AEM
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.
Agricola
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.
Applied and Environmental Microbiology, November 2005, p. 6465-6472, Vol. 71, No. 11
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.11.6465-6472.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Improvement of Galactose Uptake in Saccharomyces cerevisiae through Overexpression of Phosphoglucomutase: Example of Transcript Analysis as a Tool in Inverse Metabolic Engineering

Christoffer Bro,1 Steen Knudsen,2 Birgitte Regenberg,1 Lisbeth Olsson,1 and Jens Nielsen1*

Center for Microbial Biotechnology, BioCentrum-DTU, Building 223, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark,1 Center for Biological Sequence Analysis, Building 208, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark2

Received 4 April 2005/ Accepted 10 June 2005


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Through genome-wide transcript analysis of a reference strain and two recombinant Saccharomyces cerevisiae strains with different rates of galactose uptake, we obtained information about the global transcriptional response to metabolic engineering of the GAL gene regulatory network. One of the recombinant strains overexpressed the gene encoding the transcriptional activator Gal4, and in the other strain the genes encoding Gal80, Gal6, and Mig1, which are negative regulators of the GAL system, were deleted. Even though the galactose uptake rates were significantly different in the three strains, we surprisingly did not find any significant changes in the expression of the genes encoding the enzymes catalyzing the first steps of the pathway (i.e., the genes encoding Gal2, Gal1, Gal7, and Gal10). We did, however, find that PGM2, encoding the major isoenzyme of phosphoglucomutase, was slightly up-regulated in the two recombinant strains with higher galactose uptake rates. This indicated that PGM2 is a target for overexpression in terms of increasing the flux through the Leloir pathway, and through overexpression of PGM2 the galactose uptake rate could be increased by 70% compared to that of the reference strain. Based on our findings, we concluded that phosphoglucomutase plays a key role in controlling the flux through the Leloir pathway, probably due to increased conversion of glucose-1-phosphate to glucose-6-phosphate. This conclusion was supported by measurements of sugar phosphates, which showed that there were increased concentrations of glucose-6-phosphate, galactose-6-phosphate, and fructose-6-phosphate in the strain construct overexpressing PGM2.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In the yeast Saccharomyces cerevisiae the flux through the galactose utilization pathway is threefold lower than the rate of glucose utilization (35). From an industrial point of view galactose utilization is relevant due to the presence of galactose in several industrial media, such as cheese whey (see reference 43 for a review), molasses (40), and lignocellulose (53). Therefore, there is interest in increasing galactose utilization through the use of metabolic engineering (3, 30, 34). Besides the industrial relevance, the gene system involved in galactose metabolism in yeast (the GAL genes) has served as a eukaryotic model system for gene regulation and thus is one of the best-characterized systems of eukaryotic transcriptional control (18, 20). Moreover, the importance of the S. cerevisiae GAL system is further emphasized by its role as a model for the human hereditary disease galactosemia (42), which is caused by a disorder in galactose metabolism.

Galactose utilization in S. cerevisiae occurs through the Leloir pathway (Fig. 1), in which galactose is converted to glucose-1-phosphate via galactose-1-phosphate. Glucose-1-phosphate is then converted to glucose-6-phosphate, which may enter both the Embden-Meyerhof-Parnas pathway and the pentose phosphate pathway. The genes encoding the enzymes of the Leloir pathway are tightly regulated; they are repressed by glucose and induced up to 1,000-fold by galactose (26). When galactose is absent from the medium, Gal80 inhibits the function of the transcriptional activator Gal4, which is required for GAL gene expression. Mig1 mediates glucose repression by repressing expression of GAL1 and GAL4 (14, 19, 29), while Gal6 also exerts negative control on the GAL system, although the exact mechanism remains to be elucidated (54). In a previous study we found that overexpression of GAL4 and deletion of GAL6, GAL80, and MIG1 resulted in galactose uptake rates that were increased 26% and 41%, respectively (33). These findings were believed to result from a general up-regulation of the complete Leloir pathway. In order to study this further and to identify possible global regulatory effects of these mutations, we used DNA arrays to analyze the reference strain and two mutants. In this analysis we were particularly interested in identifying novel targets for further improvement of the flux through the pathway. Analysis of different mutants with the objective of identifying targets for metabolic engineering and subsequently pursuing a target to improve the properties of a strain has been referred to as inverse metabolic engineering (4, 8). This study is an example of this approach, which may be very useful in all cases in which a pedigree of strains with different properties is available.



View larger version (11K):
[in this window]
[in a new window]
  		FIG. 1. Galactose utilization pathway in S. cerevisiae. Extracellular galactose (Galext) is transported into the cell by galactose permease (Gal2) and then converted to glucose-6-phosphate (Glu-6P) by reactions catalyzed by galactokinase (Gal1), galactose-1-phosphate uridylyltransferase (Gal7), UDP-glucose 4-epimerase (Gal10), and phosphoglucomutase (Pgm) encoded by PGM1 and PGM2.

 

   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Yeast strains.
All S. cerevisiae strains used in this study were generated from the prototrophic CEN.PK113-7D reference strain (MATa SUC2 MAL2-8), which is a strain that is often used for physiological studies (49). A GAL4-overexpressing strain containing the 2µ high-copy-number vector pBM959 with GAL4 behind its native promoter and URA3 as a marker and a {Delta}gal6 {Delta}gal80 {Delta}mig1 strain were constructed previously (33). A strain overexpressing PGM2 was constructed by transforming CEN.PK113-5D (MATa SUC2 MAL2-8 ura3-52) with the 2µ high-copy-number vector pPGM2 (25) containing PGM2 behind the PMA1 gene promoter and URA3.

