Production of High Galacto-oligosaccharides by Pectinex Ultra SP-L

Feb 8, 2017 - ... https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... A rational optimization for the synthesis of galacto-oligosa...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JAFC

Production of High Galacto-oligosaccharides by Pectinex Ultra SP-L: Optimization of Reaction Conditions and Immobilization on GlyoxylFunctionalized Silica Isabel González-Delgado, Yolanda Segura, Gabriel Morales,* and María-José López-Muñoz Department of Chemical and Energy Technology, Chemical and Environmental Technology, Mechanical Technology and Analytical Chemistry, Universidad Rey Juan Carlos, C/Tulipán s/n, E-28933 Móstoles, Madrid, Spain S Supporting Information *

ABSTRACT: A rational optimization for the synthesis of galacto-oligosaccharides (GOS) from lactose catalyzed by βgalactosidase from Aspergillus aculeatus, included in the commercial product Pectinex Ultra SP-L, has been performed by using experimental design and surface response methodology. This accurate tool optimized empirical production of the most desired high-GOS (tri-GOS and tetra-GOS) up to 16.4% under the following reaction conditions: 59 °C, 4 U/mL free enzyme concentration, pH 6.5, 250 g/L initial lactose concentration, and 20 h of reaction. The statistical analysis revealed temperature and initial lactose concentration as critical parameters. The successful immobilization of the enzyme on a glyoxyl-functionalized porous silica support slightly increased the yield toward high-GOS (17.6%), especially tri-GOS yield (15.3%), under the optimized reaction conditions as compared to the free enzyme. Furthermore, the promotion of the transgalactosylation reaction toward tri-GOS production increased 1.5-fold the productivity of high-GOS as compared to the free enzyme. KEYWORDS: galacto-oligosaccharides (GOS), β-galactosidase, Pectinex Ultra SP-L, glyoxyl-functionalized silica, transgalactosylation, experimental design

1. INTRODUCTION Galacto-oligosaccharides (GOS) are oligosaccharides nondigestible by the human gastrointestinal tract that can act as prebiotics by stimulating the proliferation of intestinal lactic acid bacteria and bifidobacteria.1−3 Furthermore, because of their relative sweetness, low caloric content,2 and their stability at relatively high temperatures and at wide ranges of pH, GOS can be used in a variety of industrial processes. Therefore, because of all these favorable properties, as well as the current increasing demand for foods with demonstrable health benefits, the production of GOS is provided with a promising future.4−6 Nowadays, industrial production of GOS is mainly carried out by the enzyme β-galactosidase (EC 3.2.1.23) using lactose as substrate. GOS consist of a variable number of galactose units (usually in the range from 1−5) linked to a terminal glucose unit through glycosidic bonds.3,4 The enzymatic synthesis of GOS proceeds first through the hydrolysis of the substrate lactose into its constituent glucose and galactose units and then via transgalactosylation reactions where the chain of oligosaccharide grows in length as new galactose units are incorporated. The latter process occurs when the galactosyl acceptor is a saccharide (e.g., glucose, galactose, lactose, or shorter-chain GOS). If the galactosyl acceptor is a water molecule, a reaction clearly promoted in the presence of high amounts of water, then the hydrolysis takes place by not only affecting the starting lactose, but also leading to the nondesired hydrolytic cleavage of the GOS already produced. In this case, the formation of long-chain or high-GOS is hindered.3,4,7,8 Despite the fact that β-galactosidases can catalyze both hydrolysis and transgalactosylation transformations, such enzymes have been traditionally used because of their © XXXX American Chemical Society

