In silico and in vivo analysis reveal a novel gene in Saccharomyces cerevisiae trehalose metabolism

Background The ability to respond rapidly to fluctuations in environmental changes is decisive for cell survival. Under these conditions trehalose has an essential protective function and its concentration increases in response to enhanced expression of trehalose synthase genes, TPS1, TPS2, TPS3 and TSL1. Intriguingly, the NTH1 gene, which encodes neutral trehalase, is highly expressed at the same time. We have previously shown that trehalase remains in its inactive non-phosphorylated form by the action of an endogenous inhibitor. Recently, a comprehensive two-hybrid analysis revealed a 41-kDa protein encoded by the YLR270w ORF, which interacts with NTH1p. Results In this work we investigate the correlation of this Trehalase Associated Protein, in trehalase activity regulation. The neutral trehalase activity in the ylr270w mutant strain was about 4-fold higher than in the control strain. After in vitro activation by PKA the ylr270w mutant total trehalase activity increased 3-fold when compared to a control strain. The expression of the NTH1 gene promoter fused to the heterologous reporter lacZ gene was evaluated. The mutant strain lacking YLR270w exhibited a 2-fold increase in the NTH1-lacZ basal expression when compared to the wild type strain. Conclusions These results strongly indicate a central role for Ylr270p in inhibiting trehalase activity, as well as in the regulation of its expression preventing a wasteful futile cycle of synthesis-degradation of trehalose.


Background
In the yeast Saccharomyces cerevisiae, cytosolic trehalose is mobilized by hydrolysis to glucose catalyzed by neutral trehalase (EC 3.2.1.28), encoded by the NTH1 gene. NTH1p was shown to be post-translationally regulated by two different mechanisms: i) phosphorylation by cAMP dependent protein kinase (PKA) that activates the enzyme [1,2], ii) by an inhibitory protein [3]. Trehalase activity varies during growth on glucose as the result of the phosphorylation state of the enzyme [4]. At the onset of diauxie, activity undergoes a drastic decrease [4], in contrast to the high level of its mRNA [7] and the constant cAMP concentration in this phase [8]. Activation by PKA reaches a maximum value at the onset of the transition phase of growth on glucose, from fermentative to oxidative metabolism. At this point, almost 80% of the enzyme is in a cryptic non-phosphorylated form [4]. We have previously shown that at this point trehalase remains in its inactive non-phosphorylated form by the action of an endogenous inhibitory protein [3]. Recently, a comprehensive two-hybrid analysis revealed a 41 KDa NTH1p binding protein encoded by YLR270w ORF with unknown function [5,6]. In order to clarify the mechanisms of trehalase down-regulation, the involvement of Ylr270p on trehalase modulation was studied. Here we have evaluated trehalase activation by PKA, as well as the expression of the NTH1 gene promoter fused to the heterologous reporter lacZ gene in ylr270w background.
Our in silico analysis of the 5' upstream regions of the YLR270w, TPS1 and NTH1 genes revealed similar elements in their promoters. These data taken together with those found by an extensive data-mining over genomic micro-array databases point toward a coordinated expression of these genes.

