β-galactosidase stability at high substrate concentrations
© Warmerdam et al.; licensee Springer. 2013
Received: 3 April 2013
Accepted: 20 August 2013
Published: 27 August 2013
Enzymatic synthesis of galacto-oligosaccharides is usually performed at high initial substrate concentrations since higher yields are obtained. We report here on the stability of β-galactosidase from Bacillus circulans at 25, 40, and 60°C in buffer, and in systems with initially 5.0 and 30% (w/w) lactose. In buffer, the half-life time was 220 h and 13 h at 25 and 40°C, respectively, whereas the enzyme was completely inactivated after two hours at 60°C. In systems with 5.0 and 30% (w/w) lactose, a mechanistic model was used to correct the o NPG converting activity for the presence of lactose, glucose, galactose, and oligosaccharides in the activity assay. Without correction, the stability at 5.0% (w/w) lactose was overestimated, while the stability at 30% (w/w) lactose was underestimated. The inactivation constant k d was strongly dependent on temperature in buffer, whereas only a slight increase in k d was found with temperature at high substrate concentrations. The enzyme stability was found to increase strongly with the initial substrate concentrations. The inactivation energy E a appeared to be lower at high initial substrate concentrations.
Keywordsβ-galactosidase Bacillus circulans Galacto-oligosaccharides Concentrated systems Enzyme activity Enzyme stability
For enzymatic production processes it is of interest to use highly concentrated conditions, since energy, water, and material costs can be saved. However, the activity and stability of enzymes are often investigated in aqueous systems, which may lead to irrelevant data. The enzyme activity of β-galactosidases, which is used in the production of galacto-oligosaccharides (GOS), in highly concentrated systems was studied before (Warmerdam et al. 2013a) and was found to be strongly influenced by the concentration of reactants and products. The high concentration of reactants and products may not only lead to more reactions taking place, but it will also lead to molecular crowding, which can have large effects on enzyme activity (Minton 2001; Ellis 2001).
Besides the enzyme activity, their stability can as well be strongly affected by molecular crowding (Minton 2001; Ellis 2001). In 1985, Arakawa and Timasheff (1985) have already described the stabilization of the protein structure of lysozyme in the presence of osmolytes. De Cordt et al. (1994) described the influence of high concentrations of polyalcohols and carbohydrates on the enzyme stability by substrate binding or preferential hydration. They observed various situations in which the presence of inert crowding agents increases the thermo-stability of proteins (Perham et al. 2007; Stagg et al. 2007; Zhou et al. 2008). Recently, Yadav (2013) described that the presence of sucrose and trehalose strongly increased the half-life time of α-amylase.
GOS are usually produced with β-galactosidase at high temperatures and at high substrate concentrations in industry. An advantage of reactions at high temperatures is the improved solubility of the substrates which makes higher substrate concentrations possible (Bruins et al. 2001). However, the inactivation of the enzyme is faster as well (Bruins et al. 2003).
The stability of β-galactosidase from Bacillus circulans was investigated before in systems with low lactose concentration or in absence of lactose (Mozaffar et al. 1984). Vetere and Paoletti (1998), and Song et al. (2011a) studied the stability of several isoforms of β-galactosidase from Bacillus circulans in aqueous systems. They found that the enzyme preparation was (partly) stable up to 50°C. The stability of free β-galactosidase from Bacillus circulans in systems with high lactose concentrations, which are usually used in production systems, has to our knowledge never been investigated before.
When using high initial substrate concentrations, it is important to investigate the effect of reactants in the activity assay. Baks et al. (2006) found that starch and its hydrolysis products may have large effects on the Ceralpha activity assay. This assay is comparable to the activity assay used for β-galactosidases with o NPG as an artificial substrate. Lactose and (some of) its conversion products are substrate for β-galactosidase as well as o NPG: they act as acceptor molecule for the enzyme-galactose complex, and they act as inhibitors and competitors (Warmerdam et al. 2013a; Borralho et al. 2002) (Warmerdam A, Zisopoulos FK, Boom RM, Janssen AEM: Kinetic characterization of β-galactosidases, submitted). In addition, galactose and glucose are usually found to be inhibitors for β-galactosidases (Warmerdam et al. 2013a; Greenberg and Mahoney 1982; Macfarlane et al. 2008; Prenosil et al. 1987) (Warmerdam A, Zisopoulos FK, Boom RM, Janssen AEM: Kinetic characterization of β-galactosidases, submitted). Because of the interactions of these carbohydrates, it is important to correct the o NPG activity measurements for their presence.
