Trehalases are important highly conserved enzymes found in a wide variety of organisms and are responsible for the hydrolysis of trehalose that serves as a carbon and energy source as well as a universal stress protectant. Emerging evidence indicates that the enzymatic activity of the neutral trehalase Nth1 in yeast is enhanced by 14-3-3 protein binding in a phosphorylation-dependent manner through an unknown mechanism. In the present study, we investigated in detail the interaction between Saccharomyces cerevisiae Nth1 and 14-3-3 protein isoforms Bmh1 and Bmh2. We determined four residues that are phosphorylated by PKA (protein kinase A) in vitro within the disordered N-terminal segment of Nth1. Sedimentation analysis and enzyme kinetics measurements show that both yeast 14-3-3 isoforms form a stable complex with phosphorylated Nth1 and significantly enhance its enzymatic activity. The 14-3-3-dependent activation of Nth1 is significantly more potent compared with Ca2+-dependent activation. Limited proteolysis confirmed that the 14-3-3 proteins interact with the N-terminal segment of Nth1 where all phosphorylation sites are located. Site-directed mutagenesis in conjunction with the enzyme activity measurements in vitro and the activation studies of mutant forms in vivo suggest that Ser60 and Ser83 are sites primarily responsible for PKA-dependent and 14-3-3-mediated activation of Nth1.
Glycoside hydrolase family 37 (EC 184.108.40.206) of O-glycosyl hydrolases (EC 3.2.1.-) comprises enzymes which share common trehalase activity that is responsible for the hydrolysis of trehalose [α-D-glucopyranosyl-(1,1)-α-D-glucopyranoside], yielding two glucose molecules. Trehalases are important enzymes found in a wide variety of organisms and whose sequences have been highly conserved throughout evolution [1–3]. Trehalose, a naturally occurring non-reducing disaccharide, serves as a carbon and energy source as well as a universal protectant from various stress conditions such as dehydration, temperature extremes, oxidative stress and desiccation in a wide variety of organisms ranging from bacteria to invertebrates and higher plants [4,5]. In yeast and plants, it may also serve as a regulatory or signalling molecule to control certain metabolic pathways or even to affect growth.
In the budding yeast Saccharomyces cerevisiae, hydrolysis of trehalose can be carried out by three enzymes: neutral trehalases Nth1 and Nth2, and acid trehalase Ath1. Nth1 is localized in the cytosol and possesses maximal activity at pH 7.0. The trehalase activity of Nth1 is necessary for yeast cells to recover efficiently from heat shock . Neutral trehalase Nth2 shares 77% sequence homology with Nth1  and its activity was described in the presence of extracellular trehalose . Ath1, named for its lower optimum pH, displays no sequence homology with Nth1 and is important for growth on media containing trehalose as a carbon source .
The comparison of trehalase sequences from various organisms reveals that yeast enzymes (Nth1 and Nth2 from S. cerevisiae and Nth1 from Kluyveromyces lactis) possess a significantly longer N-terminal region compared with those from other organisms . This N-terminal extension seems to be an important feature governing the regulation of yeast neutral trehalases by the cAMP-dependent phosphorylation process as it contains several putative PKA (protein kinase A) phosphorylation sites [10–12]. Indeed, PKA was shown to increase trehalase activity in yeast cell extracts and the phosphorylated amino acid residue was identified as phosphoserine [10,13]. The biochemical study by Wera et al.  showed that the activation of Nth1 from S. cerevisiae, whose sequence contains 17 potential PKA phosphorylation sites, is mediated by the phosphorylation of more than one site and suggested that residues Ser20, Ser21 and Ser83 are the actual PKA sites responsible for Nth1 activation. A phosphoproteome analysis of whole-cell lysate from S. cerevisiae indicated Ser21, Ser23 and Ser83 as Nth1 sites phosphorylated in vivo, suggesting that Ser23 might be a target of a protein kinase other than PKA . Panni et al.  have suggested that phosphorylated Nth1 is regulated in a 14-3-3-protein-dependent manner, further extending the complexity of the regulation of neutral trehalase activity in yeast. In that study, it has been shown that the 14-3-3 proteins enhance the Nth1 activity and bind short synthetic phosphopeptides derived from motifs containing the PKA phosphorylation sites Ser21 and Ser23, but not Ser83. In addition, the enzymatic activity of yeast neutral trehalase is also regulated in a Ca2+-dependent manner. Franco et al.  have shown that activation of Schizosaccharomyces pombe Nth1 depends on Ca2+ binding through the conserved Ca2+-binding motif that is also present in S. cerevisiae Nth1. The mechanism of both 14-3-3-protein- and Ca2+-dependent Nth1 activation remains to be determined.
