against purified hexokinase PII, which ran at the same molecular weight as the native protein in sodium dodecyl

Transcription

1 JOURNAL OF BACTRIOLOGY, Nov. 1988, p /88/ $2./ Copyright C 1988, American Society for Microbiology Vol. 17, No. 11 Glucose Uptake in Saccharomyces cerevisiae Grown under Anaerobic Conditions: ffect of Null Mutations in the Hexokinase and Glucokinase Structural Genes CASY J. McCLLLAN AND LINDA F. BISSON* Department of Viticulture and nology, University of California, Davis, California Received 22 April 1988/Accepted 19 August 1988 Glucose uptake was investigated in a set of isogenic strains carrying a single glucose kinase structural gene, the other two kinase genes having been rendered nonfunctional through the construction of null mutations. Any one of the three kinases was sufficient for growth and glucose utilization aerobically or anaerobically. Under anaerobic conditions, substrate inhibition and regulation of carrier activity varied and depended upon the particular kinase present in the cell. Investigation of glucose uptake in Saccharomyces cerevisiae demonstrated the presence of at least two systems for hexose uptake, differing primarily in affinity for substrate (1). Low-affinity uptake (Km, 2 mm for glucose and 5 mm for fructose) is expressed constitutively during active growth and is mediated by facilitated diffusion. The activity of this transport system is lost upon continued incubation of the cells in stationary phase (3). In contrast, high-affinity uptake (Ki,m 2 mm for glucose and 5 mm for fructose) is glucose repressible, and expression of this uptake system requires the presence of at least one functional glucose kinase in the cell. S. cerevisiae possesses three enzymes capable of phosphorylating glucose at the six position: hexokinase A (PI), hexokinase B (PII), and glucokinase (7, 11, 13, 16). xhaustive mutant hunts conducted by Lobo and Maitra (8, 12) indicated that these are the only three enzymes catalyzing the phosphorylation of the hexoses glucose, fructose, and mannose in S. cerevisiae. A triple kinase mutant (hxkl hxk2 glkl) was shown to lack high-affinity uptake of glucose and fructose (1). Double kinase mutants (hxkl hxk2) which are unable to phosphorylate fructose but retain the ability to phosphorylate glucose via glucokinase display high-affinity uptake of glucose but not of fructose. Analysis of transport of 6-deoxy-D-glucose, a nonphosphorylatable analog of glucose, in the kinase mutants revealed a kinase-dependent uptake of this compound as well, suggesting some role of the kinases in uptake other than phosphorylation of the sugar (2) Ṫhese studies were all conducted by using point mutations in the kinase structural genes. As initially determined by the genetic analyses of Lobo and Maitra (12), nonsense mutations are rare in hxkl and glkl lesions and none of the 22 hxk2 mutations screened was a nonsense mutation. They suggested that this finding perhaps implied some vital physiological role of the hexokinase PII enzyme other than its catalytic function in carbon metabolism. The hexokinase PII protein is believed to play some as yet undefined role in carbon catabolite repression (6, 1), and there is evidence in mammalian systems that hexokinases may be involved in the control of Ca2+ movements within the cell (15). The strain used in our previous studies, DFY437 (hxkl hxk2 glkl) (1), contained cross-reacting material to antibody prepared * Corresponding author against purified hexokinase PII, which ran at the same molecular weight as the native protein in sodium dodecyl sulfate-polyacrylamide gels (L. Bisson, unpublished observations). The catalytically inactive protein may retain other possible functions, thus complicating the interpretation of the kinase dependency of high-affinity glucose uptake. Therefore, the kinetics of glucose uptake were examined in a set of isogenic double kinase null mutant strains, each retaining a wild-type copy of only one of the three glucose kinases. Fermentation characteristics and growth properties were also examined in these strains. A triple kinase null mutant, DFY57 (MATa lysi hxkl: LU2 hxk2::lu2 glkl::lu2) was constructed by and obtained from R. B. Walsh and D. G. Fraenkel. All three kinase genes contain an NcoI site early in the coding sequence (at position +45). The hxk2::lu2 disruption