hydrolysis capacity in the production of lactose-free dairy products.3,9,10 Nevertheless, the desired transgalactosylation activity can be enhanced over the hydrolytic activity by a proper selection of the reaction conditions. In this sense, the use of high initial lactose concentrations, moderate temperatures, and the minimum amount of water may aid in favoring the transgalactosylation reaction over the hydrolysis.3,11 On the other hand, the yield to GOS as well as the final products distribution is deeply affected by several other factors such as the enzyme source, the pH, and the time of reaction.4 Many of these parameters have been previously analyzed separately, especially for the β-galactosidase from Kluyveromyces lactis.11−13 However, this particular enzyme presents poor transgalactosylation activity, and other food-grade enzymes have been proposed as more appropriate for the synthesis of GOS, such as BbgIV from Bif idobacterium bifidum14 or the commercial enzyme preparation Pectinex Ultra SP-L, which is produced by the mesophilic organism Aspergillus aculeatus. This enzyme could work at higher temperatures than the one isolated from Kluyveromyces lactis15 and potentially lead to an increased GOS yield.16 Additionally, the complexity of the overall process of GOS production makes necessary the use of different strategies for the analysis and optimization of the variables simultaneously affecting the process. The experimental design methodology is Received: Revised: Accepted: Published: A

December 3, 2016 February 7, 2017 February 8, 2017 February 8, 2017 DOI: 10.1021/acs.jafc.6b05431 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

(Steinheim, Germany). Tetrahydrofuran (THF, 99.5%) was acquired from Fisher and potassium chloride (99%), acetone (99.8%), and sodium borohydride (98%) from Scharlab. All the reactants were used as received. Pierce 660 nm Protein Assay was supplied by Thermo Scientific Inc. (Waltham, USA) and bovine serum albumin (BSA) 2 mg/mL by BioRad (California, USA). Pectinex Ultra SP-L, a commercial enzyme preparation produced by Aspergillus aculeatus, was a gift from Novozymes A/S (Bagsvaerd, Denmark). Silica support Sipernat 50S was kindly donated by Evonik Resource Efficiency GmbH. 2.2. Preparation and Characterization of Glyoxyl-Functionalized Silica Support. Sipernat 50S silica was dried overnight by vacuuming at 55 °C and 0.2 bar. The dried silica was then grafted with epoxy groups as follows: 1 g of silica was suspended in 50 mL of dry toluene, 1 mL of GPTMS and 150 μL of triethylamine (as catalyst for the grafting reaction) were added under stirring,36 and the mixture was refluxed under nitrogen for 4 h. Then the solid was filtered off and washed thoroughly with THF. Finally, the epoxy-modified silica was dried 12 h at 55 °C and 0.2 bar. The presence of epoxy groups on the support was confirmed by FT-IR spectroscopy and TG analysis. Afterward, the hydrolysis of epoxy groups was carried out by mixing 1 g of solid with 10 mL of 0.1 M H2SO4 for 2 h at 85 °C. The resultant glyceryl-epoxide silica, after being washed with water/acetone (70:30 v/v) and dried, was subjected to a further oxidation step of the glyceryl moieties by reacting with 0.03 M NaIO4 for 2 h at RT.31,32 In this way, 1 mmol of glyoxyl groups per gram of support was generated, according to the best results reported by Bernal et al.37 The formation of glyoxyl groups on the silica surface was quantified by back-titration with NaHCO3/KI, which measured the difference in absorbance at 405 nm of the supernatant before and after the oxidation process.38 2.3. Enzyme Immobilization on Glyoxyl-Silica Support. The commercial enzyme extract Pectinex Ultra SP-L was used as received, without previous isolation of the β-galactosidase enzyme. The immobilization was carried out by suspending 0.5 g of support in 5 mL of Pectinex Ultra SP-L diluted in potassium bicarbonate buffer (100 mM, pH 10). The suspension was maintained 2 h at 4 °C under vigorous stirring. To avoid the inactivation of the enzyme at the high operational pH, required for the immobilization but far from its optimum pH of 6.5, it was necessary to add protective agents. Thus, galactose (100 mM) and glycerol (25% v/v) were added.30 Afterward, the Schiff’s bases generated on the support were reduced with a 1 mg/ mL solution of sodium borohydride in potassium phosphate buffer (50 mM, pH 6.5) with stirring for 30 min.30−32 Finally, the enzyme biocatalyst was washed with its optimum pH buffer (50 mM KH2PO4−K2HPO4) and stored at 4 °C until being used in reaction. The amount of immobilized protein was calculated as the difference of protein concentration in the supernatant before and after the immobilization procedure and was expressed in percentage of offered protein. Protein concentration was assessed at λ = 660 nm using the protein assay according to manufacturer’s instructions (Thermo Scientific Inc., Waltham, USA) using BSA as standard. The protein offered was 85 and 25 mg per gram of support of total protein and βgalactosidase, respectively. Protein concentration in Pectinex Ultra SPL was obtained from literature,39,40,34 being 17 mg/mL if total protein is considered and 5 mg/mL referred to actual β-galactosidase concentration.30 2.4. Enzyme Activity Assay. The β-galactosidase activity of Pectinex Ultra SP-L was determined by means of the oNPG method.11,16 The reaction mechanism for the oNPG hydrolysis was similar to that of the lactose hydrolysis, with the oNPG substrate being hydrolyzed into galactose and 2-nitrophenol (oNP). The production of oNP was quantified spectrophotometrically using a Varian Cary 500 Scan UV−vis NIR spectrophotometer. Different temperatures and pH values were assayed to determine the stability of the enzyme. In a typical experiment, the reaction mixture consisted of a solution of oNPG (3.48 g/L) in a phosphate buffer (50 mM KH2PO4−K2HPO4) with 0.1 mg/mL of β-galactosidase. This value was calculated from the β-galactosidase concentration of 5 mg/mL in the commercial product, Pectinex Ultra SP-L.30 Samples were withdrawn (200 μL aliquots) at given time intervals. To stop the reaction, 1 M H2SO4 (1200 μL) was