The yeast S. cerevisiae possesses two 14-3-3 genes, BMH1 and BMH2, which share a great degree of homology and are essential in most laboratory strains [17,18]. Members of the 14-3-3 protein family serve as molecular chaperones by modulating the structure, the subcellular localization or the activity of hundreds of other proteins in all eukaryotes [19,20]. They recognize specific phosphoserine/phosphothreonine-containing as well as unphosphorylated motifs (reviewed in [21,22]) and through these binding interactions, they play important roles in the regulation of signal transduction, apoptosis, cell-cycle control and nutrient-sensing pathways. If the bound partner is an enzyme, then 14-3-3 protein can regulate its enzymatic activity. It has been, for example, shown that the enzymatic activity of tryptophan and tyrosine hydroxylases [23,24] and serotonin N-acetyltransferase  is enhanced, whereas the activity of Ask1  and Yak1  kinases is suppressed upon 14-3-3 protein binding. The mechanisms behind these regulations are still unclear, but can involve either the direct structural change of bound enzyme, as has been demonstrated in the case of serotonin N-acetyltransferase , or protection against dephosphorylation, as has been suggested in the case of tyrosine hydroxylase . Another important and common feature of the 14-3-3-binding partners is the presence of multiple phosphorylation/14-3-3-binding sites, very often within disordered regions and/or bordering the functional domain, that are simultaneously used not only for the binding, but also to achieve the required function [30,31].
To better understand the mechanism of the S. cerevisiae Nth1 regulation, we investigated its interaction with yeast 14-3-3 protein isoforms in detail. The mass spectrometric analysis revealed that four residues within the disordered N-terminal segment of recombinant full-length Nth1 (Ser20, Ser21, Ser60 and Ser83) are phosphorylated by PKA in vitro. Sedimentation analysis and enzyme kinetics measurements show that both yeast 14-3-3 isoforms form a stable complex with phosphorylated Nth1 and significantly enhance its enzymatic activity. The 14-3-3-dependent activation of Nth1 is significantly more potent than Ca2+-dependent activation. Limited proteolysis confirmed that the 14-3-3 proteins interact with the N-terminal segment of Nth1 where all phosphorylation sites are located. Site-directed mutagenesis was used to decipher the importance of found phosphorylation sites for Nth1 activation. Both the enzyme activity measurements in vitro and the activation studies of mutant forms in vivo suggest that Ser60 and Ser83 are sites primarily responsible for PKA-dependent and 14-3-3-mediated activation of Nth1.
Heterologous expression, purification and phosphorylation of Nth1
The Nth1 coding sequence was PCR-amplified from the S. cerevisiae BY4741 genomic DNA. The PCR product was ligated using NcoI and BamHI sites into a modified pET-32b vector (Novagen) where 81 residues after the sequence H6SSGLVPRGS were deleted (a gift from Dr Donald Ronning, Department of Chemistry, University of Toledo, Toledo, OH, U.S.A.). The entire coding region was verified by sequencing. The Nth1 was expressed as a fusion protein, with the thioredoxin and His6 tags at the N-terminus, by IPTG (isopropyl β-D-thiogalactopyranoside) induction for 18 h at 25°C and purified from Escherichia coli Rosetta™(DE3) cells using Chelating Sepharose® Fast Flow (GE Healthcare) using the standard protocol. The thioredoxin and His6 tags were cleaved by incubation at 4°C for 12 h with 5 units of thrombin per mg of protein. After the cleavage, Nth1 was dialysed against buffer containing 50 mM sodium citrate (pH 6), 1 mM EDTA, 2 mM DTT (dithiothreitol) and purified by cation-exchange chromatography on a HiTrap SP column (GE Healthcare). The protein was eluted using a linear gradient of NaCl (50–1000 mM). Fractions containing Nth1 were concentrated to ~2 mg/ml and purified using size-exclusion chromatography on a Superdex 200 10/300 GL column (GE Healthcare) in buffer containing 20 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM DTT and 10% (w/v) glycerol.
Purified Nth1 [both WT (wild-type) and mutants] was phosphorylated by incubation at 30°C for 2 h and then overnight at 4°C with 80 units of PKA (Promega) per mg of protein in the presence of 0.75 mM ATP and 20 mM MgCl2. The completeness of the phosphorylation reaction was checked using MALDI (matrix-assisted laser-desorption ionization)–TOF (time-of-flight)-MS.
Mutants of Nth1 containing different numbers of PKA phosphorylation sites (at positions 20, 21, 60 or 83) were created by mutating other sites to alanine using the QuikChange™ approach (Stratagene). All mutations were confirmed by sequencing.
