Mutations suppressing the effects of a deletion of the phosphoglucose isomerase gene PGI1 in Saccharomyces cerevisiae

Using phosphoglucose isomerase-deficient (pgi1Δ) Saccharomyces cerevisiae to map the impact of sugar phosphate levels on D-glucose and D-xylose sensing

BACKGROUND

Despite decades of engineering efforts, recombinant Saccharomyces cerevisiae are still less efficient at converting D-xylose sugar to ethanol compared to the preferred sugar D-glucose. Using GFP-based biosensors reporting for the three main sugar sensing routes, we recently demonstrated that the sensing response to high concentrations of D-xylose is similar to the response seen on low concentrations of D-glucose. The formation of glycolytic intermediates was hypothesized to be a potential cause of this sensing response. In order to investigate this, glycolysis was disrupted via the deletion of the phosphoglucose isomerase gene (PGI1) while intracellular sugar phosphate levels were monitored using a targeted metabolomic approach. Furthermore, the sugar sensing of the PGI1 deletants was compared to the PGI1-wildtype strains in the presence of various types and combinations of sugars.

RESULTS

Metabolomic analysis revealed systemic changes in intracellular sugar phosphate levels after deletion of PGI1, with the expected accumulation of intermediates upstream of the Pgi1p reaction on D-glucose and downstream intermediates on D-xylose. Moreover, the analysis revealed a preferential formation of D-fructose-6-phosphate from D-xylose, as opposed to the accumulation of D-fructose-1,6-bisphosphate that is normally observed when PGI1 deletants are incubated on D-fructose. This may indicate a role of PFK27 in D-xylose sensing and utilization. Overall, the sensing response was different for the PGI1 deletants, and responses to sugars that enter the glycolysis upstream of Pgi1p (D-glucose and D-galactose) were more affected than the response to those entering downstream of the reaction (D-fructose and D-xylose). Furthermore, the simultaneous exposure to sugars that entered upstream and downstream of Pgi1p (D-glucose with D-fructose, or D-glucose with D-xylose) resulted in apparent synergetic activation and deactivation of the Snf3p/Rgt2p and cAMP/PKA pathways, respectively.

CONCLUSIONS

Overall, the sensing assays indicated that the previously observed D-xylose response stems from the formation of downstream metabolic intermediates. Furthermore, our results indicate that the metabolic node around Pgi1p and the level of D-fructose-6-phosphate could represent attractive engineering targets for improved D-xylose utilization.

Metabolic reconfiguration enables synthetic reductive metabolism in yeast

Cell proliferation requires the integration of catabolic processes to provide energy, redox power and biosynthetic precursors. Here we show how the combination of rational design, metabolic rewiring and recombinant expression enables the establishment of a decarboxylation cycle in the yeast cytoplasm. This metabolic cycle can support growth by supplying energy and increased provision of NADPH or NADH in the cytosol, which can support the production of highly reduced chemicals such as glycerol, succinate and free fatty acids. With this approach, free fatty acid yield reached 40% of theoretical yield, which is the highest yield reported for Saccharomyces cerevisiae to our knowledge. This study reports the implementation of a synthetic decarboxylation cycle in the yeast cytosol, and its application in achieving high yields of valuable chemicals in cell factories. Our study also shows that, despite extensive regulation of catabolism in yeast, it is possible to rewire the energy metabolism, illustrating the power of biodesign.

Functional analysis of PGI1 and ZWF1 in thermotolerant yeast Kluyveromyces marxianus

Glycolysis and the pentose phosphate pathway (PPP) are two basic metabolic pathways that are simultaneously present in yeasts. As the main pathway in most species, the glycolysis provides ATP and NADH for cell metabolism while PPP, as a complementary pathway, supplies NADPH. In this study, the performance of Kluyveromyces marxianus using glycolysis or PPP were studied through the disruption of PGI1 or ZWF1 gene, respectively. K. marxianus using glycolysis as the only pathway showed higher ethanol production ability than that of the Kluyveromyces lactis zwf1Δ mutant; K. marxianus using only PPP showed more robustness than that of Saccharomyces cerevisiae pgi1Δ mutant. Additionally, K. marxianus pgi1Δ strain accumulated much more intracellular NADPH than the wild type strain and co-utilized glucose and xylose more effectively. These findings suggest that phosphoglucose isomerase participates in the regulation of the repression of glucose on xylose utilization in K. marxianus. The NADPH/NADP⁺ ratio, dependent on the activity of the PPP, regulated the expression of multiple genes related to NADPH metabolism in K. marxianus (including NDE1, NDE2, GLR1, and GDP1). Since K. marxianus is considered a promising host in industrial biotechnology to produce renewable chemicals from plant biomass feedstocks, our research showed the potential of the thermotolerant K. marxianus to produce NADP(H)-dependent chemical synthesis from multiple feedstocks. KEY POINTS: • The function of PGI1 and ZWF1 in K. marxianus has been analyzed in this study. • K. marxianus zwf1Δ strain produced ethanol but with decreased productivity. • K. marxianus pgi1Δ strain grew with glucose and accumulated NADPH. • K. marxianus pgi1Δ strain released glucose repression on xylose utilization.