is an insertion of the LU2 gene into this NcoI site. The hxkl and glkl disruptions are substitution mutations generated by deletion of a fragment of DNA initiating at the NcoI site and extending through the end of the coding sequence, with subsequent insertion of the LU2 gene (R. B. Walsh and D. G. Fraenkel, personal communication). DFY57 was crossed with the isogenic strain DFY1-1C (MATa lysi ura3-52) to yield the set of double kinase null mutants LBY32 (MATa lysl ura3-52 HXKI hxk2::lu2 glkl::lu2), LBY33 (MATa lysi ura3-52 hxkl::lu2 HXK2 glkj:: LU2, and LBY34 (MATa lysi ura3-52 hxkl::lu2 hxk2:: LU2 GLKI). Strain genotypes were confirmed by enzyme assay and determination of fructose:glucose phosphorylation ratios (7, 12, 16) (Table 1) and by crosses against DFY57 and subsequent tetrad analysis for Mendelian (2:2) segregation of fructose and/or glucose utilization. For aerobic growth, cells (2 ml) were grown in beveled flasks (3 ml) with shaking at 2 rpm at 3 C. Anaerobic fermentations were conducted in a fermentor (5 liters; New Brunswick Scientific Co.). The vessel containing 2 liters of medium was purged by nitrogen gas flow for 2 h prior to inoculation. Fermentation temperature was 3 C, stirring was at 2 rpm, and nitrogen gas flow was 4 to 5 ml/min. The medium used for both aerobic and anaerobic fermentation consisted of (per liter) yeast extract (1 g), peptone (2 g), glucose (2 g), ergosterol (4 mg), and Tween 8 (1 ml). The ph of the medium was adjusted to 3.5. Inocula for both types of fermentations were grown aerobically in test tubes on a roller drum to approximately 2 x 17 to 5 x 17 cells per

2 VOL. 17, 1988 TABL 1. Hexokinase and glucokinase activity in null mutants' Sugar phosphorylation Fructose:glucose Strain Relevant activity genotype (U/mg of protein) phosphorylation ratio Glucose Fructose DFY1-1C HXKI HXK2 GLKI :1 LBY32 HXKI hxk2::lu :1 glkl::lu2 LBY33 hxkl:lu2 HXK :1 glkl:lu2 LBY34 hxkl::lu <.2 1 hxk2::lu2 GLKI a Cells were grown aerobically on yeast extract (1 g/liter)-peptone (2 g/liter)-glucose (2% wt/vol) rich medium overnight (16 h). The enzymatic assay was as described previously (4, 7). ml and diluted to 1 x 14 cells per ml upon inoculation of the fermentation medium. Doubling times were estimated from the exponential phase of growth, and turbidity measurements were made by using a Klett-Summerson colorimeter; samples were diluted to within the linear range of the colorimeter, with no readings taken above approximately 3 Klett units. Neither aerobic nor anaerobic growth was found to be dependent upon a particular kinase. Doubling times were similar among strains grown aerobically and anaerobically (Table 2). Only in the case of LBY34 (GLKJ) was there a significant difference between generation times, with the slower growth rate occurring anaerobically. Cell yield as indicated by maximum absorbance attained by the cultures varied somewhat among strains and was less under anaerobic conditions (dry weight per turbidity unit values were identical in these strains as expected). The major difference among the strains occurred during anaerobic growth and was the lag period length of LBY34 (GLKI). Glucose utilization rates were estimated during growth of the strains over the same change in turbidity and were found to be very similarvirtually identical in DFY1-1C, LBY32, and LBY33, with LBY34 a bit slower (Table 2). Lobo and Maitra (9) reported that anaerobic glucose utilization rates were greater for strains bearing hexokinase PI or glucokinase activity than for strains with hexokinase PII alone (9). However, these NOTS 5397 analyses were conducted using suspensions of cells in buffer, and those reported here were determined during the course of active growth and fermentation and may, therefore, not be inconsistent. It does appear, however, that loss of hexokinase PII protein does not put the cell at any obvious disadvantage in overall growth and that any one of the kinases is sufficient for good growth aerobically or anaerobically on glucose, in agreement with the studies of point mutations. The kinetics of glucose uptake were investigated in the wild-type strain DFY1-1C, in the set of double kinase null mutants, and in a triple kinase null mutant, LBY31, derived from the same cross. The results obtained under aerobic cultivation were identical to those previously