Intracellular Maltose Is Sufficient To Induce MAL Gene Expression in Saccharomyces cerevisiae

Yeast strains.The strains used in this study are listed in Table 1. CMY1001 has been described by Medintz et al. (23). It contains a single MAL1 locus at which the MAL11 maltose permease gene is replaced by hemagglutinin (HA)-tagged MAL61. No other MAL genes are present in this strain. Strain CMY1050 is a MAL61/HA-null derivative of CMY1001 and was constructed by the PCR-based gene disruption described by Medintz et al. (24). Strain CMY1061 is a MAL13 deletion disruption of CMY1050 and was constructed as follows. The appropriate upstream (5′-CCATGTAATCGCTGCATTCAGCGCAATTTGAACTGCACTCAGCTGAAGCTTCGTACGC) and downstream (5′-CGGTGCAAACAATAGTATGTCATGATTCGAAATATGTCGGCATAGGCCACTAGTGGATCTG) primers were used to amplify the G418 resistance marker gene with pFA2-kanMX2 as the template (35). Bases homologous to the template are underlined. The resulting PCR product, which has homology to the MAL13 sequence at both the 5′ and 3′ ends of the open reading frame, was then used for one-step replacement of MAL13. Candidate disruptants were confirmed by PCR analysis.

A similar PCR-based process was used to construct CMY1071 (mal13Δ::G418), CMY1072 (mal33Δ::HygB), and CMY1073 (mal13Δ::G418 mal33Δ::HygB) in strain YPH500. Strain YPH500 is isogenic to S288C, and primers for the deletion of mal13 and mal33 were determined on the basis of the sequence of S288C, which is available at the Saccharomyces Genome Database website (http://genome-www.stanford.edu/Saccharomyces/ ). The primers 5′-ACTTTAACTAAGCAAACATGCGCCAAGCAGGCATGCGACTGCTGTCGATCAGCTGAAGCTTCGTACGC and 5′-ATCAAGGGTCTATGTCTTCATTATCCTTGGGATAACCATCCAATTGTAAAGCATAGGCCACTAGTGGAT were used to amplify G418^R for mal13 disruption. Primers 5′-ACTTTAGTCAAGTATGCATGCGACTATTGTCGTGTCCGTCGAGTAAAGTGCAGCTGAAGCTTCGTACGC and 5′-AGGAATTATGTCGTCTTCATCTTTGGAATCATCATTTAGGCGCAGTGGTCGCATAGGCCACTAGTGGAT were used to amplify Hyg^R for mal33 disruption.

Plasmid construction and mutagenesis. MAL61/HA was constructed from MAL61 by inserting a 12-codon sequence containing a single of copy of the HA epitope tag at the 5′ end of the open reading frame (23). MAL61/HA was subcloned into pUN30, yielding pMAL61/HA. The native MAL61 promoter in pMAL61/HA was removed and replaced with the ADH1 promoter at the −12-bp position of MAL61, yielding pADH1-MAL61/HA, which has been described by Medintz et al. (24). Plasmids pMAL61/HA and pADH1-MAL61/HA were used as templates for in vitro mutagenesis with a Bio-Rad Mutagene Kit (Bio-Rad, Richmond, Calif.) to construct pMAL61/HA(Δ61-90) (25), pMAL61/HA(Δ571-580), pADH1-MAL61/HA(Δ61-90), pADH1-MAL61/HA(Δ571-580), and a series of C-terminal nonsense mutations (pMAL61/HA-581NS, pMAL61/HA-575NS, pMAL61/HA-570NS, and pMAL61/HA-560NS).

Plasmid pPTE18 was obtained from Norbert Sauer, University of Regensburg, Regensburg, Germany. PmSUC2 cDNA encoding a P. major sucrose transporter was ligated into the EcoRI site of vector NEV-E, yielding pPTE18 (13). Expression of PmSUC2 is controlled by the S. cerevisiae PMA1 promoter.

Maltose transport assay.Maltose transport was measured as the rate of uptake of 1 mM [¹⁴C]maltose as described by Cheng and Michels (8) and Medintz et al. (23). Assays were done in duplicate on two or three transformants. V_max was determined by Lineweaver-Burk analysis as described by Medintz et al. (23). Maltose transport activity and V_max are expressed as nanomoles of maltose transported per milligram of cells per minute.

Maltase assay.Maltase activity was determined in total cell extracts as described by Dubin et al. (11). Activity is expressed as nanomoles of p-nitrophenyl β-d-glucoside (PNPG) hydrolyzed per milligram of total protein per minute. The values reported are averages of duplicate assays obtained with extracts from at least two separate cultures.