Results and discussion
The first property observed for the Ylr270p was its ability to bind to trehalase therefore, we renamed it as Trehalase Associated Protein, Tap [9]. Later, Liu et al. [10] characterized the Ylr270p also as an enzyme with mRNA decapping activity, and introduced another name, yDcpS or DCS1.
DCS1p involvement in trehalase activity regulation was investigated by measuring basal and PKA-activated trehalase activities in both DCS1 and dcs1 cells grown on glucose and harvested at the transition phase leading to diauxie ( Figure 1).
During the onset of diauxie (2.0-2.6 mg dry weight/ml) cells lacking the DCS1 gene showed higher levels of trehalase activity than the control cells. This increase was due both to a higher basal activity in dcs1 cells and to a higher total trehalase activity revealed after in vitro activation by PKA. In the control strain, basal trehalase activity corresponded to about 20% of total trehalase, which is in agreement with the results reported for other wild type strains [3,4,9]. However, we found that in the dcs1 mutant this relation was about 50%. This fifty-fifty ratio between cryptic trehalase and basal trehalase is typically found at the exponential or stationary phase of growth when the trehalase inhibitory protein is reported as absent [3].
The trehalase inhibitory protein shows to be a Ca +2 /Calmodulin ligand [3] and possibly a substrate for the Ca +2 / Calmodulin protein kinase (CaM Kinase II) isozymes encoded by the CMK1 and CMK2 genes [9]. It has been proposed that this inhibitory protein acts as a mediator between the Ca +2 signal and trehalase activation by PKA. Indeed, when trehalase activity was measured in the presence of Ca +2 ions in crude extracts from dcs1 mutants no activation occurred in contrast to the 2-fold activation seen in the control strain (unpublished results).
These results suggest that the product of the YLR270w ORF, Dcs1p, is involved in trehalase down regulation and could be the inhibitory protein reported by De Mesquita et al (1997). These findings are supported by the observation that the maximum expression of DCS1 mRNA coincides with the lower levels of basal trehalase [3,7]. DCS1 Basal, cryptic and total trehalase activities for the YLR270w (A) and ylr270w (B) strains Figure 1 Basal, cryptic and total trehalase activities for the YLR270w (A) and ylr270w (B) strains. Cells were grown in YPD medium. Open bars represent basal trehalase activity; dashed bars, cryptic trehalase activity and filled bars, total trehalase activity (in vitro cAMP/ATP activated) determined in cells harvested at the indicated points.
mRNA expression levels were shown to be enhanced suddenly and transiently at the onset of diauxie [7], a condition where the trehalase inhibitory activity was found [3].
To determine whether the elevated total trehalase activity found in the dcs1 mutant corresponded to a change in NTH1 expression, we evaluated NTH1 expression by the activity of the lacZ reporter gene fused to 600 bp of NTH1 promoter region [10].
As shown in table 1, the expression of NTH1-lacZ fusion gene was 2.5-fold higher in dcs1 deleted than in control cells. These results raise the question whether Dcs1p could also be implicated, even in an indirect manner, in the regulation of NTH1 transcription. Indeed, this should be the case since, recently, a homologous Dcs1 in Schizosaccharomyces pombe, was predominantly found in the nucleus albeit characterized as an mRNA binding protein capable of acting as a relatively poor translation inhibitor in this organism [12].
Over the years, an intriguing paradox remained unexplained: the concomitant trehalose accumulation and the enhanced trehalase expression under stress conditions. In fact TPS1, involved in trehalose biosynthesis, NTH1 and DCS1 exhibit a coordinated induction at diauxic growth [7].
These observations led us to investigate the presence of similar regulatory elements in the promoters of the NTH1, TPS1 and DCS1 genes in order to ascertain whether the trehalose/trehalase modulation by Dcs1p transcends the post-diauxic stress.
The results in table 2 show the similar elements shared in NTH1, TPS1, and DCS1 promoters. All the putative elements found are involved in stress responses, caused by heat (STRE and HSTE), amino acid starvation (GCN4), and glicolysis/gluconeogenesis regulation (GCR1). The conserved motifs in these genes could be recognized by common factors and could mean a common regulatory program sharing similar expression profiles.
The approach to investigate the transcription profiles and characterize the DCS1 gene in order to understand its role in the yeast metabolic scenery, was to mine microarrays databases. In these databases we searched for a change in the YLR270w ORF expression using a two-fold variation cutoff. Microarray experiments provide more information than what the authors can interpret alone. This "orphan" information offers clues about gene functions. The YLR270w ORF remained, until now, with an unknown function, and its expression data was never discussed.
To analyze the transcription profiles of DCS1, NTH1, and TPS1 genes under different experimental conditions we used the Yeast Microarray Global Viewer (yMGV) [18], which allows direct comparison of the results from 1347 conditions from 75 different publications.
Heat stress at 37°C/20 minutes [19] led to a 6-fold increase in DCS1 expression in wild type cells in contrast to a 2.0-fold increase in a msn2/msn4 mutant under the same conditions. This result suggests that the STRE element found in the DCS1 promoter region could be functional and mediated by Msn2 and Msn4 transcription factors. NTH1 and TPS1 genes showed similar profiles. These three genes behave similarly also in other stress conditions [20] showing to be co-regulated during changes in oxidation, pH, and osmolarity (figure 2B).
Starvations for amino acids, purines, as well as glucose limitation induce the synthesis of the Gcn4 transcription activator of amino acids biosynthetic genes, in multiple pathways. Computational searches of the yeast genome reveal that the Gcn4 binding site is present at the promoters of numerous genes not directly connected with amino acid biosynthesis [21].
Under nitrogen depletion [19], a significant increase in the expression of DCS1, NTH1, and TPS1 genes can be observed after 8 h of nitrogen starvation. Corroborating these results, it has been demonstrated that these genes are induced up to 3.5-fold in wild type cells in response to starvation for histidine by treatment with 3-aminotriazole (3AT), a competitive inhibitor of His3p [21]. Therefore, when a gcn4 mutant strain is treated with 3AT the expression profiles of DCS1, NTH1, and TPS1 genes show only an increase of about 1.5 fold. These results reinforce the idea that the GCN4 elements found several times in GCS1, NTH1, and TPS1 promoters (see table 2) could be involved in nitrogen limitation stress response in those genes. In addition, the treatment with the alkylating agent, methyl methane sulfonate (MMS), that induces GCN4 translation [21] also induces DCS1, NTH1 and TPS1 expression [21].
Up to this day neutral trehalase has not been shown to be regulated by classical catabolite repression or derepression factors. However, NTH1 expression increases around 4.0-fold in gcr1 mutant strains after glucose addition to cells grown on lactate and glycerol. On the other hand, in the GCR1 control strain a 0.3-fold decrease was found [22]. The same occurs with DCS1 and TPS1 genes [22]. These results bring support for the GCR1 putative elements found in DCS1, NTH1, and TPS1 promoters (table  2).
The yeast gcr1 mutant grows on non-fermentative carbon sources at the same rate as wild type strains, however it exhibits a severe growth defect when grown in the presence of glucose, even when non-fermentable carbon sources are available [22]. The role of Gcr1 as an activator is best understood in relation to genes coding for glycolytic enzymes. Gcr1 is recruited to the UAS elements by Rap1 in order to activate its transcription. No changes were observed in DCS1, NTH1, and TPS1 expression in rap1 mutant [23], what indicates that the Gcr1 factor regulation for these genes does not depend on Rap1. Therefore, we might speculate that, in the presence of glucose, Gcr1 activates glycolytic genes with Rap1 and represses trehalose metabolism when either alone or complexed with another factor.
These results point toward trehalase as an alternative gluconeogenic enzyme forming glucose after stress conditions, when trehalose accumulated to protect the cell functions should be consumed as a carbon source in order to allow yeast cells to rapidly resume growth under favorable conditions.