Heterologous expression and purification of yeast 14-3-3 protein isoforms
AUC (analytical ultracentrifugation)
SV (sedimentation velocity) experiments were performed using a ProteomLab™ XL-I, Beckman Coulter analytical ultracentrifuge. SV experiments of the Bmh1, Bmh2 and Nth1 WT and Nth1 mutants were conducted at loading concentrations of 0.5–10 μM, 20°C and 42000 or 48000 rev./min rotor speed (An-50 Ti rotor, Beckman Coulter). All data were collected at 280 nm. Samples were dialysed against the buffer containing 20 mM Tris/HCl (pH 7.5), 200 mM NaCl and 2 mM 2-mercaptoethanol before analysis. Data were analysed using the SEDFIT and SEDPHAT packages [34,35].
Mass spectrometric analysis of Nth1
Samples were first separated by SDS/PAGE (8% gels), and excised protein bands were digested with trypsin endoprotease (Promega) directly in-gel with cysteine modification by iodoacetamide . Resulting peptide mixtures were extracted by 30% acetonitrile and 0.1% trifluoroacetic acid and subjected to MALDI–TOF-MS using an UltraFLEX III instrument (Bruker–Daltonics) equipped with a nitrogen laser (337 nm). Positively or negatively charged spectra were acquired with 2,5-dihydrobenzoic acid as the MALDI matrix and calibrated externally using the monoisotopic [M+H]+ or [M−H]− ions of PepMixII calibrant (Bruker–Daltonics) or internally using the corresponding monoisotopic ions of recombinant protein peptides with known sequence. Phosphorylated peptides from the peptide mixtures were enriched as described previously .
Samples containing 50 μg of purified Nth1 or pNth1 (phosphorylated Nth1) (as thioredoxin-fusion protein) and 30 μg of Bmh1 (molar ratio of Nth1/Bmh was 1:2) were incubated at 25°C with 50 ng of trypsin or chymotrypsin (protease/protein ratio was approximately 1:1000, w/w) in buffer containing 20 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1 mM DTT and 10% (w/v) glycerol, and were stopped by boiling in the presence of SDS/PAGE loading buffer at times ranging from 0 to 30 min. The resulting polypeptides were separated by SDS/PAGE (10% gels) and visualized by Coomassie Brilliant Blue R-250 staining. The resulting Nth1 and pNth1 peptides were analysed using MALDI–TOF-MS.
Trehalase activity assay
The enzyme kinetics of the hydrolysis of trehalose by Nth1 was measured using a stopped assay . The production of glucose was detected using the Amplex® Red Glucose/Glucose Oxidase Assay Kit (Invitrogen). The assay was performed at 30°C in buffer containing 20 mM Tris/HCl (pH 7.5), 150 mM NaCl, 10% (w/v) glycerol and 0–30 mM trehalose. The final concentrations of Nth1 and Bmh in in vitro experiments were 80 nM and 12 μM respectively. Aliquots of 50 μl of the reaction mixtures were taken at intervals and the reaction was stopped by boiling the aliquots for 3 min. Then, 50 μl of the Amplex® Red reagent/peroxidase/glucose oxidase solution was added, and the concentration of glucose was determined following the manufacturer's instructions. The reaction rates were determined and the data were fitted to the Michaelis–Menten equation using Origin 8.0 (OriginLab) and are given as means±S.E.M. for three experiments.
Yeast strains and plasmid construction
All in vivo experiments were carried out using the BY4741-derived deletion strain ydr001c (nth1D): genotype MATa his3Δ1 leu2Δ met15Δ ura3Δ ydr001c::kanMX4 [EUROSCARF European Saccharomyces cerevisiae Archive for Functional Analysis (Institut für Mikrobiologie, Johann Wolfgang Goethe-Universität, Frankfurt am Main, Germany)]. DNA fragments encoding the Nth1 WT and mutant forms of trehalase were cloned into the multicopy expression vector YEp352 behind the weak constitutive NHA1 (Na+/H+ antiporter 1) promoter by homologous recombination . The vectors were constructed by replacing the S. cerevisiae NHA1 gene with the S. cerevisiae NTH1 genes in the pNHA1-985 plasmid . The trehalase-coding WT and mutant regions (2253 bp) were amplified from pET-32b plasmids (using primers YEpN-NTH1-F and YEpN-NTH1-R; see Supplementary Table S1 at http://www.BiochemJ.org/bj/443/bj4430663add.htm) and the PCR products were inserted by homologous recombination into the pNHA1-985 plasmid (with the URA3 marker gene) linearized with the restriction endonuclease PstI in the NHA1 coding sequence. Yeast cells were transformed by linearized plasmid and PCR-amplified DNA fragments. The exchange of NHA1 by NTH1 in plasmids isolated from ura+ colonies was verified by restriction analysis and diagnostic PCR, and the trehalase-coding regions were sequenced.