published for the point mutants (1, 3). There was no high-affinity uptake in LBY31 and only high-affinity uptake of glucose in LBY34 (GLKJ) (data not shown). Also, LBY32 (HXKI) and LBY34 (GLKJ) showed constitutive high-level expression of high-affinity uptake, while this transport system seemed to be regulated in LBY33 (HXK2) as it is in the wild type, displaying marked glucose repression of uptake (data not shown). Again, these results are in good agreement with previous work using point mutations. Since the kinetics of glucose uptake had not been investigated in cells grown under anaerobic conditions, anaerobic fermentations of the set of double kinase null mutants and of the wild type were conducted, and samples were removed and analyzed for glucose uptake at late lag/early exponential, mid-exponential, late exponential/early stationary, and late stationary phases of growth. The patterns obtained for the wild type were very similar whether the cells had been grown aerobically or anaerobically (Fig. 1). The strain carrying the wild-type HXK2 gene, LBY33, also exhibited very similar regulation of uptake of glucose (Fig. 2B). At early log phase when the external glucose concentration was still high, high-affinity uptake was repressed, becoming derepressed as glucose was exhausted in the medium. There was significant loss of carrier activity upon continued incubation during stationary phase. The strains carrying only the wild-type gene for HXKI or GLKI displayed high-level expression of high-affinity uptake under repressing conditions (Fig. 2A and C), the level being so high in LBY34 that low-affinity uptake became obscured. The rates of glucose uptake in LBY34 were 5 to 1 times that of the wild type TABL 2. Aerobic and anaerobic growth characteristics of null mutants Maximum Strain Relevant genotype Growth Doubling Final cell anaerobic conditionsa Lag time (h) time density glucose (min) (Klett units)b utilization rate (g/h)c DFY1-1C HXKI HXK2 GLKI ,325 NDd N LBY32 HXKI hxk2::lu2 glkl:lu ,16 ND N LBY33 hxkl::lu2 HXK2 glkj:lu ,35 ND N LBY34 hxkl::lu2 hxk2::lu2 GLKI ,42 ND N ,5.17 a 2, Aerobic; N2, anaerobic. b Final cell density is expressed in Klett units. There was no significant difference in dry weight estimates versus Klett units for the four strains (data not shown). The maximum glucose utilization rate is estimated from the rate of glucose disappearance from the medium during anaerobic fermentation. Glucose concentrations were determined enzymatically by using a kit (Boehringer Mannheim Biochemicals) for sugar analysis. Since growth rates and cell densities were very similar over the most rapid phase of glucose consumption, the glucose utilization rate was not corrected for cell density, which was, of course, changing over the course of the assay. The data are therefore not directly comparable with results of previous work done using suspension of cells in buffer (9). d ND, Not determined.

3 5398 NOTS J. BACTRIOL ~~~~~~~~~~~~~~~~ U~~~~~~~~~~~~~~~ V/S (nmole/min/mg wet wt)/(mm) FIG. 1. Kinetics of glucose uptake during aerobic or anaerobic fermentation by the wild-type strain DFY1-1C. Samples were taken at the early exponential (), mid-exponential (), late exponential/early stationary (), and late stationary (O) (cells in stationary phase for 24 h) phases of growth. V is expressed as nanomoles of glucose per minute per milligram (wet weight) of cells; S is millimolar. Cells were grown as described in the text. (A) Aerobic growth; (B) anaerobic growth. under repressing conditions and two- to threefold that of the wild type under derepressing conditions. This finding is not surprising given that strains lacking hexokinase PII activity are derepressed for carbon-catabolite-repressible functions (6, 1), one of which is high-affinity hexose uptake. The HXKI strain showed loss or decay of high-affinity uptake during stationary phase, but such was not the case for LBY34 (GLKI). ither loss of hexokinase PI and PII function affects normal carrier turnover or high-affinity uptake as determined by glucokinase is insensitive to this control mechanism. The most interesting finding was that strains carrying only HXKI or GLKI showed a striking substrate inhibition under anaerobic conditions not observable in the wild-type or HXK2 strain (Fig. 3). This inhibition was most pronounced in LBY32 (HXKJ) and