Batch cultivation.
Aerobic batch cultivation was carried out in well-controlled laboratory fermentors with a working volume of 2 or 4 liters. To obtain samples for analyses of RNA, biomass, and extracellular metabolites, a defined medium (51) was used, which contained (per liter) 15 g galactose, 5.0 g (NH4)2SO4, 3.0 g KH2PO4, 0.5 g MgSO4 · 7H2O, and trace metals and vitamins as described previously (51). Separate cultivations were performed to obtain samples for measurement of intracellular metabolites, and the same medium was used except that it contained twice the amounts of galactose, salts, and trace metals. To all cultures 50 µl/liter antifoam 289 (Sigma-Aldrich, St. Louis, MO) was added to avoid foaming. Galactose was autoclaved separately from the mineral medium and then added to the fermentor together with a sterile filtered solution containing the vitamins.

Precultures were inoculated from plate cultures, which were inoculated directly from frozen stocks, and were grown at 30°C on selective media. The cultures were grown at 30°C and 150 rpm in cotton-stoppered, 500-ml Erlenmeyer flasks with baffles containing 100 ml of a medium (pH 6.5) similar to the medium in the fermentors but containing different concentrations of galactose (20 g/liter), (NH4)2SO4 (7.5 g/liter), and KH2PO4 (14 g/liter). Precultures in the exponential phase were used to inoculate the aerobic batch cultures at an initial concentration of 1 mg (dry weight)/liter. The aerobic batch cultures were incubated at 30°C with a stirrer speed of 800 rpm and with air added at a flow rate of 1 liter per liter of starting volume per min. The concentrations of dissolved oxygen in selected cultures were measured with a Mettler Toledo polarographic electrode and remained greater than 70%. The off-gas passed through a cooled condenser to avoid evaporation of ethanol from the bioreactors. The pH was kept at 5.0 by automatic addition of 4 M KOH.

Measurement of biomass and analysis of extracellular metabolites.
The dry weight of biomass was determined as described previously (33). Culture samples used for determination of glucose, ethanol, glycerol, acetate, pyruvate, and succinate concentrations were filtered through a 0.45-µm cellulose acetate filter (Osmonics, Minnetonka, MN) immediately after sampling, and the filtrate was frozen at –20°C until further analysis. The concentrations of the metabolites were determined by high-pressure liquid chromatography as described previously (28).

Sampling and RNA isolation.
Samples for RNA isolation were taken by rapidly transferring 20 ml of culture into a tube with 35 to 40 ml of crushed ice, which decreased the temperature to less than 2°C in less than 10 s. Cells were quickly pelleted by centrifugation (4,500 rpm at 0°C for 2 min), instantly frozen in liquid nitrogen, and then stored at –80°C. Total RNA was extracted by using a FastRNA kit, Red (Bio 101, Inc., Vista, CA) after the samples were thawed on ice.

Probe preparation and hybridization to arrays.
mRNA extraction, cDNA synthesis, cRNA synthesis, and labeling, as well as hybridization to Affymetrix yeast genome S98 arrays (Affymetrix, Santa Clara, CA), were performed essentially as described in the Affymetrix users' manual (1). Washing and staining of arrays were performed using a GeneChip Fluidics Station 400, and the arrays were scanned with an Affymetrix GeneArray scanner.

Data acquisition and analysis.
Acquisition and quantification of array images, as well as primary data analysis, were performed using the Affymetrix software package Microarray Suite, version 4.0.1. All arrays were globally scaled to a target value of 500 using the average signal from all gene features using Microarray Suite, version 4.0.1. The Affymetrix S98 yeast microarrays contain probe sets representing 9,335 distinct transcription features. After exclusion of all probe sets not assigned yORF abbreviations in the Saccharomyces Genome Database (http://genome-www.stanford.edu/Saccharomyces/) and all probe sets representing groups of genes, which were already represented as singletons, 6,072 probe sets remained for analysis. Statistical analyses were performed by using an analysis of variance (ANOVA) for all probe sets representing genes identified as present by the Micrarray Suite software (version 4.0.1) in at least one array. The ANOVA calculated the probability (P) that several conditions all came from the same distribution. (1 – P) gave the probability that expression of a gene was significantly changed in at least one strain (21). A Bonferroni correction for multiple testing was used to adjust the significance level for transcripts whose abundance had changed. When k genes are tested, the critical significance level ({alpha} level) is divided by k returning a new cutoff value ({alpha}/k), which on a global scale gives {alpha} probability of having one false positive (5, 21). Fold changes were calculated by using the average expression levels of each triplicate.