Activation of trehalase in vivo
The activation of trehalase in vivo was carried out as described previously . Briefly, yeast cells overexpressing the Nth1 WT or its mutant forms were grown to exponential phase (D600 of 1.5–2) in minimal YNB (yeast nitrogen base) medium with Brent supplemental mixture containing 2% (w/v) glycerol. Cells were harvested by centrifugation at 1620 g for 3 min, washed in ice-cold water and resuspended at a cell density of 25 mg/ml in a buffer containing 100 mM Mes/KOH (pH 7) and 50 μM CaCl2. The cells were pre-incubated at 30°C for 10 min before the first sample was taken. The activation of trehalase was triggered by the addition of 4% (w/v) glucose (final concentration). Aliquots (750 μl) were taken before the glucose addition and, at 10 min after the glucose addition, samples were quickly mixed with 40 ml of ice-cold water. The cells were collected by centrifugation at 1620 g for 3 min at 4°C. The pellet was resuspended in 500 μl of ice-cold buffer containing 50 mM Mes/KOH (pH 7) and 50 μM CaCl2 and broken up by vigorous mixing with glass beads. Protein concentration was determined using the Bradford procedure. Trehalase activity was determined as described previously . The specific activity of trehalase is expressed as nmol of glucose liberated per min per mg of protein.
Preparation and characterization of phosphorylated Nth1 in vitro
The main goal of the present study was to investigate in detail the 14-3-3-protein-dependent regulation of Nth1 from S. cerevisiae. Since it has been suggested that this interaction is mediated through PKA-induced phosphorylation, we first focused on the identification of phosphorylation sites essential for the interaction between Nth1 and yeast 14-3-3 isoforms Bmh1 and Bmh2. The N-terminal region of yeast Nth1 contains several putative PKA phosphorylation sites that have been implicated in the cAMP-dependent activation of this enzyme [10–12]. Furthermore, it has been suggested that 14-3-3 proteins play an important role in the regulation of Nth1 activity through binding to its phosphorylated N-terminal segment . To investigate the interaction between Nth1 and 14-3-3 proteins, we first prepared a recombinant full-length S. cerevisiae Nth1 (sequence 1–751). Purified Nth1 was then phosphorylated by PKA in vitro and the positive- or negative-ion MALDI–TOF mass spectra of the tryptic peptide mixture of pNth1 were measured in the reflection mode to confirm the amino acid sequences of phosphorylation sites and their modification. The mass spectrum of WT pNth1 revealed signals of four stoichiometrically phosphorylated serine residues: Ser20, Ser21, Ser60 and Ser83 (Supplementary Table S2 at http://www.BiochemJ.org/bj/443/bj4430663add.htm). The identity structure of all phosphorylated peptides was confirmed further by analysis of their PSD (post-source decay) spectra to authenticate all serine residues presented as phosphorylated amino acids (results not shown). As a control experiment, we prepared the Nth1 mutant where all four serine residues (at positions 20, 21, 60 and 83) were mutated to alanine. This mutant protein (denoted as pNth1-noS) was phosphorylated as Nth1 WT and no additional phosphorylated residues were found using MALDI–TOF mass spectrometric analysis.
In agreement with previous reports, all phosphorylation sites found are located within the N-terminal region of Nth1 [6,10–12,14] (Figure 1A). Although phosphorylation sites Ser20, Ser21 and Ser83 have been suggested previously to be involved in PKA-dependent regulation of Nth1 [6,14], the PKA-mediated phosphorylation of Ser60 was not observed previously. Bioinformatics analysis using the PONDR program indicates that the N-terminal region of Nth1, where all phosphorylation sites detected are located, is disordered  (Figure 1B). The protein domain features of Nth1 were analysed using the SMART program , which predicted the presence of two conserved domains upstream of the disordered N-terminal region: the trehalase domain (sequence 163–721) and the Ca2+-binding motif (sequence 114–125) similar to the EF hand domain [16,43] (Figure 1A).
