- Email: [email protected]
Characterization of α-galactosidases from germinating soybean seed and their use for hydrolysis of oligosaccharides
Phytochemistry 58 (2001) 67–73 www.elsevier.com/locate/phytochem
Characterization of a-galactosidases from germinating soybean seed and their use for hydrolysis of oligosaccharides Vale´ria Monteze Guimara˜esa,*, Sebastia˜o Tavares de Rezendeb, Maurilio Alves Moreirab, Everaldo Gonc¸alves de Barrosb, Carlos Roberto Felixa a
Departamento de Biologia Celular, Universidade de Brası´lia, Brası´lia, DF, 70.910-900, Brazil b BIOAGRO/DBB/DBG, Universidade Federal de Vic¸osa, Vic¸osa, MG, 36.571-000, Brazil Received 16 November 2000; received in revised form 22 February 2001
Abstract Raffinose oligosaccharides (RO) are the major factors responsible for flatulence following ingestion of soybean derived products. Removal of RO from seeds or soymilk would then have a positive impact on the acceptance of soy-based foods. Enzymic hydrolysis of the RO is accomplished by a-galactosidase. While the content of RO decreases during seed germination, the activity of a-galactosidase increases substantially. Two a-galactosidases were isolated from germinating seeds by partition in an aqueous two-phase system followed by ion-exchange and affinity chromatography. One of the enzyme preparations (P1) showed a single protein with Mr of 33 kDa, and the second (P2) had two proteins with Mr of 31 and 33 kDa. Maximal activities against the synthetic substrate nitrophenyl-a-d-galactopyranoside (NPGal) were detected at pH 5.0–5.5 and 45–50 C. Both enzymes were fairly stable at 40 C, but lost most of their activities after 30 min at 50 C. The Km values for hydrolysis of NPGal by the P1 and P2 enzymes were 1.55 and 0.76 mM, respectively. The Km values determined for hydrolysis of raffinose and melibiose by the P2 enzyme were 5.53 and 5.34 mM, respectively and galactose was a competitive inhibitor (Ki=0.65 mM). To different extents, both enzymes were sensitive to inhibition by galactose, melibiose, CuSO4, and SDS. Sucrose and b-mercaptoethanol showed discrete inhibitory effects on both enzymes. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Glycine max L.; Leguminosae; a-Galactosidase; Raffinose oligosaccharides; Flatulence
1. Introduction Soybean (Glycine max L.) is an excellent source of proteins for human and animal use. Nevertheless, due to its raffinose oligosaccharides (RO) content, mainly raffinose [a-d-galactopyranosyl-(1,6)-a-d-glucopyranosyl-b-d-fructofuranoside] and stachyose [a-d-galactopyranosyl-(1,6)a-d-galactopyranosyl-(1,6)-a-d-glucopyranosyl-b-d-fructofuranoside], the legume has not been used to its full potential. Since humans and other monogastric animals lack the enzyme a-galactosidase (E.C. 3.2.1.22, a-dgalactoside galactohydrolase), necessary for hydrolysis of the a-1,6 linkages present in RO (Gitzelmann and Auricchio, 1965), these sugars pass intact into the large intestine where anaerobic micro-organisms ferment then * Corresponding author. Present address: Universidade Federal de Vic¸osa, Instituto de Biotechnologia a Agropecuaria, BIOAGRO, 36.571-000 Vic¸osa, MG, Brazil. Fax: +55-31-3899-2864. E-mail address: [email protected] (V.M. Guimara˜es).
and cause flatulence. It is suggested that blocking the expression of the galactinol synthase gene which codes for the key enzyme on the synthetic route of the RO in the seeds, could decrease RO content and reduce flatulence caused by legumes (De Lumen, 1992). On the other hand, enzymic conversion of these RO in soymilk may be a rational alternative to improve the nutritional quality of this low-cost, high-quality protein supplement for humans and animals (Cruz and Park, 1982; De Rezende and Felix, 1997; Sanni et al., 1997). Hydrolysis of RO may be achieved either by a-galactosidase, invertase or both. While a-galactosidase hydrolyses the a-1,6 linkage of raffinose producing galactose and sucrose, the invertase hydrolyses the a-1,2 linkage producing melibiose and fructose. a-Galactosidase enzymes are widely distributed in microorganisms (De Rezende and Felix, 1999), plants and animals (Dey and Pridham, 1972). However, no economically efficient enzymic process using either indigenous or recombinant soybean agalactosidase to reduce RO in soymilk has yet been
0031-9422/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0031-9422(01)00165-0
68
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
proposed. Here we report on the characterization of agalactosidases from germinating soybean seeds, and on the possibility of using these enzymes for hydrolysis of the RO present in soymilk.
2. Results and discussion The RO content in germinating soybean seeds decreased during germination (Fig. 1). Consistently, the activity against the synthetic substrate (-nitrophenyl-ad-galactopyranoside) NPGal (Fig. 1), and the content of soluble fructose (not shown) increased confirming hydrolysis of the fructose-containing oligosaccharides. The results confirm previous reports that RO are used as an energy source during seed germination (Porter et al., 1990; Avigad and Dey, 1997). Most of the a-galactosidase activity (80%) present in the seed crude extract partitioned into the upper phase of an aqueous two-phase system (ATPS), which contained 12% NaCl. Chromatography of the resulting ATPS enzyme sample on a CM-Sepharose column (Fig. 2) resulted in the separation of two protein fractions (P1 and P2). SDS–PAGE analysis revealed that while the P1 active fraction contained a single protein with Mr of 33 kDa, the P2 active fraction contained two
Fig. 1. a-Galactosidase activity (&) and content of raffinose oligosaccharides (raffinose+stachyose) (*) in soybean seeds after germination.