Measurement of enzyme activities.
Cell extracts were produced with the help of a Fastprep FP120 instrument (Savant Instruments, New York) as previously described (28). In vitro enzyme activities of phosphoglucomutase and galactose-1-phosphate uridyl transferase (Gal7) were assayed at 30°C by monitoring the NADPH production in a 1-ml reaction mixture at 340 nm using a spectrophotometer (HP 8353 UV-VIS system with Chemstation software from Hewlett-Packard, Palo Alto, CA) for at least two different concentrations of cell extract. The activity of phosphoglucomutase was measured in a reaction mixture containing 80 mM triethanolamine (pH 7.6), 1.7 mM MgCl2, 20 µM glucose 1,6-diphosphate, 0.9 mM EDTA, 0.2 mM NADP+, and 0.47 U/ml glucose-6-phosphate dehydrogenase, and the reaction was started by adding glucose-1-phosphate to a concentration of 3.5 mM as described previously (6, 13). Gal7 activities were determined as described previously (6, 13) with a reaction mixture consisting of 100 mM glycine-KOH (pH 8.7), 5 mM MgCl2, 5 µM glucose-1,6-diphosphate, 5 mM dithiothreitol, 0.4 mM UDP-glucose, 0.8 mM NADP+, 0.15 U/ml phosphoglucomutase, and 0.20 U/ml glucose-6-phosphate dehydrogenase, and the reaction was started by adding galactose-1-phosphate to a concentration of 0.8 mM. Galactokinase (Gal1) activity was measured by a method modified from the method described by Mizoguchi et al. (27). The reaction mixture contained 0.1 M phosphate buffer (pH 7.5), 3.0 mM MgCl2, 1.8 mM NaF, 1.5 mM ATP, and four different concentrations of cell extract. After 5 min of incubation at 30°C, addition of 2.5 mM galactose started the reaction, and the reaction mixture was incubated for another 5 min and then boiled for 2 min. The amount of galactose consumed was determined by measuring the residual concentration of galactose enzymatically with a Boehringer Mannheim galactose kit (Boehringer Mannheim, Mannheim, Germany) after centrifugation (10 min, 1,000 x g). The protein contents of cell extracts were determined by the method of Lowry et al. (24), using essentially fatty acid-free (≥96%) bovine serum albumin (Sigma-Aldrich, St. Louis, MO) as the standard.

Analysis of intracellular metabolites.
Cell extracts were obtained and clean-up of sugar phosphates (galactose-1-phosphate, galactose-6-phosphate, glucose-6-phosphate, fructose-6-phosphate) was performed by using the method described previously (45). Cell extracts were obtained by using cold methanol as the quenching agent and chloroform as the extraction solvent. In addition, clean-up of sugar phosphates was performed using solid-phase extraction (45). Sugar phosphate contents were determined by anion-exchange chromatography with pulsed amperometric detection as described by Smits et al. (45).


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Genome-wide transcript analysis.
Previously, we determined that overexpression of GAL4 or deletion of the GAL6, GAL80, and MIG1 genes resulted in an increased rate of uptake of galactose (33). Aiming to describe the global transcriptional effects of these modifications and to identify novel targets for further improvement of galactose uptake, we examined the transcript profiles of the two strains and the corresponding reference strain, strain CEN.PK113-7D. Cells were grown under well-controlled conditions in aerobic batch cultures on a minimal medium with an initial galactose concentration of 15 g/liter. A typical fermentation profile of the reference strain is shown in Fig. 2A. All the strains had similar specific growth rates (0.17 ± 0.01 h–1), which were slightly higher than those reported previously (33). In the present study the biomass was analyzed directly, whereas in the previous study (33) samples were frozen before analysis, and this may have affected the biomass concentration measured and thus the determination of the maximum specific growth rate. All strains were cultivated in triplicate, and one whole-genome transcription analysis was performed for each replicate by using cells harvested at a residual galactose concentration of 7.5 ± 2.0 g/liter in the middle of the exponential growth phase.



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 2. Fermentation profiles of aerobic batch cultures grown on galactose. (A) Reference strain. (B) PGM2 strain. The graphs show the concentrations of residual galactose (•), the cell dry weights (DW) ({circ}), the concentrations of ethanol ({blacktriangleup}), the concentrations of glycerol (x), and the concentrations of acetate ({square}) in the cultures at different times.

 
An ANOVA was performed for transcripts of the 5,848 open reading frames (ORFs) in S. cerevisiae that were detectable on at least one of the DNA arrays. This yielded only 25 ORFs that exhibited a significant change in the transcript levels in at least one of the two mutants when we used a global likelihood of getting one false positive (cutoff at P = 1.71 x 10–4). For most of these ORFs the transcript levels were changed only in the GAL4-overexpressing strain (called the GAL4 strain) (Table 1 shows the 30 highest-scoring genes). The genome-wide analysis confirmed that the transcript levels of MIG1, GAL4, GAL6, and GAL80 were altered in the directions that would be expected from the genetic modifications of the strains. However, the limited reduction in the GAL80 transcript level in the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain was surprisingly small and unexpected as the coding region GAL80 was almost completely deleted (33). As the GAL80 transcript was down-regulated with high significance and we felt confident that GAL80 was deleted, we assumed that the relatively high transcript level measured was some kind of background.


View this table:
[in this window]
[in a new window]
  		TABLE 1. The thirty genes with the highest probability of changed transcript levels in at least one of the strains according to an ANOVA test

 
The GAL6 transcript level was reduced in the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain, but it also showed possible twofold up-regulation in the GAL4 strain. According to a Student's t test assuming unequal variances, this twofold up-regulation was significant with a 99.0% probability when we did not correct for multiple testing. In contrast to our expectations, GAL6 was the only gene encoding significantly up-regulated transcripts with a Gal4 binding site in its promoter. The hypothetical ORF YIL057C also had a putative Gal4 binding site in its promoter (17), but the level of transcript was lower in the GAL4 strain. Hence, most of the significantly changed transcript levels were likely to be the result of secondary effects of overexpression of GAL4. Gal4 is a member of the zinc cluster proteins, which is a subfamily of 54 transcriptional regulators in S. cerevisiae (2, 47), and it binds to a upstream activation sequence (UASGAL) that is characterized by the sequence 5'-CGG-N11-CCG-3' (16). Furthermore, it is able to bind a variety of sites having the general formula 5'-(A/C)GG-N10-12-CCG-3' (50). It has previously been reported that overexpression of GAL4, probably due to titration of other transcription factors, can inhibit transcription of several genes (15), and data have suggested that under appropriate circumstances Gal4 may be able to function in place of related family members to activate expression (11). Several of the genes encoding the significantly altered transcripts (MDH2, MAL genes, YGL157W, URA1, HXT10) are or might be regulated by zinc cluster proteins, while ZAP1 and YDR520C are themselves zinc protein cluster members.

