Glucose Metabolism in gcr Mutants of Saccharomyces cerevisiae

Abstract

A gcr2 null mutant of Saccharomyces cerevisiae grows well on glucose in spite of its lower level of glycolytic enzymes between triose phosphates and pyruvate. A quantitative analysis shows that these levels are adequate to the flux but glycerate phosphates are elevated.

Gcr1p and Gcr2p are elements affecting transcription of glycolytic genes in Saccharomyces cerevisiae (20, 26). Gcr1p is a DNA-binding protein interacting with a consensus sequence in the promoter, Gcr2p interacts with Gcr1p; both factors are needed for normal transcriptional activation. Null and point mutants have decreased levels of most of the glycolytic enzymes. The size of the effect depends on the enzyme, growth medium, and strain background, with levels of phosphoglycerate mutase and enolase being 5 to 10% of normal in gcr1 mutants grown in the presence of glucose and less otherwise; like strains completely blocked in single glycolytic reactions, gcr1 mutants do not grow on glucose (7). However, gcr2 mutants, which show a pattern of enzyme decreases qualitatively similar to those of gcr1 (see also below) grow well on glucose (25). The relative normality does not appear to reflect a bypassing of the most-affected steps, for the gcr2 mutation did not restore growth on glucose to a gpm1 (phosphoglycerate mutase) mutant (data not shown). Here we use a resting-cell method (3) to assess glucose flux (v_glucose), enzyme levels, and metabolite levels in gcr2 as well as gcr1 and gpm1 null mutants.

Strains.

DFY724 is a wild-type strain (MATα leu2-3,112 ura3-52 his6 [strain 2845 in reference 27]); its isogenic derivatives are DFY725 (Δgcr2-3::URA3 [strain YHU 3002-8C in reference 27]), DFY726 (Δgcr1::URA3 [strain C179-15C in reference 27]), and DFY727 (Δgpm1::LEU2). (DFY727 construction was done by cloning GPM1 as a 2.3-kbp SphI-AatI fragment from pPGM1 [29] [the open reading frame is bp 754 to 1497] into pT7/T3alpha-18 [from Bethesda Research Laboratory], replacement of its BglII-SalI fragment [bp 200 to 1647] by LEU2 from YEp13 as a BglII-XhoI fragment, and transplacement into DFY724 after digestion with SphI and SacI.) Strain DFY730 carries a plasmid with pGAL-GPM1 integrated in strain HD162-5C (MATa gpm1::URA3 trp1-289 ura3-52 leu2-3,112 his3-1 MAL2-8c SUC2 GAL) (from J. Heinisch [15], as was an isogenic GPM1 comparison strain HD162-5A, here called DFY728). (For the pGAL-GPM1 plasmid, pL834-3, the GAL1-GAL10 promoter [an EcoRI-XhoI fragment from pYEUra3 {Clontech}] was inserted in the multiple cloning site [EcoRI-SalI] of the URA3 plasmid YIp352, and then GPM1, as an EarI-AatI fragment from pGPM [EarI is 104 bp upstream of the starting codon], was inserted in the SmaI site downstream of the GAL1 promoter, giving a Gal-driven transcript with a 5′ noncoding region of 191 bp; integration was by transformation after digestion with AatI in the URA3 gene. As desired, strain DFY730 grew on galactose but not glucose, while strain HD162-5C grew on neither.)

Resting-cell incubations.

As slightly modified from reference 3, washed cells were suspended at an A₅₈₀ of 100 in buffer B61 containing cycloheximide (10 μg/ml) and antimycin A (2 μg/ml). After 30 min of preincubation at 30°C with shaking, glucose was added to 1% and the cultures were periodically sampled. For glucose use and ethanol and glycerol formation, supernatants were used; for enzyme assay, pellets from, e.g., 0.5-ml portions were frozen until use, while for intracellular metabolites at the times indicated in Table 1 1-ml portions were added to tubes containing 0.1 ml of 11.7 M perchloric acid and 20 mM Na₂EDTA, the contents of the tubes were vigorously mixed for 30 s, and after at least 30 min at −2°C and additional mixing, the contents were neutralized with 0.282 ml of 3 M KHCO₃; the supernatants were kept at −80°C until use. In about half of the experiments all three types of data were obtained.