Fig. 2. Elution profile of the a-galactosidase from germinating soybean seeds on a CM-Sepharose columm. (&) enzyme activity; (*) protein concentration and ( — ) linear gradient of NaCl.
protein forms with Mr of 31 and 33 kDa. Affinity chromatography of the enzyme fractions resulted in further purification of the enzyme forms as indicated by their higher specific activities (Table 1). The electrophoretic profile of the two enzyme preparations in SDS-polyacrylamide gel (Fig. 3) confirmed the presence of a single protein band in the P1 fraction (33 kDa), and a broad band corresponding to the two protein forms of 31 and 33 kDa in P2. It was previously reported that the cotyledonary a-galactosidases from Vigna unguiculata have Mr of 33 kDa (Oliveira-Neto et al., 1998). In addition, a recombinant protein from Glycine max produced in Pichia pastoris purified by affinity chromatography showed a relative molecular mass of 39.8 kDa under reducing conditions, and of 38 kDa under nonreducing conditions as judged by electrophoretic analysis (Davis et al., 1996). Another a-galactosidase purified to near electrophoretic homogeneity from dry soybean seeds was shown to be monomeric at pH 7.0 with a molecular mass of 40 kDa, and to be tetrameric at pH 4.0 with a molecular mass of about 160 kDa (Porter et al., 1990). Both the tetrameric and monomeric forms were enzymically active but displayed different kinetic properties (Campillo and Shannon, 1982). It is not clear whether the 31 kDa protein present in the P2 enzyme preparation reported here is also active. Phaseolus vulgaris has two a-galactosidase isoenzymes with molecular masses of 38.3 and 39.6 kDa which co-migrate after several salt precipitation and chromatographic procedures (Dhar et al., 1994). The 31 and 33 kDa proteins, which could not be separated even after affinity chromatography, might well be isoforms, but additional studies are needed to characterize the 31 kDa protein. Substantial activity against NPGal was observed for both enzyme preparations in the temperature range of 20–70 C and in the pH range of 6.5–7.0. Maximal substrate hydrolysis by the P1 and P2 enzymes were achieved at the temperature of 45 and 50 C, respectively. The optima pH for the P1 and P2 enzymes were 5.0 and 5.5, respectively. These pH optima and temperature values are close to those determined for hydrolysis of NPGal and melibiose by a crude a-galactosidase from germinating soybean seed (Cruz and Silva, 1986), by the cotyledonary a-galactosidases from V. unguiculata (Oliveira-Neto et al., 1998) and by the fungal raffinosehydrolysing enzymes (De Rezende and Felix, 1999). The Km and Vmax values calculated by the Lineweaver–Burk plot for hydrolysis of NPGal by the P2 and P1 enzymes were 0.76 mM and 1.76 mmol min 1 ml 1, and 1.55 mM and 2.43 mmol min 1 ml 1, respectively. These values are comparable to those determined for hydrolysis of the same substrate by a crude a-galactosidase from germinating soybean seed (Cruz and Silva, 1986). Both the P1 and P2 a-galactosidases were moderately thermostable. The enzymes retained about 50% of their original activity following incubation for 2 h at 40 C
69
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73 Table 1 Summary of the purification steps of the a-galactosidases from germinating soybean seeds Purification step
Total protein (mg)
Total activity (U)
Crude extract Upper layer of ATPS Dialyzed sample CM-sepharose P1 P2 Con-A (P1) Con-A (P2)
21488.0 1203.4 728.5
574.9 401.9 263.5
12.6 9.6 0.48 0.18
42.0 88.0 6.8 11.25
Specific activity (U/mg protein) 0.026 0.33 0.36 3.33 9.2 14.2 62.50
Purification (fold) – 12.7 13.9 128 352 545 2404
Recovery (%) – 69.9 45.8 7.3 15.3 1.2 2.00
Fig. 3. SDS–PAGE of different purification steps of a-galactosidases from germinating soybean seeds. Lanes are as follows: A: 1, molecular mass standards; 2, crude extract; 3, upper phase of ATPS; 4, enzyme preparation P1 from CM-Sepharose; 5, and P2, enzyme preparation P2 from CMSepharose; B: 1, molecular mass standards; 2, P1 eluted through ConA; C, 1, molecular mass standards; 2, P2 eluted through ConA.
70
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
(Fig. 4), but 90% of the activity was lost by both enzyme preparations following 45 min of incubation at 50 C (Fig. 5). The activities of the P1 and P2 a-galactosidases against other substrates are shown in Table 2. Both enzymes were able to hydrolyze melibiose, raffinose and stachyose. However, the affinities for these substrates were much lower in comparison to that against NPGal. The Km and Vmax values for hydrolysis of melibiose by the P2 enzyme were 5.34 mM and 0.13 mmol min 1 ml 1, respectively. Values for raffinose were 5.53 mM and 0.43 mmol min 1 ml 1, respectively. The Km value for hydrolysis of raffinose is close to that determined for the a-galactosidase purified from dry soybean seeds (Porter et al., 1990), but much lower than those reported for hydrolysis of raffinose by enzymes from Aspergillus fumigatus (De Rezende and Felix, 1999) and Trichoderma reesei (Zeilinger et al., 1993). Lactose and synthetic substrates containing b-linkage were hydrolyzed by the P1 enzyme but not by the P2 enzyme. None of the enzyme preparations showed substantial activity against sucrose and cellobiose and the synthetic substrates containing xylose rNP-X, arabinose NP-A and mannose NP-M (Table 2). The P1 and P2 a-galactosidases showed distinct sensitivities to simple sugars and mono and divalent ions (Table 3). The P1 enzyme showed very low or no
inhibition by galactose, glucose, sucrose, melibiose, magnesium, potassium, sodium, calcium and b-mercaptoethanol, but was highly inhibited by copper and SDS. The P2 enzyme was only partially inhibited by galactose, melibiose, copper, and SDS. This is in agreement with the results reported for the cotyledonary agalactosidases from V. unguiculata (Oliveira-Neto et al., 1998). Inhibition by galactose was found to be competitive and the Ki value was 0.65 mM as determined by the Dixon plot. In a previous work Porter et al. (1990) reported that galactose competitively inhibited (Ki of
Table 2 Hydrolysis of several substrates by the a-galactosidases from germinating soybean seeds Substrate
Concentration (mM)
NP-a-Gal Melibiose Raffinose Stachyose Lactose Maltose Cellobiose Sucrose NP-b-Gala NP-b-Gala NP-a-Glca NP-Xa NP-Aa NP-Ma
0.5 40 40 40 40 40 40 40 0.5 0.5 0.5 0.5 0.5 0.5
Relative activity (%) S.D. P1
P2
100 0.98 6.89 0.27 11.820.49 5.02 0.06 6.20 0.13 3.97 0.17 0.37 0.02 0 5.50 0.130 5.65 0.34 0 0 0 0
1000.89 10.090.20 54.310.46 10.520.15 1.290.12 1.600.12 0.550.05 0 0 0 0 0 0 0
a NP-b-Gal, para-nitrophenyl-b-d-galactopyranoside; oNP-b-Gal, ortho-nitrophenyl-b-d-galactopyranoside; NP-a-Glc, para-nitrophenyl-a-d-glycopyranoside; NP-X, para-nitrophenyl-a-d-xylopyranoside; NP-A, para-nitrophenyl-a-d-arabinopyranoside; NP-M, paranitrophenyl-a-d-mannopyranoside.
