BGTs [β-(1,3)-glucanosyltransglycosylases; EC 2.4.1.-] of the GH72 (family 72 of glycosylhydrolases) are GPI (glycosylphosphatidylinositol)-anchored proteins that play an important role in the biogenesis of fungal cell walls. They randomly cleave glycosidic linkages in β-(1,3)-glucan chains and ligate the polysaccharide portions containing newly formed reducing ends to C3(OH) at non-reducing ends of other β-(1,3)-glucan molecules. We have developed a sensitive fluorescence-based method for the assay of transglycosylating activity of GH72 enzymes. In the new assay, laminarin [β-(1,3)-glucan] is used as the glucanosyl donor and LamOS (laminarioligosaccharides) fluorescently labelled with SR (sulforhodamine) serve as the acceptors. The new fluorescent assay was employed for partial biochemical characterization of the heterologously expressed Gas family proteins from the yeast Saccharomyces cerevisiae. All the Gas enzymes specifically used laminarin as the glucanosyl donor and a SR–LamOS of DP (degree of polymerization) ≥5 as the acceptors. Gas proteins expressed in distinct stages of the yeast life cycle showed differences in their pH optima. Gas1p and Gas5p, which are expressed during vegetative growth, had the highest activity at pH 4.5 and 3.5 respectively, whereas the sporulation-specific Gas2p and Gas4p were most active between pH 5 and 6. The novel fluorescent assay provides a suitable tool for the screening of potential glucanosyltransferases or their inhibitors.

INTRODUCTION

The cell wall is an external envelope that is essential for fungal viability. It consists of a matrix of cross-linked glucose polymers (β-glucans), mannoproteins and a small amount of chitin. A balance between the biosynthesis of the constituents and hydrolysis/transglycosylation reactions is crucial for cell-wall homoeostasis, which is essential for both cell integrity and proper morphogenesis.

GH72 (family 72 of glycoside hydrolases), annotated in the CAZy (carbohydrate active enzyme; http://www.cazy.org/) database, include endo-β-(1,3)-glucanases that are endowed with significant transglycosylase activity [1,2]. Owing to this activity, these enzymes are also referred to as β-(1,3)-glucanosyltransferases. To date, 157 proteins have been identified in this family [3]. The presence of GH72 enzymes is restricted to yeast and fungal species indicating a highly specialized role in fungal wall biogenesis.

β-(1,3)-glucan is the most abundant glucose polymer of the yeast cell wall. It is synthesized by plasma membrane enzyme complexes, glucan synthases with Fks proteins as their catalytic components [4]. On the cytoplasmic side of the cell membrane, UDP-glucose serves as a substrate for the vectorial synthesis of the polymer. In the extracellular space, GH72 enzymes catalyse the linear elongation of the polymers. Moreover, they contribute to the formation of lateral branches by elongating branching points, β-(1,6)-linked glucose residues, that are created by branching enzymes [5]. All these reactions occur through transglycosylation due to the lack of availability of small energy-exchange molecules such as ATP in the extracellular space.

In addition to GH72 and branching enzymes active on β-(1,3)-glucan, other transglycosylases are involved in fungal cell-wall formation [6]. Remarkably, in Saccharomyces cerevisiae it has been demonstrated that Crh enzymes (Crh1, Crh2 and Crr1 of GH16 family) cross-link chitin both to the β-(1,6)-glucan, which in turn tethers GPI (glycosylphosphatidylinositol)-trimmed mannoproteins to the wall, and to the β-(1,3)-glucan [7,8]. Crh1p and Crh2p are required to covalently link a pool of the yeast Gas1 protein molecules to the bud scars [9]. The in vivo role of other glycanases/transglycosylases such as Dcw1p and Dfg5p (putative mannanases of family GH5) and Scw10/Scw4 proteins (putative glu-canases of family GH17) is still unclear.

Hydrolases/transglycosylases also play a role in plant cell-wall growth and modification and are the object of intense biochemical investigation. An example is provided by the XET/XTH (xyloglucan endotransglycosylase/hydrolase), a ubiquitous plant enzyme involved in the loosening and reformation of xyloglucan tethers between cellulose microfibrils in the cell walls [1013]. Thus the creation of an interwoven texture of polymers in the cell wall shares some common principles both in fungal and plant cells.

The first clues as to the activity of the GH72 enzymes came from biochemical studies on Gel1p from Aspergillus fumigatus [14]. When Gel1p was incubated with borohydride-reduced LamOS (laminarioligosaccharides) of 10 or more glucose residues (rG≥10), both shorter and longer products were obtained. This was consistent with a double-inversion transglycosylating reaction mechanism of the so-called retaining glycoside hydrolases [15]. In the first step, the enzyme cleaves an internal β-(1,3)-glycosidic linkage in the donor substrate molecule, the non-reducing part is released and the newly created reducing-end portion of the donor molecule is linked via an ester bond to the catalytic acid residue in the enzyme, thus forming a covalent glycosyl-enzyme intermediate. In the next step, the glycan portion is transferred from the intermediate to a molecule of water (hydrolysis) or to C3(OH) of the non-reducing end glucose residue in the acceptor molecule with formation of a β-(1,3)-glycosidic linkage (transglycosylation). Consequently, a disproportionation of the substrate takes place and a series of oligosaccharide products with DP (degree of polymerization) both higher and lower than that of the donor are formed. HPAEC (high-performance anion-exchange chromatography) was used to analyse the reaction mixture and to detect the transglycosylation activity [14,16].

The GH72 enzymes are not pure transglycosylases, but also possess a significant hydrolytic activity. The ratio of transferase to hydrolytic activity depended on the substrate concentration; for Gel1p from A. fumigatus it was 100% at high-substrate concentration but decreased to 35% at low concentration [14]. The term β-(1,3)-glucan elongase was also coined to indicate the ability of these enzymes to elongate the substrate molecules [17].

