Discovery and functional characterization of a yeast sugar alcohol phosphatase

Abstract

Sugar alcohols (polyols) exist widely in nature. While some specific sugar alcohol phosphatases are known, there is no known phosphatase for some important sugar alcohols (e.g., sorbitol-6-phosphate). Using liquid chromatography-mass spectrometry-based metabolomics, we screened yeast strains with putative phosphatases of unknown function deleted. We show that the yeast gene YNL010W, which has close homologues in all fungi species and some plants, encodes a sugar alcohol phosphatase. We term this enzyme, which hydrolyzes sorbitol-6-phosphate, ribitol-5-phosphate, and (D)-glycerol-3-phosphate, polyol phosphatase 1 or PYP1. Polyol phosphates are structural analogs of the enediol intermediate of phosphoglucose isomerase (Pgi). We find that sorbitol-6-phosphate and ribitol-5-phosphate inhibit Pgi and that Pyp1 activity is important for yeast to maintain Pgi activity in the presence of environmental sugar alcohols. Pyp1 expression is strongly positively correlated with yeast growth rate, presumably because faster growth requires greater glycolytic and accordingly Pgi flux. Thus, yeast express the previously uncharacterized enzyme Pyp1 to prevent inhibition of glycolysis by sugar alcohol phosphates. Pyp1 may be useful for engineering sugar alcohol production.

Graphical Abstract

Introduction

Polyols, the reduced forms of sugars, are an important natural family of carbohydrates ^{1, 2} Glycerol, the simplest polyol, is the backbone of phospholipids and is excreted by many microbes in response to stress ^{3, 4} Longer chain polyols include erythritol, ribitol, xylitol, arabitol, sorbitol, and mannitol, all of which exist only in the (D)-form in nature and are usually found in plants. Due to the inability of most organisms to assimilate long chain polyols into glycolysis, they are often regarded as inert solutes with unclear physiological functions. While the biological function of polyols has remained obscure, they have gained commercial interest in the last decade due to their increased usage as a sugar substitute by the food industry ⁵

Many fungi species are able to produce polyols via reduction of the corresponding sugar ⁶ To engineer polyol production, the standard approach is to first dephosphorylate the sugar phosphate and then reduce the resulting sugar to the polyol ^7–10. This engineering approach differs from the natural glycerol production pathway, where (L)-glycerol-3-phosphate (a.k.a. sn-glycerol-3-phosphate) is first made from the reduction of dihydroxyacetone phosphate, followed by the dephosphorylation of the phosphorylated polyol ^{11, 12}. This natural approach avoids accumulation of the potentially toxic metabolite dihydroxyacetone (DHA). Moreover, it avoids making a dephosphorylated sugar that might escape from the cell. However, this series of reactions has not been feasible as an engineering strategy due to lack of a suitable polyol phosphatase.

Because many enzymes remain unannotated even in the best studied organisms such as Baker’s yeast ¹³ it is possible that various polyol phosphate phosphatases exist but have yet to be characterized. One approach to metabolic enzyme annotation is to knock out the corresponding gene and assess the metabolome of the resulting strain. In general, substrates of the enzyme are likely to increase and products to decrease. This approach has now been successfully used to assign function to a variety of enzymes^14–22.

Using this approach and subsequent biochemical studies, here we show that YNL010W, a gene conserved across all fungi species and some plants, encodes a polyol phosphatase (Pyp1). We further show that the enzyme prevents polyol phosphate accumulation in yeast, and that this is physiologically important due to polyols being Pgi transition state analogues (Scheme 1). Thus, through assigning function to this previously unannotated yeast gene, we identify both a new glycolytic regulatory mechanism and a promising new enzyme for polyol metabolic engineering.

Scheme 1.