HXT10, which encodes a hexose transporter capable of transporting galactose (52), was up-regulated in both mutants, which could explain the increased galactose uptake in the two mutants. However, as HXT10 was expressed at a low level compared to the level of the main galactose transporter gene, GAL2, and since overexpression of GAL2 results in a decreased galactose uptake rate (32), we decided not to pursue this possible target for further increasing galactose utilization. The transcript level of the hypothetical ORF YKL031W was also higher in both mutants. This ORF is predicted to encode a 137-amino-acid protein which binds phosphoinositol 4,5-biphosphate in vitro (55), and it contains two predicted transmembrane segments (39); however, it has no obvious role in carbon metabolism, and therefore, we did not pursue this possible target.

Analysis of gene subset.
The genome-wide transcript analysis of the two mutants did not indicate any clear reason for the mutants' increased galactose uptake rates. Therefore, we focused on the transcription results for the GAL genes, which are the most obvious candidates for key enhancers. Besides GAL4, GAL6, and GAL80, only the PGM2 transcript, encoding the major isoform of phosphoglucomutase, had a high probability of being changed (Table 2). Taking multiple testing of the 10 GAL genes into consideration, there was a 94% probability of a changed PGM2 transcript level in at least one of the two mutant strains. The PGM2 transcript level increased 1.7- and 1.3-fold in the GAL4 strain and the {Delta}gal6 {Delta}gal80 {Delta}mig1 mutant, respectively. In order to verify that the concentration of active phosphoglucomutase indeed increased and that the concentrations of other Gal enzymes did not increase, in vitro enzyme activities of Gal1, Gal7, and phosphoglucomutase from the same cultures were measured. The results supported the conclusion drawn from the transcript analysis (Table 2); i.e., only phosphoglucomutase activity (encoded by both PGM1 and PGM2) was consistently increased in the two mutants. As Pgm2 was the isoform assumed to have higher activities in the mutants, we could use Kacser's theory of large deviations (44) to calculate the flux control coefficient for phosphoglucomutase. The flux control coefficient described the relative contribution of the enzyme to the control of flux in the pathway, using the data from the two mutants. We found that this enzyme has a flux control coefficient of 0.47 to 0.87 in the reference strain (0.47 was found by using the data for the GAL4 strain, and 0.87 was found by using the data for the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain). Even though these calculations are relatively uncertain, they indicate that phosphoglucomutase activity exerted substantial control on the flux through the Leloir pathway in the reference strain.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Transcript analysis of the GAL genes and analysis of in vitro enzyme activities

 
Overexpression of PGM2.
On the basis of the transcript analysis, PGM2 was found to be up-regulated in the mutants with a higher flux through the Leloir pathway. To evaluate whether Pgm2 could be a target for improving the flux through the galactose utilization pathway, a strain overexpressing PGM2 (called the PGM2 strain) was constructed. This strain was cultivated in triplicate under the same conditions that were used for the other strains. The PGM2 strain had a significantly increased maximum specific galactose uptake rate compared to the rates for both the reference strain and the two other mutants (Fig. 3). Overexpression of PGM2 resulted in a 70% increase in the maximum specific galactose uptake rate and a threefold-higher maximum specific ethanol production rate compared to the rates for the reference strain. Detailed physiological data for all the strains analyzed are shown in Table 3, and a fermentation profile for the PGM2 strain is shown in Fig. 2B. The maximum specific growth rate on galactose of the PGM2 strain was 0.23 ± 0.02 h–1, compared to 0.17 ± 0.01 h–1 for the reference strain. Due to the increase in both the specific galactose uptake rate and the specific growth rate, the volumetric galactose uptake was much greater in the PGM2 strain, and the exponential growth phase was 30% shorter than that of the reference strain. In vitro enzyme activity measurements verified that the concentration of active phosphoglucomutase indeed increased in the PGM2 strain, which exhibited 17-fold-higher activity. However, the in vitro enzyme activity of Gal7 also increased approximately threefold in this strain, while the activity of Gal1 was unchanged.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 3. Maximum specific ethanol production rates and maximum specific galactose uptake rates determined in aerobic batch cultures grown on galactose for the reference strain, the GAL4 strain, the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain, and the PGM2 strain. Each point is the average of three replicates, and the error bars indicate standard deviations for the replicates. DW, dry weight; EtOH, ethanol.