The HPAEC method used in the previous studies to detect transglycosylation activity of the Gas proteins enables both qualitative and quantitative characterization of the reaction products, but determination of overall transglycosylation activity is complicated. Moreover, the HPAEC assay is rather cumbersome and does not permit simultaneous measurements of multiple enzyme samples. Of considerable importance may also be the fact that the HPAEC instrumentation is quite sophisticated and hence expensive. It was the aim of the present study to design a simple, rapid and cost-effective method to test the transglycosylating activity of GH72 enzymes. The new assay of BGT [β-(1,3)-glucanosyltransglycosylase] activity exploits laminarin as the donor and fluorescently labelled LamOS as the acceptors. The assay was used to determine the basic biochemical properties of heterologous yeast Gas proteins and to map the distribution of BGT activity in the yeast-cell fractions.

EXPERIMENTAL

Substrates and chemicals

Laminarin [β-(1,3)-glucan] was purchased from Sigma and pustulan [β-(1,6)-glucan] from Calbiochem. LamOS and N-acetyl chito-oligosaccharides were from Megazyme. Oligosaccharides derived from pustulan (DP 2–7), were prepared by partial acid hydrolysis of pustulan with 4 M trifluoroacetic acid and fractionated on a Biogel P4 column (1.7×120 cm) eluted with water. Lissamine™ rhodamine B sulfonyl chloride was purchased from Anaspec or from Acros Organics. All other chemicals used were of chemical grade quality, mostly from Merck or Fluka.

Protein purification

Pichia pastoris strains expressing and secreting tail-less and His6-tagged Gas1p, Gas2p, Gas3p, Gas4p and Gas5p proteins from S. cerevisiae (UniProt KB codes: Gas1, P22146; Gas2, Q06135; Gas3, Q03655; Gas4, Q08271; and Gas5, Q08193) were described previously [18,19]. Recombinant proteins were purified from the supernatants of P. pastoris cultures as described previously [18,19]. Protein concentration was determined by Quant-iT™ Protein Assay Kit (Invitrogen) using BSA as a standard. Fractions of the purified preparations were checked for purity by SDS/PAGE (10% gels). Proteins were stained with Coomassie Blue R-250 or silver nitrate.

Cell fractionation

Diploid strain S. cerevisiae CCY 21-4-13 from the Culture Collection of Yeasts, Institute of Chemistry, Slovak Academy of Sciences, Bratislava, Slovakia was grown in YPD medium [1% (w/v) yeast extract/2% (w/v) peptone/2% (w/v) glucose] at 30 °C on an orbital shaker. The cells were harvested in middle logarithmic phase of growth by centrifugation, washed twice with cold water, suspended in the homogenization buffer (0.1 M citrate, pH 5.0, containing 1 mM each EDTA, PMSF and dithiothreitol) and disintegrated with Ballotini beads (0.2 mm diameter) in a rotatory disintegrator cooled from the outside with an ice-bath. All subsequent steps were carried out at 0–4 °C. The progress of the cell disintegration was checked microscopically. When more than 97% cells had ruptured, the beads were decanted with the homogenization buffer and the suspension was centrifuged at 1500 g for 10 min. The sediment containing the cell walls and unbroken cells was washed twice with the buffer and suspended in 0.1 M citrate buffer, pH 5.0 (buffer A), in a volume corresponding to one-tenth of the wet mass of the starting cells. The supernatant from the first centrifugation was centrifuged at 20000 g for 15 min, the sediment was washed twice with the homogenization buffer, suspended in a small volume of the buffer A so as to obtain a thick suspension, and used as the plasma membrane-enriched fraction. The 20000 g supernatant was spun at 100000 g for 30 min, the sediment washed once with the buffer A, suspended in an equal volume of the same buffer and used as the microsomal fraction. The 100000 g supernatant was considered as the cytosolic fraction. The prepared subcellular fractions could be stored at −70 °C for more than 1 week before the assays, without an appreciable loss of activity.

Synthesis of SR (sulforhodamine)-labelled oligosaccharides

The oligosaccharides were labelled with SR B according to a previously published protocol [20]. The purity of individual SR-labelled oligosaccharides derived from laminarin and from pustulan was checked by TLC on silica gel plates using the solvent system 1-butanol:ethanol:water (5:3:2, by vol.), whereas for the resolution of SR-N-acetyl-D-chitooligosaccharides the solvent system 2-propanol:water:25% aqueous ammonia (7:2:1, by vol.) was used. The chromatograms were viewed under an UV lamp. Molar concentrations of the labelled oligosaccharides were calculated on the basis of the molar absorption coefficient for LRSC (lissamine rhodamine B sulfonyl chloride), ϵ566=85000 M−1·cm−1 (Certificate of Analysis for Lissamine rhodamine B sulfonyl chloride, Lot No. CM16-027 Anaspec).

Assay of BGT activity

The standard assay mixture contained 2.5 mg/ml laminarin, 30–45 μM appropriate SR-labelled oligosaccharide, 50 mM citrate-phosphate buffer, pH 2.5–7.5, and 0.9–3.0 μg of the respective purified Gas protein (or 0.05–0.12 mg of protein in the case of subcellular fractions) in a total volume of 20 μl. Incubations were carried out at 30–37 °C for 30–60 min. The reaction was stopped by addition of 20 μl of 40% (v/v) formic acid. Aliquots of 5 μl were applied in quintuplicate from the stopped reaction mixture on to a chromatographic paper (Whatman 3MM) template into positions exactly fitting the positions of wells in a 96-well ELISA microtitre plate and dried. The paper was then washed with several changes of 66% (v/v) aqueous ethanol containing 5% (v/v) formic acid to remove the unreacted SR-oligosaccharides, and left overnight in 66% ethanol. After drying, the paper was sandwiched between two glass plates and fluorescence was measured in a Synergy HT Multi-Detection Microplate Reader (BioTek Instruments) at λex 530 nm and λem 575 nm.