Results

A yeast Pyp1 deletion mutant accumulates octulose-8-phosphate and polyol phosphates

We screened yeast strains, with putative phosphatases of unknown function deleted, for changes in metabolite concentrations. Yeast were grown in glucose minimal media, and metabolites were extracted into 40:40:20 methanol: acetonitrile: water, followed by metabolome analysis by reversed-phase ion-pairing liquid chromatography – high resolution mass spectrometry (LC-MS). We found that the deletion of PYP1, formerly known as the uncharacterized gene YNL010W, while not significantly altering the concentrations of most metabolites, led to the statistically significant (false discovery rate < 0.05) accumulation of three compounds in negative ion mode (Figure 1A). The metabolites’ formulae were obtained by labeling cells with ¹³C and ¹⁵N and observing no shift from nitrogen labeling and a shift of +8, +5, or +4 daltons from carbon labeling (Figure 1B)²³ Exact masses of these compounds matched putative formulae of C₈H₁₇O₁₁P₁, C₅H₁₃O₈P₁ and C₄H₁₁O₇P₁. Searching for these formulae in the KEGG database returned the metabolites octulose-8-phosphate (C₈H₁₇O₁₁P₁); ribitol-5-phosphate, arabitol-5-phosphate, and xylitol-5-phosphate (C₅H₁₃O₈P₁); and erythritol-4-phosphate (C₄H₁₁O₇P₁) (Figure 1C).

Figure 1. Metabolomic phenotype of pyplΔ cells.

(A). Metabolite profiles from glucose-grown wild type and pyp1Δ yeast cells detected by untargeted negative ion mode LC-high resolution MS. 1, 2, and 3 represent compounds with significantly different signal intensity between wild type and pyp1Δ cells. The x-axis represents the average of signal intensity in WT cells (N = 3). The y-axis represents the average of signal intensity in pyp1Δ cells (N = 3). Adduct and isotopic peaks were excluded. (B). Impact of ¹³C- and ¹⁵N-labeling on peaks of interest. Compounds 1, 2 and 3 were assigned formula as C₈H₁₇O₁₁P₁, C₅H₁₃O₈P₁ and C₄H₁₁O₇P₁ respectively. (C). Chemical structures of octulose-8P, ribitol-5P and erythritol-4P. (D). Extracted ion chromatograms of polyol phosphates produced from ¹³C-glucose (above) or by phosphorylation of ¹²C-polyols fed to the pyp1Δ cells (below). pyp1Δ cells growing exponentially on 2% ¹³C-glucose was switched to 2% ¹²C- polyols. Yeast metabolome right before the switch and four hours after the switch were analyzed by LC- MS. Retention time identifies the glucose-derived five carbon polyol phosphate as ribitol-5-phosphate and the four carbon polyol phosphate as erythritol-4P. (E and F). Extracted ion chromatogram for endogenous ribitol-5P (E) and octulose-8P (F) compared to synthetic standards.

To identify the five- and four-carbon polyol phosphates, we performed an isotope labeling experiment involving switching pyp1Δ cells from U-¹³C-glucose to unlabeled ribitol, arabitol, xylitol or erythritol. Although baker’s yeast cannot effectively utilize these polyols as carbon sources, in the absence of glucose, they slowly transport and phosphorylate them. We found that feeding ribitol and erythritol resulted in the build-up of intracellular metabolites with exact mass and LC retention time matching the C₅H₁₃O₈P₁ and C₄H₁₁O₇P₁ that accumulated with Pyp1 deletion (Figure 1D). Feeding of arabitol and xylitol resulted in the accumulation of polyol phosphates with different retention times (Figure 1D). Based on these results, we synthesized ribitol-5P and confirmed that exact mass and retention time matched to the endogenous 5-carbon sugar alcohol (Figure 1E). Octulose-8P has been reported in yeast previously18, and the chromatographic retention time of the accumulated C₈H₁₇O₁₁P₁ exactly matched the synthetic octulose-8P standard (Figure 1F). Thus, the compounds that accumulate in the pyp1Δ strain are octulose-8P, ribitol-5P and erythritol-4P.