The aim of this study is therefore to investigate the stability of β-galactosidase from Bacillus circulans at various temperatures both in buffer, and in systems with initially 5.0 and 30% (w/w) lactose. The remaining enzyme activity is measured via the o NPG activity assay. The activity measurements are corrected for the effect of the carbohydrates present in the reaction mixture.
Materials and methods
Lactose monohydrate (Lactochem), Vivinal-GOS and a β-galactosidase from Bacillus circulans called Biolacta N5 (Daiwa Kasei K. K., Japan) were gifts from FrieslandCampina (Beilen, The Netherlands). Biolacta N5 was previously found to have a total protein content of 19 ± 3% (Warmerdam et al. 2013b). In all calculations, the total enzyme concentration was assumed to be equal to the total protein concentration, because the actual enzyme concentration is not known. Sulphuric acid, sodium hydroxide, o-nitrophenyl β-D-galactopyranoside (o NPG), o-nitrophenol (o NP), D(+)-glucose, D(+)-galactose, maltotriose, maltotetraose, maltopentaose, maltohexaose, and maltoheptaose were purchased from Sigma-Aldrich (Steinheim, Germany). Sodium carbonate, citric acid monohydrate, and disodium hydrogen phosphate were purchased from Merck (Darmstadt, Germany).
McIlvaine’s buffer was prepared by adding together 0.1 M citric acid and 0.2 M disodium hydrogen phosphate in the right ratio to achieve a pH of 6.0.
The stability of Biolacta N5 was investigated in a 0, 5.0, and 30% (w/w) lactose-in-buffer solution in a temperature controlled batch reactor with an anchor stirrer at 150 rpm. A certain mass of lactose monohydrate and a certain mass of buffer were weighted, so that a final concentration of lactose was obtained on a weight basis of 5% and 30% (w/w). 30% (w/w) lactose is close to the solubility at 50°C. The lactose was dissolved at approximately 60°C prior to cooling the solution to the desired temperature. The initial reaction volume was 25 mL. Temperatures were kept at 25, 40, or 60°C. A volume of 1.0 mL of 2.0 g∙L-1 Biolacta N5 was added once the temperature was constant. Samples were taken at 30 s, 5, 10, 15, 30, 60, 120, 240, 360 minutes and 22, and 24 hours for determination of the carbohydrate composition (100 μL sample) and for determination of the enzyme activity (210 μL sample). The final reaction volume was 21 mL.
Sample handling for determination of the carbohydrate composition
The sample (100 μL) taken from the reactor for determination of the carbohydrate composition was directly added into an Eppendorf tube with 50 μL of 5% (w/w) H2SO4 to inactivate the enzyme. Subsequently, the samples were stored at −20°C until further preparation.
Before HPLC analysis, the enzyme was removed from the samples by filtering the samples at 14,000 × g at 18°C for 30 minutes using pretreated Amicon® ultra-0.5 centrifugal filter devices (Millipore Corporation, Billerica, MA, United States) with a cut-off of 10 kDa in a Beckman Coulter Allegra X-22R centrifuge. The pretreatment of the filters consisted out of two centrifugation steps: first, 500 μL of Milli-Q water was centrifuged at 14,000 × g at 18°C for 15 minutes; and second, the filters were placed up-side-down in the tube and centrifuged at 14,000 × g at 18°C for 5 minutes. After filtration, the samples were neutralized with 5% (w/w) sodium hydroxide.
Measurement of the carbohydrate composition
The filtered samples were analysed with HPLC using a Rezex RSO oligosaccharide column (Phenomenex, Amstelveen, the Netherlands) at 80°C. The column was eluted with Milli-Q water at a flow rate of 0.3 mL/min. The eluent was monitored with a refractive index detector.
The standards that were used for calibration of the column were lactose, glucose, galactose, maltotriose, maltotetraose, maltopentaose, maltohexaose, and maltoheptaose. Galacto-oligosaccharides up to a degree of polymerization of 7 were assumed to have the same response as the glucose-oligomers with an equal degree of polymerization. This was confirmed with mass balances.
Enzyme activity measurements
The o NPG activity measurements, adapted from Nakanishi et al. (Nakanishi et al. 1983), were performed immediately after the sample was taken from the reaction mixture. An Eppendorf tube with 790 μL of 0.25% (w/w) o NPG-in-buffer was preheated in an Eppendorf Thermomixer at 40°C and 600 rpm for 10 minutes. Subsequently, 210 μL of sample was added and these mixtures were incubated for another 10 minutes at 40°C and 600 rpm. A volume of 1.0 mL of 10% (w/w) Na2CO3 solution was added to stop the reaction and, afterwards, the absorbance of o NP was measured at 420 nm. The o NP concentration was determined using the law of Lambert-Beer of which the extinction coefficient was determined to be 4576 M-1∙cm-1. The o NP formation was found to be linear during the first 10 minutes of the reaction. This initial rate of o NP formation was expressed in mmol∙min-1∙g protein-1. Measurements were performed in duplicate and the average enzyme activity was used.
