Heterologous Expression of the Saccharomyces cerevisiae PGU1 Gene in Schizosaccharomyces pombe Yields an Enzyme with More Desirable Properties for the Food Industry

Abstract

The Saccharomyces cerevisiae PGU1 gene was successfully expressed in Schizosaccharomyces pombe. The optimum pH and temperature for the recombinant enzyme were 5 and 40°C, respectively, these being around 0.5 U higher and 5°C lower than those shown by the native enzyme. The K_m value was about fourfold higher than that of the S. cerevisiae enzyme. The recombinant endopolygalacturonase was more efficient in reducing the viscosity of polygalacturonic acid and was also more stable at different pHs and temperatures than the native enzyme.

Pectic substances are plant polysaccharides that contain a large proportion of galacturonic acid subunits linked by α-1,4-glycosidic bonds. Pectin and pectate are pectic substances that are either partially methyl esterified or free of methyl groups, respectively. The enzymes able to degrade pectic substances are pectin esterases, which release the methyl ester groups from pectin, and depolymerases, including pectate or pectin lyases and polygalacturonases (4).

Pectinases have many applications in industry, particularly in fruit and vegetable processing. These applications include such procedures as extraction and clarification of fruit juices (e.g., grape and apple juice), as well as textile processing, paper making, coffee and tea fermentation, and pectic wastewater treatment. Additionally, pectolytic enzymes can be used in complex mixtures with other hydrolases in the formulation of animal feeds (7).

To date, commercial pectinase preparations consist of enzymatic mixtures of fungal origin, where, in addition to having different pectinolytic activities, certain enzymes with undesirable properties (methylesterases, arabinofuranosidases, and amylases) may also be found. In view of this, it would be of advantage to produce the different pectic enzymes separately and then mix them in the downstream process as required. It is interesting to note that total hydrolysis of certain substrates often requires the concerted action of several enzymes in the correct proportions (2).

In Saccharomyces cerevisiae, the PGU1 gene codes for an endopolygalacturonase (1). In the present paper, we report the cloning, expression, and efficient secretion of the PGU1 gene in Schizosaccharomyces pombe and the characterization of the recombinant enzyme as a product with better properties for industrial applications.

Expression of the PGU1 gene in S. pombe.

A XhoI-BamHI fragment containing the S. cerevisiae PGU1 gene was recovered from plasmid pBSK-PGU1 (1) and ligated to S. pombe pREP3X vector (9) by standard procedures (13), thus generating plasmid pREP3X-PGU1. In this plasmid, the PGU1 gene was cloned under the control of the thiamine-repressible nmt1 promoter (9). The recombinant plasmid was amplified in Escherichia coli DH5α and purified according to previously described methods (13). Cells of S. pombe (h⁺ ade6-M210 leu1-32 ura 40-18 PG⁻) grown in YEPD medium (1% yeast extract, 1% peptone, 2% glucose) were transformed with the pREP3X-PGU1 plasmid by the lithium acetate protocol (6). After 4 days at 30°C, several recombinant colonies appeared on SD−ade/ura selective medium plates (0.7% yeast nitrogen base without amino acids, 2% glucose, 20 mg of adenine per liter, 20 mg of uracil per liter). One of these clones (USC10) was grown in 3-liter flasks containing 500 ml of selective liquid medium supplemented with thiamine (5 μg/ml) for 30 h at 30°C and 200 rpm. Then the cells were harvested and resuspended in the same medium without thiamine. Endopolygalacturonase activity in the supernatant fluid was determined. Samples for enzymatic reactions were filtered through 0.22-μm-pore-diameter membranes and dialyzed for 48 h in 0.05 M sodium acetate buffer (pH 4.5 to 5). Typical reaction mixtures containing 0.5 ml of sample and 0.5 ml of 0.5% polygalacturonic acid in 0.05 M sodium acetate buffer (pH 4.5 to 5) were incubated at 37°C for 1 h. Polygalacturonase activity was measured by evaluating the reducing power according to the method of Somogyi (15), as modified by Nelson (11). One unit of activity was defined as the amount of enzyme that released 1 nmol of glucose or its equivalent in reducing power per hour at 37°C. The results are shown in Fig. 1. Maximum enzymatic activity (7,200 U/ml) was found at 35 h after derepression. The level of expression found in the culture medium was about 40-fold higher than that shown by the wild-type S. cerevisiae strains and 4-fold higher than that found when the same gene was overexpressed in S. cerevisiae (1). When the location of the enzyme was tested according to the procedure described by Villa et al. (16), it was found that endopolygalacturonase activity occurred mainly in the growth medium, and only certain levels of intracellular activity (ca. 1%) were detected. This observation shows that the enzyme is efficiently secreted into the medium when the native signal peptide of S. cerevisiae is used in S. pombe. The secretion of gene products in S. cerevisiae is mediated by a hydrophobic signal/leader sequence. The signal/leader sequence of Pgu1p, MISANSLLISTLCAFAIATPLSKR, has the KR²⁴ sequence, which is probably recognized and cleaved by the S. cerevisiae KEX2-like endopeptidase, krp, of S. pombe (3).