Ribitol-5P and erythritol-4P are polyol phosphates, but octulose-8P is a sugar phosphate. We were curious why Pyp1 deletion resulted in accumulation of metabolites from these different structural families. The most common conformer of the seven carbon sugar sedoheptulose and its derivatives in solution is β-furanose ²⁴. Octulose-8P likely has a similar β-furanose ring structure (Figure 1C), which has three carbons (6’−8’) of octulose out of the ring, forming a tail that resembles (D)-glycerol-3P, a polyol phosphate. Note that (D)-glycerol-3P is the same compound as (L)-glycerol-IP, which is the enantiomer of the common central carbon metabolite (L)-glycerol-3P (a.k.a. sn-glycerol-3P). We use the (D)-glycerol-3P nomenclature to emphasize the structural similarity to longer (D)-polyol phosphates, including ribitol-5P and sorbitol-6P (Figure 1C).

Recombinant Pyp1 dephosphorylates ribitol-5-phosphate, sorbitol-6-phosphate, and (D)- glycerol-3-phosphate

Although polyols are excretion products of fungi and plants, polyol phosphates were not previously thought to be an intermediate in this pathway (sugar phosphate → sugar → polyol). To determine if the accumulated compounds were indeed the substrates of Pyp1, we analyzed the biochemical activity of the purified recombinant Pyp1 on these compounds. Incubation with Pyp1 led to the depletion of ribitol- 5P and the accumulation of ribitol (Figure 2A), whereas no detectable phosphatase activity was found against octulose-8P or erythritol-4P (Figure 2B). Because flux through octulose-8P and erythritol-4P is very low in cells, minimal Pyp1 activity may nevertheless be sufficient to alter the cellular concentration ¹⁸. Alternatively, Pyp1 may need to bind an activator or other substrate in order to hydrolyze these species, or the accumulation of octulose-8P and erythritol-4P may be an indirect consequence of Pyp1 loss (e.g. occurring secondary to build-up of a more preferred substrate).

Figure 2. Pyp1 hydrolyzes (D)-polyol phosphates.

(A). Purified Pypl’s phosphatase activity was assayed against 0.5 mM ribitol-5P by LC-MS. Incubation with Pyp1 led to the depletion of ribitol-5P (left panel) and the accumulation of ribitol (right panel). (B). Pyp1’s (D)-polyol phosphatase activity was measured in the presence of 0.5 mM substrate and 5 mM Mg²⁺ (pH = 7.0, 30°C) by monitoring the appearance the polyol. See Experimental Procedures for details. The y-axis represents specific phosphatase activity (μmoles product per min per mg of protein) (mean ± range, N = 2).

To better define the range of substrates of Pyp1, we also measured its activity against erythrose-4P, sorbitol-6P, racemic glycerol-3P, and (L)-glycerol-3P. Incubation with Pyp1 led to the hydrolysis of erythrose-4P, sorbitol-6P and racemic glycerol-3P, but not (L)-glycerol-3P (Figure 2B). Thus, unlike the known glycerol-3-phosphatases (Hor2 and Rhr2) in the glycerol biosynthetic pathway, which act on both (L) and (D)-glycerol-3P ¹², Pyp1 is specific to (D)-glycerol-3P. No other phosphatase activity was found for other common phosphorylated metabolites (see Supplementary Table 1). The best biochemical substrates for Pyp1 do not perfectly match those that accumulated in cells, likely because cellular accumulation depends on the absence of other routes of metabolizing the compounds.

Source of ribitol-5P in glucose-grown yeast

Although polyol phosphates have not been previously described in Baker’s yeast, they have been noted in other fungi ²⁵. There are two likely routes for their cellular biosynthesis. The first involves the reduction of the corresponding sugar phosphate and the second involves phosphorylation of polyols. Both activities have been described in fungi but the responsible genes remain unknown25. In Baker’s yeast grown on glucose, the only known polyol is glycerol, and thus polyol phosphates are likely made from the reduction of sugar phosphates (as in Haemophilus influenza, Staphylococcus aureus26, and Escherichia coli27). Consistent with this, deletion of Zwf1, the first step in the oxidative pentose phosphate pathway (PPP), eliminated the endogenous peak for ribitol-5P, presumably due to decreased levels of ribose-5P and ribulose-5P (Figure 3). This is consistent with ribitol-5P being made via the reduction of ribose-5P or ribulose-5P in yeast grown on glucose.