TABLE 1.

Glucose metabolism by resting cells

Item	Data seta:
	1	2	3	3A	4
Strain	DFY724 (wild type)	DFY725 (gcr2)	DFY724 (wild type)	DFY724 (wild type)	DFY725 (gcr2)
Substrate employed for growth	Glucose	Glucose	Lactic acid	Lactic acid + glucose (0.75 h)	Lactic acid
V of enzyme (μmol/[min × mg of protein])b
Zwf	0.22 ± 0.034 (4)	0.44 ± 0.17 (3)	0.33 ± 0.06 (5)	0.31	0.34 ± 0.04 (4)
Pgi	12.5 ± 4.92 (5)	9.58 ± 3.42 (3)	8.50 ± 2.18 (4)	8.95	1.68 ± 0.15 (3)
Tpi	41.1 ± 19.2 (3)	21.1 ± 9.76 (2)	19.4	10.4	—e
Gap	9.90 ± 2.96 (4)	8.85 ± 4.08 (3)	7.82 ± 3.30 (5)	8.68	3.84 ± 1.52 (4)
Pgk	95.9 ± 39.6 (4)	29.6 ± 8.10 (3)	54.7 ± 14.9 (5)	62.9	5.16 ± 0.98 (4)
Gpm	10.8 ± 3.84 (5)	2.35 ± 0.28 (6)	4.11 ± 2.00 (5)	5.55	0.51 ± 0.11 (4)
Eno	7.93 ± 3.20 (4)	2.42 ± 0.92 (3)	2.10 ± 0.56 (5)	2.22	0.16 ± 0.07 (4)
Pyk	10.8 ± 4.81 (4)	5.40 ± 2.30 (3)	3.01 ± 0.70 (5)	4.11	0.93 ± 0.21 (4)
vglucose (μmol/[min × mg of protein])d	0.30 ± 0.03 (9)	0.30 ± 0.03 (10)	0.09 ± 0.02 (5)	0.18	0.02 ± 0.01 (5)
Products (μmol/μmol of glucose used)
Ethanol	1.40 ± 0.36 (8)	1.28 ± 0.20 (8)	1.28 ± 0.16 (5)	1.41	0.94 ± 0.37 (4)
Glycerol	0.14 ± 0.05 (7)	0.25 ± 0.05 (7)	0.20 ± 0.06 (5)	0.27	0.72 ± 0.30 (4)
Intracellular metabolite concn (mM)e
Glucose-6-phosphate	0.06 ± 0.06 (3), 6.36 ± 1.36 (5)	0.12 ± 0.14 (4), 5.78 ± 1.75 (6)	1.23 ± 0.24 (3)	2.20	0.79 ± 0.28 (2)
Fructose-1,6-bisphosphate	0.08 ± 0.04 (5), 15.2 ± 3.26 (6)	0.10 ± 0.10 (5), 12.2 ± 1.97 (6)	2.88 ± 0.86 (3)	6.08	1.10 ± 0.61 (2)
Glycerate-3-phosphate	0.94 ± 0.24 (5), 0.97 ± 0.19 (6)	0.92 ± 0.65 (4), 4.31 ± 1.13 (7)	0.39 ± 0.01 (2)	1.06	7.52 ± 2.71 (2)
Glycerate-2-phosphate	0.18 ± 0.06 (5), 0.15 ± 0.05 (5)	0.22 ± 0.09 (5), 1.36 ± 0.49 (5)	0.05 ± 0.02 (3)	0.08	0.64 ± 0.29 (2)
Phosphoenolpyruvate	0.34 ± 0.09 (5), 0.31 ± 0.08 (5)	0.30 ± 0.14 (5), 0.52 ± 0.22 (5)	0.04 ± 0.01 (3)	0.06	0.05 ± 0.04 (2)