FIG. 1.

Growth and endopolygalacturonase secretion in the recombinant S. pombe strain USC10 in shake flasks (A and B) and in a 30-liter fermentor (C and D). (A) Time course (□) and endopolygalacturonase activity (○). (B) Enzymatic activity after induction (○). (C) Time course in selective (□) or low-cost (▪) medium and enzymatic activity in selective (○) or low-cost (•) medium. (D) Enzymatic activity after induction in selective (○) and low-cost (•) medium. Triangles in panel C show the percentage of cells that retain the plasmid.

Biochemical characterization of the enzyme produced by S. pombe.

The enzyme produced by the recombinant S. pombe strain was partially purified by gel filtration chromatography and biochemically compared with the native enzyme produced by S. cerevisiae strain USC4 (MATα leu2-3 leu2-112 his4 PG⁺). Concentrated supernatants (5 ml) were applied to a Sephacryl S-200 column (LKB 80 by 3.5 cm), equilibrated with 0.05 M sodium acetate buffer (pH 5). Chromatography was carried out with the same buffer, and 3.5-ml fractions were collected at a flow rate of 0.5 ml/min. Bovine serum albumin (67 kDa), ovalbumin (45 kDa), chymotrypsinogen A (25 kDa), and lysozyme (14 kDa) were used as molecular mass standards. The recombinant enzyme eluted as a single peak with a elution volume/void volume (V_e/V₀) value of 1.36, indicating the presence of only one pectic enzyme. The molecular mass, determined as described by Whitaker (17), was estimated at 38.8 kDa. The molecular mass of the enzyme produced by S. cerevisiae is 37.3 kDa. This suggests that the glycosylation that occurs during PGU1 expression in S. pombe might differ from that taking place in S. cerevisiae, resulting in different N-glycosylated enzymatic forms (10). The values of the apparent kinetic parameters K_m and V_max (8) obtained for polygalacturonic acid were 2.19 mg/ml and 30.4 nmol/min, respectively, for the recombinant enzyme and 0.55 mg/ml and 0.23 nmol/min, respectively, for the native enzyme. The apparent K_m of the recombinant enzyme was roughly fourfold higher than that found for the native enzyme, indicating a lower affinity for polygalacturonic acid in the case of the enzyme produced by S. pombe.

The optimum temperature and pH were also determined for the recombinant endopolygalacturonase and compared with those of the enzyme produced by S. cerevisiae USC4. The optimal temperature was tested in the range 10 to 60°C, and the optimal pH was determined in the range 4.0 to 9.0. Sodium acetate buffer (0.05 M) was used for the pH range 4.0 to 6.0, and 0.05 M phosphate buffer or Tris-HCl was used for the pH range 6.0 to 9.0. All assays were carried out in triplicate; the results are shown in Fig. 2A and B. The optimum temperature for the recombinant enzyme (40°C) was around 5°C lower than that found for the enzyme produced by S. cerevisiae, whereas the optimum pH was around 0.5 U higher for the recombinant enzyme (pH 5) than for the native enzyme. In both cases, the enzyme was active in a pH range of between 4 and 6. Interestingly, the enzyme produced by the USC10 strain was stable at higher pHs. Thus, more than 90 and 65% of the activity remained after 60 min of incubation at pH 7 and 8, respectively, and even at pH 9, the enzyme retained 40% of its activity after 40 min of treatment (Fig. 2D). In contrast, the native enzyme was completely inactivated after 5 min at pH 8. The recombinant enzyme was also more stable at higher temperatures, retaining 100% of activity after 10 min of incubation at 60°C, whereas the native enzyme was almost completely denatured under the same conditions. At 50°C, the temperature used in some industrial processes (5), the enzyme retained 80% of its activity after 30 min of treatment (Fig. 2C). These differences in the biochemical enzymatic properties could be due to overglycosylation of the recombinant enzyme when produced in S. pombe. In this sense, similar results have been found by other authors (12) upon expression of a cellobiohydrolase II from Trichoderma reesei in S. pombe.

FIG. 2.

Effect of temperature and pH on the activity and stability of the endopolygalacturonase. Effect of temperature (A) and pH (B) on the activity of the native (□) and the recombinant (○) enzyme. The highest activity was designated as 100%. (C) Decay in activity over time at 60°C for the native enzyme (□) and at 50°C (▵) and 60°C (○) for the recombinant endopolygalacturonase. (D) Decay in activity over time at pH 8 for the native enzyme (□) and at pH 7 (•), 8 (▪), and 9 (▴) for the recombinant enzyme. The endopolygalacturonase activity prior to preincubations at different pHs and temperatures was taken as 100%.

Hydrolytic behavior of the recombinant enzyme.