Figure 3. Intracellular levels of ribitol-5-phosphate and ribulose-5-phosphate/xylulose-5-phosphate in wild type and zwf1Δ strain.

The metabolomes of wild type and zwf1Δ strains growing on 2% glucose were measured by LC-MS. The y-axis represents the relative level for select metabolites (mean ± range, N = 2).

Growth on sorbitol requires Pyp1

When polyols are present in the growth environment, polyol phosphates can be made by their phosphorylation. Of the natural long-chain polyols, the only one known to support yeast growth as the sole carbon source is sorbitol ²⁸.Catabolism of sorbitol begins with its dehydrogenation to fructose, which then enters central metabolism.. Due to the structural similarity of sorbitol to glucose, we hypothesized that yeast might sometimes erroneously phosphorylate sorbitol into sorbitol-6P, which could be toxic in excess. To explore this possibility, we fed both wild type and pyp1Δ strains minimal media containing sorbitol as the sole carbon source. While wild type cells grew to saturation after a long lag phase, pyp1Δ cells never grew (Figure 4A). Metabolome profiling revealed that sorbitol-6P levels in the pyp1Δ strain were much higher than the wild type strain (Figure 4B). These results are consistent with Pyp1 being required to dephosphorylate sorbitol-6P, which otherwise accumulates to toxic levels, thereby precluding growth on sorbitol.

Figure 4. Pyp1 accelerates yeast growth on or in the presence of polyols.

(A). Growth of wild type and pyp1Δ yeast on sorbitol. Yeast cells grown on YPD medium were switched to minimal media containing 2% sorbitol as the carbon source. The x-axis represents days after the switch and the y-axis represents optical density (A₆₀₀). (B). Absolute concentration of sorbitol-e-phosphate in wild type and pyp1Δ cells in the experiment shown in (A). The x-axis represents days after the switch and the y-axis represents absolute intracellular concentration (mean ± range, N = 2). (C). Growth of wild type and pyp1Δ cells on glycerol and mannitol. Yeast cells grown on YPD medium were switched to medium containing complete supplement mixture (CSM) and 3% glycerol as the carbon source or minimal medium containing 2% mannitol as the carbon source. The x-axis represents time after the switch and the y-axis represents optical density (A₆₀₀). (D). Impact of ribitol in the presence of trehalose as the carbon source. Absolute concentration of ribitol-5P in wild type and pyp1Δ cells (left panel) and their growth rates (right panel) on minimal media containing 1% trehalose +/− 1% ribitol.

To investigate whether Pyp1 is important for growth on other polyol substrates, we fed pyp1Δ yeast glycerol or mannitol. The pyp1Δ cells grew poorly on glycerol. While wild type cells managed to reach saturation after a long lag phase on mannitol, pyp1Δ· cells were unable to grow on mannitol (Figure 4C). Thus, Pyp1 contributes to polyol phosphate detoxification, and such detoxification is required for growth on diverse polyol substrates.

To more directly ascertain whether the impaired growth was a result of polyol phosphate toxicity, we fed yeast trehalose (a dimer of glucose that can be slowly utilized by yeast resulting in carbon-limited growth^{29, 30}) with or without addition of ribitol. In wild type yeast, addition of ribitol to the medium resulted in accumulation of intracellular ribitol-5P, without impacting growth rate (Figure 4D). In pyp1 yeast, the ribitol-5P levels rose yet higher, resulting in a ~20% decrease in the growth rate (Figure 4D). Thus, buildup of polyol phosphate compounds impairs yeast growth, including on substrates other than polyols.