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Key physiological data for the strains evaluated in this study during exponential growth

 
Intracellular hexose-6-phosphates.
To clarify the mechanisms which cause the effects of overexpression of PGM2, we measured the intracellular levels of glucose-6-phosphate, galactose-1-phosphate, galactose-6-phosphate, and fructose-6-phosphate. In previous studies we consistently found that galactose-6-phosphate was present in yeast cells grown on galactose (unpublished data), and we assume that this metabolite is formed by the action of phosphoglucomutase on galactose-1-phosphate, as phosphoglucomutase previously has been reported to catalyze this reaction in vitro, although at a low rate (23, 38). Cultivation with twice the amounts of galactose, salts, and trace metals compared to the previous cultivation was performed in order to obtain a higher concentration of biomass, and samples were collected from the exponential growth phase to measure the levels of three intracellular metabolites downstream of phosphoglucomutase in the metabolism. The results showed that there were increased intracellular concentrations of glucose-6-phosphate, fructose-6-phosphate, and galactose-6-phosphate in the strain overexpressing PGM2, while the data for galactose-1-phosphate were inconclusive due to high standard deviations for some of the strains (Table 4). This clearly indicates that there was increased phosphoglucomutase activity, and the increased levels of glucose-6-phosphate and fructose-6-phosphate may play an important role in ensuring a higher glycolytic flux in the PGM2 strain.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Intracellular concentrations of hexose phosphates during exponential growth

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In the current study we found that overexpression of phosphoglucomutase results in increased flux through the Leloir pathway. This finding is surprising as this enzyme has generally not been considered to be an enzyme that exerts any control over the flux through the galactose utilization pathway, since the basal level of expression of phosphoglucomutase is quite high and expression increased only three- to fourfold in response to the presence of galactose (31). Furthermore, phosphoglucomutase has high affinities for its substrates (12), which often indicates that an enzyme has low flux control. Moreover, increased expression of only one enzyme-encoding gene is often insufficient to increase the flux of an entire pathway that the enzyme is part of, as illustrated by the efforts to increase the flux through glycolysis (41, 46). The positive effect of overexpression of PGM2 found here, however, may well be due to a combination of a direct effect of a higher Pgm2 activity and secondary effects, such as increased activity of Gal7 or a decreased level of a toxic intermediate. The slow growth on galactose of the reference strain indicates a limitation or the presence of a toxic intermediate, such as galactose-1-phosphate, an effect that may be reduced in the PGM2 strain, as Gal7 and possibly also phosphoglucomutase use galactose-1-phosphate as a substrate.

PGM2 is induced by different stress conditions via the four stress response elements present in its promoter (10, 48), illustrating its dual role in metabolism. Its many functions in galactose metabolism, synthesis of UDP-glucose and glycogen, and trehalose metabolism explain the high basal level of expression of PGM2 in the absence of galactose. However, PGM2 also contains one putative Gal4-binding site in its promoter (17), suggesting that PGM2 is transcriptionally regulated by Gal4, resulting in the three- to fivefold up-regulation of PGM2 upon galactose induction (31). Therefore, the higher levels of PGM2 transcript identified in the GAL4 strain and the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain were not in themselves surprising, but what was surprising were the unchanged transcript levels of GAL1, GAL2, GAL7, and GAL10. Apparently, these genes were expressed at their optimal levels in the reference strain, while PGM2 still had room for increased expression, presumably due to its lesser dependence on Gal4 for its regulation. The transcript level of PGM2 and in vitro phosphoglucomutase activity were higher in the strain with GAL4 overexpression than in the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain, whereas the specific galactose uptake rate was higher in the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain. This shows that increased Pgm2 activity is not the sole cause for increased galactose uptake rates; other factors must be important as well. One such candidate is Gal6, which affects galactose metabolism negatively (33). The transcript level of GAL6 increased twofold in the strain with GAL4 overexpression, and this might have had an unidentified negative effect that counteracted the higher transcript level of PGM2 compared to that in the {Delta}gal6 {Delta}gal80 {Delta}mig1 strain. Deletion of GAL6 alone results in a 24% increase in the galactose uptake rate (33) and has also been reported to result in increased amounts of GAL1, GAL2, and GAL7 mRNA (54). In agreement with the previously observation of Ideker and coworkers (17), we did not observe any effect on the amounts of GAL1, GAL2, or GAL7 mRNA due to deletion of GAL6 during growth on galactose. This may indicate that Gal6 has a function different from an effect on the concentration of GAL gene mRNA (e.g., it may affect translation).

Overexpression of PGM2 resulted in increased conversion of glucose-1-phosphate to glucose-6-phosphate, thereby enabling more efficient shunting of sugar from the Leloir pathway into glycolysis. This may be seen in the higher intracellular concentrations of glucose-6-phosphate, fructose-6-phosphate, and maybe also galactose-6-phosphate in the strain overexpressing PGM2. The increased intracellular levels of glucose-6-phosphate and fructose-6-phosphate may ensure that the flux through glycolysis could be increased, and this may have a positive effect on galactose utilization through more efficient regeneration of ATP required for phosphorylation of galactose. Galactose-6-phosphate can be formed from galactose-1-phosphate in a reaction catalyzed by phosphoglucomutase (23, 38). Thus, the increased level of galactose-6-phosphate in the PGM2-overexpressing strain may also reflect increased activity of Pgm2.