Fig. 4. pH (A) and temperature (B) influences on the activity of agalactosidases (P1, ~; P2, *) from germinating soybean seeds.
Table 3 Effect of simple sugars, ions, sodium dodecylsulphate and b-mercaptoethanol on a-galactosidases from germinating soybean seeds Effectora
Fig. 5. Temperature influences on the stability of the a-galactosidase from germinating soybean seeds. Enzyme preparations were pre-incubated at 40 C P1 (&), P2 * or 50 C P1 (*), P2 (~) for up to 2 h and the assays run at 37 C.
– d-Galactose d-Glucose Sucrose Melibiose CuSO4 MgCl2 KCl NaCl CaCl2 SDS b-Mercaptoethanol
Relative activity (%) S.D. P1
P2
1000.92 83.021.02 108.350.79 89.481.25 84.171.23 5.520.47 95.770.95 98.712.00 99.361.5 95.110.70 1.940.11 88.711.25
1000.100 31.520.18 97.160.93 89.661.86 64.080.66 51.681.48 96.471.70 103.361.01 98.710.66 102.151.72 62.640.15 88.370.55
a The final concentration of SDS was 1 mM. The final concentrations of all the other effectors were 2 mM.
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
0.12 mM) the a-galactosidase purified from dry soybean seeds with raffinose as substrate. As the enzymes were able to hydrolyze also melibiose, raffinose and stachyose, the ability of P1 and P2 to hydrolyse RO present in the soybean aqueous extract (soymilk) was tested. Sucrose, raffinose and stachyose were present in dried soymilk at the concentrations of 10.25, 1.33 and 8.47 mg% (w/w), respectively. No change was observed for the amount of the RO in the control reactions. The P1 enzyme could not hydrolyze the RO present in soymilk, in contrast the P2 enzyme reduced considerably the amount of raffinose (by 73.3%) and stachyose (by 40.6%) after an incubation period of 8 h at 30 C. Hydrolysis of RO in soymilk by enzymic treatment was tested in 1982 by Cruz and Park (1982), and later by other researchers (Mulinami et al., 1997; Sanni et al., 1997; Scalabrini et al., 1998). However, no reliable, inexpensive and efficient enzymic process with indigenous or recombinant enzymes is available so far. In general, the enzymes suggested for this purpose are of microbial origin and therefore present the disadvantage of having no GRAS (generally regard as safe) status. The results presently reported indicate that the soybean seed a-galactosidases or the corresponding genes may be used for establishment of a process to improve the nutritional value of soymilk. Cloning of the genes encoding the P1 and P2 a-galactosidases is underway.
3. Experimental 3.1. Plant material and enzyme preparation Soybean seeds (G. max L. Merr. cv. Doko) were allowed to germinate for 54 h, on a water-soaked filter paper (Germitest) at 27 C, and then frozen at 20 C. The frozen pre-germinated seeds (400 g fresh weight) were powdered in a blender and then resuspended in 800-ml of 0.1 M citric acid containing 0.05 M sodium phosphate, pH 5.5, and incubated with agitation for 1 h at 4 C. The suspension was then filtered through cheesecloth, the filtrate centrifuged (15,300 g) for 40 min at 4 C, and the supernatant used as a crude enzyme preparation. 3.2. -Galactosidase assay The enzyme was assayed using a reaction system (1 ml final volume) containing 650–750 ml of 0.1 M sodium acetate buffer, pH 5 (except when stated), 0–100 ml of enzyme preparation and 250ml of 2 mM -nitrophenyl a-d-galactopyranoside (NPGal) or, other synthetic substrates. Reaction was conducted for 15 min at 37 C and stopped by the addition of 1 ml of 0.5 M sodium carbonate. The reaction was shown to be a first order
71
reaction over a 60-min period. One unit (U) of enzyme was defined as the amount of protein necessary to produce one mmol of -nitrophenol per min. For determination of thermal stability the enzyme fractions were pre-incubated for several time periods at either 40 or 50 C, and the residual activity was determined using the standard assay. The activities against raffinose, stachyose and sucrose were assayed for 20 min at 37 C using a reaction mixture containing 500 ml of 0.1 mM sodium acetate buffer, pH 5.0, 100ml of enzyme extract, and 400 ml of a 0.1 M substrate solution. The amount of reducing sugar produced was determined by adding 1 ml of 3,5-dinitrosalicylate reagent according to Miller (1956). The activities against melibiose, lactose, maltose and cellobiose were assayed using the same reaction conditions and substrate concentrations used for raffinose, stachyose and sucrose. In this case, the amount of glucose formed was determined by the glucose–oxidase method (Bergmeyer and Bernt, 1974). The data presented for all a-galactosidase activity determinations are mean values of triplicate assays in which the standard deviations were always smaller than 10%. 