RESULTS

Setup of a fluorescent assay for BGT

In designing the novel fluorescence-based approach to test the BGT activity, we took advantage of the previous finding that the LamOS of seven glucose residues or shorter were able to function as the acceptor molecules but not as donors [14]. Therefore LamOS of DP 2–7 (L2–L7) were labelled with SR (Figure 1) and used as glucanosyl acceptors with laminarin as the donor substrate. To prove their transglycosylating activity, individual Gas proteins were incubated with laminarin and L6-SR (sulforhodamine-labelled laminarihexaose) and at time intervals aliquots were taken and applied on to the cellulose filter paper Whatman 3MM and dried. Washing the paper with 66% aqueous ethanol removed the low-Mr reaction products and the unused fluorescent substrate, whereas the high-Mr transfer products remained adsorbed on the paper. The binding of the labelled L6-SR to the polymer fraction was indicated by an increase in fluorescence adsorbed to the filter paper and resistant to washing with 66% ethanol (Figure 2). The transfer of the label was due specifically to the enzyme activity, since the sample of L6-SR incubated under the same conditions, but without enzyme showed no changes (results not shown). Gas3p was not tested since it is an inactive member of the Gas family [21].

Structure of SR-labelled laminaritriose (L3-SR)

Figure 1
Structure of SR-labelled laminaritriose (L3-SR)

SR-substituted laminaritriose is shown as a representative of the fluorescently labelled oligosaccharides used in this work.

Figure 1
Structure of SR-labelled laminaritriose (L3-SR)

SR-substituted laminaritriose is shown as a representative of the fluorescently labelled oligosaccharides used in this work.

Transfer of the fluorescent label from L6-SR into high-Mr fractions adherent to filter paper and insoluble in 66% ethanol

Figure 2
Transfer of the fluorescent label from L6-SR into high-Mr fractions adherent to filter paper and insoluble in 66% ethanol

Standard reaction mixtures (see the Experimental section) were incubated with individual Gas proteins under the respective optimal conditions. After indicated time intervals, 5 μl aliquots were taken from the reaction mixtures, applied on to the chromatographic paper Whatman 3MM and dried. The paper was washed overnight in 66% ethanol and photographed under UV light. In the case of reactions catalysed by Gas4p, the incubation times were four times longer than indicated.

Figure 2
Transfer of the fluorescent label from L6-SR into high-Mr fractions adherent to filter paper and insoluble in 66% ethanol

Standard reaction mixtures (see the Experimental section) were incubated with individual Gas proteins under the respective optimal conditions. After indicated time intervals, 5 μl aliquots were taken from the reaction mixtures, applied on to the chromatographic paper Whatman 3MM and dried. The paper was washed overnight in 66% ethanol and photographed under UV light. In the case of reactions catalysed by Gas4p, the incubation times were four times longer than indicated.

In order to analyse multiple protein samples simultaneously, the incubations were performed in 1.5 ml conical Eppendorf tubes and multiple 5 μl aliquots from the stopped reaction mixtures were applied as dot blots on to the filter paper, dried and subsequently washed with 66% ethanol. The quantification of the transglycosylase activity was accomplished in a high-throughput mode using a fluorescent ELISA plate reader. High sensitivity of the fluorescence measurement (less than picomole amounts of SR could be determined) enabled miniaturization of the assay and measurement of multiple parallel applications from one sample ensured high precision and reproducibility of the results. When required, a calibration curve constructed by measuring the fluorescence of a series of known quantities of the oligosaccharide acceptor was used to convert the F.u. (fluorescence units) into the absolute amounts of the products.

The novel fluorescent method could also be applied for the determination of BGT activity in isolated cell walls and membrane fractions prepared from vegetative yeast cells. The highest-specific activity was found in the fraction of plasma membranes, although a significant BGT activity was also detected in microsomes and in the cell walls (Table 1).

Table 1
Distribution of BGT activity in subcellular fractions

The incubation mixtures contained 0.05 M citrate buffer, pH 5, 10 mg/ml laminarin, 20 μM L7-SR and 0.05–0.1 mg of protein in a total volume of 60 μl. The incubation was carried out at 37 °C for 3 h. The results are from a representative experiment.

Fraction Specific activity (pkat/mg of protein) 
Cell walls 0.21 
Plasma membrane 0.46 
Microsomes 0.22 
Cytoplasm 0.04 
Fraction Specific activity (pkat/mg of protein) 
Cell walls 0.21 
Plasma membrane 0.46 
Microsomes 0.22 
Cytoplasm 0.04 

Time-course of the reaction and the effect of enzyme concentration on activity

Progress curves of the reactions catalysed by Gas proteins exhibited typical hyperbolic shapes characteristic for pseudo-first-order reactions (Figure 3). With Gas4p, the reaction rate was constant throughout the 2-h assay, presumably due to the low initial activity of the enzyme. Furthermore, the dependence of activity on the enzyme concentration was linear with all the Gas proteins (Figure 4).

Progress curves of reactions catalysed by individual Gas proteins

Figure 3
Progress curves of reactions catalysed by individual Gas proteins

Standard reaction mixtures were incubated under conditions optimal for the respective Gas protein. The transglycosylation was monitored by the fluorescent method as described in the Experimental section, F.u., arbitrary fluorescence units. The error bars denote the S.D. calculated from five dot blots.