Modeling the effect of carbohydrates on the activity assay
where v0, oNP is the initial rate of o NP formation in mM o NP∙s-1, E0 is the initial enzyme concentration in g protein∙L-1 or in mmol protein∙L-1 with the reaction rate constants k 1 , k a1 , k a2 , k 3 , k a3 , k a4 , k a5 , and k 6 in mmol o NP∙L∙(mmol X∙g protein∙s)-1 or in mmol o NP∙L∙(mmol X∙mmol protein∙s)-1, respectively, with X being the corresponding reactant. The inhibition constant K i is in mM.
Parameters for Biolacta N5 in the conversion of o NPG, lactose, glucose, galactose, and oligosaccharides
k1 (mmol o NP∙L∙(mmol o NPG∙g protein∙s)-1)
k3 (mmol o NP∙L∙(mmol lactose∙g protein∙s)-1)
k6 (mmol o NP∙L∙(mmol oligos∙g protein∙s)-1)
ka 1 (mmol o NP∙L∙(mmol o NPG∙g protein∙s)-1)
ka 2 (mmol o NP∙L∙(mmol H2O∙g protein∙s)-1)
ka 4 (mmol o NP∙L∙(mmol glucose∙g protein∙s)-1)
ka 5 (mmol o NP∙L∙(mmol galactose∙g protein∙s)-1)
K i (mM)
where is the initial rate of o NP formation without addition of carbohydrates C. At each time point, the concentrations of reactants used in this equation is the concentration that has been measured with HPLC.
Activity measurements corrected for the presence of carbohydrates
where A measured and A corrected are the enzyme activity calculated directly from the absorbance measurements (see “Enzyme activity measurements”), and the enzyme activity corrected for the presence of lactose, glucose, galactose, and oligosaccharides, respectively.
For each sample made with Vivinal-GOS, the concentration of lactose, glucose, galactose, and total oligosaccharide was calculated. The concentrations of lactose, galactose, glucose and total oligosaccharides are 19 dm%, 1 dm%, 21 dm%, and 59 dm% in Vivinal-GOS. Oligosaccharides were assumed to be mainly trisaccharides with a molecular weight of 504 g/mol.
Determination of enzyme stability
where k 0 and k t are the reaction rates at time zero and time t in h, k d is the enzyme inactivation constant in h-1, and t is the running time at which the sample was taken in hours. The enzyme inactivation constant k d and the reaction rate at time zero k0 were determined by linearization of equation 4.
where k d and k∞ (the Arrhenius constant) are in s-1, R is the gas constant in J∙mol-1∙K-1, and T is the temperature in K.
Results and discussion
Effect of temperature and initial lactose concentration on enzyme stability
The initial activity in buffer was approximately 13 mmol∙min-1∙g enzyme-1, while the initial activities were approximately 10 and 4 mmol∙min-1∙g enzyme-1 in 5.0 and 30% (w/w) lactose, respectively. The reduction in the initial activity with an increasing lactose concentration is caused by the competition of lactose (that is present in the samples) with o NPG in the activity assay, as will be discussed later.
In buffer, the enzyme was stable at 25°C, but lost 84% of its activity after 24 hours of incubation at 40°C, and was completely inactivated after two hours at 60°C. This complete inactivation in buffer at 60°C was expected: Mozaffar et al. (1984), Vetere and Paoletti (1998), and Song et al. (2011b) described that its isoforms are stable up to at most 50°C for one hour. The stability at elevated temperatures improved considerably in the presence of lactose. In a 5.0% (w/w) lactose solution, it took six hours of incubation at 60°C before most of the activity was lost, while in a 30% (w/w) lactose solution, 27% of the enzyme activity was left after 24 hours at 60°C.
The measured activity in Figure 1B and C after 24 hours of reaction at 25°C was lower than at 40°C. One would expect a better stability at a lower temperature. These unexpected stability values are the result of the presence of reactants during the activity assay. These reactants interfere with the activity measurements similarly as was described by (Baks et al. 2006) (Warmerdam A, Zisopoulos FK, Boom RM, Janssen AEM: Kinetic characterization of β-galactosidases, submitted). Therefore, the carbohydrate composition in the samples was determined and the effect of these reactants on the activity assay was determined with equation 2.