Besides providing further insight into galactose metabolism, the present study is an example of how genome-wide transcript analysis can be used in an inverse metabolic engineering strategy. DNA arrays are very efficient for genome-wide screening of variations in transcript levels, but as illustrated here, changes in fluxes may be associated with very small changes in gene transcripts, and it is inherently difficult to identify genes that have small changes in transcription. Thus, even for carefully replicated cultures more false positives are observed with changes of less than twofold than with greater changes when a statistical analysis is performed (37). Changes less than twofold are often not considered significant in transcript analyses (7, 18, 22, 37) as a result of the increasing false discovery rate with decreasing fold changes. Some approaches to overcome this problem are increasing the number of replicates and applying the strategy described here of zooming in on specific genes and also analyzing several strains with different degrees of the desired phenotype, thereby increasing the statistical confidence in possible gene targets with low, but significant, fold changes. This study shows that it is advantageous to analyze two or more strains with improved properties in order to obtain a list of candidate genes for which there are similar trends in the transcript profiles. However, the transcript level of PGM2 did not appear to be significantly changed from the genome-wide analysis, and it changed only when a relevant subset of genes was tested. Increasing the number of replicates would have increased the likelihood of finding PGM2 in a whole-genome test as the number of false negatives drops substantially with more replicates (21), but with the approach of analyzing a gene subset more replicates were unnecessary. A key factor for success of this strategy is, however, the correct choice of genes in the analysis. The genes may be chosen either based on physiological knowledge, though the use of metabolic information, as demonstrated recently (9), or through the use of model-guided analysis of transcript data (36).


   ACKNOWLEDGMENTS
 
We thank Lene Christiansen for her help with transcript analysis and Wian de Jongh for his work on measuring enzyme activities and intracellular metabolites. Plasmid pPGM2 was a kind gift from Monica Montero-Lomeli, Universidade Federal Do Rio Janeiro, Brazil.

This work was financed in part by the Danish Biotechnology Instrument Center (DABIC).