Polyol phosphates are inhibitors of phosphoglucose isomerase (Pgi)

One potential mechanism by which polyol phosphates could impair cell growth is through metabolic enzyme inhibition. Phosphoglucose isomerase (Pgi1) catalyzes the interconversion between glucose-6P and fructose-6P in glycolysis and is essential for yeast to grow on either glucose or fructose as the sole carbon source ³¹. Sorbitol-6P is structurally similar to the enediol transition state of the Pgi reaction (Scheme 1) and sorbitol-6P and other structural mimics of the enediol intermediate are known Pgi inhibitors ³². We confirmed that sorbitol-6P and less potently ribitol-5P inhibit Pgi (Figure 5A). Motivated by these observations, we sought to determine whether polyol phosphates significantly and selectively inhibit Pgi in yeast cells. We hypothesized that two metabolic hallmarks of Pgi inhibition would be increased glucose-6P and decreased glycolytic intermediates downstream of fructose-6P, with fructose-1, 6-bisphosphate (FBP) a convenient marker compound. To evaluate whether decreased Pgi activity indeed induces these metabolic changes, we constructed a yeast strain with PGI under control of an estradiol-inducible promoter. In the low induction condition (1 nM estradiol, Pgi protein level ~1/7 of WT cells), we observed both increased glucose-6P and decreased FBP (Figure 5B).

Figure 5. Polyol phosphates inhibit phosphoglucose isomerase in vitro and in growing yeast.

(A). Activity of the purified Pgi1 was assayed in the presence of 0.6 mM fructose-6-phosphate (substrate), 5 mM Mg²+ (pH = 7.0, 30°C) and a range of 0.05–5 mM of sorbitol-6P or ribitol-5P by monitoring the appearance of glucose-6-phosphate. The x-axis represents polyol phosphate concentration and the y-axis represents relative Pgi1 activity (mean ± range of N = 2 replicates). (B). Metabolome profiling of cells with high (100 nM estradiol, comparable to WT protein level, shown in black) and low (1 nM estradiol, ~1/7 of WT protein level, shown in grey) Pgi1 induction. This experiment is used to define a characteristic low-Pgi metabolome. (C). Metabolome profiling of pyp1Δ (dark red) and wild type (blue) cells at t = 1 h after switching from trehalose + ribitol to glucose + ribitol. This experiment shows that PYP1 deletion results, in the presence of ribitol, in a characteristic low-Pgi metabolome.

We then tested whether ribitol-5P accumulation results in these metabolic hallmarks of Pgi deficiency. Cells were initially grown in trehalose + ribitol and then switched to glucose + ribitol to enhance glycolytic flux. Relative to wild-type yeast, the pyp1Δ strain accumulated dramatically more ribitol-5P (Figure 5C). Critically, the yeast lacking Pyp1 also manifested both hallmarks of physiological Pgi inhibition, increased glucose-6P and decreased FBP (Figure 5C).

Discussion

Sorbitol-6-phosphatase activity has been found in apple leaves ³³ and silk worms ³⁴. The gene encoding this enzyme, however, has not been discovered. In engineered fungi, sorbitol production has been achieved by expressing bacterial sorbitol-6P dehydrogenase, but again the gene encoding the required phosphatase activity remained missing ³⁵. Here we identify the previously unannotated yeast gene YNL010W, which we now term PYP1, as a polyol phosphate phosphatase. Pyp1 dephosphorylates a variety of compounds with the common structural motif of a (D)-glycerol-3-phosphate tail, including (D)- glycerol-3P, erythrose-4P, ribitol-5P, and sorbitol-6P (Figure 2B). It is likely that Pyp1 would also dephosphorylate erythritol-4P, arabitol-5P, and xylitol-5P. The ability of Pyp1 to hydrolyze a variety of sugar alcohol phosphates gives rise to an intriguing opportunity to employ Pyp1 in the production of diverse sugar alcohol consumables.