Data seta:
4A	5	5A	6	7	8
DFY725 (gcr2)	DFY726 (gcr1)	DFY726 (gcr1)	DFY727 (gpm1)	DFY728 (wild type)	DFY730 (gpm1 pGAL-GPM)
Lactic acid + glucose (3 h)	Lactic acid	Lactic acid + glucose (6 h)	Lactic acid	Galactose	Galactose
0.27	0.37 ± 0.01 (2)	0.39	0.33	—	—
4.82	0.7 ± 0.18 (2)	2.58	6.52	—	—
16.3	0.54	5.38	—	—	—
7.59	0.92 ± 0.28 (2)	11.96	4.17	—	—
14.1	7.92 ± 1.6 (2)	30.7	58.12	—	—
1.99	0.09 ± 0.15 (3)	0.41	0.003 ± 0.004 (2)	10.1 ± 7.59 (3)	0.30 ± 0.42 (3)
1.74	0.16 ± 0.02 (3)	0.82	2.21 ± 0.16 (2)	3.86	5.03
4.51	0.51 ± 0.01 (2)	2.08	3.77	—	—
0.16	0.03	0.10	0.0	0.25 ± 0.07 (3)	0.03 ± 0.02 (4)
1.35	0.76	0.87	0.0	1.19	0.72
0.33	0.92	0.33	0.0	0.12	0.64
3.84	0.92	2.80	0.19	7.32 ± 0.96 (2)	0.78 ± 0.33 (2)
9.72	0.56	5.60	0.08	7.24	5.72
6.96	8.08	13.52	17.0	0.87 ± 0.24 (2)	8.36 ± 0.84 (2)
0.80	0.15	0.04	—	0.048 ± 0.02 (2)	0.85 ± 0.20 (2)
0.16	0.01	0.06	—	0.17 ± 0.18 (2)	0.27 ± 0.13 (2)

^aThroughout the table, values are means ± standard deviations, based on the number of separate experiments indicated in parentheses. 

^bEnzyme abbreviations are as follows: Zwf, glucose-6-phosphate dehydrogenase; Pgi, phosphoglucose isomerase; Tpi, triose phosphate isomerase; Gap, glyceraldehyde-3-phosphate dehydrogenase; Pgk, phosphoglycerate kinase; Gpm, phosphoglycerate mutase; Eno, enolase; and Pyk, pyruvate kinase. Protein was determined by the biuret assay using bovine albumin (Sigma Protein Standard). With the substrate concentrations employed, most of the V values are near to V_max. Because most of the assays employ the reverse direction to glycolysis, those values were corrected by the indicated V_f/V_r factors (V_f and V_r are the V_max values for the forward and reverse reactions, respectively): Pgi 2.5 (3, 32); Tpi 3.2 (22); Gap, 1.5 (not yeast [14]); Pgk, 8.9 (5); and Gpm, 0.6 (12). 

^c—, not determined. 

^dExpressed in the same units as enzyme activity, using a value of 79 ± 4.3 μg of protein per ml of cell suspension with an A₅₈₀ value of 1.0. (The value was obtained with eight strains or conditions corresponding to data sets 1 to 7 in a calibration experiment using trichloroacetic acid precipitation of cells followed by 30 min at 100°C in 1 N NaOH and biuret assay [16]. The same experiment gave a dry weight [15 h at 102°C] of 197 ± 33 μg and a wet weight of 868 ± 212 μg, all per unit A₅₈₀.) 

^eIntracellular concentrations are given as 40 times that measured in the final neutralized acid extracts (1.382 ml/100 A₅₈₀ units of cells, as described above), using an internal volume value of 2 μl per mg (dry wt); values of 2 to 2.5 have been reported (1, 13, 18, 32). For data sets 1 and 2, the first and second entries separated by a comma correspond to sample times of −2 min (before the addition of glucose; v_glucose = 0.0 μmol/[min × mg of protein]) and 10 min, respectively. The remainder of the data sets had one sampling point each, as follows: 3, 20 min; 3A, 22 min; 4, 20 min; 4A, 22 min; 5, 22 min; 5A, 24 min; 6, 30 min; 7, 10 min; 8, 40 min. 

Metabolite assays.