Western blot analysis.Western blot analysis was carried out as described previously (23). Mal61/HA protein was detected by using anti-HA-specific antibody, the Vistra-ECF kit (Amersham), and the Storm 860 image analyzer (Molecular Dynamics). Relative protein levels were determined by using the Storm 860 image capture software. Western blot analysis was done in duplicate on extracts prepared from duplicate experiments carried out with at least two independent transformants.

PmSUC2 suppresses the noninducible phenotype of a maltose permease deletion.CMY1001 is a maltose-fermenting strain containing MAL1 as the sole MAL locus (23). The MAL1 locus of CMY1001 encodes maltose permease (MAL61/HA), maltase (MAL12), and the MAL activator (MAL13). Charron et al. (5) showed that loss of either maltose permease or the MAL activator blocks maltose induction of MAL gene expression but loss of maltase has no obvious effect on induction. MAL61/HA was deleted from strain CMY1001 to create CMY1050 (mal61Δ::HIS3), and maltose induction of maltase expression was determined (Table 2). Under uninduced conditions (3% glycerol, 2% lactate), CMY1001 expresses a low but significant level of maltose transport activity (0.17 nmol/min/mg [dry weight] of cells). Deletion of the maltose permease gene reduces this rate to a background rate of 0.09 nmol/min/mg that probably represents the nonspecific low-affinity binding activity described by Benito and Lagunas (2) and not true transport. Table 2 also confirms that maltose induction is dependent on maltose permease. A 6-h induction period is sufficient for full induction of maltase expression in CMY1001, while permease deletion strain CMY1050 exhibits no induction during the same period.

PmSUC2 encodes a sucrose transporter from P. major that is a member of the 12-transmembrane domain family of sugar transporters (13). Blast analysis comparing the amino acid sequence of PmSUC2 to that of Mal61p indicates that these transporters do not share significant sequence homology (only 10% identity largely in transmembrane domains). When expressed in Saccharomyces, PmSUC2 was found to be capable of transporting sucrose and this sucrose transport was inhibited by maltose, indicating that PmSUC2 could also be capable of transporting maltose (13). Plasmid pPTE18, which carries the PmSUC2 gene under the control of the constitutive Saccharomyces PMA1 promoter, was introduced into maltose permease deletion strain CMY1050, and the ability to support maltose induction was tested. As shown in Table 2, growth in glycerol-lactate allows expression of PmSUC2 and produces a low but significant level of maltose transport activity. The maltose transport rate in cells expressing PMA1promoter-PmSUC2 is approximately 10-fold lower than that in cells carrying ADH1-MAL61, but this reduced transport rate is nevertheless sufficient to cause wild-type levels of maltase induction. CMY1050 carrying constitutively expressed ADH1promoter-MAL61 is presented as a control.

To test whether PmSUC2-mediated induction is dependent on the MAL-activator, MAL13 was deleted from CMY1050, creating CMY1061 (mal61Δ mal13Δ) and maltose induction of maltase was assayed in transformants constitutively expressing either Mal61p or PmSUC2. The results in Table 2 demonstrate that loss of the MAL13 activator gene blocks maltose-induced maltase expression by both the PmSUC2 transporter and Mal61/HA permease. Thus, PmSUC2-dependent maltose induction requires the MAL-activator. Taken together, theses results indicate that intracellular maltose is sufficient to induce MAL gene expression independently of the mechanism of maltose transport and that induction by intracellular maltose is dependent on the MAL-activator.

The concentration of intracellular maltose correlates with the level of the induction.Medintz et al. (24) constructed a series of deletions of the N-terminal cytoplasmic domain of Mal61/HA permease as part of a study done to localize the target site of glucose-induced endocytosis and proteolysis. One interesting mutation, a deletion of residues 61 to 90, exhibited little or no transport activity but nonetheless was able to induce MAL gene expression. Further analysis of this mutant is shown in Table 3. The maltose transport activity of strains expressing mal61/HA(Δ61-90) from the native promoter, when measured at a substrate concentration of 1 mM maltose (standard assay condition), does not differ significantly from that of the vector control. Measurement of V_max by Lineweaver-Burk analysis found an approximately threefold increase in V_max compared to that of the vector control, and this suggests that the low level of maltose induction exhibited by this strain could be a function of this very low rate of maltose transport (Table 3). The K_m of the Mal61/HA(Δ61-90) permease is unaffected (approximately 1.1 mM for both alleles). Expression of mal61/HA(Δ61-90) from the constitutive ADH1 promoter allowed a modest increase in maltose transport activity and a slight increase in maltase induction levels. These activities were still too low to allow rapid maltose fermentation. As a result, the strain grows slowly on maltose as the sole carbon source and growth is not associated with carbon dioxide bubble formation.