The carbohydrate content changed in time, and varied considerably between the initially different lactose concentrations. At an initial lactose concentration of 5.0% (w/w), the carbohydrate concentrations hardly changed anymore after 6 hours of reaction at 60°C. At 25 and 40°C the GOS content showed an optimum around 6 hours of incubation, indicating hydrolysis of the desired product at longer incubation times. Also a considerable amount of galactose was present after 24 hours of reaction. At an initial lactose concentration of 30% (w/w), GOS synthesis continued at all temperatures, including 60°C, until at least 22 hours of reaction (Figure 3B). Only small amounts of galactose were formed at all temperatures. The galactose production (indicating hydrolysis) was substantial at an initial lactose concentration of 5.0% (w/w) because of a high availability of water molecules, whereas no significant amounts of galactose were observed at an initial lactose concentration of 30% (w/w). It is clear that both the initial lactose concentration as well as the reaction temperature had a strong effect on the carbohydrate composition.
Correction for the presence of carbohydrates in stability experiments
The initial o NP formation rate k 0 of Biolacta N5 in mmol∙min -1 ∙g protein -1 at various initial lactose concentrations and temperatures, together with its 95% confidence interval
9.1 ± 1.0
11 ± 1
12 ± 1
9.0 ± 1.4
9.9 ± 1.2
13 ± 1
12 ± 1
8.9 ± 0.6
12 ± 0
The inactivation constant k d of Biolacta N5 in h -1 at various initial lactose concentrations and temperatures, together with its 95% confidence interval
0.0032 ± 0.0112
0.043 ± 0.023
0.024 ± 0.011
0.054 ± 0.040
0.041 ± 0.025
0.024 ± 0.015
15 ± 3
0.85 ± 0.19
0.043 ± 0.006
The half-life time t ½ in h of Biolacta N5 in hours at various initial lactose concentrations and temperatures
Inactivation energy E a for various lactose concentrations
A higher thermostability at high substrate concentrations is very favourable in the production of GOS by β-galactosidases from B.circulans. At high substrate concentrations, the reaction temperature can be higher than the enzyme’s stable ranges that were reported before in aqueous solutions, and it can be equal/closer to their optimal temperatures (Mozaffar et al. 1984; Vetere and Paoletti 1998; Song et al. 2011a), which will result in a higher enzyme stability.
β-Galactosidase from Bacillus circulans was found to be quite stable against temperature at high substrate concentrations. For a proper conclusion on the remaining enzyme activity versus time it was important to correct the enzyme activity measurements for the presence of various reactants.
Without correcting the enzyme activity at 5.0% (w/w) lactose, the actual stability was overestimated, whereas not correcting the enzyme activity at 30% (w/w) lactose resulted in an underestimation of the actual stability of β-galactosidase from Bacillus circulans. A high initial lactose concentration had a large positive effect on the enzyme stability.
The improved stability in more concentrated systems is very interesting for production conditions. The utilization of more concentrated systems for enzymatic conversions is economically more interesting in order to avoid the unnecessary use of water, to save energy as a smaller volume needs to be heated, and to save on capital expenditures as less equipment is necessary.
The authors would like to thank Eric Benjamins, Linqiu Cao, Ellen van Leusen, Albert van der Padt, and Jan Swarts of FrieslandCampina for the valuable scientific discussions.
This project is jointly financed by the European Union, European Regional Development Fund and The Ministry of Economic Affairs, Agriculture and Innovation, Peaks in the Delta, the Municipality of Groningen, the Provinces of Groningen, Fryslân and Drenthe as well as the Dutch Carbohydrate Competence Center (CCC WP9).