3.3. -Galactosidase purification The soybean enzyme was purified by separation on an ATPS (Hustedt et al., 1985) followed by two chromatography procedures. The ATPS consisted of 224 g of a 50% (w/w) polyethyleneglycol (PEG 1500) solution, 280 g of a 40% (w/w) monobasic and dibasic sodium phosphate solution, pH 5.0, 84 g of sodium chloride, and 112 g of crude enzyme extract. Samples (50 g each) of the aqueous mixture was transferred to centrifuge tubes which were vortex mixed for 1 min and then centrifuged for 5 min at 4100 g and 8 C. The volumes of the upper and lower phases were determined in graduated tubes, and collected separately. Protein and enzyme partition coefficients in the ATPS were calculated by measuring the amount of protein and enzyme in each phase. The effect of NaCl on the partition coefficient was investigated using 0, 3, 6, 9 and 12% NaCl in the ATPS. The upper phase resulting from the ATPS containing the enzyme was dialyzed overnight against 50 mM sodium acetate buffer, pH 4.0, and loaded onto a CM-Sepharose Fast Flow column (313 cm) equilibrated with 0.1 M sodium acetate buffer, pH 4.0. The proteins were eluted at 4 C at a flow rate of 40 ml/h, with a linear gradient formed with 300 ml of the acetate buffer and 300 ml of the same buffer containing 0.8 M NaCl. Fiveml fractions were collected. Fractions containing agalactosidase activity were pooled and mixed with equal volume of sodium acetate buffer, pH 5.0, containing 1 mM each of MnCl2, MgCl2 and CaCl2, and 0.9% NaCl. The resulting enzyme preparations where then chromatographed on a concanavalin-A Sepharose column (215.5 cm) equilibrated with the acetate ion-contain-
72
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
ing buffer. Proteins were recovered with 150 ml of equilibrating buffer containing 0.1 M methyl a-D mannoside. 3.4. Electrophoresis analysis Enzyme preparations were analyzed by SDS–PAGE (12.5% gels) as described by Laemmli (1970). Proteins were silver-stained according to Blum et al. (1987).
Acknowledgements V.M.G. gratefully acknowledges financial support from the Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior (CAPES, Brasil) and from the Universidade Federal de Goia´s. This work was partially supported by a grant to CRF (FAPDF, No. 193.000.237/96).
References 3.5. Enzyme characterization Kinetic experiments were performed at 37 C and pH 5.0. The Michaelis–Menten constant (Km) and Vmax for substrate hydrolysis and the inhibition constant (Ki) for galactose were calculated by the Lineweaver–Burk plot and Dixon plot, respectively. The substrate concentrations were: 0.025, 0.05, 0.1, 0.15, 0.25, 0.50, 0.75, 1.0, 1.5, and 2.0 mM in the case of NPGal, 0.25, 0.5, 1.5, 2.5, 5.0, 10.0, 20.0, and 40.0 mM for melibiose and raffinose. For Ki determination for galactose, the following NPGal concentrations were used: 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, and 1.0 mM. The galactose concentrations used were 0.25, 0.5, and 1.0 mM. The effect of ions, simple sugars and reducing agent on the enzyme activity was tested by the standard assay with enzyme samples preincubated with each of the compounds tested for 15 min at 37 C. 3.6. Treatment of soybean milk with -galactosidase Soymilk was prepared from dry seeds (50 g). The seeds were chopped, homogenized in 400 ml water at 80 C, incubated for 10 min at 85 C, and filtered through cheese-cloth. Soymilk samples (2 ml) were then incubated with either water or 2 U of purified a-galactosidase for zero or 8 h with shaking (100 rpm) at 30 C. Each reaction mixture was dried and the soluble sugars extracted from 20 mg of dried powder with organic solvents according to Saravitz et al. (1987). The solvent was evaporated at 40 C and the sugars resuspended in 1 ml water. The sugars were analyzed by HPLC on a Shimadzu series 10A chromatograph using an analytical column [aminopropil (-NH2)] eluted with an acetonitrile–water isocratic mixture (80:20 v/v). Individual sugars were automatically identified and quantified by comparison with retention times and known concentrations of standard sugars. Gentiobiose was used as an internal standard as it does not interfere with the other sugars and it is not found in soybean seeds. 3.7. Protein determination Protein was quantified by the Coomassie Blue binding method (Bradford, 1976) with bovine serum albumin as standard.