Conclusions
Our results strongly indicate a central role for Dcs1p in inhibiting trehalase activation and preventing a wasteful futile cycle of synthesis and degradation of trehalose. These conclusions provide a clear explanation for the above-mentioned paradox and broaden our understanding of metabolic fluxes in yeast cells.
Under stress conditions trehalose is accumulated in order to provide protection for membranes and proteins. During recovery from stress, trehalase is activated. On the other hand, should trehalose be required as a carbon source, activation of trehalase leads it through the gluconeogenic pathway.
Escherichia coli DH5α was used for plasmid manipulations.

Growth conditions
Yeast cells were grown either in a rich medium (YP) containing 1% yeast extract, 2% peptone or in a mineral medium (YNB, Difco yeast nitrogen base, with the necessary supplements); 2% glucose was used as carbon source. The pNL1 plasmid, a gift from C. Gancedo, was constructed inserting the NTH1 promoter in frame to the lacZ gene in the Yep353 [11].

Enzymatic activity measurements
Yeast extracts were prepared by shaking with glass beads, and protein was determined by the standard method of Stickland [24].

Trehalase activity determination
The assay of basal neutral trehalase was performed in 50 mM maleate buffer pH 6.0, 100 mM trehalose in a total volume of 200 µL. After incubation for 15 min. at 30°C the reaction was stopped in boiling water bath (3 min.). Glucose was determined by the glucose oxidase-peroxidase method. To perform the activation of cryptic trehalase, 50 µL of the activation cocktail containing 2 mM ATP, 20 mM MgCl 2 , 50 µM cAMP, 50 mM NaF and 5 mM theophylline prepared in 50 mM phosphate buffer, pH 7.5, was added to a suitable sample of the crude extract in a final volume of 100 µL. After incubation at 30°C for 15 min., the reaction was stopped by dilution with 400 µL of ice-cold 50 mM maleate buffer pH 6.0. Blanks were performed with omission of the activation cocktail. Basal trehalase activity is measured in cell free extracts prior to activation. Cryptic activity is the difference between totally activated trehalase and its basal activiy. One unit of trehalase is defined as the amount of enzyme that catalyses the hydrolysis of trehalose under the assay conditions giving rise to 1 µmol of glucose / minute.

Authors' contributions
JFM carried out the experimental and in silico studies, and prepared the draft of the manuscript. ADP and PSA participated in the design of the study, performed the critical Comparison of DCS1, NTH1, and TPS1 expression profiles Figure 2 Comparison of DCS1, NTH1, and TPS1 expression profiles. The 'alignment transcription profile for several genes' tool was used to search yMGV [18] for DCS1 (gray bars), NTH1 (striped bars) and TPS1 (black bars) genes. The main similarities in their expression patterns are represented. (A) Induction by heat treatment [19]. (B) Induction by changes in oxidation, pH, and osmolarity [20]. (C) Induction by nitrogen starvation [19]. (D) Effect of gcr1 mutation in the presence of glucose [22].
reading of the manuscript and coordination. All authors read and approved the final manuscript.