One functional role of Pyp1 is to limit the polyol phosphate concentrations in cells. High levels of polyol phosphates impair yeast growth, at least in part by inhibition of the upper glycolytic enzyme phosphoglucose isomerase (Pgi), for which sorbitol-6-phosphate is a transition state analogue. Other sugar phosphate isomerases, such as triose-phosphate isomerase and ribose-5-phosphate isomerase, have similar enediol reaction intermediates ^36,37, and thus may also be inhibited by polyol phosphates. Similar to strong inhibition of Pgi by sorbitol-6 phosphate, triose-phosphate isomerase may be inhibited by (D)-glycerol-3P, and ribose-5-phosphate isomerase by ribitol-5P or xylitol-5P. Flux through each of these enzymes tends to increase with faster yeast growth rate. For example, ribose-5-phosphate isomerase is required to feed ribosome biogenesis. Thus rapidly growing yeast cells may be particularly sensitive to enzyme inhibition by polyol phosphates. To determine whether Pyp1 function is associated with growth rate, we analyzed its expression as a function of growth rate across 25 different chemostat conditions ³⁸. Interestingly, Pypl’s expression was strongly positively correlated with growth rate, regardless of the limiting nutrient used (top 6% of all transcripts in genome) (Figure 6). Thus, unlike most protective or detoxification genes (e.g. against heat, osmolarity or oxidative stress), which are highly expressed under slower growth conditions ^38,39, PYP1 is a fast-growth gene that maintains high Pgi flux by dephosphorylating polyol phosphates.

Figure 6. Pyp1 is highly expressed and functionally important in rapid growth.

Relative expression level of PYP1 at different growth rates in C-, N-, P- and S-limited chemostats ³⁸. The x-axis represents different growth rates and the y-axis represents relative expression level of PYP1.

The transition state inhibition by polyol phosphates and their derivatives is not limited to sugar phosphate isomerase. In plants, ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) also has an enediol intermediate and is known to be inhibited by 2-carboxy-D-arabitol-1-phosphate (CA1P) ⁴⁰. The corresponding 2-carboxy-D-arabitol-1-phosphatase is activated by light and controls cellular level of CA1P. In darkness, RuBisCO is inhibited and also protected by CA1P from proteolysis ⁴¹.

Pyp1 is conserved across fungi species and some plants and bacteria, including Ricinus communis, Dehalococcoides ethenogenes, and Bacillus anthracis42. The highly conserved nature of Pyp1 in fungi indicates the importance of clearing polyol phosphates. It is likely that the homologous enzymes play the same role in bacteria and plants. Although we have not identified a Pyp1 homolog in mammals, nor are linear polyol phosphates longer than three carbons known mammalian metabolites, polyols are also produced in humans and can accumulate in disease43. For example, during hyperglycemia (e.g., due to diabetes), sorbitol accumulates in a number of organs due to the action of aldose reductase on glucose, which may contribute to diabetic complications including diabetic retinopathy, peripheral neuropathy and diabetic kidney disease^44–46. It is unclear whether such sorbitol sometimes becomes phosphorylated to sorbitol-6P and, if so, whether sorbitol-6P contributes to disease pathology.

Beyond their potential cellular toxicity, polyol phosphates may also have a regulatory role in central carbon metabolism. Pgi sits at the branch point between glycolysis and the oxidative pentose phosphate pathway. Despite being ideally situated to regulate the branching ratio between these pathways (Figure 5), Pgi is not an allosteric enzyme, with no physiological regulators known^{32, 47}. It is tempting to speculate that polyol phosphates might serve as endogenous Pgi regulators, with Pgi an enzyme controlled by active site competition^48–53. Further investigation is merited to see whether polyol phosphates contribute to physiological metabolic flux control.

Experimental Procedures

Chemicals and reagents

Most chemicals, reagents and media components in the study, including (D)-sorbitol-6-phosphate (S1753), (D)-ribose-5-phosphate (83875), (D)-erythritol-4-phosphate (08897), (D)-erythrose-4- phosphate (E0377), (DL)-glycerol-3-phosphate (61668), (L)-glycerol-3P (94124), (D)-glycerol-3P (92034), NaBH₄ (452882), (D)-sorbitol (S1876), ribitol (adonitol, A5502), (D)-arabitol (A3381), xylitol (X3375), erythritol (PHR1479), tributylamine (90780), and acetic acid (695092), were obtained from Sigma-Aldrich (St. Louis, MO). LCMS-grade acetonitrile (A955), methanol (A456), water (W6) were obtained from Fisher Scientific (Waltham, MA). ¹³C₆-glucose (CLM-1396) was from Cambridge Isotope Laboratories (Tewksbury, MA).