Avigad, G., Dey, P.M., 1997. Carbohydrate metabolism: storage carbohydrates, In: Dey, P.M., Harborne, J.B. (Ed.), Plant Biochemistry, Academic Press, San Diego, pp. 143–204. Bergmeyer, H.U., Bernt, E., 1974. Determination of glucose with oxidase and peroxidase. In: Bergmeyer, H.U. (Eds.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, pp. 1205–1215. Blum, H., Beier, H., Gross, H., 1987. Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8, 93–99. Bradford, M.M., 1976. A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle dye binding. Anal. Biochem. 72, 680–685. Campillo, H., Shannon, L.M., 1982. An a-galactosidase with hemagglutinin properties from soybean seeds. Plant Physiol. 69, 628– 631. Cruz, R., Park, Y.K., 1982. Production of a-galactosidase and its application to the hydrolysis of galactoligosaccharides in soybean milk. J. Food Sci. 47, 1973–1975. Cruz, R., Silva, A.L., 1986. Soybean (Glycine max), endogenous alpha-galactosidase and invertase during the germinative process. Arq. Biol. Technol. 29, 435–443. Davis, M.O., Hata, D.J., Johnson, S.A., Walker, J.C., Smith, D.S., 1996. Cloning, expression and characterization of a blood group B active recombinant a-d-galactosidase from soybean (Glycine max). Biochem. Mol. Biol. Int. 39, 471–485. De Lumen, O.B., 1992. Molecular strategies to improve protein quality and reduce flatulence in legumes: a review. Food Structure 11, 33–46. De Rezende, S.T., Felix, C.R., 1997. Raffinose-hydrolyzing activity of Aspergillus fumigatus. Biotech. Letters 19, 217–220. De Rezende, S.T., Felix, C.R., 1999. Production and characterization of raffinose-hydrolysing and invertase activities of Aspergillus fumigatus. Folia Microbiol. 44, 191–195. Dey, P.M., Pridham, J.B., 1972. Biochemistry of a-galactosidases. Adv. Enzymol. 36, 91–130. Dhar, M., Mitra, M., Hata, J., Butnariu, O., Smith, D., 1994. Purification and characterization of Phaseolus vulgaris a-d-galactosidase isozymes. Biochem. Mol. Biol. Int. 34, 1055–1062. Gitzelmann, R., Auricchio, S., 1965. The handling of soy a-galactoside by a normal and galactosemic child. Pediatrics 36, 231–232. Hustedt, H., Kroner, K.K., Kula, M.R., 1985. Applications of phase partitioning in biotechnology. In: Walter, H., Brooks, D.E., Fisher, D. (Eds.), Partitioning in Aqueous Two-Phase Systems, Academic Press, Orlando. Laemmli, U.K., 1970. Cleavage of structural of structural protein during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Miller, G.L., 1956. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31, 426–428. Mulimani, V.H., Thippeswamy, S., Ramalingam, 1997. Effect of soaking, cooking and crude a-galactosidase treatment on the oligosaccharide content of soybean flour. Food Chem. 59, 279–282. Oliveira-Neto, O.B., Gomes-Filho, E., Prisco, J.T., Ene´as-Filho, J., 1998. Partial purification and properties of cotyledonary a-galacto
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73 sidase from three cultivars of Vigna unguiculata. R. Bras. Fisiol. Veg. 10, 91–96. Porter, J.E., Herrmann, K.M., Ladish, M.R., 1990. Integral kinetics of a-galactosidase purified from Glycine max for simultaneous hydrolysis of stachyose and raffinose. Biotechnol. Bioeng. 35, 15–22. Sanni, A.I., Onilude, A.A., Ogundoye, O.R., 1997. Effect of bacterial galactosidase treatment on the nutritional status of soybean seeds and its milk derivative. Nahrung 41, 18–21. Saravitz, D.M., Pharr, D.M., Carter Jr., T.E., 1987. Galactinol syn-
73
thase activity and soluble sugars developing seeds of four soybean genotypes. Plant Physiol. 83, 185–189. Scalabrini, P., Rossi, M., Spettoli, P., Matteuzzi, D., 1998. Characterization of Bifidobacterium strains for use in soymilk fermentation. Int. J. Food Microbiol. 39, 213–219. Zeilinger, S., Kristufek, D., Arisan-Atac, I., Hodits, R., Kubicek, C.P., 1993. Conditions of formation, purification, and characterization of an a-galactosidase of Trichoderma reesei RUT C-30. Appl. Environ. Microbiol. 59, 1347–1353.
Characterization of a-galactosidases from germinating soybean seed and their use for hydrolysis of oligosaccharides Vale´ria Monteze Guimara˜esa,*, Sebastia˜o Tavares de Rezendeb, Maurilio Alves Moreirab, Everaldo Gonc¸alves de Barrosb, Carlos Roberto Felixa a
Departamento de Biologia Celular, Universidade de Brası´lia, Brası´lia, DF, 70.910-900, Brazil b BIOAGRO/DBB/DBG, Universidade Federal de Vic¸osa, Vic¸osa, MG, 36.571-000, Brazil Received 16 November 2000; received in revised form 22 February 2001
Abstract Raffinose oligosaccharides (RO) are the major factors responsible for flatulence following ingestion of soybean derived products. Removal of RO from seeds or soymilk would then have a positive impact on the acceptance of soy-based foods. Enzymic hydrolysis of the RO is accomplished by a-galactosidase. While the content of RO decreases during seed germination, the activity of a-galactosidase increases substantially. Two a-galactosidases were isolated from germinating seeds by partition in an aqueous two-phase system followed by ion-exchange and affinity chromatography. One of the enzyme preparations (P1) showed a single protein with Mr of 33 kDa, and the second (P2) had two proteins with Mr of 31 and 33 kDa. Maximal activities against the synthetic substrate nitrophenyl-a-d-galactopyranoside (NPGal) were detected at pH 5.0–5.5 and 45–50 C. Both enzymes were fairly stable at 40 C, but lost most of their activities after 30 min at 50 C. The Km values for hydrolysis of NPGal by the P1 and P2 enzymes were 1.55 and 0.76 mM, respectively. The Km values determined for hydrolysis of raffinose and melibiose by the P2 enzyme were 5.53 and 5.34 mM, respectively and galactose was a competitive inhibitor (Ki=0.65 mM). To different extents, both enzymes were sensitive to inhibition by galactose, melibiose, CuSO4, and SDS. Sucrose and b-mercaptoethanol showed discrete inhibitory effects on both enzymes. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Glycine max L.; Leguminosae; a-Galactosidase; Raffinose oligosaccharides; Flatulence
1. Introduction Soybean (Glycine max L.) is an excellent source of proteins for human and animal use. Nevertheless, due to its raffinose oligosaccharides (RO) content, mainly raffinose [a-d-galactopyranosyl-(1,6)-a-d-glucopyranosyl-b-d-fructofuranoside] and stachyose [a-d-galactopyranosyl-(1,6)a-d-galactopyranosyl-(1,6)-a-d-glucopyranosyl-b-d-fructofuranoside], the legume has not been used to its full potential. Since humans and other monogastric animals lack the enzyme a-galactosidase (E.C. 3.2.1.22, a-dgalactoside galactohydrolase), necessary for hydrolysis of the a-1,6 linkages present in RO (Gitzelmann and Auricchio, 1965), these sugars pass intact into the large intestine where anaerobic micro-organisms ferment then * Corresponding author. Present address: Universidade Federal de Vic¸osa, Instituto de Biotechnologia a Agropecuaria, BIOAGRO, 36.571-000 Vic¸osa, MG, Brazil. Fax: +55-31-3899-2864. E-mail address: [email protected] (V.M. Guimara˜es).