Metabolically engineered recombinant Saccharomyces cerevisiae for the production of 2-Deoxy-scyllo-inosose (2-DOI)

Saccharomyces cerevisiae is a versatile industrial host for chemical production and has been engineered to produce efficiently many valuable compounds. 2-Deoxy-scyllo-inosose (2-DOI) is an important precursor for the biosynthesis of 2-deoxystreptamine-containing aminoglycosides antibiotics and benzenoid metabolites. Bacterial and cyanobacterial strains have been metabolically engineered to generate 2-DOI; nevertheless, the production of 2-DOI using a yeast host has not been reported. Here, we have metabolically engineered a series of CEN.PK yeast strains to produce 2-DOI using a synthetically yeast codon-optimized btrC gene from Bacillus circulans. The expression of the 2-Deoxy-scyllo-inosose synthase (2-DOIS) gene was successfully achieved via an expression vector and through chromosomal integration at a high-expression locus. In addition, the production of 2-DOI was further investigated for the CEN.PK knockout strains of phosphoglucose isomerase (Δpgi1), D-glucose-6-phosphate dehydrogenase (Δzwf1) and a double mutant (Δpgi1, Δzwf1) in a medium consisting of 2% fructose and 0.05% glucose as a carbon source. We have found that all the recombinant strains are capable of producing 2-DOI and reducing it into scyllo-quercitol and (-)-vibo-quercitol. Comparatively, the high production of 2-DOI and its analogs was observed for the recombinant CEN.PK-btrC carrying the multicopy btrC-expression vector. GC/MS analysis of culture filtrates of this strain showed 11 times higher response in EIC for the m/z 479 (methyloxime-tetra-TMS derivative of 2-DOI) than the YP-btrC recombinant that has only a single copy of btrC expression cassette integrated into the genomic DNA of the CEN.PK strain. The knockout strains namely Δpgi1-btrC and Δpgi1Δzwf1-btrC, that are transformed with the btrC-expression plasmids, have inactive Pgi1 and produced only traces of the compounds. In contrast, Δzwf1-btrC recombinant which has intact pgi1 yielded relatively higher amount of the carbocyclic compounds. Additionally, ¹H-NMR analysis of samples showed slow consumption of fructose and no accumulation of 2-DOI and the quercitols in the culture broth of the recombinant CEN.PK-btrC suggesting that S. cerevisiae is capable of assimilating 2-DOI.

The effects of disruption of phosphoglucose isomerase gene on carbon utilisation and cellulase production in Trichoderma reesei Rut-C30

Background

Cellulase and hemicellulase genes in the fungus Trichoderma reesei are repressed by glucose and induced by lactose. Regulation of the cellulase genes is mediated by the repressor CRE1 and the activator XYR1. T. reesei strain Rut-C30 is a hypercellulolytic mutant, obtained from the natural strain QM6a, that has a truncated version of the catabolite repressor gene, cre1. It has been previously shown that bacterial mutants lacking phosphoglucose isomerase (PGI) produce more nucleotide precursors and amino acids. PGI catalyzes the second step of glycolysis, the formation of fructose-6-P from glucose-6-P.