Primary structure of Nth1
Yeast 14-3-3 protein isoforms (Bmh1 and Bmh2) form stable complexes with phosphorylated pNth1 and enhance its enzymatic activity in vitro
Next, we investigated whether the pNth1 forms a stable complex with the 14-3-3 proteins using AUC. SV measurements revealed that pNth1 alone exists in solution as monomers of molecular mass 86 kDa, whereas Bmh proteins form stable dimers, as has been shown previously . Both unphosphorylated Nth1 WT and pNth1-noS mutant showed no significant interaction with Bmh1 or Bmh2, whereas phosphorylated pNth1 WT formed a stable complex with both Bmh isoforms (Figure 2, only data for Bmh1 are shown). Analysis of SV data revealed that the binding stoichiometries of pNth1–Bmh complexes are 1:2 (a dimer of Bmh protein binds one molecule of pNth1). The apparent equilibrium dissociation constants (Kd) for pNth1 binding to Bmh1 and Bmh2 were determined to be 0.15×10−6 and 0.25×10−6 M respectively (Table 1).
|Kd (×106 M)|
|Nth1 variant||Number of phosphosites||Bmh1||Bmh2|
|Kd (×106 M)|
|Nth1 variant||Number of phosphosites||Bmh1||Bmh2|
To investigate the effect of phosphorylation and 14-3-3 binding on the enzymatic activity of Nth1, we measured the trehalase activity of pNth1 and Nth1 both in the presence and in the absence of Bmh proteins. The trehalase activity was measured by estimating the glucose produced by hydrolysis of trehalose. In agreement with a previous study by Panni et al. , both Nth1 and pNth1 show very low enzymatic activity in the absence of 14-3-3 proteins (Figure 3A). The presence of 14-3-3 proteins dramatically enhances only the activity of phosphorylated pNth1, whereas the enzymatic activity of unphosphorylated Nth1 remains unchanged. We did not observe any significant difference between the two yeast 14-3-3 isoforms in their abilities to activate pNth1, thus only the data for Bmh1 are presented. In addition, Franco et al.  has shown that the enzymatic activity of Nth1 from S. pombe is significantly enhanced upon Ca2+ binding through the conserved Ca2+-binding motif and suggested that Ca2+ binding affects the oligomeric state of the enzyme. The Ca2+-binding site is conserved among yeast neutral trehalases and, in the sequence of Nth1 from S. cerevisiae, is located between residues 114 and 125 (sequence D114TDKNYQITIED125) [16,43]. This sequence closely resembles the Ca2+-binding site present in the EF hand motif of numerous Ca2+-binding proteins . Therefore we also measured the Nth1 activity in the presence of Ca2+ that has been shown to enhance Nth1 activity . The presence of 5 mM Ca2+ somewhat increased the activity of pNth1, but the 14-3-3-dependent activation is significantly more potent. Interestingly, in the presence of both Ca2+ and 14-3-3, the activity of pNth1 is significantly lower compared with the sample without Ca2+.
Activation of Nth1 in vitro and enzyme kinetics of activated Nth1
The Michaelis–Menten kinetics for the hydrolysis of trehalose by pNth1 in the presence of Bmh1 and/or 5 mM Ca2+ are shown in Figure 3(B). At pH 7.5 and 30°C, the full-length pNth1 in the presence of Bmh1 has a Michaelis constant, Km, of 8±1 mM and a turnover number, kcat, of 71±4 s−1 (kcat/Km of 8.9 s−1·mM−1). These values are comparable with those published in the literature, as several authors have reported Km values of 5–10 mM [45–47]. In the presence of 5 mM Ca2+ (in the absence of Bmh1) pNth1 has a Km value of 6.0±0.4 mM and a kcat value of 6.2±0.1 s−1 (kcat/Km of 1.0 s−1·mM−1), whereas, in the presence of both Bmh1 and 5 mM Ca2+, we obtained for pNth1 a Km value of 3.6±0.7 mM and a kcat value of 42±2 s−1 (kcat/Km of 11.7 s−1·mM−1). These data show that the 14-3-3 protein binding significantly increases (~11-fold) turnover number of pNth1 compared with the Ca2+ only-induced activation. We were unable to compare these data with the kinetic parameters of pNth1 or Nth1 alone as in the absence of the 14-3-3 protein binding or Ca2+ their enzymatic activities are very low for any analysis.
Furthermore, AUC was used to study the effect of Ca2+ on the oligomeric state of Nth1 from S. cerevisiae. These experiments revealed that Nth1 exists in all three solutions tested containing 5, 10 and 15 mM Ca2+ as monomers of molecular mass 86 kDa (Supplementary Figure S1 at http://www.BiochemJ.org/bj/443/bj4430663add.htm), and hence the Ca2+-dependent activation does not result from the change of the oligomeric state of Nth1.