Figure 3
Progress curves of reactions catalysed by individual Gas proteins

Standard reaction mixtures were incubated under conditions optimal for the respective Gas protein. The transglycosylation was monitored by the fluorescent method as described in the Experimental section, F.u., arbitrary fluorescence units. The error bars denote the S.D. calculated from five dot blots.

Dependence of BGT activity on enzyme concentration

Figure 4
Dependence of BGT activity on enzyme concentration

Standard incubation mixtures were incubated at 37 °C for 30 min except the reaction with Gas4p where the incubation was extended to 2 h. Conditions were as in Figure 3, but the amounts of enzyme proteins in the reaction mixtures varied as indicated. The error bars denote the S.D. calculated from five dot blots.

Figure 4
Dependence of BGT activity on enzyme concentration

Standard incubation mixtures were incubated at 37 °C for 30 min except the reaction with Gas4p where the incubation was extended to 2 h. Conditions were as in Figure 3, but the amounts of enzyme proteins in the reaction mixtures varied as indicated. The error bars denote the S.D. calculated from five dot blots.

pH and temperature optima of transglycosylase reactions

The dependence of BGT activity of Gas proteins on pH was tested in the pH range 2.5–7.5. The amount of transfer products produced after 30 min reaction time (after 3 h in the reaction with Gas4p) was determined. As shown in Figure 5, all the Gas proteins showed approximately symmetrical bell-shaped dependences of activity on pH. There was a rather broad maximum with Gas1p; the curve showed a plateau in the pH range between 3.5 and 6 and the Gas1 enzyme still retained over 60% of its maximal activity at pH 6. Gas5p activity was maximal at pH 3.5–4.5, decreased sharply at pH values higher than 4.5 and was <10% at pH 6. By contrast, Gas2p activity reached its maximum at pH 6 and was still over 50% at pH 7. Gas4p had a similar curve but with its maximum shifted towards pH values 5.0–6.0. In conclusion, the pH optimum of Gas5p protein was acidic, whereas the pH optima for Gas2p and Gas4p were closer to neutral (Figure 5). Gas1p had the broader range of pH optimum. Interestingly, these values are consistent with the functional interplay among the Gas isoforms during the yeast life cycle (see the Discussion).

Dependence of BGT activity of Gas proteins on pH

Figure 5
Dependence of BGT activity of Gas proteins on pH

Standard incubation mixtures were buffered with 0.05 M McIlvaine citrate-phosphate buffers and incubated at their respective optimal temperatures for 30–180 min. The error bars denote the S.D. calculated from five dot blots.

Figure 5
Dependence of BGT activity of Gas proteins on pH

Standard incubation mixtures were buffered with 0.05 M McIlvaine citrate-phosphate buffers and incubated at their respective optimal temperatures for 30–180 min. The error bars denote the S.D. calculated from five dot blots.

In the temperature assay, the incubation temperature varied between 20 and 55 °C. The temperature optima for the individual Gas proteins were found to be in the range between 30 and 37 °C (results not shown).

Donor and acceptor substrate specificities of the Gas proteins

In order to determine the preferred donor substrate, individual Gas proteins were incubated with polysaccharides similar to or identical with those present in the yeast cell wall. Of the polysaccharides tested, only laminarin was active as the donor substrate with all the Gas enzymes. Pustulan, carboxymethyl-chitin and/or yeast α-mannan did not yield detectable amounts of transfer products.

The acceptor specificity was tested using equimolar concentrations of SR-labelled hexamers derived from laminarin, pustulan and chitin representing the potential inter-polymeric linkage pairs occurring in the yeast cell wall [22]. It was found that with all Gas proteins, only the glucooligosaccharides of the β-(1,3)-series were active as acceptors, whereas the formation of products with SR-pustulooligosaccharides or SR-N-acetyl-D-chitooligosaccharides was negligible.

To determine the effect of oligosaccharide chain length (expressed as DP) on the acceptor efficiency, the purified Gas proteins were incubated with a series of SR-labelled LamOS containing from 1 to 7 glucose residues (L1-SR–L7-SR). As can be seen from Figure 6, their capability to serve as the acceptor increased with increasing number of glucosyl units in the β-(1,3)-oligosaccharide moiety, the smallest active acceptor being laminaripentaose L5-SR with all Gas proteins.

Effect of oligosaccharide chain length on acceptor effectivity

Figure 6
Effect of oligosaccharide chain length on acceptor effectivity

The reaction mixtures contained equimolar (40 μM) concentrations of the respective SR–LamOS containing 1–7 glucosyl units, and the reactions were carried out with the individual Gas proteins under the respective optimal conditions for 1 h (5 h in the case of the reaction with Gas4p).

Figure 6
Effect of oligosaccharide chain length on acceptor effectivity

The reaction mixtures contained equimolar (40 μM) concentrations of the respective SR–LamOS containing 1–7 glucosyl units, and the reactions were carried out with the individual Gas proteins under the respective optimal conditions for 1 h (5 h in the case of the reaction with Gas4p).

DISCUSSION

The assay

In the initial studies reporting the enzymatic activity of recombinant fungal BGTs, the activities were assayed using a rG13 (reduced G13) LamOS as the substrate. A short incubation generated products compatible with the following reaction scheme:

 
formula
 
formula

where 4<x<7 and E denotes the enzyme.

An incubation time of several hours produced longer oligosaccharides that themselves became either donors or acceptors, resulting in a wide range of products with increasing chain length (DP>30) that gradually became water-insoluble. Monitoring the progress of the reaction and analysis of the reaction products were accomplished by HPAEC [14,16,19]. Although the HPAEC is suitable for characterization of the reaction products, its usefulness for quantitative expression of transglycosylase activity is limited.

In the present study, we describe a highly sensitive fluorescent method to assay the BGT activity of yeast Gas proteins and of GH72 enzymes in general. The method employs principles previously applied successfully to the study of plant transglycosylases such as XET [2325].