Results

We deleted the gene pgi1, encoding PGI, in the T. reesei strain Rut-C30 and we introduced the cre1 gene in a Δpgi1 mutant. Both Δpgi1 and cre1⁺Δpgi1 mutants showed a pellet-like and growth as well as morphological alterations compared with Rut-C30. None of the mutants grew in media with fructose, galactose, xylose, glycerol or lactose but they grew in media with glucose, with fructose and glucose, with galactose and fructose or with lactose and fructose. No growth was observed in media with xylose and glucose. On glucose, Δpgi1 and cre1⁺Δpgi1 mutants showed higher cellulase activity than Rut-C30 and QM6a, respectively. But in media with lactose, none of the mutants improved the production of the reference strains. The increase in the activity did not correlate with the expression of mRNA of the xylanase regulator gene, xyr1. Δpgi1 mutants were also affected in the extracellular β-galactosidase activity. Levels of mRNA of the glucose 6-phosphate dehydrogenase did not increase in Δpgi1 during growth on glucose.

Conclusions

The ability to grow in media with glucose as the sole carbon source indicated that Trichoderma Δpgi1 mutants were able to use the pentose phosphate pathway. But, they did not increase the expression of gpdh. Morphological characteristics were the result of the pgi1 deletion. Deletion of pgi1 in Rut-C30 increased cellulase production, but only under repressing conditions. This increase resulted partly from the deletion itself and partly from a genetic interaction with the cre1-1 mutation. The lower cellulase activity of these mutants in media with lactose could be attributed to a reduced ability to hydrolyse this sugar but not to an effect on the expression of xyr1.

Two mechanisms for oxidation of cytosolic NADPH by Kluyveromyces lactis mitochondria

Null mutations in the structural gene encoding phosphoglucose isomerase completely abolish activity of this glycolytic enzyme in Kluyveromyces lactis and Saccharomyces cerevisiae. In S. cerevisiae, the pgi1 null mutation abolishes growth on glucose, whereas K.lactis rag2 null mutants still grow on glucose. It has been proposed that, in the latter case, growth on glucose is made possible by an ability of K. lactis mitochondria to oxidize cytosolic NADPH. This would allow for a re-routing of glucose dissimilation via the pentose-phosphate pathway. Consistent with this hypothesis, mitochondria of S. cerevisiae cannot oxidize NADPH. In the present study, the ability of K. lactis mitochondria to oxidize cytosolic NADPH was experimentally investigated. Respiration-competent mitochondria were isolated from aerobic, glucose-limited chemostat cultures of the wild-type K. lactis strain CBS 2359 and from an isogenic rag2Delta strain. Oxygen-uptake experiments confirmed the presence of a mitochondrial NADPH dehydrogenase in K.lactis. This activity was ca. 2.5-fold higher in the rag2Delta mutant than in the wild-type strain. In contrast to mitochondria from wild-type K. lactis, mitochondria from the rag2Delta mutant exhibited high rates of ethanol-dependent oxygen uptake. Subcellular fractionation studies demonstrated that, in the rag2Delta mutant, a mitochondrial alcohol dehydrogenase was present and that activity of a cytosolic NADPH-dependent 'acetaldehyde reductase' was also increased. These observations indicate that two mechanisms may participate in mitochondrial oxidation of cytosolic NADPH by K. lactis mitochondria: (a) direct oxidation of cytosolic NADPH by a mitochondrial NADPH dehydrogenase; and (b) a two-compartment transhydrogenase cycle involving NADP(+)- and NAD(+)-dependent alcohol dehydrogenases.

Starting up yeast glycolysis

Successfully igniting the yeast glycolytic flux during the transition from gluconeogenic to fermentative growth seems to be a matter of balance and coordination between a multitude of events. The contours of the sugar sensing and signalling pathways that regulate this transition are only beginning to emerge.

Open reading frames in the antisense strands of genes coding for glycolytic enzymes in Saccharomyces cerevisiae

Open reading frames longer than 300 bases were observed in the antisense strands of the genes coding for the glycolytic enzymes phosphoglucose isomerase, phosphoglycerate mutase, pyruvate kinase and alcohol dehydrogenase I. The open reading frames on both strands are in codon register. It has been suggested that proteins coded in codon register by complementary DNA strands can bind to each other. Consequently, it was interesting to investigate whether the open reading frames in the antisense strands of glycolytic enzyme genes are functional. We used oligonucleotide-directed mutagenesis of the PGI1 phosphoglucose isomerase gene to introduce pairs of closely spaced base substitutions that resulted in stop codons in one strand and only silent replacements in the other. Introduction of the two stop codons into the PGI1 sense strand caused the same physiological defects as already observed for pgil deletion mutants. No detectable effects were caused by the two stop codons in the antisense strand. A deletion that removed a section from -31 bp to +109 bp of the PGI1 gene but left 83 bases of the 3' region beyond the antisense open reading frame had the same phenotype as a deletion removing both reading frames. A similar pair of deletions of the PYK1 gene and its antisense reading frame showed identical defects. Our own Northern experiments and those reported by other authors using double-stranded probes detected only one transcript for each gene. These observations indicate that the antisense reading frames are not functional. On the other hand, evidence is provided to show that the rather long reading frames in the antisense strands of these glycolytic enzyme genes could arise from the strongly selective codon usage in highly expressed yeast genes, which reduces the frequency of stop codons in the antisense strand.