   FOOTNOTES
 
* Corresponding author. Mailing address: Center for Microbial Biotechnology, BioCentrum-DTU, Building 223, Technical University of Denmark, DK-2800 Kgs Lyngby, Denmark. Phone: 45 45252696. Fax: 45 45884148. E-mail: jn{at}biocentrum.dtu.dk. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Affymetrix. 2000. Affymetrix GeneChip expression analysis technical manual. Affymetrix, Santa Clara, CA.
  2. Akache, B., K. Wu, and B. Turcotte. 2001. Phenotypic analysis of genes encoding yeast zinc cluster proteins. Nucleic Acids Res. 29:2181-2190.[Abstract/Free Full Text]
  3. Bailey, J. E. 1991. Toward a science of metabolic engineering. Science 252:1668-1675.[Abstract/Free Full Text]
  4. Bailey, J. E., A. Shurlati, V. Hatzimanikatis, K. Lee, W. A. Renner, and P. S. Tsai. 1996. Inverse metabolic engineering: a strategy for directed genetic engineering of useful phenotypes. Biotechnol. Bioeng. 52:109-121.
  5. Bender, R., and S. Lange. 2001. Adjusting for multiple testing—when and how? J. Clin. Epidemiol. 54:343-349.[CrossRef][Medline]
  6. Bergmeyer, H. U., J. Bergmeyer, and M. Grassl. 1983. Methods of enzymatic analysis, vol. II. Samples, reagents, assessment of results. Verlag Chemie, Weinheim, Germany.
  7. Boer, V. M., J. H. De Winde, J. T. Pronk, and M. D. Piper. 2003. The genome-wide transcriptional responses of Saccharomyces cerevisiae grown on glucose in aerobic chemostat cultures limited for carbon, nitrogen, phosphorus, or sulfur. J. Biol. Chem. 278:3265-3274.[Abstract/Free Full Text]
  8. Bro, C., and J. Nielsen. 2004. Impact of ‘ome’ analyses on inverse metabolic engineering. Metab. Eng. 6:204-211.[CrossRef][Medline]
  9. Bro, C., B. Regenberg, and J. Nielsen. 2004. Genome-wide transcriptional response of a Saccharomyces cerevisiae strain with an altered redox metabolism. Biotechnol. Bioeng. 85:269-276.[CrossRef][Medline]
  10. Burchett, S. A., P. Flanary, C. Aston, L. Jiang, K. H. Young, P. Uetz, S. Fields, and H. G. Dohlman. 2002. Regulation of stress response signaling by the N-terminal dishevelled/EGL-10/pleckstrin domain of Sst2, a regulator of G protein signaling in Saccharomyces cerevisiae. J. Biol. Chem. 277:22156-22167.[Abstract/Free Full Text]
  11. D'Alessio, M., and M. C. Brandriss. 2000. Cross-pathway regulation in Saccharomyces cerevisiae: activation of the proline utilization pathway by Ga14p in vivo. J. Bacteriol. 182:3748-3753.[Abstract/Free Full Text]
  12. Daugherty, J. P., W. F. Kraemer, and J. G. Joshi. 1975. Purification and properties of phosphoglucomutase from Fleischmann's yeast. Eur. J. Biochem. 57:115-126.[Medline]
  13. Elsevier, J. P., L. Wells, B. B. Quimby, and J. L. Fridovich-Keil. 1996. Heterodimer formation and activity in the human enzyme galactose-1-phosphate uridylyltransferase. Proc. Natl. Acad. Sci. USA 93:7166-7171.[Abstract/Free Full Text]
  14. Frolova, E., M. Johnston, and J. Majors. 1999. Binding of the glucose-dependent Mig1p repressor to the GAL1 and GAL4 promoters in vivo: regulation by glucose and chromatin structure. Nucleic Acids Res. 27:1350-1358.[Abstract/Free Full Text]
  15. Gill, G., and M. Ptashne. 1988. Negative effect of the transcriptional activator GAL4. Nature 334:721-724.[CrossRef][Medline]
  16. Giniger, E., S. M. Varnum, and M. Ptashne. 1985. Specific DNA binding of GAL4, a positive regulatory protein of yeast. Cell 40:767-774.[CrossRef][Medline]
  17. Ideker, T., V. Thorsson, J. A. Ranish, R. Christmas, J. Buhler, J. K. Eng, R. Bumgarner, D. R. Goodlett, R. Aebersold, and L. Hood. 2001. Integrated genomic and proteomic analyses of a systematically perturbed metabolic network. Science 292:929-934.[Abstract/Free Full Text]
  18. Johnston, M., and M. Carlson. 1992. Regulation of carbon and phosphate utilization, p. 193-281. In E. W. Jones, J. R. Pringel, and J. Broach (ed.), The molecular and cellular biology of the yeast Saccharomyces: gene expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
  19. Johnston, M., J. S. Flick, and T. Pexton. 1994. Multiple mechanisms provide rapid and stringent glucose repression of GAL gene expression in Saccharomyces cerevisiae. Mol. Cell. Biol. 14:3834-3841.[Abstract/Free Full Text]
  20. Klein, C. J., L. Olsson, and J. Nielsen. 1998. Glucose control in Saccharomyces cerevisiae: the role of Mig1 in metabolic functions. Microbiology 144:13-24.[Medline]
  21. Knudsen, S. 2002. A biologist guide to analysis of DNA microarray data. John Wiley & Sons, New York, N.Y.
  22. Lascaris, R., H. J. Bussemaker, A. Boorsma, M. Piper, S. H. van der, L. Grivell, and J. Blom. 2003. Hap4p overexpression in glucose-grown Saccharomyces cerevisiae induces cells to enter a novel metabolic state. Genome Biol. 4:R3.[CrossRef][Medline]
  23. Lowry, O. H., and J. V. Passonneau. 1969. Phosphoglucomutase kinetics with the phosphates of fructose, glucose, mannose, ribose, and galactose. J. Biol. Chem. 244:910-916.
  24. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.[Free Full Text]
  25. Masuda, C. A., M. A. Xavier, K. A. Mattos, A. Galina, and M. Montero-Lomeli. 2001. Phosphoglucomutase is an in vivo lithium target in yeast. J. Biol. Chem. 276:37794-37801.[Abstract/Free Full Text]
  26. Melcher, K. 1997. Galactose metabolism in Saccharomyces cerevisiae: a paradigm for eukaryotic gene regulation, p. 235-269. In F. K. Zimmermann and K. D. Entian (ed.), Yeast sugar metabolism: biochemistry, genetics, biotechnology, and applications. Technomic Publishing Company, Inc., Lancaster, Pa.
  27. Mizoguchi, N., T. Eguchi, N. Sakura, and K. Ueda. 1993. Erythrocyte galactokinase assay with high performance liquid chromatography. Clin. Chim. Acta 216:145-151.[CrossRef][Medline]
  28. Møller, K., C. Bro, J. Piskur, J. Nielsen, and L. Olsson. 2002. Steady-state and transient-state analysis of aerobic fermentation in Saccharomyces kluyveri. FEMS Yeast Res. 2:233-244.[Medline]
  29. Nehlin, J. O., M. Carlberg, and H. Ronne. 1991. Control of yeast GAL genes by MIG1 repressor: a transcriptional cascade in the glucose response. EMBO J. 10:3373-3377.[Medline]
  30. Nielsen, J. 2001. Metabolic engineering. Appl. Microbiol. Biotechnol. 55:263-283.[CrossRef][Medline]
