Saccharomyces cerevisiae continues to prove its potential as an excellent microbial production platform of many bulk chemicals [1–4]. While traditionally S. cerevisiae has mainly been used for its high speed and capacity to convert sugars into ethanol and CO2, presently its robustness and genetic accessibility are also much appreciated in many metabolic engineering efforts for production of bio-based fuels [5–7] and chemicals [8–11]. In fact, in several industrial processes, including those centered around pyruvate-derived products such as malate [12, 13] or lactate [14–16], ethanol is now considered an undesired by-product.
Even under fully aerobic conditions, S. cerevisiae converts part of its sugar substrate to ethanol when confronted with high sugar concentrations . Conversion of glucose to ethanol yields much less ATP than complete conversion to CO2 and H2O via respiratory dissimilation, which is a drawback in ATP-requiring production processes . The strong tendency of S. cerevisiae towards alcoholic fermentation is thought to have evolved as a mechanism to outcompete other organisms by the resulting fast glucose uptake and build-up of growth-inhibiting ethanol concentrations [19, 20]. Although beneficial in natural environments, in many applied contexts this phenomenon lowers product yields. Therefore, several metabolic engineering studies have sought to disrupt aerobic fermentation of sugars by S. cerevisiae[21–26].
A powerful approach to prevent alcoholic fermentation in S. cerevisiae is elimination of pyruvate decarboxylase, which catalyzes the first step in the conversion of pyruvate to ethanol. S. cerevisiae strains in which all three structural genes encoding pyruvate decarboxylase (PDC1
PDC5 and PDC6) were deleted, did not produce ethanol, but were unable to grow in the presence of high glucose concentrations and, when grown in glucose-limited cultures, required the addition of ethanol or acetate to growth media, due to their inability to synthesize cytosolic acetyl-CoA from pyruvate [23–25]. To overcome these deficiencies, a Pdc- yeast was selected for growth on glucose as the sole carbon source in an evolutionary engineering experiment . First, C2-carbon source prototrophic mutants were selected by prolonged cultivation in glucose-limited chemostat cultures, in which the acetate concentration in the medium gradually decreased to zero. Subsequently, a mutant able to grow at high glucose concentrations was selected by cultivation in serial shake flask cultures. The resulting evolved mutant could grow at a growth rate of 0.20 h-1 on synthetic medium with glucose as the sole carbon source and proved to be an efficient pyruvate producer .
Elucidation of the genetic background of glucose tolerance in Pdc-
S. cerevisiae is not only of fundamental interest, but is also required to enable its fast introduction in metabolic engineering strategies. The process of elucidating and subsequent reconstruction of a desired phenotypic trait is known as reverse metabolic engineering [28, 29]. Reverse engineering of phenotypes obtained by laboratory evolution has the added benefit that potential detrimental effects of random mutations obtained during evolution can be eliminated. Identification of relevant mutations is an essential step in reverse metabolic engineering. Transcriptional profiling of the evolved Pdc- mutant during growth in nitrogen-limited chemostat cultures revealed the altered expression of many hexose transporters (Hxt) in this evolved strain compared to a wild type strain . It was found that the summed transcript abundance of all HXT genes represented on the arrays (HXT1 to HXT10
HXT14, and HXT16) was four-fold lower in the evolved Pdc- strain than in a Pdc+ reference strain .
Transcription of HXT genes in S. cerevisiae is predominantly regulated via the transcriptional regulator Rgt1 [30–33], which also regulates MIG2 and STD1 expression [34–36]. MIG2 and STD1 are both down-regulated in the evolved Pdc- strain . Rgt1 is regulated by the concerted action of the glucose sensors Rgt2 and Snf3, which relay the extracellular glucose signal via the paralogous repressors Mth1 and Std1 to Rgt1 [33, 36–40]. In the absence of extracellular glucose, Mth1 and Std1 are in a complex with Rgt1, Ssn6 and Tup1 resulting in the transcriptional repression or activation of Rgt1 targets [41–43]. In the presence of glucose, the conformation of the glucose sensors Rgt2 and Snf3 is thought to change, which facilitates the phosphorylation of Mth1 and Std1 by Yck1 . When phosphorylated, Mth1 and Std1 are targeted for degradation . The absence of Mth1 or Std1 enables phosphorylation of Rgt1 [30, 42, 44], which is subsequently released from the promoters of, amongst others, the Hxt transporters [30–33]. The altered transcript profiles of HXT genes in the evolved, glucose-tolerant Pdc-
S. cerevisiae strain might therefore be explained by mutations in this regulatory network. For a comprehensive review and graphical representation of the regulation of the HXT transporters see Gancedo et al. 2008 .
The goal of the present study was to identify the mutation(s) responsible for the ability of the evolved Pdc- strain isolated by Van Maris et al. (2004) to grow on high concentrations of glucose as sole carbon source. Our results identified a mutation in MTH1, whose impact on growth on glucose in the absence of added C2-compounds was investigated after reintroduction in an ancestral Pdc-