a valid strategy for multivariable analysis, which provides useful information on the influence of contemporary operating variables. It has been applied to evaluate the GOS production from skim milk or lactose solutions with Bif idobacterium sp.,17,18 from whey permeate or skim milk with different enzymes,19,20 or from lactose by β-galactosidase from Kluyveromyces lactis.13 In the context of GOS production, besides the overall yield achieved, the final distribution of the produced GOS is important in determining the achieved health benefits. In particular, in terms of digestibility, tri- and tetra-GOS are not hydrolyzed in vitro by several real and simulated salivary and gastric juices.3,21 Other studies have revealed that tri- and tetraGOS are completely metabolized by the intestinal gut flora.22−25 Therefore, the prebiotic properties of this product can be associated with the so-called high-GOS (mainly comprising tri-GOS and tetra-GOS).3,4 On the other hand, the immobilization of the enzymes improves its application as industrial biocatalysts. Usual benefits include enzyme recovery and reutilization, continuous operation, enzyme stability, potential improvement of selectivity, and reduction of inhibition phenomena.26,10 Among the several immobilization techniques, multipoint covalent immobilization via glyoxyl moieties on agarose or silica supports provides a fair enzyme stabilization and reuse without significant enzyme leaching.26,27 Agarose-based supports have been widely used with successful results.28−30 Alternatively, porous silica supports offer several advantages when used as immobilization support among them high surface area, thermal and mechanical stability, and absence of toxicity. Furthermore, they are unaffected by the presence of microbes and organic solvents, are easy to handle, and remain stable under the relatively severe conditions of high flow rates in continuous reactors, etc. For instance, Bernal et al. analyzed the immobilization of β-galactosidases from Bacillus circulans and Aspergillus oryzae on glyoxyl-hierarchical porous silica and obtained high immobilization yields and a significant improvement on thermal stability compared to the free systems.31,32 In this work, the optimization of the main reaction parameters affecting the production of high-GOS using a βgalactosidase-containing commercial crude extract from Aspergillus aculeatus is presented. This enzyme preparation has a food-grade status and has been used for GOS synthesis from lactose in a few previous reports.16,30,33−35 The optimization includes the application of the surface response methodology to assess the simultaneous influence of the two critical reaction parameters, that is, temperature and initial lactose concentration. Additionally, we present the immobilization of the enzyme on a glyoxyl-functionalized silica support and the application of the resultant biocatalyst under the optimized reaction conditions by showing an improvement of both the GOS yield and the selectivity toward the most prebiotic GOS, as compared to the use of free enzyme.