The hydrolytic behavior of the recombinant endopolygalacturonase was monitored by viscosimetry and compared to that of the native enzyme. Viscosimetry assays were performed in triplicate as described previously (14). Figure 3 shows that the endopolygalacturonase produced by the recombinant S. pombe strain (8 U/ml) reduced the viscosity of a 0.5% polygalacturonic acid solution by 50% in 35 min, whereas the same amount of the native enzyme did the same in about 75 min. As in the case of the biochemical parameters, these differences could be due to the different glycosylation pattern of the recombinant enzyme or even to differences in protein folding between the native and the recombinant enzyme, rendering the latter more efficient at hydrolyzing polygalacturonic acid even with a higher apparent K_m.

FIG. 3.

Reduction in viscosity of 0.5% polygalacturonic acid solution at 37°C by the native (□) and the recombinant (•) enzyme. ▪ and ○, native and recombinant inactivated enzymes, respectively.

Scaling up of enzyme production in a fermentor.

In order to optimize a low-cost medium for enzyme production, eight different media prepared by combination of different concentrations of beet molasses and ammonium sulfate were tested. In all cases, the medium was buffered with 0.05 M sodium acetate (pH 5) and supplemented with thiamine. SD−ade/ura medium supplemented with thiamine was used as a control. Three-liter flasks containing 500 ml of each medium were inoculated with the recombinant S. pombe strain and incubated at 30°C with stirring (200 rpm) for 2 days. Then, the cells were harvested, washed, and resuspended in 100 ml of the same medium without thiamine in order to allow gene expression. Enzymatic activity was measured in the expression medium over 48 h. As shown in Table 1, the highest enzymatic activity (5,500 U/ml) was found 25 h after derepression in the medium containing 6% beet molasses and 0.5% ammonium sulfate.

TABLE 1.

Effect of medium composition on enzyme production by strain USC10

% Ammonium sulfate	Enzymatic activity (U/ml) on beet molassesa
	4%	6%	8%	10%
0.5	3,250	5,500	2,100	2,600
1	5,200	4,090	2,750	1,000

^aNumbers indicate enzymatic activity after derepression.

In order to scale up the production of the recombinant endopolygalacturonase, two fermentation processes were set up using the optimized low-cost medium in comparison with SD−ade/ura medium. Twenty liters of medium in a Biostat C (B. Braun Biotech International) 30-liter-capacity fermentor were inoculated at an initial cell density of 1.5 × 10⁵ cells per ml. During the entire process, the temperature was kept at 30°C (pH 5) with stirring at 200 rpm. The air supply (measured with a A-10360-30 oxygen analyzer from Cole-Parmer) for maximum enzymatic activity was previously optimized in shake flasks and estimated to lie between 5.4 and 5.7 ppm. The airflow in the fermentor was increased from 10 to 25 liters/min during the process to maintain this value constant. At the beginning of the exponential phase, fermentation was stopped, and the cells were harvested, washed, and resuspended in the expression medium. Samples were taken at different times in order to estimate cell density (CFU on YEPD and/or SD−ade/ura plates), plasmid loss (by the differences in CFU between YEPD and SD−ade/ura plates), and enzymatic activity. The results for strain growth and enzyme production are shown in Fig. 1C and D. When growth was carried out in a fermentor, the recombinant strain exhibited a 15- to 20-h delay in growth compared to the pattern obtained in shake flasks (in both SD−ade/ura and low-cost medium), with the stationary phase being reached 50 h after inoculation. Regarding enzyme production, maximum enzyme activity appeared 23 h after derepression in both media (about 15 h earlier than in shake flasks), the amount of enzyme produced in the fermentor being very similar to that in shake flasks. The recombinant S. pombe strain produced a smaller amount of endopolygalacturonase in the low-cost medium (5,400 U/ml) than in SD−ade/ura medium (7,350 U/ml), probably due to plasmid loss under the nonselective conditions. Plasmid stability was estimated during the fermentation process, and the results revealed that the cells were gradually losing the plasmid (Fig. 1C). At the end of the fermentation, only 20% of cells had retained the plasmid. However, the amount of recombinant product in the low-cost medium was three times higher than when the PGU1 gene was overexpressed in S. cerevisiae (1).

The S. pombe recombinant strain that expresses the PGU1 gene produces and secretes an endopolygalacturonase in an efficient manner, both in shake-flasks and in a fermentor, with either selective or low-cost medium. This means that the process is industrially feasible. The recombinant enzyme produced by S. pombe showed small but important differences from the native enzyme in certain biochemical properties, such as optimum temperature and pH, as well as stability. However, these do not affect its possible use in food applications. Indeed, the recombinant endopolygalacturonase exhibits improved properties for the food industry, being more stable at different temperatures and pHs than the native enzyme. It also shows a better ability to reduce the viscosity of the substrate, hydrolyzing the same amount of polygalacturonic acid in half the time required by the native enzyme. Due to its efficiency and its thermostability at 50°C, the recombinant enzyme is expected to be attractive for a wide-range of industrial applications, especially for pomace liquefaction (5), a newer and alternative technique for the production of apple juice.