An additional series of MAL61/HA mutants containing deletions of the C-terminal cytoplasmic domain were constructed by creation of a translation stop site at codon 560, 570, 575, or 581 and by removal of codons 571 to 580 as described in Materials and Methods. Plasmid-borne copies of these mutant alleles were transformed into strain CMY1050, and the transformants were tested for the ability to induce maltase activity. The results are shown in Table 4. Also shown are the levels of maltose transport activity and permease protein expressed by the transformants. A clear correlation can be observed between the levels of maltose transport activity and maltase induction. Additionally, residues 571 to 580 appear to play an important role in maltose transport activity but not an essential role in induction. Truncation of Mal61p to residue 580 has only a modest effect on transport activity, while truncation to residue 574 or 559 or deletion of residues 571 to 580 severely reduces or eliminates transport (Table 4). Deletion of residues 571 to 580 reduces transport activity almost 20-fold, but maltase induction is reduced to about one-third of the wild-type level (Table 4). The results reported in Tables 3 and 4 suggest that extremely low levels of intracellular maltose are able to trigger MAL gene induction and that, at these limiting intracellular maltose concentrations, the level of induction correlates with the presumed level of intracellular maltose.

Low rates of maltose transport activity cause a delay in induction by very low extracellular concentrations of maltose.The finding that, at very low levels of maltose transport activity, induction correlates with transport rates suggests that the threshold level of intracellular maltose needed to induce the MAL genes is very low. If a strain expressing mal61/HA(Δ61-90) from its native promoter were grown in medium containing a very low concentration of maltose, we would expect a delay in induction because greater time would be needed to accumulate sufficient intracellular maltose to reach this threshold level. We wished to test this prediction. Strains expressing either mal61/HA(Δ61-90) or MAL61/HA were grown to early log phase under uninduced conditions and transferred to medium containing either 0.5 or 0.01% maltose. The time courses of maltase induction at these two concentrations are compared in Fig. 1. When transferred to medium containing 0.5% maltose, the strain expressing wild-type maltose permease initiates induction at approximately 30 min while induction is delayed to about 1 to 1.5 h in the strain expressing mutant permease. Moreover, the rate of increase in maltase activity is far more rapid in cells expressing wild-type permease. In contrast, when the time course of induction is followed in medium containing 0.01% maltose, induction in the wild-type strain still initiates at about 30 min, although the rate of increase in maltase activity is slowed, but no significant induction is observed in the strain expressing mutant maltose permease, even after 4 h. These results are consistent with the proposal that a threshold level of intracellular maltose is needed in order to trigger MAL gene induction and that this threshold level is rather low.

• Open in new tab
• Download powerpoint

FIG. 1.

Time course of maltase induction in strains expressing either MAL61/HA or mal61/HA(Δ61-90) maltose permease. CMY1050 strains expressing either mal61/HA(Δ61-90) or MAL61/HA were grown to early log phase under uninduced conditions (3% glycerol and 2% lactate) and transferred to inducing medium containing 3% glycerol-2% lactate plus either 0.5% (13.9 mM) (A) or 0.01% (0.3 mM) (B) maltose. Cells were collected at the indicated time points, and maltase activity was assayed as described in Materials and Methods. The data represent the averages and standard deviations of results obtained in at least two independent assays for cultures grown under both conditions.

Overexpression of Mal61 maltose permease rescues the defect in MAL induction caused by the mal13, but not the mal33, mutant activator.Strain YPH500, which is essentially isogenic to S288C, contains two MAL loci mapping to the right telomeres of chromosomes VII (MAL1) and II (MAL3) but is unable to ferment maltose and is defective for maltose induction (Table 5, line 3). This induction defect is complemented by a plasmid-borne copy of MAL63 encoding the MAL-activator from MAL6 (Table 5, line 1), indicating that the noninducible phenotype of YPH500 results from a lack of a functional MAL activator. The amino acid sequences of the defective mal13p and mal33p MAL activators encoded by YPH500 are 70.7 and 71.2% identical to Mal63p, respectively.

As part of a separate study of the MAL activator, we introduced a plasmid carrying the ADH1promoter-MAL61 gene into YPH500. Much to our surprise, we found that constitutive high-level expression of Mal61 permease is able to restore maltose inducibility of maltase expression to wild-type levels in strain YPH500 despite its defective MAL-activators (Table 5, lines 1, 2, and 4).

To determine whether rescue by ADH1promoter-MAL61/HA is dependent on mal13, mal33, or both, single and double deletions of both genes were constructed in strain YPH500, yielding CMY1071, CMY1072, and CMY1073, and maltose induction of maltase was assayed. Loss of mal13 blocked the suppressing effect of constitutive Mal61 permease (Table 5, lines 6 and 10). The strain lacking only mal33 was still rescued by overexpressed Mal61 permease, as well as strain YPH500 (Table 5, line 8). These results indicate that constitutive, high-level Mal61 permease expression rescues the induction defect caused by mal13 but not that caused by mal33.