  31. Oh, D., and J. E. Hopper. 1990. Transcription of a yeast phosphoglucomutase isozyme gene is galactose inducible and glucose repressible. Mol. Cell. Biol. 10:1415-1422.[Abstract/Free Full Text]
  32. Ostergaard, S. 2000. Metabolic engineering of the galactose metabolism in Saccharomyces cerevisiae. Ph.D. thesis. Technical University of Denmark, Kgs Lyngby, Denmark.
  33. Ostergaard, S., L. Olsson, M. Johnston, and J. Nielsen. 2000. Increasing galactose consumption by Saccharomyces cerevisiae through metabolic engineering of the GAL gene regulatory network. Nat. Biotechnol. 18:1283-1286.[CrossRef][Medline]
  34. Ostergaard, S., L. Olsson, and J. Nielsen. 2000. Metabolic engineering of Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 64:34-50.[Abstract/Free Full Text]
  35. Ostergaard, S., C. Roca, B. Rønnow, J. Nielsen, and L. Olsson. 2000. Physiological studies in aerobic batch cultivations of Saccharomyces cerevisiae strains harboring the MEL1 gene. Biotechnol. Bioeng. 68:252-259.[CrossRef][Medline]
  36. Patil, K. R., and J. Nielsen. 2005. Uncovering transcriptional regulation of metabolism by using metabolic network topology. Proc. Natl. Acad. Sci. USA 102:2685-2689.[Abstract/Free Full Text]
  37. Piper, M. D., P. Daran-Lapujade, C. Bro, B. Regenberg, S. Knudsen, J. Nielsen, and J. T. Pronk. 2002. Reproducibility of oligonucleotide microarray transcriptome analyses. An interlaboratory comparison using chemostat cultures of Saccharomyces cerevisiae. J. Biol. Chem. 277:37001-37008.[Abstract/Free Full Text]
  38. Posternak, T., and J. P. Rosselet. 1954. Action de la phosphoglucomutase du muscle sur des acides aldose-1-phosphoriques. Transformation de l'acide galactose-1-phosphorique. Helv. Chim. Acta 37:246-250.[CrossRef]
  39. Purnelle, B., J. Skala, L. van Dyck, and A. Goffeau. 1994. Analysis of an 11.7 kb DNA fragment of chromosome XI reveals a new tRNA gene and four new open reading frames including a leucine zipper protein and a homologue to the yeast mitochondrial regulator ABF2. Yeast 10:125-130.[CrossRef][Medline]
  40. Rosen, K. 1987. Production of baker's yeast, p. 471-500. In D. R. Berry, I. Russell, and G. G. Stewart (ed.), Yeast biotechnology, part V. Allen & Unwin, London, United Kingdom.
  41. Schaaff, I., J. Heinisch, and F. K. Zimmermann. 1989. Overproduction of glycolytic enzymes in yeast. Yeast 5:285-290.[CrossRef][Medline]
  42. Segal, S. 1995. Disorders of galactose metabolism, p. 967-1000. In C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (ed.), The metabolic and molecular basis of inherited disease. McGraw-Hill, New York, N.Y.
  43. Siso, M. I. G. 1996. The biotechnological utilization of cheese whey: a review. Bioresour. Technol. 57:1-11.
  44. Small, J. R., and H. Kacser. 1993. Responses of metabolic systems to large changes in enzyme activities and effectors. 1. The linear treatment of unbranched chains. Eur. J. Biochem. 213:613-624.[Medline]
  45. Smits, H. P., A. Cohen, T. Buttler, J. Nielsen, and L. Olsson. 1998. Cleanup and analysis of sugar phosphates in biological extracts by using solid-phase extraction and anion-exchange chromatography with pulsed amperometric detection. Anal. Biochem. 261:36-42.[CrossRef][Medline]
  46. Smits, H. P., J. Hauf, S. Müller, T. J. Hobley, F. K. Zimmermann, B. Hahn-Hägerdal, J. Nielsen, and L. Olsson. 2000. Simultaneous overexpression of enzymes of the lower part of glycolysis can enhance the fermentative capacity of Saccharomyces cerevisiae. Yeast 16:1325-1334.[CrossRef][Medline]
  47. Todd, R. B., and A. Andrianopoulos. 1997. Evolution of a fungal regulatory gene family: the Zn(II)2Cys6 binuclear cluster DNA binding motif. Fungal Genet. Biol. 21:388-405.[CrossRef][Medline]
  48. Treger, J. M., A. P. Schmitt, J. R. Simon, and K. McEntee. 1998. Transcriptional factor mutations reveal regulatory complexities of heat shock and newly identified stress genes in Saccharomyces cerevisiae. J. Biol. Chem. 273:26875-26879.[Abstract/Free Full Text]
  49. van Dijken, J. P., J. Bauer, L. Brambilla, P. Duboc, J. M. Francois, C. Gancedo, M. L. Giuseppin, J. J. Heijnen, M. Hoare, H. C. Lange, E. A. Madden, P. Niederberger, J. Nielsen, J. L. Parrou, T. Petit, D. Porro, M. Reuss, N. van Riel, M. Rizzi, H. Y. Steensma, C. T. Verrips, J. Vindelov, and J. T. Pronk. 2000. An interlaboratory comparison of physiological and genetic properties of four Saccharomyces cerevisiae strains. Enzyme Microb. Technol. 26:706-714.[CrossRef][Medline]
  50. Vashee, S., H. Xu, S. A. Johnston, and T. Kodadek. 1993. How do "Zn2 cys6" proteins distinguish between similar upstream activation sites? Comparison of the DNA-binding specificity of the GAL4 protein in vitro and in vivo. J. Biol. Chem. 268:24699-24706.[Abstract/Free Full Text]
  51. Verduyn, C., E. Postma, W. A. Scheffers, and J. P. van Dijken. 1992. Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation. Yeast 8:501-517.[CrossRef][Medline]
  52. Wieczorke, R., S. Krampe, T. Weierstall, K. Freidel, C. P. Hollenberg, and E. Boles. 1999. Concurrent knock-out of at least 20 transporter genes is required to block uptake of hexoses in Saccharomyces cerevisiae. FEBS Lett. 464:123-128.[CrossRef][Medline]
  53. Wiselogel, A., S. Tyson, and D. Johnson. 1996. Biomass feedstock resources and composition, p. 105-118. In C. E. Wyman (ed.), Handbook on bioethanol: production and utilization. Taylor & Francis, Washington, D.C.
  54. Zheng, W., H. E. Xu, and S. A. Johnston. 1997. The cysteine-peptidase bleomycin hydrolase is a member of the galactose regulon in yeast. J. Biol. Chem. 272:30350-30355.[Abstract/Free Full Text]
  55. Zhu, H., M. Bilgin, R. Bangham, D. Hall, A. Casamayor, P. Bertone, N. Lan, R. Jansen, S. Bidlingmaier, T. Houfek, T. Mitchell, P. Miller, R. A. Dean, M. Gerstein, and M. Snyder. 2001. Global analysis of protein activities using proteome chips. Science 293:2101-2105.[Abstract/Free Full Text]

Applied and Environmental Microbiology, November 2005, p. 6465-6472, Vol. 71, No. 11
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.11.6465-6472.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.
Agricola
Right arrow Articles by Bro, C.
Right arrow Articles by Nielsen, J.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
J. Bacteriol. Microbiol. Mol. Biol. Rev. Eukaryot. Cell All ASM Journals