The fluorescent assay has several advantages over the HPAEC method applied previously to assess transglycosylating activity of GH72 proteins. Among them is the high sensitivity, the ability to express the overall reaction rates in quantitative terms, the possibility to perform multiple assays simultaneously and finally the simplicity that results in low price-per-assay. Theoretically, every catalytic event leads to a transfer of the label from the pool of acceptor molecules to the pool of donor molecules. It should be taken into account, however, that the fluorescent assay records only those labelled transfer products that are insoluble in 66% ethanol. As the solubility of SR–LamOS in 66% ethanol decreases with increasing oligosaccharide chain length, it is advantageous to use acceptors having their carbohydrate moiety composed of at least five glucosyl units. In such a case, any lengthening of the sugar chain caused by the oligoglucosyl transfer would result in the formation of a product insoluble in 66% ethanol. In contrast with HPAEC, the new fluorescence-based method does not yield information about the qualitative composition of the reaction products.

Using the new fluorescence-based method, it was possible to detect BGT activity also in yeast subcellular fractions. The highest-specific activity of BGT was found in the plasma membrane-enriched fraction, but the presence of significant activity was also detected in microsomes and cell-wall fractions (Table 1). Such a distribution seems to map the typical maturation and transport pathway of GPI proteins from the site of their synthesis in the ER (endoplasmic reticulum) to their final destination. It should be kept in mind, however, that the measured quantities may be distorted by the presence of β-1,3-glucan hydrolysing enzymes that could degrade the product as it is being synthesized. As for the isolation of subcellular fractions, the cells were taken from the logarithmic growth phase, the detected activity presumably belonged to Gas1 and Gas5 proteins that are expressed during vegetative growth, although we cannot exclude the contribution of other enzymes catalysing a transglycosylase reaction in the conditions used [26].

It is noteworthy to point out to a potential versatility of the described fluorescent assay method. Namely, by selecting suitable combinations of donor/acceptor pairs, similar fluorescent assays could also be devised for other types of transglycosylases that possibly are present in fungal cell walls. An effective way for performing such screenings may be the use of polysaccharide microarrays that would enable detection of any type of transglycosylase [24].

Enzymatic properties of Gas proteins

From previous studies, Gas1p, Gas2p, Gas4p and Gas5p were known to be endowed with a similar type of enzymatic activity, but no further analysis of their biochemical properties had been performed [14,16,19]. The fluorescent assay described in the present paper enabled us to ascertain the basic biochemical parameters of the Gas proteins and to establish optimal conditions for determination of their activity.

All of the reactions that were catalysed by Gas proteins possessed attributes characteristic for an enzymatic reaction. The progression curves of the reactions were typical for pseudo-first-order reactions (Figure 3), and there was a linear dependence between the enzyme concentration and the transglycosylating activity (Figure 4).

Substrate specificity

The results obtained using the fluorescent method essentially confirmed the previous findings that all Gas proteins are specific to β-(1,3)-linked glucan and β-(1,3)-oligoglucosides [14,16,19]. The acceptor effectivity of SR–LamOS increased with increasing oligoglucoside chain length, starting with DP 5. This result corroborates well the finding of Hartland et al. [14] that laminaripentaose was the smallest possible acceptor when using purified BGT from A. fumigatus. Interestingly, the recently published three-dimensional structure of Gas2p revealed binding of laminaripentaose in the active site both in the donor and acceptor position [27]. With regard to the donor specificity, our results are in agreement with previous data indicating that β-(1,3)-glucan is the preferred substrate for Gel proteins of A. fumigatus [14] and further extend this conclusion to the complete Gas protein family of S. cerevisiae.

pH optima and their biological significance

Our previous studies on the biological role of Gas proteins revealed that a Gas1/Gas5 protein pair is required in proper cell-wall formation during vegetative growth, whereas Gas2 and Gas4 proteins are essential in spore-wall formation [26]. In accordance with this, GAS1–GAS5 and GAS2–GAS4 pairs have different expression patterns. GAS1–GAS5 pair is expressed during vegetative growth and repressed during sporulation, whereas the GAS2–GAS4 pair exhibits the reverse pattern. As the Gas proteins are anchored through GPI in the outer layer of the yeast plasma membrane or spore membrane, they are exposed to changes of external pH. This probably is the reason why Gas1p and Gas5p, which are expressed in vegetative cells, are most active between pH 3.5 and 5. This is consistent with the fact that the vegetative growth-specific isoforms face the culture medium that is usually acidic due to the proton efflux occurring during yeast growth. The external pH drops from approx. 5.5 to approx. 3 in late exponential phase. Interestingly, the defective phenotype of cells lacking GAS1 gets worse in media buffered at pH 6.5 and pH 6.7 is a semi-lethal condition for growth of the mutant [28]. On the basis of the results described in the present paper, we can hypothesize that the auxiliary function of Gas5p is less effective at pH 6.5 since its activity decreases to less than 10% and therefore cell-wall defects could be aggravated due to the lack of sufficient BGT activity. Thus during vegetative growth the more extended optimum pH range of Gas1p compared with Gas5p could guarantee the presence of BGT activity over a wide range of extracellular pH, whereas the auxiliary function of Gas5p could be effective only in a narrow range of pH values which is more typical of the initial phases of vegetative growth in unbuffered medium. On the other hand, the sporulation-specific Gas2 and Gas4 proteins, which are synthesized in a narrow temporal window coincidentally with the assembly of the spore wall, face the cytoplasmic pH of the diploid cell inside which the spores develop [29]. The optima between pH 5 and 6 suggest that Gas2 and Gas4 proteins work better at pH values closer to the intracellular pH of the yeast cytoplasm, which is between pH 6.5 and 6.9 depending on the strain [30]. It should be also taken into account that alkalization of the medium by accumulation of bicarbonate is essential for sporulation (together with respiratory competency and high cell density) and may induce an increase in intracellular pH [31].