Yeast 14-3-3 protein isoforms interact with the N-terminal part of phosphorylated pNth1
Both AUC and native TBE (Tris/borate/EDTA)/PAGE revealed that Bmh proteins interact with phosphorylated pNth1 (Figure 2 and Table 1, and Supplementary Figure S2 at http://www.BiochemJ.org/bj/443/bj4430663add.htm). Since all PKA phosphorylation sites are located within the N-terminal region of pNth1, it is reasonable to assume that Bmh proteins make extensive contacts with this part of pNth1. In order to verify this hypothesis, the limited proteolysis was used to check whether 14-3-3 protein binding protects the N-terminal region of pNth1 against the proteolytic degradation. The results of limited proteolysis of Nth1, pNth1 and Bmh1 alone and in mixtures by low levels of trypsin or chymotrypsin (the protease/protein ratio was 1:1000, w/w) are presented in Figure 4 and Supplementary Figure S3 (at http://www.BiochemJ.org/bj/443/bj4430663add.htm). It can be noticed that Bmh1 protein alone was resistant to trypsin or chymoptrypsin within the time course of this experiment (Supplementary Figure S3). Mass spectrometric analysis of protein bands formed after 30 min chymotrypsin or trypsin partial digestions of Nth1 and pNth1 both in the absence and in the presence of Bmh1 revealed clear 14-3-3-mediated protection of the N-terminal sequence containing all PKA phosphorylation sites. No proteolytic degradation from the C-terminus of either Nth1 or pNth1 was observed under the conditions of partial proteolysis used as the analysis of Nth1 alone, pNth1 alone and Nth1+Bmh1 revealed the sequence coverage of Nth1 protein from His168 to the C-terminal Leu751 using both trypsin or chymotrypsin protease (the first peptide found from the N-terminal segment was m/z signal 1128.56 corresponding to the amino acid sequence 168–176). On the other hand, the N-terminal segment of pNth1 was, in the presence of Bmh1, protected against the chymotrypsin partial proteolysis as confirmed by the determination of pNth1 protein sequence coverage from Ala−1 to Leu751. A signal at m/z 1630.78 that corresponds to pNth1 N-terminal peptide from Ala−1 to Arg15 was detected and its peptide sequence was confirmed using PSD analysis. All phosphorylated serine residues (at positions 20, 21, 60 and 83) were detected in the pNth1 protein bands obtained. Similarly, the use of trypsin for 30 min of partial proteolysis of pNth1 in the presence of Bmh1 generated peptides with sequence coverage determined from Thr58 to Leu751 and the first peptide found from the N-terminus was m/z signal 1579.70 corresponding to amino acid sequence Thr58–Leu70 with confirmed phosphorylation on Ser60 using PSD analysis. Taken together, these data confirm that the 14-3-3 proteins interact with the N-terminal segment of pNth1.
Limited proteolysis of Nth1
The importance of individual phosphorylation sites of pNth1 for its interaction with 14-3-3 proteins in vitro
Mass spectrometric analysis of pNth1 revealed that PKA stoichiometrically phosphorylates four serine residues (at positions 20, 21, 60 and 83) located within the N-terminal segment of the enzyme (Supplementary Table S2). Our data also show that the yeast 14-3-3 proteins bind pNth1 with high affinity, interact with its phosphorylated N-terminal segment and significantly enhance its enzymatic activity (Figures 2–4). To identify which phosphorylation sites are responsible for the interaction between pNth1 and the 14-3-3 proteins, we prepared four mutants containing a single phosphorylation site (denoted as S20, S21, S60 and S83) by mutating other serine residues to alanine. We also prepared six mutants containing all possible combinations of two phosphorylation sites (S20+21, S20+60, S20+83, S21+60, S21+83 and S60+83).
The differences in the 14-3-3-protein-binding affinity among all prepared singly and doubly phosphorylated pNth1 mutants were tested using both SV measurements and native TBE/PAGE analysis (Table 1, and Supplementary Figures S2 and S4 at http://www.BiochemJ.org/bj/443/bj4430663add.htm). The binding affinities of singly and doubly phosphorylated pNth1 mutants containing either Ser60 or Ser83 for both Bmh isoforms (with Kd values ranging from 0.2×10−6 to 1×10−6 M) were found to be fully comparable with the binding affinity of pNth1 WT, which binds Bmh1 and Bmh2 with Kd values of 0.15×10−6 and 0.25×10−6 M respectively (Table 1). On the other hand, pNth1 mutants containing phosphorylation sites pSer20 and pSer21 only (pS20, pS21 and pS20+21) exhibited very weak binding with Kd values of 10×10−6 M. This suggests that phosphorylation sites Ser60 and Ser83 are essential for pNth1 binding to yeast 14-3-3 proteins.