2. MATERIALS AND METHODS 2.1. Materials. D-Lactose monohydrate (≥99%), D-(+)-galactose (≥99%), D-(+)-melezitose monohydrate (≥99%), sodium periodate (≥99.8%), magnesium chloride (≥98%), potassium phosphate dibasic trihydrate (≥99%), potassium phosphate monobasic (≥98%), 2nitrophenyl β-D-galactopyranoside (oNPG) (≥98%), 2-nitrophenol (oNP) (≥99%), 3-glycidyloxypropyltrimethoxy-silane (GPTMS, ≥ 98%), triethylamine (≥99%), toluene anhydrous (99.8%), potassium carbonate (≥99%), potassium bicarbonate (99.7%), sodium bicarbonate (≥99.8%), potassium iodide (≥99.5%), potassium hydroxide (≥85%), and sulfuric acid (98%) were purchased from Sigma-Aldrich B

DOI: 10.1021/acs.jafc.6b05431 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 1. Effect of (A) pH and (B) temperature on the activity of the enzyme β-galactosidase from Aspergillus aculeatus in the hydrolysis of oNPG.

Figure 2. Influence on tri- and tetra-galactooligosaccharides (Yhigh‑GOS) production of: (A) pH at 2 U/mL of enzyme, and (B) enzyme concentration at pH 6.5. Experimental conditions: initial lactose concentration 285 g/L; temperature 55 °C. were diluted and injected into a 300 × 7.8 mm Rezex RCMMonosaccharide Ca2+ (8%) column (Phenomenex). The mobile phase consisted of Milli-Q water (0.5 mL/min). Column and detector cell temperatures were 80 and 35 °C, respectively. Standard solutions of glucose, galactose, and melezitose (as a reference for trigalactooligosaccharides, tri-GOS) of known concentrations were used for quantitative analysis. R2 values for each standard were over 0.999. Response factor for tetra-GOS was extrapolated considering an additional galactose unit from tri-GOS. An example of chromatogram obtained could be seen in Figure S1 (Supporting Information) were the maximum polymerization grade in GOS synthesized by Pectinex Ultra SP-L was tetra-GOS as previously reported Frenzel et al.35 Data acquisition and processing were performed with the Galaxie Chromatography Data System 1.9.3.2 software (Varian, Inc.). Carbohydrate yields were referred to the starting lactose content (wt/vol) in the reaction media and calculated using eq 1:

added. Then the acidity of each sample was neutralized with 1 M Na2CO3 (1800 μL), which also favored the development of a yellowish color. Quantitative production of 2-nitrophenol was determined by measuring the absorbance at 420 nm. Afterward, the enzyme activity was calculated as enzyme units (U/mL), defined as the amount of enzyme that produces 1 μmol of oNP per mL per min. Additionally, a specific activity (U/mg), taking into account the β-galactosidase concentration in the Pectinex Ultra SP-L commercial preparation (5 mg/mL),30 was calculated. 2.5. GOS Synthesis Experiments. In a typical experiment, a lactose solution was prepared in potassium phosphate buffer (50 mM) at the desired pH value. The reactions were carried out in a 100 mL round-bottomed flask with magnetic stirring, at 600 rpm in the experiments with free enzyme and 60 rpm in the experiments with the immobilized enzyme. Temperature was controlled using a water bath with a thermocouple probe. At selected time intervals, samples were withdrawn and immediately placed in a thermoblock (5 min, 90 °C) to stop the reaction by thermal inactivation of the enzyme. To avoid removing immobilized enzyme from reaction media, the agitation of the reaction was stopped 1 min earlier to get the sample, which allowed the support to be deposited at the bottom of the flask. Thereafter, the sample was centrifuged at 2200 × g for 2 min at room temperature to separate possible remaining support particles. Afterward, supernatant was inactivated at 90 °C for 5 min using the method previously described for the free enzyme. Samples were stored at −20 °C until analysis. Finally, reaction samples were diluted in Milli-Q water, to properly quantify the reaction products, and analyzed by means of high-performance liquid chromatography (HPLC). 2.6. Chromatographic Analysis and Quantification of Products. Carbohydrates were determined using a Varian ProStar 500 HPLC chromatograph equipped with a Varian 356-LC Refractive Index Detector. Analysis of both carbohydrates and GOS standards was performed using a method adapted from the literature.41 Samples

Yi =

i produced (g/L) × 100 Laco (g/L)

(1)

3. RESULTS AND DISCUSSION 3.1. Enzyme Activity in oNPG Hydrolysis Assay. To evaluate both the stability and the maximum activity of the enzyme β-galactosidase from Aspergillus aculeatus under different reaction conditions, a number of preliminary oNPG hydrolysis experiments were carried out. This hydrolysis reaction is chemically similar to the hydrolysis of lactose into galactose and glucose, and therefore, it can be properly used as a rapid prospective assay to estimate the activity and stability of the enzyme. C

DOI: 10.1021/acs.jafc.6b05431 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Table 1. Matrix of Values and Coded Levels and Experimental Results for the Production of GOS from Lactose by βGalactosidase from Aspergillus aculeatusa run

IT

IL

T (°C)

1 2 3 4 5 6 7 8 9 10 11 12

−1 −1 −1 0 0 0 1 1 1 0 0 0

−1 0 1 −1 0 1 −1 0 −1 0 0 0

55 55 55 60 60 60 65 65 65 60 60 60 mean ± s.d.

[L]o (g L−1)

Yhigh‑GOS (%)

[high-GOS] (g L−1)

Ytri‑GOS (%)

[tri-GOS] (g L−1)

Ytetra‑GOS (%)

[tetra-GOS] (g L−1)

Glu/Gal

50 150 250 50 150 250 50 150 250 150 150 150

10.4 12.5 14.0 11.6 13.3 15.9 7.2 7.0 12.1 13.4 13.9 14.0 13.7 ± 0.3

6.6 18.0 35.3 7.4 23.3 43.0 4.4 11.3 32.5 22.5 22.7 22.7 22.8 ± 0.3

7.1 9.0 11.5 7.2 9.5 12.5 5.6 4.8 10.0 10.2 10.2 10.1 10.0 ± 0.3

4.5 13.0 28.9 4.6 16.6 33.7 1.6 7.7 27.0 17.0 16.7 16.4 16.7 ± 0.3

3.3 3.5 2.5 4.4 3.8 3.4 4.7 2.2 2.1 3.3 3.7 3.9 3.7 ± 0.3

2.1 5.1 6.4 2.8 6.7 9.3 2.8 3.6 5.5 5.4 6.0 6.3 6.1 ± 0.5

2.1 2.8 3.0 2.4 2.6 2.7 3.1 3.9 4.3 2.7 2.8 2.8 2.7 ± 0.1

a Experimental conditions: pH 6.5; 4 U/mL enzyme concentration; 20 h reaction time. Note: columns 2 and 3 represent the 0 and ±1 encoded factor levels on a dimensionless scale, whereas columns 4 and 5 represent the factor levels on a natural scale. [L]o, initial concentration of lactose; T, temperature; I, coded value; Yhigh‑GOS, yield to high-GOS; Ytri‑GOS, yield to tri-GOS; Ytetra‑GOS, yield to tetra-GOS; [high-GOS], concentration of highGOS; [tri-GOS], concentration of tri-GOS; [tetra-GOS], concentration of tetra-GOS; Glu/Gal, molar ratio free glucose/free galactose.