Although factors other than pH of the epiplasm and spore wall, such as different types of β-(1,3)-glucan chains or the presence of small molecules, may explain the requirement for a specialized pair of enzymes during sporulation, pH appears to be an important parameter that affects the activity of Gas2–Gas4 proteins. In previous studies, GAS2 and GAS4 genes were ectopically expressed in vegetative gas1Δ cells by fusion to the GAS1 promoter. GAS4 fully complemented the defective phenotype of gas1Δ mutant in vegetative growth even in medium buffered at pH 5.5, whereas GAS2 could partially suppress the defects of the mutant only when the ambient pH was 6.5 [32]. These findings are consistent with the pH optima for the isolated enzymes reported in the present paper. The character of dependence of activity on the ambient pH may help explain the presence of a specialized pair of Gas enzymes during sporulation. It should be noted that in the human pathogen Candida albicans the GAS1-homologous genes PHR1 and PHR2 are differently regulated by ambient pH with PHR1 expressed at pH values of 5.5 and higher and PHR2 at pH values of 5.5 or lower [33,34]. C. albicans colonizes niches of the human body with very different pHs, slightly alkaline in the bloodstream (pH 7.3) close to neutrality in organs such as the kidney, liver and duodenum, or acidic as in the stomach (pH~2) or vagina (pH 4.5). Thus we speculate that the existence of these two genes may meet the requirement to express at the cell surface glucanosyltransferase activities that work optimally in a wide range of ambient pH's. As pH affects the ionization of the amino acid residues in the active site, the molecular models built on the basis of the recently resolved three-dimensional structure of Gas2p may provide further information [27].

Several studies indicate that the GH72 family of fungal enzymes plays a crucial role in fungal morphogenesis [35,36]. The lack of these enzymes reduces β-(1,3)-glucan incorporation into the expanding cell wall and induces cell-wall changes aimed at counteracting the damage and prevent lysis [37]. The evoked changes cause defects in morphology and impairment in developmental processes such as sporulation or conidiogenesis [26,35]. The fundamental role of GH72 in fungal biology is further testified by the recent finding that GEL4 is an essential gene for A. fumigatus [36], a major human fungal pathogen. In addition, GAS4+ and the GAS2–GAS4 pair are essential for spore viability in Schizosaccharomyces pombe and S. cerevisiae respectively [26,32]. Furthermore, fungal pathogens lacking GH72 enzymes exhibit reduced virulence. An example is provided by the PHR1 null mutant of C. albicans that is avirulent in a mouse model of systemic infection, but uncompromised in its ability to cause vaginal infections in rats, whereas the virulence phenotype of the Phr2 null mutant is the inverse, consistent with the pH-dependent expression of PHR1 and PHR2 genes [38]. Moreover, the PHR1 null mutant manifests defects in adhesion to and invasion of reconstituted human epithelia, two processes required for the establishment of infections [38,39]. Despite the existence of a high degree of redundancy, it is envisaged that the active site of most GH72 members could have sufficient similarity to be targeted by inhibitors of the same type.

The armamentarium of drugs to fight fungal infections is still limited to a few compounds. The new fluorescent assay for BGTs could provide a simple tool for the detection, identification and characterization of GH72 family enzymes in diverse fungal species as well as for screening and identification of potential inhibitors of BGT activity, which may lead to the development of new antifungal drugs and to new therapeutic strategies.

Abbreviations

     
  • BGT

    β-(1,3)-glucanosyltransglycosylase

  •  
  • DP

    degree of polymerization

  •  
  • F.u.

    fluorescence units

  •  
  • GH72

    family 72 of glycosylhydrolases

  •  
  • GPI

    glycosylphosphatidylinositol

  •  
  • HPAEC

    high-performance anion-exchange chromatography

  •  
  • LamOS

    laminarioligosaccharides

  •  
  • L6-SR

    sulforhodamine-labelled laminarihexaose

  •  
  • SR

    sulforhodamine

  •  
  • XET

    xyloglucan endotransglycosylose

AUTHOR CONTRIBUTION

Marián Mazáň performed most of the biochemical experiments and drew the Figures. Enrico Ragni prepared the recombinant P. pastoris strains used in the study. Vladimír Farkaš and Laura Popolo designed and supervised the research and wrote the manuscript with contributions from all authors.

We thank L. Fischerová and T. Lipka for technical assistance. We thank Eva Stratilová for drawing Figure 1 and Michael J. Bailey for correcting the English style.

FUNDING

This work was supported, in part, by the Grant Agency for Science VEGA (Slovakia) [grant number 2/0011/09] to V.F. and by Universitá degli Studi di Milano [grant number P.U.R. 2009] to L.P. Enrico Ragni was a recipient of a type A contract from Universitá degli Studi di Milano. The support to the Centre of Excellence for Glycomics [ITMS 26240120031] from the Research and Development Operational Programme of the ERDF is gratefully acknowledged.