Next, we studied the ability of both yeast 14-3-3 protein isoforms to enhance the enzymatic activity of all pNth1 mutants prepared (Figure 5). Trehalase activity was measured as glucose production in the presence of either 14 or 30 mM trehalose (data for 30 mM trehalose only are shown). These measurements revealed results similar to those of binding experiments. The activities of mutants containing only phosphorylation sites Ser20 and Ser21 (pS20, pS21 and pS20+21) are very low and comparable with activities of unphosphorylated Nth1, pNth1-noS mutant and pNth1 WT in the absence of 14-3-3 proteins. On the other hand, mutants containing phosphorylation sites Ser60 and Ser83 show significantly higher levels of activation. Singly phosphorylated pS60 and pS83 as well as doubly phosphorylated mutants pS60+83, pS20+83 and pS21+83 exhibit activities comparable with the activity of pNth1 WT. Surprisingly, doubly phosphorylated mutants pS20+60 and pS21+60 show significantly reduced activities although possessing comparable binding affinities for 14-3-3 proteins as pNth1 WT or pS60 (Table 1). We may speculate that, in these two cases, the 14-3-3 protein binding affects the conformation of trehalase differently compared with other active complexes containing either phosphorylated Ser60 alone or phosphorylated Ser83 (both alone or in combination with the second phosphorylation site). In addition, no significant differences between Bmh1 and Bmh2 were observed for most samples, in good agreement with our binding experiments.
The role of individual phosphorylation sites of pNth1 for its interaction with 14-3-3 proteins in vitro
Activation of Nth1 mutants in vivo
To evaluate further the role of Ser60 and Ser83 in the activation of Nth1, we performed the activation studies in vivo. Five nth1Δ transformants expressing Nth1 WT, S60 (containing only one phosphorylation site, Ser60), S83 (containing only one phosphorylation site, Ser83), S20+21 (containing two phosphorylation sites, Ser20 and Ser21) and Nth1-noS (where all PKA sites were removed) trehalases were prepared. S. cerevisiae cells were grown in glycerol-containing medium and the activation of trehalase was triggered by the addition of glucose to a final concentration of 4% (w/v) . The specific trehalase activities of extracts prepared from yeast cells before and after the activation are shown in Figure 6. The statistically significant activation of trehalase activity was observed only in cells expressing Nth1 WT, S60 and S83. The highest level of activation was observed for Nth1 WT (2.4-fold), whereas S60 and S83 mutants containing a single phosphorylation site were activated moderately (1.2-fold and 1.5-fold respectively). Cells expressing Nth1-noS and S20+21 showed no significant change in the specific trehalase activity upon the incubation with glucose. The level of trehalase activation observed in these experiments, however, is lower compared with our in vitro measurements (Figures 3 and 5). We assume that this difference might reflect an incomplete phosphorylation of Nth1 in vivo and/or a low concentration of Bmh proteins, whereas in vitro experiments were performed with stoichiometrically phosphorylated Nth1 and with an excess of Bmh proteins. In addition, all samples taken before the activation possessed a significant trehalase activity that probably reflects the Ca2+-dependent activity and/or the 14-3-3-dependent activity of Nth1 that was partially phosphorylated before activation. Taken together, the activation experiments in vivo, in agreement with in vitro data, suggest that Ser60 and Ser83, but not Ser20 and Ser21, are important for the activation of Nth1.
Activation of Nth1 mutants in vivo
In the present study, our main aim was to investigate further so far unresolved details governing the 14-3-3 protein binding and activation of Nth1. We identified four residues of Nth1 that are phosphorylated by PKA in vitro (serine residues at positions 20, 21, 60 and 83), of which just Ser60 and Ser83 are important for the 14-3-3-protein-dependent activation of Nth1 (Table 1 and Figure 5). Both the binding and the enzymatic activity measurements revealed that the phosphorylation of either of these two sites is sufficient for high-affinity binding to 14-3-3 and full activation of pNth1 in vitro. In addition, limited proteolysis in conjunction with MS analysis confirmed that the 14-3-3 proteins interact with the N-terminal segment of pNth1 where these sites are located (Figure 4). Panni et al.  suggested that the interaction between yeast Nth1 and 14-3-3 proteins is mediated through phosphorylated Ser21 and Ser23. The possible reason for the phosphorylation sites Ser60 and Ser83 not being detected could be the different methodological approach. In the present study, the singly (or doubly) phosphorylated mutants of full-length recombinant pNth1 were used to study the interaction with the 14-3-3 proteins, whereas Panni et al.  used short synthetic phosphopeptides to identify the 14-3-3-protein-binding sites of Nth1.