occurred throughout the time course of fermentation (data shown for mid-exponential-phase growth only). The phenomenon of substrate inhibition is believed to reflect more than one point of attachment between the substrate and the substrate-binding site of the carrier (14). At very high substrate concentrations, two substrate molecules may attempt to bind to the carrier simultaneously through different points of attachment. This type of a substrate-carrier interaction would be abortive; no net transport would occur, resulting in a decrease in the measurable rate of uptake. As demonstrated here, glucose transport in anaerobically grown cells in the absence of HXK2 gene product displayed marked substrate inhibition not observable in the wild-type strain under these growth conditions. Absence of HXK2 gene product is not a normal physiological condition for the cells, and it could be argued that substrate inhibition is an aberration of loss of hexokinase PII protein. However, substrate inhibition is observable in wild-type cells during fermentation of grape must (C. J. McClellan, A. L. Does, and L. F. Bisson, Am. J. nol. Vitic., in press), conditions in which substrate concentrations during growth are much higher than those used here (2 versus 2%). It is perhaps more correct to view loss of hexokinase PII function not as unmasking substrate inhibition but as shifting the sugar concentration range at which substrate inhibition becomes apparent. Whether this reflects a direct role of hexokinase PII in regulation or modulation of the substrate sensitivity of glucose transport or is a normal characteristic of glucokinase- and hexokinase PI-dependent high-affinity uptake remains to be determined. The factors leading to the appearance of substrate inhibition of glucose transport are obviously complex, also involving the presence or absence of oxygen in the growth medium. In interpretation of these findings, it is also important to consider that glucose uptake is normally assayed in restingphase cells. This is necessary in order to ensure uniformity of the cell suspension during the assay itself. Although samples are rapidly processed by filtration, resuspended in buffer, and equilibrated to 3 C, this procedure requires several minutes and cells are exposed to oxygen during manipulation. Thus, substrate inhibition is readily observed in resting-phase cells from an anaerobic fermentation subsequently exposed to oxygen and high glucose concentration. Substrate inhibition may be a mechanism to allow rapid response and adaptation to changing environmental conditions. The role of the hexokinase PII protein in modulating the appearance of substrate inhibition is being investigated. The pattern of glucose uptake in the wild-type strain does not appear to be the simple summation of the individual patterns obtained for the strains each carrying a single kinase gene. It is tempting to speculate that these results imply

4 VOL. 17, 1988 NOTS S ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~ v/s (nmole/ m in/mg wet wt) /(mm) FIG. 2. Kinetics of glucose uptake during the course of anaerobic utilization of glucose in the double kinase null mutants. Cells were grown as described in the text. Samples were removed at the early exponential (), mid-exponential (), late exponential/early stationary (U), and late stationary (l) (24 h in stationary phase) phases of growth. V is expressed as nanomoles of glucose per minute per milligram (wet weight) of cells; S is millimolar. (A) LBY32 (HXKI hxk2::lu2 glkl::lu2); (B) LBY33 (hxkl::lu2 HXK2 glkl::lu2); (C) LBY34 (hxkl::lu2 hxk2::lu2 GLKI). some sort of interaction between the kinases or a signal molecule produced by the kinases and the carrier protein or some competition between the kinases for a binding site on the carrier protein, all of which is under the control of glucose concentration and stage of growth. The SNF3 gene appears to encode the high-affinity carrier or a component 6 4. c 1 2 S (mm) FIG. 3. Substrate inhibition of glucose uptake in anaerobically grown double kinase null mutants. Data are taken from the midexponential phase of growth. V is expressed as nanomoles of glucose per minute per milligram (wet weight) of cells; S is millimolar and represents the concentration range from 2 to 2 mm. DFY1-1C (HXKI HXK2 GLKI) (U); LBY32 (HXKI hxk2::lu2 