and cause flatulence. It is suggested that blocking the expression of the galactinol synthase gene which codes for the key enzyme on the synthetic route of the RO in the seeds, could decrease RO content and reduce flatulence caused by legumes (De Lumen, 1992). On the other hand, enzymic conversion of these RO in soymilk may be a rational alternative to improve the nutritional quality of this low-cost, high-quality protein supplement for humans and animals (Cruz and Park, 1982; De Rezende and Felix, 1997; Sanni et al., 1997). Hydrolysis of RO may be achieved either by a-galactosidase, invertase or both. While a-galactosidase hydrolyses the a-1,6 linkage of raffinose producing galactose and sucrose, the invertase hydrolyses the a-1,2 linkage producing melibiose and fructose. a-Galactosidase enzymes are widely distributed in microorganisms (De Rezende and Felix, 1999), plants and animals (Dey and Pridham, 1972). However, no economically efficient enzymic process using either indigenous or recombinant soybean agalactosidase to reduce RO in soymilk has yet been
0031-9422/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0031-9422(01)00165-0
68
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
proposed. Here we report on the characterization of agalactosidases from germinating soybean seeds, and on the possibility of using these enzymes for hydrolysis of the RO present in soymilk.
2. Results and discussion The RO content in germinating soybean seeds decreased during germination (Fig. 1). Consistently, the activity against the synthetic substrate (-nitrophenyl-ad-galactopyranoside) NPGal (Fig. 1), and the content of soluble fructose (not shown) increased confirming hydrolysis of the fructose-containing oligosaccharides. The results confirm previous reports that RO are used as an energy source during seed germination (Porter et al., 1990; Avigad and Dey, 1997). Most of the a-galactosidase activity (80%) present in the seed crude extract partitioned into the upper phase of an aqueous two-phase system (ATPS), which contained 12% NaCl. Chromatography of the resulting ATPS enzyme sample on a CM-Sepharose column (Fig. 2) resulted in the separation of two protein fractions (P1 and P2). SDS–PAGE analysis revealed that while the P1 active fraction contained a single protein with Mr of 33 kDa, the P2 active fraction contained two
Fig. 1. a-Galactosidase activity (&) and content of raffinose oligosaccharides (raffinose+stachyose) (*) in soybean seeds after germination.
Fig. 2. Elution profile of the a-galactosidase from germinating soybean seeds on a CM-Sepharose columm. (&) enzyme activity; (*) protein concentration and ( — ) linear gradient of NaCl.
protein forms with Mr of 31 and 33 kDa. Affinity chromatography of the enzyme fractions resulted in further purification of the enzyme forms as indicated by their higher specific activities (Table 1). The electrophoretic profile of the two enzyme preparations in SDS-polyacrylamide gel (Fig. 3) confirmed the presence of a single protein band in the P1 fraction (33 kDa), and a broad band corresponding to the two protein forms of 31 and 33 kDa in P2. It was previously reported that the cotyledonary a-galactosidases from Vigna unguiculata have Mr of 33 kDa (Oliveira-Neto et al., 1998). In addition, a recombinant protein from Glycine max produced in Pichia pastoris purified by affinity chromatography showed a relative molecular mass of 39.8 kDa under reducing conditions, and of 38 kDa under nonreducing conditions as judged by electrophoretic analysis (Davis et al., 1996). Another a-galactosidase purified to near electrophoretic homogeneity from dry soybean seeds was shown to be monomeric at pH 7.0 with a molecular mass of 40 kDa, and to be tetrameric at pH 4.0 with a molecular mass of about 160 kDa (Porter et al., 1990). Both the tetrameric and monomeric forms were enzymically active but displayed different kinetic properties (Campillo and Shannon, 1982). It is not clear whether the 31 kDa protein present in the P2 enzyme preparation reported here is also active. Phaseolus vulgaris has two a-galactosidase isoenzymes with molecular masses of 38.3 and 39.6 kDa which co-migrate after several salt precipitation and chromatographic procedures (Dhar et al., 1994). The 31 and 33 kDa proteins, which could not be separated even after affinity chromatography, might well be isoforms, but additional studies are needed to characterize the 31 kDa protein. Substantial activity against NPGal was observed for both enzyme preparations in the temperature range of 20–70 C and in the pH range of 6.5–7.0. Maximal substrate hydrolysis by the P1 and P2 enzymes were achieved at the temperature of 45 and 50 C, respectively. The optima pH for the P1 and P2 enzymes were 5.0 and 5.5, respectively. These pH optima and temperature values are close to those determined for hydrolysis of NPGal and melibiose by a crude a-galactosidase from germinating soybean seed (Cruz and Silva, 1986), by the cotyledonary a-galactosidases from V. unguiculata (Oliveira-Neto et al., 1998) and by the fungal raffinosehydrolysing enzymes (De Rezende and Felix, 1999). The Km and Vmax values calculated by the Lineweaver–Burk plot for hydrolysis of NPGal by the P2 and P1 enzymes were 0.76 mM and 1.76 mmol min 1 ml 1, and 1.55 mM and 2.43 mmol min 1 ml 1, respectively. These values are comparable to those determined for hydrolysis of the same substrate by a crude a-galactosidase from germinating soybean seed (Cruz and Silva, 1986). Both the P1 and P2 a-galactosidases were moderately thermostable. The enzymes retained about 50% of their original activity following incubation for 2 h at 40 C
69
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73 Table 1 Summary of the purification steps of the a-galactosidases from germinating soybean seeds Purification step
Total protein (mg)
Total activity (U)
Crude extract Upper layer of ATPS Dialyzed sample CM-sepharose P1 P2 Con-A (P1) Con-A (P2)
21488.0 1203.4 728.5
574.9 401.9 263.5
12.6 9.6 0.48 0.18
42.0 88.0 6.8 11.25
Specific activity (U/mg protein) 0.026 0.33 0.36 3.33 9.2 14.2 62.50
Purification (fold) – 12.7 13.9 128 352 545 2404
Recovery (%) – 69.9 45.8 7.3 15.3 1.2 2.00
Fig. 3. SDS–PAGE of different purification steps of a-galactosidases from germinating soybean seeds. Lanes are as follows: A: 1, molecular mass standards; 2, crude extract; 3, upper phase of ATPS; 4, enzyme preparation P1 from CM-Sepharose; 5, and P2, enzyme preparation P2 from CMSepharose; B: 1, molecular mass standards; 2, P1 eluted through ConA; C, 1, molecular mass standards; 2, P2 eluted through ConA.