Glucose, ethanol, glycerol, glucose-6-phosphate, fructose-1,6-bisphosphate, and glycerate-3-phosphate assays employed a Roche Cobas Bio Autoanalyzer (as in reference 8) at 340 nm (for ethanol, 365 nm) and standard endpoint techniques (23); prepared reagents were used for ethanol (Roche OnLine) and glycerol (Boehringer; samples were pretreated 5 min at 85°C to remove ATPase activity.) For improved sensitivity, glycerate-2-phosphate and phosphoenolpyruvate were assayed as ATP by a method modified from reference 19; the samples were pretreated with activated charcoal (Sigma catalog no. C 4386), and the charcoal was removed with spin filters (0.4-ml size; 0.22 μl porosity; Millipore MC).

Enzyme assays.

The cell pellets were suspended in a solution containing 0.5 ml of 50 mM potassium phosphate (pH 7.4), 2 mM Na₂EDTA, 2 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride; treated 50 s with a Mini Beadbeater (Biospec Products); and centrifuged for 10 min. Three-microliter portions (undiluted and 1/10 and 1/100 dilutions) were assayed with an Autoanalyzer (as described above) at 30°C, with 0.35-ml incubation volumes. The pH 7.1 buffer (see reference 30) (25 mM Tris, 20 mM KCl, 10 mM MgSO₄) also contained 0.2 mM NADH (or, for glucose-6-phosphate dehydrogenase or phosphoglucose isomerase, 0.2 mM NADP), 1 mM substrate(s) unless specified otherwise, and auxiliary enzymes from Boehringer (amounts [in micrograms per milliliter of assay] are given in brackets) as follows: for glucose-6-phosphate dehydrogenase, glucose-6-phosphate; for phosphoglucose isomerase, fructose-6-phosphate and glucose-6-phosphate dehydrogenase [2]; for triose-phosphate isomerase (0.25 mM), glyceraldehyde-3-phosphate and glycerol-3-phosphate dehydrogenase [1]; for glyceraldehyde-3-phosphate dehydrogenase, 0.5 mM glycerate-3-phosphate, ATP, and phosphoglycerate kinase [2]; for phosphoglycerate kinase, 0.5 mM glycerate-3-phosphate, ATP, and glyceraldehyde-3-phosphate dehydrogenase [20]; for phosphoglycerate mutase, glycerate-2-phosphate, 0.2 mM glycerate-2,3-bisphosphate, ATP, phosphoglycerate kinase [2], and glyceraldehyde-3-phosphate dehydrogenase [20]; for enolase, glycerate-2-phosphate, ADP, pyruvate kinase [6], and lactate dehydrogenase [2]; and for pyruvate kinase, phosphoenolpyruvate, ADP, and lactic dehydrogenase [10].

Growth.

Enriched medium R61 (9) was supplemented with uracil, 50 μg/ml, and 2% glucose or galactose with harvest at an A₅₈₀ of ca. 1 or on 0.2% lactic acid with harvest at an A₅₈₀ of ca. 10; as specified, in some of the latter experiments, the medium was supplemented with 1% glucose for a limited period before harvest. For gcr1 and gpm1 strains the medium also contained 0.01% glycerol.

For growth, the basic parameters are growth rate and rate of substrate utilization. For the parental wild-type strain DFY724 and gcr2 null mutant DFY725 in enriched medium with glucose at 30°C, the growth rates were 0.53 and 0.31 h⁻¹, respectively, and the corresponding rates of glucose utilization (v_glucose values) were 0.49 and 0.28 μmol/(min × mg of protein). Thus, the mutant does grow more slowly than the parent but with similar yield. The data in Table 1 were obtained with nongrowing cells. (The protocol [from reference 3] uses cells grown as desired [not necessarily with glucose], suspended in a buffer with cycloheximide and antimycin to prevent new protein synthesis and respiration at a high-cell density so metabolites can be obtained by direct acidification without a concentration step.) Under such conditions the rate of glucose use is one-half or more of that in growth. (By contrast with reference 3, however, in the present experiments metabolite levels did not attain a plateau before glucose exhaustion.) The main data sets, 1 and 2, are for wild-type and gcr2 mutant strains grown on glucose. Both strains used glucose at a similar rate and with an ethanolic fermentation; slightly more glycerol was made in the mutant than in the wild type. Enzyme assay values show that the later reactions of glycolysis in the mutant have levels one-fourth to one-half those of the wild type; with improved assays these values are somewhat higher than those previously reported (25). Since in Table 1 enzyme activities (Vs) are expressed in the forward glycolytic direction and in the same units as v_glucose, it can be seen that even the most-affected steps in the mutant have capacities (values of ca. 2.4) in excess of the rate of ethanol formation (ca. 0.4), the latter being a measure of the net in vivo rate of those two reactions.