References

References
1
Henrissat
B.
Davies
G.
Structural and sequence-based classification of glycoside hydrolases
Curr. Opin. Struct. Biol.
1997
, vol. 
7
 (pg. 
637
-
644
)
2
Henrissat
B.
Davies
G. J.
Glycoside hydrolases and glycosyltransferases. Families, modules, and implications for genomics
Plant Physiol.
2000
, vol. 
124
 (pg. 
1515
-
1519
)
3
Cantarel
B. L.
Coutinho
P. M.
Rancurel
C.
Bernard
T.
Lombard
V.
Henrissat
B.
The carbohydrate-active enzymes database (CAZy): an expert resource for glycogenomics
Nucleic Acids Res.
2009
, vol. 
37
 (pg. 
233
-
238
)
4
Mazur
P.
Morin
N.
Baginsky
W.
el-Sherbeini
M.
Clemas
J. A.
Nielsen
J. B.
Foor
F.
Differential expression and function of two homologous subunits of yeast 1,3-beta-D-glucan synthase
Mol. Cell. Biol.
1995
, vol. 
15
 (pg. 
5671
-
5681
)
5
Gastebois
A.
Mouyna
I.
Simenel
C.
Clavaud
C.
Coddeville
B.
Delepierre
M.
Latgé
J. P.
Fontaine
T.
Characterization of a new beta(1–3)-glucan branching activity of Aspergillus fumigatus
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
2386
-
2396
)
6
Latgé
J. P.
The cell wall: a carbohydrate armour for the fungal cell
Mol. Microbiol.
2007
, vol. 
66
 (pg. 
279
-
290
)
7
Cabib
E.
Blanco
N.
Grau
C.
Rodrígues-Peňa
J. M.
Arroyo
J.
Crh1p and Crh2p are required for the cross-linking of chitin to beta(1–6)glucan in the Saccharomyces cerevisiae cell wall
Mol. Microbiol.
2007
, vol. 
63
 (pg. 
921
-
935
)
8
Cabib
E.
Farkaš
V.
Kosík
O.
Blanco
N.
Arroyo
J.
McPhie
P.
Assembly of the yeast cell wall. Crh1p and Crh2p act as transglycosylases in vivo and in vitro
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
29859
-
29872
)
9
Rolli
E.
Ragni
E.
Calderon
J.
Porello
S.
Fascio
U.
Popolo
L.
Immobilization of the glycosylphosphatidylinositol-anchored Gas1 protein into the chitin ring and septum is required for proper morphogenesis in yeast
Mol. Biol. Cell
2009
, vol. 
20
 (pg. 
4856
-
4870
)
10
Fry
S. C.
Smith
R. C.
Renwick
K. F.
Martin
D. J.
Hodge
S. K.
Mathews
K. J.
Xyloglucan endotransglycosylase: a new wall-loosening enzyme activity from plants
Biochem. J.
1992
, vol. 
282
 (pg. 
821
-
826
)
11
Nishitani
K.
Tominaga
R.
Endo-xyloglucan transferase, a novel class of glycosyltransferase that catalyzes transfer of a segment of xyloglucan molecule to another xyloglucan molecule
J. Biol. Chem.
1992
, vol. 
267
 (pg. 
21058
-
21064
)
12
Farkaš
V.
Sulová
Z.
Stratilová
E.
Hanna
R.
Maclachlan
G.
Cleavage of xyloglucan by nasturtium seed xyloglucanase and transglycosylation to xyloglucan subunit oligosaccharides
Arch. Biochem. Biophys.
1992
, vol. 
298
 (pg. 
365
-
370
)
13
Hrmova
M.
Farkaš
V.
Lahnstein
J.
Fincher
G. B.
A barley xyloglucan xyloglucosyl transferase covalently links xyloglucan, cellulosic substrates, and (1,3;1,4)-beta-D-glucans
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
12951
-
12962
)
14
Hartland
R. P.
Fontaine
T.
Debeaupuis
J.-P.
Simenel
C.
Delepierre
M.
Latgé
J.-P.
A novel beta-(1–3)-glucanosyltransferase from the cell wall of Aspergillus fumigatus
J. Biol. Chem.
1996
, vol. 
271
 (pg. 
26843
-
26849
)
15
Sinnott
M. L.
Catalytic mechanism of enzymatic glycosyl transfer
Chem. Rev.
1990
, vol. 
90
 (pg. 
1171
-
1202
)
16
Mouyna
I.
Fontaine
T.
Vai
M.
Monod
M.
Fonzi
W. A.
Diaquin
M.
Popolo
L.
Hartland
R. P.
Latgé
J. P.
Glycosylphosphatidylinositol-anchored glucanosyltransferases play an active role in the biosynthesis of the fungal cell wall
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
14882
-
14889
)
17
Lesage
G.
Bussey
H.
Cell wall assembly in Saccharomyces cerevisiae
Microbiol. Mol. Biol. Rev.
2006
, vol. 
70
 (pg. 
317
-
343
)
18
Carotti
C.
Ragni
E.
Palomares
O.
Fontaine
T.
Tadeschi
G.
Rodrígez
R.
Latgé
J. P.
Vai
M.
Popolo
L.
Characterization of recombinant forms of the yeast Gas1 protein and identification of residues essential for glucanosyltransferase activity and folding
Eur. J. Biochem.
2004
, vol. 
271
 (pg. 
3635
-
3645
)
19
Ragni
E.
Fontaine
T.
Gissi
C.
Latgé
J. P.
Popolo
L.
The Gas family of proteins of Saccharomyces cerevisiae: characterization and evolutionary analysis
Yeast
2007
, vol. 
24
 (pg. 
297
-
308
)
20
Kosík
O.
Farkaš
V.
One-pot fluorescent labeling of xyloglucan oligosaccharides with sulforhodamine
Anal. Biochem.
2008
, vol. 
375
 (pg. 
232
-
236
)
21
Rolli
E.
Ragni
E.
Rodrígues-Peňa
J. M.
Arroyo
J.
Popolo
L.