The role of the NAD-dependent glutamate dehydrogenase in restoring growth on glucose of a Saccharomyces cerevisiae phosphoglucose isomerase mutant

Phosphoglucose isomerase pgi1-deletion mutants of Saccharomyces cerevisiae cannot grow on glucose as the sole carbon source and are even inhibited by glucose. These growth defects could be suppressed by an over-expression on a multi-copy plasmid of the structural gene GDH2 coding for the NAD-dependent glutamate dehydrogenase. GDH2 codes for a protein with 1092 amino acids which is located on chromosome XII and shows high sequence similarity to the Neurospora crassa NAD-glutamate dehydrogenase. Suppression of the pgi1 deletion by over-expression of GDH2 was abolished in strains with a deletion of the glucose-6-phosphate dehydrogenase gene ZWF1 or gene GDH1 coding for the NADPH-dependent glutamate dehydrogenase. Moreover, this suppression required functional mitochondria. It is proposed that the growth defect of pgi1 deletion mutants on glucose is due to a rapid depletion of NADP which is needed as a cofactor in the oxidative reactions of the pentose phosphate pathway. Over-expression of the NAD-dependent glutamate dehydrogenase leads to a very efficient conversion of glutamate with NADH generation to 2-oxoglutarate which can be converted back to glutamate by the NADPH-dependent glutamate dehydrogenase with the consumption of NADPH. Consequently, over-expression of the NAD-dependent glutamate dehydrogenase causes a substrate cycling between 2-oxoglutarate and glutamate which restores NADP from NADPH through the coupled conversion of NAD to NADH which can be oxidized in the mitochondria. Furthermore, the requirement for an increase in NADPH consumption for the suppression of the phosphoglucose isomerase defect could be met by addition of oxidizing agents which are known to reduce the level of NADPH.

Glucose uptake and catabolite repression in dominant HTR1 mutants of Saccharomyces cerevisiae

Growth and carbon metabolism in triosephosphate isomerase (delta tpi1) mutants of Saccharomyces cerevisiae are severely inhibited by glucose. By using this feature, we selected for secondary site revertants on glucose. We defined five complementation groups, some of which have previously been identified as glucose repression mutants. The predominant mutant type, HTR1 (hexose transport regulation), is dominant and causes various glucose-specific metabolic and regulatory defects in TPI1 wild-type cells. HTR1 mutants are deficient in high-affinity glucose uptake and have reduced low-affinity transport. Transcription of various known glucose transporter genes (HXT1, HXT3, and HXT4) was defective in HTR1 mutants, leading us to suggest that HTR mutations affect a negative factor of HXT gene expression. By contrast, transcript levels for SNF3, which encodes a component of high-affinity glucose uptake, were unaffected. We presume that HTR1 mutations affect a negative factor of HXT gene expression. Multicopy expression of HXT genes or parts of their regulatory sequences suppresses the metabolic defects of HTR1 mutants but not their derepressed phenotype at high glucose concentrations. This suggests that the glucose repression defect is not a direct result of the metabolically relevant defect in glucose transport. Alternatively, some unidentified regulatory components of the glucose transport system may be involved in the generation or transmission of signals for glucose repression.