70
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
(Fig. 4), but 90% of the activity was lost by both enzyme preparations following 45 min of incubation at 50 C (Fig. 5). The activities of the P1 and P2 a-galactosidases against other substrates are shown in Table 2. Both enzymes were able to hydrolyze melibiose, raffinose and stachyose. However, the affinities for these substrates were much lower in comparison to that against NPGal. The Km and Vmax values for hydrolysis of melibiose by the P2 enzyme were 5.34 mM and 0.13 mmol min 1 ml 1, respectively. Values for raffinose were 5.53 mM and 0.43 mmol min 1 ml 1, respectively. The Km value for hydrolysis of raffinose is close to that determined for the a-galactosidase purified from dry soybean seeds (Porter et al., 1990), but much lower than those reported for hydrolysis of raffinose by enzymes from Aspergillus fumigatus (De Rezende and Felix, 1999) and Trichoderma reesei (Zeilinger et al., 1993). Lactose and synthetic substrates containing b-linkage were hydrolyzed by the P1 enzyme but not by the P2 enzyme. None of the enzyme preparations showed substantial activity against sucrose and cellobiose and the synthetic substrates containing xylose rNP-X, arabinose NP-A and mannose NP-M (Table 2). The P1 and P2 a-galactosidases showed distinct sensitivities to simple sugars and mono and divalent ions (Table 3). The P1 enzyme showed very low or no
inhibition by galactose, glucose, sucrose, melibiose, magnesium, potassium, sodium, calcium and b-mercaptoethanol, but was highly inhibited by copper and SDS. The P2 enzyme was only partially inhibited by galactose, melibiose, copper, and SDS. This is in agreement with the results reported for the cotyledonary agalactosidases from V. unguiculata (Oliveira-Neto et al., 1998). Inhibition by galactose was found to be competitive and the Ki value was 0.65 mM as determined by the Dixon plot. In a previous work Porter et al. (1990) reported that galactose competitively inhibited (Ki of
Table 2 Hydrolysis of several substrates by the a-galactosidases from germinating soybean seeds Substrate
Concentration (mM)
NP-a-Gal Melibiose Raffinose Stachyose Lactose Maltose Cellobiose Sucrose NP-b-Gala NP-b-Gala NP-a-Glca NP-Xa NP-Aa NP-Ma
0.5 40 40 40 40 40 40 40 0.5 0.5 0.5 0.5 0.5 0.5
Relative activity (%) S.D. P1
P2
100 0.98 6.89 0.27 11.820.49 5.02 0.06 6.20 0.13 3.97 0.17 0.37 0.02 0 5.50 0.130 5.65 0.34 0 0 0 0
1000.89 10.090.20 54.310.46 10.520.15 1.290.12 1.600.12 0.550.05 0 0 0 0 0 0 0
a NP-b-Gal, para-nitrophenyl-b-d-galactopyranoside; oNP-b-Gal, ortho-nitrophenyl-b-d-galactopyranoside; NP-a-Glc, para-nitrophenyl-a-d-glycopyranoside; NP-X, para-nitrophenyl-a-d-xylopyranoside; NP-A, para-nitrophenyl-a-d-arabinopyranoside; NP-M, paranitrophenyl-a-d-mannopyranoside.
Fig. 4. pH (A) and temperature (B) influences on the activity of agalactosidases (P1, ~; P2, *) from germinating soybean seeds.
Table 3 Effect of simple sugars, ions, sodium dodecylsulphate and b-mercaptoethanol on a-galactosidases from germinating soybean seeds Effectora
Fig. 5. Temperature influences on the stability of the a-galactosidase from germinating soybean seeds. Enzyme preparations were pre-incubated at 40 C P1 (&), P2 * or 50 C P1 (*), P2 (~) for up to 2 h and the assays run at 37 C.
– d-Galactose d-Glucose Sucrose Melibiose CuSO4 MgCl2 KCl NaCl CaCl2 SDS b-Mercaptoethanol
Relative activity (%) S.D. P1
P2
1000.92 83.021.02 108.350.79 89.481.25 84.171.23 5.520.47 95.770.95 98.712.00 99.361.5 95.110.70 1.940.11 88.711.25
1000.100 31.520.18 97.160.93 89.661.86 64.080.66 51.681.48 96.471.70 103.361.01 98.710.66 102.151.72 62.640.15 88.370.55
a The final concentration of SDS was 1 mM. The final concentrations of all the other effectors were 2 mM.
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
0.12 mM) the a-galactosidase purified from dry soybean seeds with raffinose as substrate. As the enzymes were able to hydrolyze also melibiose, raffinose and stachyose, the ability of P1 and P2 to hydrolyse RO present in the soybean aqueous extract (soymilk) was tested. Sucrose, raffinose and stachyose were present in dried soymilk at the concentrations of 10.25, 1.33 and 8.47 mg% (w/w), respectively. No change was observed for the amount of the RO in the control reactions. The P1 enzyme could not hydrolyze the RO present in soymilk, in contrast the P2 enzyme reduced considerably the amount of raffinose (by 73.3%) and stachyose (by 40.6%) after an incubation period of 8 h at 30 C. Hydrolysis of RO in soymilk by enzymic treatment was tested in 1982 by Cruz and Park (1982), and later by other researchers (Mulinami et al., 1997; Sanni et al., 1997; Scalabrini et al., 1998). However, no reliable, inexpensive and efficient enzymic process with indigenous or recombinant enzymes is available so far. In general, the enzymes suggested for this purpose are of microbial origin and therefore present the disadvantage of having no GRAS (generally regard as safe) status. The results presently reported indicate that the soybean seed a-galactosidases or the corresponding genes may be used for establishment of a process to improve the nutritional value of soymilk. Cloning of the genes encoding the P1 and P2 a-galactosidases is underway.