glkl::lu2) (); LBY33 (hxkl::lu2 HXK2 glkl::lu2) (); LBY34 (hxkl::lu2 hxk2::lu2 GLKI) (). thereof (5), and snf3 mutants do not display any kinasedependent glucose uptake (4), even though all three glucose kinases are present, suggesting that they might function through a common carrier mechanism. To summarize, any one of the three glucose-phosphorylating enzymes, hexokinases PI, PII, or glucokinase, is sufficient to allow good growth both anaerobically and aerobically on glucose, and the loss of hexokinase PII gene product through the construction of a null mutation does not appear deleterious to the cell. The regulation and character of kinase-dependent uptake is dependent upon the particular kinase present in the cell. In strains bearing HXK2, glucose uptake is regulated as it is in the wild type. Strains bearing only HXKJ or GLKI wild-type genes show constitutive expression of kinase-dependent uptake, with dramatic substrate inhibition at high (2 mm glucose or greater) substrate concentrations. This inhibition is not observed during aerobic growth (either for point mutations [1] or for null mutations [L. Bisson, unpublished observations]). Glucokinasedetermined kinase-dependent uptake did not decrease in activity in the stationary phase of growth and may represent the stable component of high-affinity uptake observable in the wild-type strain under these conditions. Thus, the three kinases do not confer identical properties to the high-affinity uptake system in S. cerevisiae, although only one of the three enzymes seems essential for glucose utilization. This research was supported in part by funds from the University of California, Davis, and by a grant from Winegrowers of California. Casey J. McClellan was a recipient of the August Sebastiani Memorial Research Assistantship and of scholarships from the University of California, Wine Spectator, and American Society for nology and Viticulture.

5 54 NOTS Dan Fraenkel is thanked for the gift of strain DFY57, the triple kinase null mutant. Jill Frommelt is thanked for preparation of the manuscript. LITRATUR CITD 1. Bisson, L. F., and D. G. Fraenkel Involvement of kinases in glucose and fructose uptake by Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 8: Bisson, L. F., and D. G. Fraenkel Transport of 6-deoxy- D-glucOse in Saccharomyces cerevisiae. J. Bacteriol. 155: Bisson, L. F., and D. G. Fraenkel xpression of kinasedependent uptake in Saccharomyces cerevisiae. J. Bacteriol. 159: Bisson, L. F., L. Neigeborn, M. Carlson, and D. G. Fraenkel The SNF3 gene is required for high-affinity glucose transport in Saccharomyces cerevisiae. J. Bacteriol. 169: Celenza, J. L., L. Marshall-Carlson, and M. Carlson The yeast SNF3 gene encodes a glucose transporter homologous to the mammalian protein. Proc. Natl. Acad. Sci. USA 85: ntian, K.-D., and K.-U. Frohlich Saccharomyces cerevisiae mutants provide evidence of hexokinase Pll as a bifunctional enzyme with catalytic and regulatory domains for triggering carbon catabolite repression. J. Bacteriol. 158: Gancedo, J. M., D. Clifton, and D. G. Fraenkel Yeast J. BACTRIOL. hexokinase mutants. J. Biol. Chem. 252: Lobo, Z., and P. K. Maitra Genetics of yeast hexokinase. Genetics 86: Lobo, Z., and P. K. Maitra Physiological role of glucosephosphorylating enzymes in Saccharomyces cerevisiae. Arch. Biochem. 182: Ma, H., and D. Botstein ffects of null mutations in the hexokinase genes of Saccharomyces cerevisiae on catabolite repression. Mol. Cell. Biol. 6: Maitra, P. K A glucokinase from Saccharomyces cerevisiae. J. Biol. Chem. 245: Maitra, P. K., and Z. Lobo Genetics of glucose phosphorylation in yeast, p In G. G. Stewart and I. Russell (ed.), Current developments in yeast research, Pergamon Press, Inc., lmsford, N.Y. 13. Muratsubake, H., and T. Katsume Distribution of hexokinase isoenzymes depending on carbon source in Saccharomyces cerevisiae. Biotech. Bioeng. 88: Neame, K. D., and T. B. Richards lementary kinetics of membrane carrier transport. John Wiley & Sons, Inc., New York. 15. Panfili,., and G. Sandri The role of hexokinase as a possible modulator of Ca2" movements in isolated rat brain mitochondria. Biochem. Biophys. Res. Commun. 131: Womack, F. C., M. K. Welch, J. Nielsen, and S. P. Colowick Purification and serological comparison of the yeast hexokinases PI and PII. Arch. Biochem. Biophys. 158: