The initial step in glucose metabolism is uptake of glucose, which is carried out by specific hexose transporter (Hxt) proteins that are localized in the plasma membrane (Bisson et al., 1993; Boles and Hollenberg, 1997; Kruckeberg, 1996; Ozcan and Johnston, 1999). Uptake of glucose is mediated by facilitated diffusion. A multigene family of 20 genes, which encode different putative HXTs, namely HXT1–HXT17, GAL2, SNF3 and RGT2, has been identified. HXT1–4 and HXT6–7 encode the major hexose transporters, because expression of each of these genes in the MC996 background strain deleted for HXT1–7 allows growth on glucose (Reifenberger et al., 1995). Overexpression of HXT8–17 (except HXT12) individually in an hxt1–17 gal2 deletion strain in the CEN.PK background restored growth on glucose, indicating that Hxt8–11p and Hxt13–17p are also able to transport glucose (Wieczorke et al., 1999), but expression is low under normal growth conditions (Ozcan and Johnston, 1999). Recently, it was shown that Hxt5p also has glucose transport capacity (Diderich et al., 2001). Expression of the major HXTs is mainly regulated by extracellular glucose concentrations (Ozcan and Johnston, 1999). SNF3 and RGT2 encode plasma membrane proteins that are involved in sensing the amount of glucose to induce expression of specific HXTs (Ozcan et al., 1996, 1998).
HXTs encode highly homologous proteins with 12 putative transmembrane segments, with amino- and carboxy-terminal domains localized in the cytoplasm (Kruckeberg, 1996). The intracellular amino-terminal domains of Hxt proteins show little homology amongst each other in contrast to the remaining domains. The HXT5 gene encodes a protein of 592 amino acids, which is approximately 20 amino acids larger than the major hexose transporters. The intracellular amino-terminal domains of Hxt1p, Hxt6p and Hxt7p contain 59 amino acids, those of Hxt2p, Hxt3p and Hxt4p 50, 56 and 65 amino acids respectively, but the amino-terminal intracellular domain of Hxt5p is 82 amino acids. Because Hxt5p has a longer amino-terminal domain, it is tempting to speculate that Hxt5p may have a specific function in addition to glucose transport in yeast. Furthermore, HXT5 has a different expression pattern compared to the major hexose transporters, as determined by DNA micro-array experiments. In glucose-grown batch culture experiments, expression of HXT5 was highly induced upon glucose depletion (DeRisi et al., 1997), which was confirmed at the protein level by monitoring expression of Hxt5–GFP (Diderich et al., 2001). Other studies indicated that expression of HXT5 is induced by increasing the osmolarity of the growth medium after addition of NaCl or sorbitol (Gasch et al., 2000; Posas et al., 2000; Rep et al., 2000; Yale and Bohnert, 2001), or by increasing the temperature (Gasch et al., 2000). Increased expression of HXT5 upon glucose depletion, temperature and osmotic upshift suggests a specific role for Hxt5p in adaptation of cells to these new conditions. To test whether Hxt5p was essential for growth, HXT5 was deleted, which did not result in a clear phenotype. Inoculation of stationary phase cells, which normally would have expressed HXT5, into fresh medium containing high levels of glucose, resulted in slightly slower growth of the HXT5 deletion strain compared to wild-type cells (Diderich et al., 2001).
In this study we determined the expression of the hexose transporter HXT5 to obtain insight in the regulation of expression and to obtain clues about a possible function for Hxt5p in addition to glucose transport. HXT5 expression was determined under different conditions, including batch growth, fed-batch growth and well-defined growth conditions in continuous cultures. It was shown that expression of HXT5 mRNA and Hxt5p is not regulated by extracellular glucose concentrations, as is the case for major Hxt proteins, but merely by the growth rates of cells. Based on the unique expression profile of HXT5 and the presence of an extended amino-terminal domain, a possible function for Hxt5p will be discussed.
Materials and methods
Strains, media and growth conditions
The Saccharomyces cerevisiae strains used in this study include CEN.PK 113-7D (MATa, SUC2, MAL2-8c) and was kindly provided by P. Kötter (Institut für Mikrobiologie, Johann Wolfgang Goethe-Universität, Frankfurt am Main, Germany), KY98 (MATa, SUC2, MAL2-8c, HXT5::GFP) by A. Kruckeberg (E. C. Slater Institute, University of Amsterdam, Amsterdam, The Netherlands) and strain JBY20 (MATa, SUC2, MAL2-8c, ura3, HXT5::HA) by J. Becker (Institut für Mikrobiologie, Heinrich-Heine-Universität, Düsseldorf, Germany). During batch culture experiments yeast cells were grown on 0.67% (w/v) yeast nitrogen base without amino acids (Difco) and 2% (w/v) of the carbon source as indicated in the text. Cells were grown at 30 °C at 180 rpm in a shaking incubator (New Brunswick Scientific). For temperature upshift experiments, cells were grown to OD600 = 1.2 (±0.1). The culture was subdivided into 15 ml Falcon tubes and incubated in a waterbath at 42 °C, while shaking the tubes every 10 min. For the osmotic upshift experiments the cells were also grown to OD600 = 1.2 (±0.1) and concentrated 5 M NaCl was added to a final concentration of 0.7 M. Samples were directly frozen in liquid nitrogen and stored at −80 °C.
Fed-batch growth conditions
Experiments with synchronous cultures in fed-batch experiments were performed as described earlier, with some modifications (Sillje et al., 1997). Cells were grown in YNB medium without amino acids with glucose as carbon source at a cell density of 1.2 × 107 cells/ml, and an initial extracellular glucose concentration of 0.05 mM. Glucose dissolved in YNB medium was continuously added at rates of 10 and 50 fmol/cell/h, respectively. The number of cells and the external glucose concentration was monitored throughout the experiment.
Centrifugal elutriation was performed as described previously, with some modifications (Woldringh et al., 1993). Cells were grown until the exponential phase in YNB medium containing 1% galactose at 30 °C. 2 × 1010 cells were loaded in a 40 ml chamber of a Beckman J-6MI centrifuge with a JE-5.0 rotor at 30 °C and 2000 rpm. Cells were grown in the elutriation chamber on YNB 1% galactose medium. Newborn daughter cells were collected on ice, centrifuged and kept overnight on ice in YNB 1% galactose medium. The cell size was monitored with a Coulter Multizer II, and the flow rate of the elutriation was adapted to maintain a constant cell size.
Cells were grown in a 2 L BiofloIII chemostat (New Brunswick Scientific) connected to a computer controller unit with Advanced Fermentation Software (New Brunswick Scientific). Cells were inoculated in the EGLI culture medium, as described previously (Meijer et al., 1998) and continuous feed was connected after overnight batch growth. During the different dilution rates, the NH4+ concentration was adapted to 1.5 g/l to maintain nitrogen limitation, the glucose concentration in the feed was 200 mM. Steady-state samples were taken as described (Meijer et al., 1998).
Extracellular glucose concentrations were determined as described (Bergmeyer, 1974). Samples were mixed with hexokinase/glucose-6-phosphate dehydrogenase (Boehringer Mannheim), NADP+ (Roche) in a 100 mM imidazole, 10 mM MgCl2 buffer, pH 7.0, and the conversion of NADP+ into NADPH was determined using a spectrophotometer (Pharmacia Biotech).
Northern blot analysis
To isolate total RNA, yeast cells were broken with 0.45 mm glass beads in a Bead Beater (Biospec Products Inc.) in phenol and RNA extraction buffer (1 mM EDTA, 100 mM LiCl, 100 mM Tris–HCl, pH 7.5, 0.5% (w/v) lithium dodecylsulphide). A phenol/chloroform extraction was performed, and total RNA was precipitated by adding 3 M NaAc, pH 5.6. Samples were washed with ethanol, air-dried and suspended in DMPC-treated water. Equal amounts of total RNA (10 µg) were loaded on a 1% denaturing formamide/formaldehyde gel and RNA was separated by electrophoresis. RNA was transferred to Hybond-N membrane (Amersham Pharmacia Biotech) and cross-linked using UV light in a UV stratalinker (Stratagene). 15 pmol HXT5-specific oligonucleotides (5′-TCCCAAGGGGCCTTGATGAGCGTT-3′) was labelled with T4 polynucleotide kinase (USB) and 50 µCi γ-32P-ATP (Amersham Pharmacia Biotech), and purified using the QIAquick nucleotide removal kit (Qiagen). The blots were washed once in 2× SSC at room temperature, incubated for prehybridization in hybridization mixture (1 mM EDTA, 7% SDS, 0.5 M NaPO4, pH 7.5) for at least 1 h at 45 °C in a Micro-4 hybridization oven (Biozym). Labelled oligonucleotides were added and hybridized overnight at 45 °C. After hybridization the blots were washed for 2 min in 2× SSC at room temperature, twice for 20 min in 2× SSC, 0.1% SDS, 0.1% NaPPi at 45 °C, and once for 20 min with 0.5× SSC, 0.1% SDS, 0.1% NaPPi at 45 °C. After a final wash for 10 min in 2× SSC at room temperature, the membrane was wrapped in Saran (Dow Chemicals), and autoradiograms were developed using hyperfilm MP (Amersham). To control whether equal amounts of RNA was loaded, the gels were checked for ethidium bromide staining by UV light and the membranes were probed with an oligonucleotide against ACT1 (5′-TGTCTTGGTCTACCGACGATAGATGGGAAG-3′).
Western blot analysis
JBY20 cells were collected in Falcon tubes, frozen immediately in liquid nitrogen and stored at −80 °C. The cells were thawed on ice, collected by centrifugation, washed in ice-cold water, and resuspended in PBS containing Complete protease inhibitors (Boehringer). Equal amounts of cells were lysed by shaking vigorously with 0.45 mm glass beads in a Bead Beater. Equal amounts of protein were loaded on a 10% SDS–PAGE gel, and transferred to PVDF membrane (Roche) after electrophoresis. The membranes were blocked in 5% Protifar (Nutricia) in TBST buffer (50 mM Tris–HCl, pH 7.4, 10 mM NaCl and 0.1% Tween 20) for 1 h at room temperature. The membranes were incubated with 12CA5 antibody (Roche) with TBST/0.5% Protifar for 1 h at room temperature. The primary antibody was detected using peroxide-conjugated rabbit anti-mouse (Jackson Immunoresearch). Proteins were visualized by Enhanced Chemoluminescence (Renaissance).
Expression of HXT5 mRNA and Hxt5 protein in glucose-grown batch cultures
To determine whether HXT5 expression was related to the extracellular glucose concentration, overnight-grown JBY20 cells were inoculated into fresh medium containing 2% glucose and grown in batch cultures. In JBY20 cells, HXT5 mRNA expression was induced 9 h after inoculation (Figure 1a). Hxt5-HA protein was expressed after 9 h after inoculation (Figure 1b). Yeast cells had different growth rates during growth on glucose in batch cultures (Figure 1c). JBY20 cells exhibited fast growth during the exponential growth phase (µ = 0.36/h) and slow growth upon entry in the stationary phase approximately 9 h after inoculation (µ = 0.028/h). No HXT5 mRNA or Hxt5-HA was expressed during the exponential growth phase. Expression of Hxt5-HA was observed in slowly growing cells until 2 h after inoculation. This was probably a remainder of the Hxt5-HA that was expressed in the overnight-grown cells used for inoculation, and probably not the result of de novo synthesis, because HXT5 mRNA was not expressed at these time points. Furthermore, JBY20 cells grown exponentially overnight did not show expression of Hxt5-HA at the same optical densities (data not shown).
To establish whether the increased expression was related to the extracellular glucose concentration, extracellular glucose levels in the medium were measured. Extracellular glucose levels decreased, starting from 105 mM at inoculation and decreasing to 0.55 mM after 15 h of growth, and to 0.03 mM after 18 h of growth (Figure 1c). HXT5 expression was already initiated 9 h after inoculation, when 48 mM glucose was still present. However, when glucose was depleted (<1 mM) 15 h after inoculation, HXT5 expression was maximally induced. However, Hxt5-HA expression remained constant from approximately 12 h after inoculation. Therefore, regulation of HXT5 expression by glucose alone seems unlikely, but a parameter that correlated with increased expression of HXT5 after 9 h of growth was a decrease in growth rate of the cells.
Expression of HXT5 on different carbon sources
To investigate whether HXT5 expression was dependent on growth on glucose, CEN.PK 113-7D cells were grown in batch cultures on YNB media containing different carbon sources. On easily fermentable carbon sources, including glucose, galactose, fructose, mannose, sucrose and maltose, cells exhibited high growth rates during the exponential phase of cell growth. Cells growing on non-easily fermentable carbon sources, e.g. ethanol and glycerol, exhibited low growth rates during the exponential phase of cell growth (Table 1). Independently of the carbon source that was used, HXT5 was expressed after 24 h of growth when cells entered the stationary phase and exhibited low growth rates (Figure 2). Cells growing exponentially 6 h after inoculation on medium containing the carbon sources glucose, galactose, fructose, mannose, sucrose and maltose did not express HXT5. Expression of HXT5 was observed during the exponential phase of cell growth, when cells were grown on ethanol or glycerol (Figure 2). These results indicate that glucose is not a determining factor in the regulation of HXT5 expression, but a decrease in growth due to the carbon source was accompanied by induction of HXT5 expression.
Table 1. Growth rates of CEN.PK 113-7D cells growing exponentially on YNB medium with 2% of different carbon sources in batch cultures
Exponential growth (µ/h)
Expression of HXT5 during environmental changes leading to low growth rates
Changing growth conditions by increasing the temperature or osmolarity creates conditions that influence the growth rate of cells. When the temperature of the medium of cells growing exponentially on YNB 2% glucose was changed from 30 °C to 42 °C over a 90 min period, growth rates decreased from µ = 0.36/h to µ = 0.14/h. Cells growing exponentially at 30 °C did not express HXT5. Increasing the temperature to 42 °C induced expression of HXT5 (Figure 3a). To determine whether expression of Hxt5 protein followed the expression pattern of HXT5 mRNA, expression of Hxt5-HA during temperature upshift was determined. Hxt5-HA was expressed 20 min after temperature upshift and was present throughout the treatment (Figure 3b). Throughout the treatment the extracellular glucose concentration remained higher than 60 mM, indicating that increased expression of HXT5 was not a result of glucose depletion.
Stress and hence a lower growth rate was also introduced by adding NaCl to a final concentration of 0.7 M to cells growing exponentially on YNB 2% glucose (µ = 0.35/h for the non-treated and µ = 0.25/h for the treated cells). This switch in external conditions was accompanied by a transient expression of HXT5 (Figure 4a). This transient induction is not due to glucose limitation, as the glucose concentration remained higher than 60 mM throughout the experiment. During the switch of external conditions the pattern on the Hxt5-HA protein level was different from that of HXT5 mRNA, as the protein was present throughout the experiment, and HXT5 mRNA was expressed transiently (Figure 4b). Again, the results of these experiments indicate that a decrease in the growth rate of cells correlates with increased expression of HXT5.
Expression of HXT5 during low growth rates in fed-batch cultures
To study the effect of G1 phase elongation, and hence low growth rates, on expression of HXT5, Hxt5-GFP tagged cells that were synchronized early in the G1 phase were grown in fed-batch cultures. Low growth rates were induced by growing the cells on 10 fmol glucose/cell/h, while 50 fmol glucose/cell/h was used to induce high growth rates (Sillje et al., 1997). At a consumption rate of 10 fmol glucose/cell/h, HXT5 mRNA was detected. No HXT5 mRNA was detected at a consumption rate of 50 fmol glucose/cell/h (Figure 5a). Fluorescence studies confirmed that at the consumption rate of 10 fmol glucose/cell/h, Hxt5–GFP was incorporated in the plasma membrane within 2 h, whereas at the consumption rate of 50 fmol/cell/h, even after 8 h of growth, no fluorescence was observed (data not shown).
Even under these conditions the external glucose concentration did not have a regulatory function in HXT5 expression. During the experiment extracellular glucose levels of the culture growing on 10 fmol glucose/cell/h remained at a concentration of approximately 0.1 mM (Figure 5b). The fast growing cells with a consumption rate of 50 fmol glucose/cell/h initially have higher extracellular glucose levels (0.2–0.4 mM). After 5 h of growth, the extracellular glucose concentration diminishes to 0.1–0.15 mM. These concentrations were also found in the culture growing at a consumption rate of 10 fmol glucose/cell/h (Figure 5b). The cells growing at higher growth rates did not express HXT5 and Hxt5-GFP after 5 h, even when the extracellular glucose concentration is comparable to concentrations of the slowly growing culture. During slow growth in fed-batch cultures the parameter that correlated with increased expression of HXT5 was again a decrease of the growth rate, and the growth rate per se might therefore regulate the expression of HXT5.
Expression of HXT5 in a nitrogen-limited continuous culture
To determine whether the growth rate was the only parameter that determines expression of HXT5, CEN.PK 113-7D cells were grown in a nitrogen-limited continuous culture. This experimental set-up allowed modulation of only the growth rate of yeast, by changing the dilution rate of the culture (Meijer et al., 1998). Other parameters, e.g. temperature, agitation, pH and oxygen availability, were constant. Also the concentration of intracellular metabolites, including glucose-6-phosphate, fructose-6-phosphate, glucose-1-phosphate and ATP, were constant at different dilution rates (data not shown). The concentration of the limiting compound, in this case nitrogen, was also constant, namely virtually 0 mM at all dilution rates (Herbert et al., 1956; Tempest, 1970). HXT5 was only expressed at dilution rates lower than 0.10/h, whereas no HXT5 expression was observed at dilution rates higher than 0.13/h (Figure 6). At the dilution rate of 0.10/h, HXT5 was expressed to a lower extent compared with expression at dilution rates of 0.068/h and 0.071/h. The dilution rate is the only parameter that is changed in the continuous culture experiments. Low dilution rates, and hence low growth rates, result in increased expression of HXT5. Therefore, these results clearly indicate that the growth rate determines expression of HXT5 in Saccharomyces cerevisiae.
To obtain clues about the function of Hxt5 and to find mechanisms that are involved in expression of HXT5, its expression was determined in different experimental set-ups. Batch culture experiments revealed that HXT5 is expressed at both the mRNA and protein level when ample glucose is still available in the medium, and remained present after glucose depletion. These results are largely in agreement with earlier observations (DeRisi et al., 1997); however, HXT5 is also expressed prior to glucose depletion in our experiments. This suggests that Hxt5p might contribute to glucose transport, which is supported by the observation that Hxt5p is indeed able to transport glucose across the plasma membrane (Diderich et al., 2001).
Independent of the carbon source in which cells are inoculated, HXT5 is expressed after 24 h of growth, and cells grown in ethanol or glycerol already expressed HXT5 in the exponential phase of batch growth. These results are in agreement with earlier observations, where expression of Hxt5–GFP in cells growing on different carbon sources in batch cultures was studied (Diderich et al., 2001). Increasing the temperature or osmolarity of the growth medium of exponentially growing cells resulted in increased expression of HXT5 at both mRNA and protein levels. These results confirm the results of various DNA micro-array experiments, which show increased expression of HXT5 after increasing the temperature (Gasch et al., 2000) or osmolarity (Gasch et al., 2000; Posas et al., 2000; Rep et al., 2000; Yale and Bohnert, 2001). In fed-batch cultures HXT5 was expressed when cells were grown at 10 fmol glucose/cell/h, whereas no HXT5 expression was observed at 50 fmol glucose/cell/h. Taken together, our results indicate that during all experiments one parameter that results in induction of HXT5 expression is in common, viz. a decrease in the growth rate. These results were confirmed by continuous culture experiments, which were used for modulation of only the growth rate of cells under well-defined growth conditions. HXT5 is expressed only at dilution rates lower than 0.10/h, and expression of HXT5 is increased even more when the growth rate is further diminished to dilution rates of 0.068/h. The concentration of the growth-limiting substrate, in this case nitrogen, is extremely low at virtually all growth rates and is the only substrate that determines the growth rate (Herbert et al., 1956; Tempest, 1970).
Our results suggest that expression of HXT5 is not regulated by glucose and not subjected to glucose repression. Furthermore, in a hexokinase II deletion mutant, a protein known to be involved in the regulation of glucose repression, HXT5 expression is not derepressed at high extracellular glucose concentrations in batch cultures (Petit et al., 2000). This result indicates that HXT5 expression is not regulated by glucose repression. Interestingly, expression of HXT7 is derepressed at high extracellular glucose concentrations, showing that expression of certain HXTs can be repressed by glucose (Petit et al., 2000). Also, Snf3 and Rgt2 are not involved in the regulation of HXT5 expression, as cells deleted for snf3 or rgt2 still expressed HXT5 in glucose-grown batch cultures (data not shown). The Snf3/Rgt2 pathway does regulate expression of the major HXTs (Ozcan and Johnston, 1999). Because glucose repression and the Snf3/Rgt2 pathway are not involved in regulation of HXT5 expression, another mechanism probably regulates expression of HXT5. Indeed, our results indicate that expression of HXT5 is induced upon a decrease in growth rates of cells.
To obtain further insight into how HXT5 expression is regulated, the promoter region of the HXT5 gene was analysed to reveal elements that might be involved in regulation of expression (http://embnet.cifn.unam.mx/∼jvanheld/rsa-tools). The HXT5 promoter contains two stress-responsive elements (STREs; −472 bp and −304 bp relative to the translation initiation site, respectively), two HAP2/3/4/5 binding sites (−845 bp and −785 bp, respectively) and one PDS element (−544 bp) (Boorstein and Craig, 1990; Estruch, 2000; Olesen et al., 1987). Surprisingly, the promoter of HXT5 appears to be homologous with the promoter of GSY2, encoding glycogen synthase 2, which is involved in glycogen synthesis (Farkas et al., 1991). The promoter of GSY2 also contains two STREs, two putative HAP2/3/4/5 complex binding sites and one PDS element. Furthermore, expression of GSY2 exhibits a similar expression pattern as HXT5 (Parrou et al., 1999), indicating that HXT5 and GSY2 expression could be regulated in a similar matter. The involvement of low growth rates in HXT5 expression does not exclude the involvement of these elements in the promoter of HXT5 in determining expression of HXT5 during low growth rates. Furthermore, the transcriptional elements may even be activated under conditions that cause low growth rates to induce expression of HXT5.
Cell cycle duration is a well-organized process of which environmental conditions are the main regulators. It was shown that cell cycle duration could be greatly elongated by growing cells in fed-batch cultures on low amounts of carbon source. Concomitantly, trehalose and glycogen were accumulated in these cells (Sillje et al., 1997). Similar results were obtained from cells that were grown in continuous cultures, where an increase in cell cycle duration is also accompanied with elevated trehalose and glycogen levels (Paalman, 2001). Surprisingly, we observed that HXT5 was expressed whenever trehalose was accumulated during growth in batch cultures, fed-batch cultures and continuous cultures (data not shown). Furthermore, genome-wide analysis of stressful conditions, including temperature upshift, adding chemical compounds that are hazardous for cells and increased osmolarity, revealed that expression of HXT5 is induced concomitantly with genes involved in reserve carbohydrate metabolism (Gasch et al., 2000). None of the other hexose transporters has this specific expression pattern. Furthermore, HXT5 is structurally different compared to the major HXTs, because it contains a larger intracellular amino-terminal domain. Taken together, these observations suggest a specific role for Hxt5p in the accumulation or metabolism of reserve carbohydrates. The precursor for trehalose is glucose and Hxt5p may specifically regulate uptake of glucose that is designated for production of trehalose during conditions that induce low growth rates. Furthermore, it was postulated that Tps1p, a protein involved in trehalose synthesis, might function as a direct regulator of glucose transport, probably by interacting with a hexose transporter (Thevelein and Hohmann, 1995). In our opinion, Hxt5p seems a good candidate to interact with Tps1p, thereby regulating accumulation of trehalose.
In conclusion, our results indicate that expression of HXT5 is determined by the growth rate of cells and is not dependent on the extracellular glucose concentration. The promoter of HXT5 contains putative regulatory elements, which may contribute to expression of HXT5 during low growth rates. The extended amino-terminal domain of Hxt5p and the unique expression pattern of HXT5 during various kinds of conditions leading to low growth rates that are concomitant to accumulation of trehalose, suggest a role for Hxt5p in accumulation of this reserve carbohydrate besides glucose transport.
We thank Jessica Becker for providing strain JBY20 and Arle Kruckeberg for providing strain KY98 and oligonucleotides for analysing HXT5 expression. We are grateful to Sjoukje Slofstra for technical assistance.