The last five lines of Table 1 show intracellular metabolite concentrations, both before and 10 min after glucose addition. For the hexose phosphates the patterns were similar in the wild type and the gcr2 mutant: marginal levels during starvation and high levels afterwards. For later metabolites of glycolysis, the pattern was more complicated, with significant amounts in the starved cells (as known, and perhaps related to insufficient fructose-1,6-bisphosphate to activate pyruvate kinase [2, 21]). During glucose metabolism, in the wild-type strain those levels were unchanged, while by contrast in the gcr2 mutant, as might be expected in view of its reduced enzyme levels, they were considerably elevated.

The other data sets are less complete. With lactic acid instead of glucose as the carbon source, in the wild-type strain (data set 3) the levels of lower enzymes of glycolysis were approximately one-third of those during growth on glucose and, perhaps coincidentally, so was the glucose flux under the resting-cell conditions; metabolite levels were low. Forty-five minutes of exposure to glucose prior to harvest (data set 3A) did not clearly increase levels of the assayed enzymes, but flux and metabolites significantly increased in the test situation. For both gcr2 (data set 4) and gcr1 (data set 5) strains, enzyme levels from gluconeogenic growth were in the 2 to 10% range, glucose use was marginal, levels of hexose phosphates were low, and levels of glycerate-3-phosphate were high; several hours of prior exposure to glucose (data sets 4A and 5A) increased enzyme levels and glucose flux, more for gcr2 than for gcr1, with glycerate-3-phosphate levels remaining high.

The most notable perturbations in gcr2 and gcr1 strains were of glycerate-3-phosphate, implying significant in vivo impediment at phosphoglycerate mutase; for comparison purposes, data set 6 is for a gpm1 structural gene mutant, which with normal levels of other enzymes and completely unable to grow on glucose, gave a v_glucose of 0, formed no ethanol, but likewise accumulated high levels of glycerate-3-phosphate (see also reference 6). And (not shown) a 3-h prior exposure to glucose did not confer on this mutant the ability to use glucose in the resting-cell situation either. Data set 8 is for a strain with GPM1 expression dependent on galactose (data set 7 is for the isogenic wild type). Unfortunately, the phosphoglycerate mutase level was low even in cells grown on galactose, so apart from showing that the need for phosphoglycerate mutase may be less on galactose than glucose, the strain was not useful as a means of providing a range of levels of a single enzyme. Nontheless, as expected, the low enzyme level was associated with low but not zero v_glucose and a high level of glycerate-3-phosphate. There was also a large relative increase in the amount of glycerol made (as also seen for the marginal glucose use by gcr mutants [data sets 4A and 5A]; glycerol formation is an alternative route for NADH reoxidation when acetaldehyde is unavailable [11]). It is of interest that this strain with nominally only a low phosphoglycerate mutase level (the enolase level was normal) contained unusually high levels of glycerate-2-phosphate; perhaps this finding is related to the inhibition of enolase by high-level glycerate-3-phosphate (31).