Different signals control the activation of glycolysis in the yeast Saccharomyces cerevisiae

The glycolytic pathway in Saccharomyces cerevisiae is activated by fermentable sugars at several steps. Mutants with deletions of genes coding for enzymes of the upper part of glycolysis were used to characterize the triggering mechanisms. Synthesis of fructose-2,6-bisphophate is catalysed by two 6-phosphofructo-2-kinase isoenzymes, one of which is activated by fermentable sugars while synthesis of the second enzyme is induced (Kretschmer and Fraenkel, 1991). Increase in the level of fructose-2,6-bisphosphate is demonstrated to depend on an internal metabolite upstream of the phosphoglucose isomerase reaction. The signalling process correlates with distinct temporal changes in the concentration of glucose-6-phosphate but not with its absolute level, indicating an adaptational mechanism. It is independent of the uptake and phosphorylation systems used by different sugars. Interestingly, this increase, although delayed, could also be observed in strains lacking the rapid cAMP increase after sugar addition which is thought to be responsible for the activating process. Synthesis of glucose-6-P and fructose-6-P is needed for the complete induction of pyruvate kinase and inactivation of fructose-1,6-bisphosphatase. On the other hand, induction of pyruvate decarboxylase depends mainly on a signal in the lower part of glycolysis.

Saccharomyces cerevisiae phosphoglucose isomerase and fructose bisphosphate aldolase can be replaced functionally by the corresponding enzymes of Escherichia coli and Drosophila melanogaster

Two glycolytic enzymes, phosphoglucose isomerase and fructose-1,6-bisphosphate aldolase, of Saccharomyces cerevisiae could be replaced by their heterologous counterparts from Escherichia coli and Drosophila melanogaster. Both heterologous enzymes, which show respectively little and no sequence homology to the corresponding yeast enzymes, fully restored wild-type properties when their genes were expressed in yeast deletion mutants. This result does not support notions of an obligatory formation of glycolytic multi-enzyme aggregates in yeast; nor does it support possible regulatory functions of yeast phosphoglucose isomerase.

Characterization of mutations that overcome the toxic effect of glucose on phosphoglucose isomerase less strains of Saccharomyces cerevisiae

Glucose inhibits growth of yeast phosphoglucose isomerase mutants in permissive media. Mutants insensitive to this effect were isolated by selection on media containing 2% fructose + 2% glucose. A nuclear, monogenic, recessive mutation named rgl was responsible for this phenotype. The mutants isolated belonged to two complementation groups and have been termed rgl1 and rgl2. When the double mutants were grown on fructose, fermentation of fructose or glucose was similar to that of the parental pgi strain but was not measurable when grown on fructose+glucose. Under these conditions, respiration of glucose and to a lesser extent of fructose was enhanced. The double mutants pgi rgl did not grow on fructose+glucose in the presence of antimycin A or ethidium bromide and their cytochrome oxidase was no longer sensitive to glucose repression. The results are interpreted as an indication that in the double mutants the glucose may be channeled through the pentose phosphate pathway to respiration.

Molecular analysis of the structural gene for yeast transaldolase

We have cloned the structural gene for yeast transaldolase. Transformants carrying the TAL1 gene on a multicopy plasmid over-produced transaldolase. A deletion mutant which was constructed using the cloned gene did not show any detectable transaldolase activity in vitro. Furthermore, both transaldolase isoenzymes which were detected in wild-type crude extracts by immunoblotting were missing in the deletion mutants. Thus, TAL1 is the only transaldolase structural gene in yeast. TAL1 is not an essential gene. Deletion of the transaldolase gene did not affect growth on complete media with different carbon sources or on synthetic media. However, the transaldolase-deficient strains accumulated sedoheptulose 7-phosphate, an intermediate of the pentose-phosphate pathway. Mutants lacking both transaldolase and phosphoglucose isomerase grew more slowly than the single mutants. They accumulated more sedoheptulose 7-phosphate on medium containing fructose than on glucose medium. This shows that fructose 6-phosphate and glyceraldehyde 3-phosphate, metabolites of glycolysis, can enter the nonoxidative part of the pentose-phosphate pathway.

Regulation of sugar and ethanol metabolism in Saccharomyces cerevisiae

This review briefly surveys the literature on the nature, regulation, genetics, and molecular biology of the major energy-yielding pathways in yeasts, with emphasis on Saccharomyces cerevisiae. While sugar metabolism has received the lion's share of attention from workers in this field because of its bearing on the production of ethanol and other metabolites, more attention is now being paid to ethanol metabolism and the regulation of aerobic metabolism by fermentable and nonfermentable substrates. The utility of yeast as a highly manipulable organism and the discovery that yeast metabolic pathways are subject to the same types of control as those of higher cells open up many opportunities in such diverse areas as molecular evolution and cancer research.