GAS3, a developmentally regulated gene, encodes a highly mannosylated and inactive protein of the Gas family of Saccharomyces cerevisiae
Yeast
2010
, vol. 
27
 (pg. 
597
-
610
)
22
Kollár
R.
Reinhold
B. B.
Petrakova
E.
Yeh
H. J.C.
Ashwell
G.
Drgonova
J.
Kapteyn
J. C.
Klis
F. M.
Cabib
E.
Architecture of the yeast cell wall: β(1-6)-glucan interconnects mannoprotein, β(1-3)-glucan and chitin
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
17762
-
17775
)
23
Fry
S. C.
Novel ‘‘dot-blot’’ assays for glycosyltransferases and glycosylhydrolases: optimization for xyloglucan endotransglycosylase (XET) activity
Plant J.
1997
, vol. 
11
 (pg. 
1141
-
1150
)
24
Kosík
O.
Auburn
R.
Russell
S.
Stratilová
E.
Garajová
S.
Hrmova
M.
Farkaš
V.
Polysaccharide microarrays for high-throughput screening of transglycosylase activities in plant extracts
Glycoconj. J.
2010
, vol. 
27
 (pg. 
79
-
87
)
25
Kosík
O.
Garajová
S.
Matulová
M.
Řehulka
P.
Stratilová
E.
Farkaš
V.
Effect of the label of oligosaccharide acceptors on kinetic parameters of nasturtium seed xyloglucan endotransglycosylase
Carbohydr. Res.
2011
, vol. 
346
 (pg. 
357
-
361
)
26
de Medina-Redondo
M.
Arnáiz-Pita
Y.
Fontaine
T.
Del Rey
F.
Latgé
J. P.
De Aldana
C. R.V.
The beta-1,3-glucanosyltransferase gas4p is essential for ascospore wall maturation and spore viability in Schizosaccharomyces pombe
Mol. Microbiol.
2008
, vol. 
68
 (pg. 
1283
-
1299
)
27
Hurtado-Guerrero
R.
Schüttelkopf
A. W.
Mouyna
I.
Ibrahim
A. F. M.
Shepherd
S.
Fontaine
T.
Latgé
J.-P.
van Aalten
D. M.F.
Molecular mechanisms of yeast cell wall glucan remodelling
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
8461
-
8469
)
28
Vai
M.
Orlandi
I.
Cavadini
P.
Alberghina
L.
Popolo
L.
Candida albicans homologue of GGP1/GAS1 is functional in Saccharomyces cerevisiae and contains determinants for glycosylphosphatidylinositol attachment
Yeast
1996
, vol. 
12
 (pg. 
361
-
368
)
29
Rolli
E.
Ragni
E.
de Medina-Redondo
M.
Arroyo
J.
Vazquez de Aldana
C.
Popolo
L.
Expression, stability and replacement of glucan-remodeling enzymes during the developmental transitions in Saccharomyces cerevisiae
Mol. Biol. Cell
2011
, vol. 
22
 (pg. 
1585
-
1598
)
30
Marešová
L.
Hošková
B.
Urbánková
E.
Chaloupka
R.
Sychrová
H.
New applications of pHluorin-measuring intracellular pH of prototrophic yeasts and determining changes in the buffering capacity of strains with affected potassium homeostasis
Yeast
2010
, vol. 
27
 (pg. 
317
-
325
)
31
Ohkuni
K.
Hayashi
M.
Yamashita
I.
Bicarbonate-mediated social communication stimulates meiosis and sporulation of Saccharomyces cerevisiae
Yeast
1998
, vol. 
14
 (pg. 
623
-
631
)
32
Ragni
E.
Coluccio
A.
Rolli
E.
Rodrígues-Peňa
J. M.
Colasante
G.
Arroyo
J.
Neiman
A. M.
Popolo
L.
GAS2 and GAS4, a pair of developmentally regulated genes required for spore wall assembly in Saccharomyces cerevisiae
Eukaryot. Cell
2007
, vol. 
6
 (pg. 
302
-
316
)
33
Saporito-Irwin
S. M.
Birse
C. E.
Sypherd
P. S.
Fonzi
W. A.
PHR1, a pH-regulated gene of Candida albicans, is required for morphogenesis
Mol. Cell. Biol.
1995
, vol. 
15
 (pg. 
601
-
611
)
34
Fonzi
W. A.
PHR1 and PHR2 of Candida albicans encode putative glycosidases required for proper cross-linking of beta-1,3- and beta-1,6-glucans
J. Bacteriol.
1999
, vol. 
181
 (pg. 
7070
-
7079
)
35
Mouyna
I.
Morelle
W.
Vai
M.
Monod
M.
Léchenne
B.
Fontaine
T.
Beauvais
A.
Sarfati
J.
Prévost
M.-C.
Henry
C.
, et al. 
Deletion of GEL2 encoding for a beta(1–3)glucanosyltransferase affects morphogenesis and virulence in Aspergillus fumigatus
Mol. Microbiol.
2005
, vol. 
56
 (pg. 
1675
-
1688
)
36
Gastebois
A.
Fontaine
T.
Latgé
J.-P.
Mouyna
I.
beta(1–3)Glucanosyltransferase Gel4p is essential for Aspergillus fumigatus
Eukaryot. Cell
2010
, vol. 
9
 (pg. 
1294
-
1298
)
37
Popolo
L.
Gualtieri
T.
Ragni
E.
The yeast cell-wall salvage pathway
Med. Mycol.
2001
, vol. 
39
 (pg. 
111
-
121
)
38
De Bernardis
F.
Mühlschlegel
F. A.
Cassone
A.
Fonzi
W. A.
The pH of the host niche controls gene expression in and virulence of Candida albicans
Infect. Immun.
1998
, vol. 
66
 (pg. 
3317
-
3325
)
39
Calderon
J.
Zavrel
M.
Ragni
E.
Fonzi
W. A.
Rupp
S.
Popolo
L.
PHR1, a pH-regulated gene of Candida albicans encoding a glucan-remodeling enzyme, is required for adhesion and invasion
Microbiology
2010
, vol. 
156
 (pg. 
2484
-
2494
)