Qualitatively, the results show (i) how rapid glucose metabolism in a wild-type strain goes with high levels of glycolytic enzymes; (ii) the compatibility of substantial decreases in certain enzymes with normal v_glucose (e.g., gcr2 as grown on glucose), as also known in specific cases: 2% level of phosphoglycerate kinase and normal glucose flux (24), and 0.7% level of phosphoglycerate mutase activity allowing half the normal growth rate on glucose (30); and (iii) the apparent absence of a major bypass of the lower reactions of glycolysis. It is pertinent that two other putative phosphoglycerate mutase genes as tested have marginal or no function (15). Lower glycolytic enzyme levels are associated, however, with accumulations of their substrates, and one speculation is that although such accumulation is not always clearly detrimental (e.g., normal flux with high glycerate-3-phosphate in the gcr2 mutant; see also reference 10), it may be that glycolytic enzyme activities are normally in such apparent excess in part to avoid effects like the inhibition of enolase by glycerate-3-phosphate cited above.

It is tempting to ask whether the metabolite concentrations also make quantitative sense in terms of the measured amounts and known parameters of the enzymes and the observed fluxes. There is one case in yeast for a reversible reaction where the fit was adequate (phosphoglucose isomerase [32]). As shown in Table 2 calculations with the present data show a poor fit of measured values versus predicted values, although, as also shown, relatively small changes in certain individual entries would remove the differences. Unfortunately, aside from the cancellation of errors, each entry in such calculations has limitations, including (i) the large standard deviations of items in Table 1; (ii) the fact that enzyme parameters are not all well known, or have a wide range of reported values, or need correction for pH or Mg²⁺ (17); (iii) the presence of isoenzymes for several reactions and limited knowledge of their individual parameters; (iv) for phosphoglycerate mutase, the absence of data on glycerate-2,3-bisphosphate concentration; (v) the use of normalized values for protein content and internal cell volume; (vi) the fact of certain glycolytic enzymes having a nominal active-site concentration within a 0.1 factor of substrate concentration, requiring additional correction (4, 28); and (vii) for reversible reactions the limitation of calculating net flux as a difference between two much larger numbers (Table 2). We also remark that in the present work emphasis was on gcr mutant strains with several enzymes affected; although strain DF730 (Table 1, data set 8) was not useful for the purpose, varying a single enzyme in one strain would nonetheless be preferred.

TABLE 2.

Calculated net flux through phosphoglycerate mutase^a

Item	Data set 1 (DFY724 [wild type])		Data set 2 (DFY725 [gcr2])
	−2 min	10 min	−2 min	10 min
Vf	11.6 (11.1)	11.6 (9.72)	2.52 (3.02)	2.52 (4.75)
Vr	19.4 (20.2)	19.4 (25.6)	4.21 (3.51)	4.21 (1.85)
Glycerate-3-phosphate (mM)	0.94 (0.90)	0.97 (0.74)	0.92 (1.10)	4.31 (8.79)
Glycerate-2-phosphate (mM)	0.18 (0.19)	0.15 (0.20)	0.22 (0.18)	1.36 (0.66)
Ks (mM)	0.24 (0.25)	0.24 (0.31)	0.24 (0.20)	0.24 (0.11)
Kp (mM)	0.08 (0.076)	0.08 (0.06)	0.08 (0.095)	0.08 (0.16)
vf	6.34	7.36	1.27	1.25
vr	6.09	5.26	1.52	1.99
Net v (calculated)	0.25	2.10	−0.25	−0.73
Net v (measured)b	0.0	0.42	0.0	0.38

^aThe rate equations v_f = (V_fS/K_s)/D and v_r = (V_rP/K_p)/D, with D = 1 + S/K_s + P/K_p (where V_f and V_r are the V_max values for the forward and reverse reactions, respectively, and K_s and K_p are the K_ms for glycerate-3-phosphate and glycerate-2-phosphate, respectively), were applied, by using Microsoft EXCEL, to data from data sets 1 and 2 (Table 1), employing the K_ms from reference 12 and its V_r/V_f value of 1.66. The values in parentheses are recalculations using the ethanol flux (last line) as one of the six input values and allowing just one of the other six inputs to vary: i.e., for set 1, 10 min, a net v of 0.42 would have been obtained with a V_f of 9.72 instead of 11.6 μmol/min × mg of protein, or a V_r of 25.6 instead of 19.4 μmol/min × mg of protein, etc. V and v units are micromoles per minute per milligram of protein. 

^bEthanol flux = v_glucose × (ethanol/glucose).