Rescue of the defective mal13 MAL-activator is not specifically dependent on Mal61 maltose permease.The results shown in Table 5 may be interpreted in two ways. First, if the maltose permease protein itself were to play a direct role in induction, simple restoration of maltose transport activity by an unrelated transport protein such as PmSUC2 permease should not be able to rescue the defective mal13 MAL-activator. An example of such is the direct binding of the Escherichia coli MalT transcription activator to the maltose transport complex (MalEFGK₂) (29). Alternatively, if rapid delivery of an initially very high concentration of intracellular maltose were capable of triggering MAL gene induction by the defective mal13 MAL-activator, different permeases with varied transport activities should vary in the ability to rescue mal13. Thus, even the PmSUC2-encoded transporter should be capable of rescuing mal13 despite its lack of sequence homology to Mal61 permease. To test this, plasmids constitutively expressing different versions of maltose permease genes [MAL61/HA, mal61/HA(Δ61-90), and PmSUC2] were introduced into YPH500 and maltase induction was assayed. The results in Table 6 indicate that PmSUC2 permease is able to rescue the defective mal13 MAL-activator. Moreover, the significantly reduced level of maltose transport by PmSUC2 has a relatively modest impact on mal13 suppression, suggesting that the suppression is solely a function of the maltose transport activity of PmSUC2 permease and is unrelated to any specific sequence feature of this protein.

Maltose permease does not appear to function as a maltose sensor.Demonstrating that intracellular maltose is sufficient to stimulate MAL gene induction does not exclude the possibility that the Saccharomyces maltose permease also has a sensor-like role in induction. Our finding that constitutive high-level expression of Mal61 maltose permease suppresses the defective mal13-encoded MAL-activator suggested a possible role in induction for maltose permease similar to that of the E. coli maltose transport complex in the bacterial maltose utilization system. In the E. coli maltose-maltodextrin system, uptake is mediated by a periplasmic binding protein-dependent ABC transporter (the MalEFGK₂ complex). MalT, a transcriptional activator, is sequestered at the plasma membrane in the inactive monomeric form by binding to the idling transporter complex via MalK. Transport of maltose by the transporter complex releases MalT from MalK-MalT interaction, enabling it to form active oligomeric MalT in the presence of the effectors ATP and maltotriose (3).

One possible mechanism by which constitutive Mal61 permease expression could suppress the defective mal13 gene is specific binding of the activator. Under uninduced growth conditions, the permease might sequester the MAL-activator at the plasma membrane and thereby prevent the MAL-activator from entering the nucleus and activating MAL gene transcription. Conformational changes in maltose permease induced by maltose transport might destabilize the putative interaction between maltose permease and the MAL-activator, resulting in MAL-activator release, nuclear entry, and MAL gene activation. Binding of the defective mal13 activator to the constitutive elevated levels of maltose permease might protect it from degradation or other forms of inactivation, thereby suppressing this mutant form of the activator.

Our finding that constitutive expression of PmSUC2 also suppresses the defective mal13 MAL-activator argues against the specific-binding model and supports a second model. In this model, the large bolus of maltose transported into the cell via the abundant maltose permease suppresses the defect of mal13p, suggesting that the defect of the mutant activator protein is in its ability to respond to the maltose signal.

On the basis of these results, we propose that maltose permease serves solely as a maltose transporter. It accumulates intracellular concentrations of maltose to levels sufficient to both induce MAL gene expression and provide an energy source for growth. In cells grown under uninduced conditions, the basal low level of maltose permease expression is sufficient to transport enough maltose into the cell to allow the accumulation of intracellular maltose to levels that are adequate to promote the activation of the MAL activator and induce further structural gene expression. Wykoff and O'Shea (37) similarly concluded that the phosphate transporter Pho84p, which is essential during phosphate starvation, does not serve as the phosphate sensor and suggested that the intracellular phosphate concentration regulates PHO gene expression.

So, what is the maltose sensor? One likely possibility is that the MAL activator itself binds to maltose directly. In work to be reported elsewhere, we have shown that maltase induction is dependent on the Hsp90 molecular chaperone complex (B. Zhang, M. Bali, K. Morano, and C. A. Michels, unpublished data). Several other transcriptional activators are known to associate with Hsp90 (16). Members of the steroid receptor family of activators are released from the chaperone complex following direct binding of the cognate steroid (31). Similarly, Hap1 activator release from the Hsp90 chaperone complex is dependent on heme binding (14, 40). An alternate possibility is that a Gal3-like protein serves as the maltose sensor. We do not favor this hypothesis because, despite extensive genetic analysis, specific maltose-nonfermenting mutants with mutations in genes unlinked to the MAL loci have not been isolated. Currently, we are using genetic approaches to identify the maltose-binding sensor.