GOS (ca. 1%) were produced, even after the reaction time increased up to 24 h. At pH 4.5, it must be noted, however, that after a maximum value was reached, the yield to high-GOS progressively decreased with the time until a value of 7.0% after 24 h. This might be explained in terms of hydrolysis of the initially formed high-GOS into shorter GOS once the presence of the starting substrate, that is, lactose, is strongly reduced. By contrast, at pH 6.5, the yield to high-GOS keeps increasing after 24 h (16.6%) hence indicating that as the pH decreases from 6.5 to 4.5, the hydrolysis rate increases. Therefore, at the lower pH value, the hydrolysis would predominate over the transgalactosylation reaction, in agreement with previous published results.16 Moreover, this can also be confirmed by the ratio of free glucose to free galactose in the reaction medium since it can be used as a parameter indicative of the degree of hydrolysis. Thus, low Glu/Gal ratios reveal that the hydrolysis reaction takes predominance over the transgalactosylation reaction (with the theoretical minimum of 1/1, resulting from the complete hydrolysis of lactose). On the contrary, high Glu/Gal ratios report the increasing contribution of transgalactosylation reactions (galactosyl transfer giving rise to disaccharides, trisaccharides, and tetrasaccharides, and eventually longer GOS), while Glu units remain in the reaction medium. Accordingly, in this case, after 20 h of reaction, Glu/ Gal ratios of 1.6 and 3.5 at pH of 4.5 and 6.5, were respectively obtained. The same 3.5 ratio has been previously reported for Pectinex Ultra SP-L under similar reaction conditions using UF-skimmed milk permeate with 42.8% initial lactose concentration as substrate.35 On the other hand, Figure 2B displays the kinetic curves of the yield toward high-GOS at pH 6.5 using different enzyme concentrations (2, 4, and 8 U/mL). The figure shows that the evolution of high-GOS production over the reaction time increases by increasing the concentration of the β-galactosidase during the first hours of reaction (below 5 h). However, at long reaction times (24 h), the trend changes, with the highest enzyme concentration resulting in the lowest yield to high-GOS (15.5% for 8 U/mL as compared to 16.9 and 17.8% for 2 and 4 U/mL, respectively). These results indicate that, at pH 6.5, the

The effect of critical parameters such as temperature and pH was examined in this system using 2 U/mL of enzyme, equivalent to a concentration of 0.1 mg/mL. The results are depicted in Figure 1. The effect of pH in the range 4.5−7.5 was assayed at 60 °C, whereas the influence of temperature was analyzed at pH 4.5 in the range 40−70 °C. As displayed in Figure 1A, the enzyme showed a maximum activity at pH 4.5 with 83% yield to oNP in 5 min and a calculated specific activity of 19.4 U/mg. The increase of pH to 6.5 led to a lower activity (32% yield and 6.4 U/mg), whereas remarkably, at pH 7.5, no production of oNP could be detected. Therefore, it is evidenced that the value of pH has a significant effect due to the high sensitivity of the enzyme to the pH of the medium.16 On the other hand, Figure 1B shows that this enzyme is highly active up to moderate temperatures, in the range between 40 and 70 °C, and provides a maximum performance at 60 °C. An increase of reaction temperature to 70 °C resulted in negligible differences in terms of yield to oNP, which can be related to the lack of thermal stability of the β-galactosidase from Aspergillus aculeatus at temperatures over 60 °C. This is in agreement with previous reports, wherein optimal temperature for the purified enzyme was found to be 55−60 °C.42 Hence, the optimum hydrolytic performance for this enzyme, within the studied conditions, is achieved at 60 °C and pH 4.5. 3.2. Study of pH and Enzyme Concentration in the Synthesis of GOS. To establish the appropriate range of pH and enzyme concentration for the optimization analysis, a preliminary set of experiments on the production of GOS from lactose focused on the yield to high-GOS (tri and tetra-GOS) was carried out. The pH value was analyzed in the same range previously studied for determination of the hydrolytic activity, that is, pH 4.5−7.5, using an initial lactose concentration of 285 g/L, 2 U/mL of enzyme concentration, and 55 °C. Figure 2A depicts the kinetic curves at pH 4.5, 6.5, and 7.5. As it can be seen, within the first 3 h, the trend observed is in agreement with the previous results of maximum specific activity. The higher yield to tri- and tetra-GOS was achieved at pH 4.5 (10.3%), whereas at pH 7.5, similarly to the lack of activity of the enzyme for oNP production, negligible amounts of highD