- Arakawa T, Timasheff SN: The stabilization of proteins by osmolytes. Biophys J 1985, 47: 411-414. 10.1016/S0006-3495(85)83932-1View ArticleGoogle Scholar
- Baks T, Janssen AEM, Boom RM: The effect of carbohydrates on α-amylase activity measurements. Enzyme Microb Technol 2006, 39: 114-119. 10.1016/j.enzmictec.2005.10.005View ArticleGoogle Scholar
- Borralho T, Chang Y, Jain P, Lalani M, Parghi K: Lactose induction of the lac operon in Escherichia coli B23 and its effect on the o-nitrophenyl β-galactoside assay. Journal of Experimental Microbiology and Immunology 2002, 2: 117-123.Google Scholar
- Bruins M, Janssen A, Boom R: Thermozymes and their applications. Appl Biochem Biotechnol 2001, 90: 155-186. 10.1385/ABAB:90:2:155View ArticleGoogle Scholar
- Bruins ME, Van Hellemond EW, Janssen AEM, Boom RM: Maillard reactions and increased enzyme inactivation during oligosaccharide synthesis by a hyperthermophilic glycosidase. Biotechnol Bioeng 2003, 81: 546-552. 10.1002/bit.10498View ArticleGoogle Scholar
- de Cordt S, Hendrickx M, Maesmans G, Tobback P: The influence of polyalcohols and carbohydrates on the thermostability of α-amylase. Biotechnol Bioeng 1994, 43: 107-114. 10.1002/bit.260430202View ArticleGoogle Scholar
- Ellis RJ: Macromolecular crowding: obvious but underappreciated. Trends Biochem Sci 2001, 26: 597-604. 10.1016/S0968-0004(01)01938-7View ArticleGoogle Scholar
- Greenberg NA, Mahoney RR: Production and characterization of β-galactosidase from Streptococcus thermophilus . J Food Sci 1982, 47: 1824-1835. 10.1111/j.1365-2621.1982.tb12891.xView ArticleGoogle Scholar
- Macfarlane GT, Steed H, Macfarlane S: Bacterial metabolism and health-related effects of galacto-oligosaccharides and other prebiotics. J Appl Microbiol 2008, 104: 305-344.Google Scholar
- Minton AP: The influence of macromolecular crowding and macromolecular confinement on biochemical reactions in physiological media. J Biol Chem 2001, 276: 10577-10580. 10.1074/jbc.R100005200View ArticleGoogle Scholar
- Mozaffar Z, Nakanishi K, Matsuno R, Kamikubo T: Purification and properties of β-galactosidases from Bacillus circulans . Agric Biol Chem 1984, 48: 3053-3061. 10.1271/bbb1961.48.3053Google Scholar
- Nakanishi K, Matsuno R, Torii K, Yamamoto K, Kamikubo T: Properties of immobilized β-galactosidase from Bacillus circulans . Enzyme Microb Technol 1983, 5: 115-120. 10.1016/0141-0229(83)90044-3View ArticleGoogle Scholar
- Perham M, Stagg L, Wittung-Stafshede P: Macromolecular crowding increases structural content of folded proteins. FEBS Lett 2007, 581: 5065-5069. 10.1016/j.febslet.2007.09.049View ArticleGoogle Scholar
- Prenosil JE, Stuker E, Bourne JR: Formation of oligosaccharides during enzymatic lactose: part I: state of art. Biotechnol Bioeng 1987, 30: 1019-1025. 10.1002/bit.260300904View ArticleGoogle Scholar
- Song J, Abe K, Imanaka H, Imamura K, Minoda M, Yamaguchi S, Nakanishi K: Causes of the production of multiple forms of β-galactosidase by Bacillus circulans . Biosci Biotechnol Biochem 2011, 75: 268-278. 10.1271/bbb.100574View ArticleGoogle Scholar
- Song J, Imanaka H, Imamura K, Minoda M, Katase T, Hoshi Y, Yamaguchi S, Nakanishi K: Cloning and expression of a beta-galactosidase gene of Bacillus circulans . Biosci Biotechnol Biochem 2011, 75: 1164-1167.Google Scholar
- Stagg L, Zhang S-Q, Cheung MS, Wittung-Stafshede P: Molecular crowding enhances native structure and stability of α/β protein flavodoxin. Proc Natl Acad Sci 2007, 104: 18976-18981. 10.1073/pnas.0705127104View ArticleGoogle Scholar
- Vetere A, Paoletti S: Separation and characterization of three β-galactosidases from Bacillus circulans . Biochimica et Biophysica Acta (BBA) - General Subjects 1998, 1380: 223-231. 10.1016/S0304-4165(97)00145-1View ArticleGoogle Scholar
- Warmerdam A, Wang J, Boom RM, Janssen AEM: Effects of carbohydrates on the o NPG converting activity of β-galactosidases. J Agric Food Chem 2013, 61: 6458-6464. 10.1021/jf4008554View ArticleGoogle Scholar
- Warmerdam A, Paudel E, Jia W, Boom RM, Janssen AEM: Characterization of β-galactosidase isoforms from Bacillus circulans and their contribution to GOS production. Appl Biochem Biotechnol 2013, 170: 340-358. 10.1007/s12010-013-0181-7View ArticleGoogle Scholar
- Yadav JK: Macromolecular crowding enhances catalytic efficiency and stability of α-amylase. ISRN Biotechnology 2013, 2013: Article ID 737805.View ArticleGoogle Scholar
- Zhou HX, Rivas GN, Minton AP: Macromolecular crowding and confinement: biochemical, biophysical, and potential physiological consequenses. Annu Rev Biophys 2008, 37: 375-397. 10.1146/annurev.biophys.37.032807.125817View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.