Yeast strains and media

Yeast strains were derived from prototrophic S288C. Prototrophic deletions were created by homologous recombination using the allele amplified by PCR from the synthetic genetic analysis (SGA) deletion set ⁵⁴. The Pgi1 inducible strain was generated as described:55 uman estradiol was used to induce PGI1 via insertion of a synthetic promoter involving chimeric transcription factor-estrogen receptor in front of PGI1. 1 nM and 5 nM estradiol were added to achieve low and high expression levels of yeast Pgi1, resulting in a 2.5-fold difference in Pgi1 protein level ⁵⁵.

Cells were grown in minimal media comprising 6.7g/L Difco Yeast Nitrogen Base without amino acids plus 2% (w/v) glucose, sorbitol or mannitol. Glycerol medium was composed of 6.7g/L Difco Yeast Nitrogen Base without amino acids, 0.79 g/L Complete Supplement Mixture (Sunrise Science, San Diego, CA) and 3% (v/v) glycerol. Trehalose minimal medium was prepared by mixing 6.7g/L Difco Yeast Nitrogen Base without amino acids and 1% (w/v) trehalose, and adjusting pH to 4.8 by adding succinic acid.

Yeast culture conditions and extraction

The metabolome of batch culture Saccharomyces cerevisiae was characterized as described previously ⁵⁶: Saturated overnight cultures were diluted 1:30 and grown in liquid media in a shaking flask to A₆₀₀ of ~ 0.6. A portion of the cells (3 mL) were filtered onto a 50 mm nylon membrane filter (Millipore, Billerica, MA), which was immediately transferred into −20 °C extraction solvent (40:40:20 acetonitrile/methanol/water). For carbon upshift, 100 mL of cell culture grown on trehalose at A600 of ~0.6 were poured onto a 100 mm cellulose acetate membrane filter (Sterlitech, Kent, WA) resting on a vacuum filter holder with a 1000 mL funnel (Kimble Chase, Vineland, NJ) and were washed with 100 mL pre-warmed (30°C) glucose minimal medium. Immediately after the wash media went through, the filter was taken fromthe holder and the cells were washed into a new flask containing 100 mL pre-warmed (30 °C) glucose minimal medium. Samples were then taken at the indicated time points after the switch, and filtered and quenched as described above.

LC-MS metabolite measurement

Cell extracts were analyzed by reversed phase ion-pairing liquid chromatography (LC) coupled by electrospray ionization (ESI) (negative mode) to a high-resolution, high-accuracy mass spectrometer (Exactive; Thermo Fisher Scientific, Waltham, Massachusetts) operating in negative ion mode scanning m/z 70–1000with 100,000 resolution at m/z 200. Liquid chromatography separation was achieved on a Synergy Hydro-RP column (100 mm × 2 mm, 2.5 μm particle size, Phenomenex, Torrance, CA) with the following gradient: 0 min, 0% B; 2.5 min, 0% B; 5 min, 20% B; 7.5 min, 20% B; 13 min, 55% B; 15.5 min, 95% B; 18.5 min, 95% B; 19 min, 0% B; 25 min, 0% B. Solvent A is 97:3 water/methanol with 10 mM tributylamine and 15 mM acetic acid; solvent B is methanol. Other LC parameters are autosampler temperature 4 °C, injection volume 10 μL, and column temperature 25 °C. Peaks differing between wild-type and pyplΔ strain were determined using the in-house developed, open-source software MAVEN ^57,58. Compounds’ identities were verified by mass and retention time match to authenticated standards. Differences in metabolome between wild type and pyp1Δ strains were tested for significance using Student’s T-test. The resulting P-values were then corrected using the Benjamini-Yekutieli False Discovery Rate (FDR) model ⁵⁹.