DOI: 10.1021/acs.jafc.6b05431 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 3. Correlation between model predicted values versus experimental values of (A) Yhigh‑GOS, (B) Ytri‑GOS, (C) Ytetra‑GOS, and (D) Glu/Gal.

high-GOS, whereas the glucose/galactose ratio revealed the predominant transformation (hydrolysis vs transgalactosylation). The selection of the levels for each factor was based on either preliminary experimental results or previously reported data. In the case of the temperature, the optimal value for triGOS production with this β-galactosidase previously isolated and purified has been reported to be in the range of 55−60 °C.16,42 On the other hand, the initial lactose concentration is limited by the maximum lactose solubility in water at the temperature of reaction, for example, 250 g/L at 40 °C,44 to ensure a homogeneous medium. Table 1 summarizes the code levels and selected values for each factor: temperature (55, 60, and 65 °C) and initial lactose concentration (50, 150, and 250 g/L), together with the experimental results. All the experiments were carried out in a randomized order to minimize the effect of unexplained variability in the observed responses. Furthermore, the central point was repeated three times to determine the variability of the results and to assess the experimental error. An empirical multiple quadratic model was assumed to fit the experimental results. Multiple regression analysis was performed by using the Statgraphics software to fit a second-order polynomial equation for each response:

increase of enzyme concentration appears to enhance the hydrolytic activity thus slowing the formation of high-GOS at long reaction times. A similar trend has been reported previously,16 and in particular, more clearly with the βgalactosidase from Kluyveromyces lactis.11,13 Considering the above results, 4 U/mL was considered as the most appropriate enzyme concentration for the following optimization experiments. 3.3. Optimization of Temperature and Lactose Concentration. The influence of temperature and starting lactose concentration was investigated by using a factorial experimental design43 to assess the simultaneous combined effect thereof, as well as the possible interactions. It must be pointed out that temperature is a key factor in this system. The increase of temperature enables the possibility of using higher lactose concentrations (up to the point of solubilization), leading to an improvement of the transgalactosylation activity and the corresponding increase in the yield to high-GOS.3,11 On the basis of the results shown in the previous section, in all experiments the value of pH, enzyme concentration and reaction time were fixed at pH 6.5, 4 U/mL and 20 h, respectively. A 32 factorial experimental design (two factors at three levels) was carried out. The selected responses for the analysis were the yield to high-GOS (trisaccharides plus tetrasaccharides; Yhigh‑GOS), the yield to trisaccharides (Ytri‑GOS), the yield to tetra-saccharides (Ytetra‑GOS), and glucose/galactose ratio (Glu/Gal). Specifically, the optimization of the reaction parameters aimed to increase the yield to high-GOS, as they are the products with the highest prebiotic effect. The different GOS yields revealed the amount and profile of the produced

2

YA = β0 +

2

2

∑ βi Xi + ∑ βiiXi 2 + ∑ ∑ βijXiXj i=1

i=1

i