It is now well established that the dimeric nature of the 14-3-3 proteins with its two ligand-binding grooves arranged in an antiparallel fashion is very important for 14-3-3 protein functions as it allows simultaneous binding of two motifs [31,48]. Many 14-3-3-protein-binding partners contain two or more 14-3-3-protein-binding motifs, and our data strongly suggest that S. cerevisiae Nth1 is such an example. It has been shown that a minimal linker sequence of approximately ten residues is required between the two motifs to generate a tandem 14-3-3-protein-binding motif . Therefore it is entirely possible that Ser60 and Ser83 of Nth1 form such a doubly phosphorylated 14-3-3-binding motif. On the other hand, it is very likely that Ser20 and Ser21, the other two Nth1 sites phosphorylated by PKA in vitro (Table 1 and Figure 1A), do not form the 14-3-3-protein-binding motif as it has never been observed so far that the 14-3-3 proteins would recognize the peptide containing two subsequent or very closely located phosphorylated residues . Consistently, neither singly phosphorylated pS20 and pS21 nor doubly phosphorylated pS20+21 bind the yeast 14-3-3 isoforms with high affinity (Table 1). At the same time, all three mutants show very low enzymatic activities in the presence of 14-3-3 proteins (Figure 5). Similar results were obtained from the activation studies of Nth1 mutant forms in vivo that revealed significant activation only for Nth1 WT, S60 and S83, but not for S20+21 (Figure 6).
The 14-3-3 proteins have been shown to regulate the activity of a number of various enzymes from both yeasts and higher eukaryotes, including serotonin N-acetyltransferase, tyrosine and tryptophan hydroxylases, Raf kinases, Ask1 kinase and Yak1 kinase (reviewed in [17,18,21,22]). The mechanisms behind these regulations are mostly elusive, but can involve the direct structural change of bound enzyme, as has been demonstrated in the case of serotonin N-acetyltransferase . Since all PKA phosphorylation sites/14-3-3-protein-binding motifs are located within the presumably disordered N-terminal segment outside the catalytic domain of Nth1 and the PKA-induced phosphorylation does not increase the enzyme activity of Nth1 by itself, it is reasonable to speculate that the 14-3-3 protein binding directly affects the structure of Nth1, probably of its catalytic trehalase domain, and hence increases its enzymatic activity.
Franco et al.  suggested that Ca2+ ions are likely to be an integral part of active trehalase. We did observe a moderate activation of pNth1 in the presence of Ca2+; however, its effect was significantly weaker compared with the Bmh-induced activation (Figure 3B). A simultaneous presence of Ca2+ and Bmh1 led to a suppression of both kcat and Km values for the hydrolysis of trehalose by pNth1 compared with the presence of Bmh1 only. As a result, the catalytic efficiency, defined as the kcat/Km ratio, remained relatively unchanged. However, it suggests that the mechanism of pNth1 activation by Ca2+ differs from the 14-3-3-protein-dependent activation. The Ca2+-binding EF-hand-like motif is located between the segment containing all sites phosphorylated by PKA and the catalytic domain (Figure 1A), implying that Ca2+ binding might also affect the interaction between pNth1 and the 14-3-3 protein and thus its effect on the enzymatic activity of pNth1.
In conclusion, our results show that S. cerevisiae Nth1 is phosphorylated by PKA at multiple sites, of which Ser60 and Ser83 are important for the 14-3-3-protein-mediated activation of Nth1. In addition, the 14-3-3-protein-dependent activation of phosphorylated Nth1 is significantly more potent compared with Ca2+-only-dependent activation.
acid trehalase 1
matrix-assisted laser-desorption ionization
Na+/H+ antiporter 1
protein kinase A
Dana Veisova carried out the cDNA cloning, prepared recombinant proteins, performed all enzyme kinetics experiments and performed in vivo expression experiments. Eva Macakova performed limited proteolysis. Eva Macakova and Petr Vacha prepared recombinant proteins. Lenka Rezabkova carried out the SV analysis. Miroslav Sulc generated and interpreted MS data. Hana Sychrova supervised the preparation of constructs for in vivo expression experiments. Veronika Obsilova conceived the study. Veronika Obsilova and Tomas Obsil designed experiments, analysed the data and wrote the paper. All authors have approved the final paper.
We thank Pavla Herynkova for help with the construction of Nth1 transformants used for in vivo experiments.
This work was supported by the Czech Science Foundation [grant number P207/11/0455], Grant Agency of the Charles University [grant number 350111], Grant Agency of the Academy of Sciences of the Czech Republic [grant number IAA500110801], Academy of Sciences of the Czech Republic [research project AV0Z50110509 of the Institute of Physiology and AV0Z50200510 of the Institute of Microbiology] and Ministry of Education, Youth and Sports of the Czech Republic [research project MSM0021620857].