The number of C and N atoms in each accumulated compound was determined by the method described ²³. Yeast batch cultures were grown with uniformly labeled glucose or ammonium sulfate (Cambridge Isotopes, Andover, MA) for > 20 generations to ensure complete labeling of the metabolome.

Absolute intracellular metabolite concentrations in steadily growing wild type and pyplΔ cells were determined as described previously ⁶⁰. Metabolite concentrations after perturbations were computed based on fold-change in ion counts relative to steadily-growing cells (grown and analyzed in parallel) multiplied by the known absolute concentration in the steadily growing cells, as determined using an isotope ratio-based approach 61.

Synthesis of ribitol-5-phosphate and octulose-8-phosphate

Synthesis of ribitol-5-phosphate was performed as described previously through the reduction of ribose-5-phosphate using NaBH₄ ⁶². Synthesis of (D)-glycero-D-altro-octulose-8-phosphate was performed enzymatically as described ⁶³.

Protein purification and enzymatic assays

For Pyp1’s polyol phosphatase activity assay, a yeast Open Reading Frame (ORF) strain with an expression vector containing C-terminal His-tagged PYP1 (Open Biosystems, Thermo Fisher Scientific, San Jose, CA) was grown on galactose to induce Pyp1 expression. The resulting cells were lysed using glass beads and His-tagged Pyp1 was purified using Qiagen Ni-NTA spin columns according to the manufacturer’s instructions. Phosphatase activity against ribitol-5-phosphate, sorbitol-6-phosphate, erythritol-4-phosphate and octulose-8-phosphate was determined by monitoring the increase of ribitol, sorbitol, erythritol or octulose using LC-MS. The reaction mixture contained 100 mM Tris-HCl at the pH 8.0, 10 mM MgCl₂ and a range of substrate concentrations (0.01 to 5 mM).

Purified Pyp1 was also assayed against other compounds with similar structures and 90 common phosphorylated metabolites (see Supplementary Table for the list of compounds). Briefly, the gene encoding Pyp1 was amplified by PCR using S. cerevisiae genomic DNA. The amplified fragments were cloned into a modified pET15b vector (Novagen, Darmstadt, Germany) and overexpressed in the E. coli BL21 (DE3) Gold strain (Stratagene, La Jolla, California) as previously described ⁶⁴. The recombinant protein was purified using metal ion affinity chromatography on nickel chelate resin (Qiagen, Hilden, Germany) to high homogeneity and stored at −80 °C. Purified Pyp1 was then screened for phosphatase activity as described previously65. Due to the difficulty of obtaining (D)-glycerol-3P, the activity of Pyp1 on (D)-glycerol-3P was determined using racemic glycerol-3P as the substrate. Because Pyp1 has minimal activity on (L)-glycerol-3P, its activity on racemic glycerol-3P was taken as the activity on (D)-glycerol-3P. Compounds with specific activity higher than 0.1 μmol/mg/min, as well as erythritol-4P, are shown in Figure 2B.

To measure the inhibition of Pgi by sorbitol-6-phosphate and ribitol-5-hosphate, yeast Pgi1 was purchased from Sigma Aldrich (St. Louis, MO). Phosphoglucose isomerase activity was determined by adding fructose-6-phosphate and monitoring the appearance of glucose-6-phosphate using LC/MS. We found such LC/MS based assay is consistently more sensitive and accurate than the more typical colorimetric-based assay, which involves coupling the Pgi1 activity with glucose-6-phosphate dehydrogenase activity and monitoring the appearance of NADH. The reaction mixture contained 100 mM Tris-HCl at pH 8.0, 10 mM MgCl₂, 0.6 mM fructose-6-phosphate (the physiological concentration in cells grown exponentially on glucose), and a range of sorbitol-6-phosphate and ribitol-5-phosphate concentrations (0.05 to 5 mM). The resulting data were fitted to the Hill equation using the GraphPad Prism Software.