucose-6-phosphate (Glu6P) and fructose-1,6-bisphosphate (Fru1,6bisP) hyperaccumulation. It also reduces the glucose uptake rate, but a similar reduction in glucose uptake rate in a tps1Δ hxt2,4,5,6,7Δ strain does not prevent glucose sensitivity and Fru1,6bisP hyperaccumulation. Hence, our results suggest that the glucose-induced intracellular acidification in tps1Δ cells may explain, at least in part, the apparent glycolytic bottleneck at GAPDH but does not appear to fully explain the extreme glucose sensitivity of the tps1Δ mutant. IMPORTANCE Glucose catabolism is the backbone of metabolism in most organisms. In spite of numerous studies and extensive knowledge, major controls on glycolysis and its connections to the other metabolic pathways remain to be discovered. A striking example is provided by the extreme glucose sensitivity of the yeast tps1Δ mutant, which undergoes apoptosis in the presence of just a few millimolar glucose. Previous work has shown that the conspicuous glucose-induced hyperaccumulation of the glycolytic metabolite fructose-1,6-bisphosphate (Fru1,6bisP) in tps1Δ cells triggers apoptosis through activation of the Ras-cAMP-protein kinase A (PKA) signaling pathway. However, the molecular cause of this Fru1,6bisP hyperaccumulation has remained unclear. We now provide evidence that the persistent drop in intracellular pH upon glucose addition to tps1Δ cells likely compromises the activity of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a major glycolytic enzyme downstream of Fru1,6bisP, due to its unusually high pH optimum. Our work highlights the potential importance of intracellular pH fluctuations for control of major metabolic pathways.

Development of cell factories for conversion of lignocellulosic biomass hydrolysates into biofuels or bio-based chemicals faces major challenges, including the presence of inhibitory chemicals derived from biomass hydrolysis or pretreatment. Extensive screening of 2526 Saccharomyces cerevisiae strains and 17 non-conventional yeast species identified a Candida glabrata strain as the most 5-hydroxymethylfurfural (HMF) tolerant. Whole-genome (WG) transformation of the second-generation industrial S. cerevisiae strain MD4 with genomic DNA from C. glabrata, but not from non-tolerant strains, allowed selection of stable transformants in the presence of HMF. Transformant GVM0 showed the highest HMF tolerance for growth on plates and in small-scale fermentations. Comparison of the WG sequence of MD4 and GVM1, a diploid segregant of GVM0 with similarly high HMF tolerance, surprisingly revealed only nine non-synonymous SNPs, of which none were present in the C. glabrata genome. Reciprocal hemizygosity analysis in diploid strain GVM1 revealed AST2N406I as the only causative mutation. This novel SNP improved tolerance to HMF, furfural and other inhibitors, when introduced in different yeast genetic backgrounds and both in synthetic media and lignocellulose hydrolysates. It stimulated disappearance of HMF and furfural from the medium and enhanced in vitro furfural NADH-dependent reducing activity. The corresponding mutation present in AST1 (i.e. AST1D405I) the paralog gene of AST2, also improved inhibitor tolerance but only in combination with AST2N406I and in presence of high inhibitor concentrations. Our work provides a powerful genetic tool to improve yeast inhibitor tolerance in lignocellulosic biomass hydrolysates and other inhibitor-rich industrial media, and it has revealed for the first time a clear function for Ast2 and Ast1 in inhibitor tolerance.