3. Experimental 3.1. Plant material and enzyme preparation Soybean seeds (G. max L. Merr. cv. Doko) were allowed to germinate for 54 h, on a water-soaked filter paper (Germitest) at 27 C, and then frozen at 20 C. The frozen pre-germinated seeds (400 g fresh weight) were powdered in a blender and then resuspended in 800-ml of 0.1 M citric acid containing 0.05 M sodium phosphate, pH 5.5, and incubated with agitation for 1 h at 4 C. The suspension was then filtered through cheesecloth, the filtrate centrifuged (15,300 g) for 40 min at 4 C, and the supernatant used as a crude enzyme preparation. 3.2. -Galactosidase assay The enzyme was assayed using a reaction system (1 ml final volume) containing 650–750 ml of 0.1 M sodium acetate buffer, pH 5 (except when stated), 0–100 ml of enzyme preparation and 250ml of 2 mM -nitrophenyl a-d-galactopyranoside (NPGal) or, other synthetic substrates. Reaction was conducted for 15 min at 37 C and stopped by the addition of 1 ml of 0.5 M sodium carbonate. The reaction was shown to be a first order
71
reaction over a 60-min period. One unit (U) of enzyme was defined as the amount of protein necessary to produce one mmol of -nitrophenol per min. For determination of thermal stability the enzyme fractions were pre-incubated for several time periods at either 40 or 50 C, and the residual activity was determined using the standard assay. The activities against raffinose, stachyose and sucrose were assayed for 20 min at 37 C using a reaction mixture containing 500 ml of 0.1 mM sodium acetate buffer, pH 5.0, 100ml of enzyme extract, and 400 ml of a 0.1 M substrate solution. The amount of reducing sugar produced was determined by adding 1 ml of 3,5-dinitrosalicylate reagent according to Miller (1956). The activities against melibiose, lactose, maltose and cellobiose were assayed using the same reaction conditions and substrate concentrations used for raffinose, stachyose and sucrose. In this case, the amount of glucose formed was determined by the glucose–oxidase method (Bergmeyer and Bernt, 1974). The data presented for all a-galactosidase activity determinations are mean values of triplicate assays in which the standard deviations were always smaller than 10%. 3.3. -Galactosidase purification The soybean enzyme was purified by separation on an ATPS (Hustedt et al., 1985) followed by two chromatography procedures. The ATPS consisted of 224 g of a 50% (w/w) polyethyleneglycol (PEG 1500) solution, 280 g of a 40% (w/w) monobasic and dibasic sodium phosphate solution, pH 5.0, 84 g of sodium chloride, and 112 g of crude enzyme extract. Samples (50 g each) of the aqueous mixture was transferred to centrifuge tubes which were vortex mixed for 1 min and then centrifuged for 5 min at 4100 g and 8 C. The volumes of the upper and lower phases were determined in graduated tubes, and collected separately. Protein and enzyme partition coefficients in the ATPS were calculated by measuring the amount of protein and enzyme in each phase. The effect of NaCl on the partition coefficient was investigated using 0, 3, 6, 9 and 12% NaCl in the ATPS. The upper phase resulting from the ATPS containing the enzyme was dialyzed overnight against 50 mM sodium acetate buffer, pH 4.0, and loaded onto a CM-Sepharose Fast Flow column (313 cm) equilibrated with 0.1 M sodium acetate buffer, pH 4.0. The proteins were eluted at 4 C at a flow rate of 40 ml/h, with a linear gradient formed with 300 ml of the acetate buffer and 300 ml of the same buffer containing 0.8 M NaCl. Fiveml fractions were collected. Fractions containing agalactosidase activity were pooled and mixed with equal volume of sodium acetate buffer, pH 5.0, containing 1 mM each of MnCl2, MgCl2 and CaCl2, and 0.9% NaCl. The resulting enzyme preparations where then chromatographed on a concanavalin-A Sepharose column (215.5 cm) equilibrated with the acetate ion-contain-
72
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73
ing buffer. Proteins were recovered with 150 ml of equilibrating buffer containing 0.1 M methyl a-D mannoside. 3.4. Electrophoresis analysis Enzyme preparations were analyzed by SDS–PAGE (12.5% gels) as described by Laemmli (1970). Proteins were silver-stained according to Blum et al. (1987).
Acknowledgements V.M.G. gratefully acknowledges financial support from the Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior (CAPES, Brasil) and from the Universidade Federal de Goia´s. This work was partially supported by a grant to CRF (FAPDF, No. 193.000.237/96).
References 3.5. Enzyme characterization Kinetic experiments were performed at 37 C and pH 5.0. The Michaelis–Menten constant (Km) and Vmax for substrate hydrolysis and the inhibition constant (Ki) for galactose were calculated by the Lineweaver–Burk plot and Dixon plot, respectively. The substrate concentrations were: 0.025, 0.05, 0.1, 0.15, 0.25, 0.50, 0.75, 1.0, 1.5, and 2.0 mM in the case of NPGal, 0.25, 0.5, 1.5, 2.5, 5.0, 10.0, 20.0, and 40.0 mM for melibiose and raffinose. For Ki determination for galactose, the following NPGal concentrations were used: 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, and 1.0 mM. The galactose concentrations used were 0.25, 0.5, and 1.0 mM. The effect of ions, simple sugars and reducing agent on the enzyme activity was tested by the standard assay with enzyme samples preincubated with each of the compounds tested for 15 min at 37 C. 3.6. Treatment of soybean milk with -galactosidase Soymilk was prepared from dry seeds (50 g). The seeds were chopped, homogenized in 400 ml water at 80 C, incubated for 10 min at 85 C, and filtered through cheese-cloth. Soymilk samples (2 ml) were then incubated with either water or 2 U of purified a-galactosidase for zero or 8 h with shaking (100 rpm) at 30 C. Each reaction mixture was dried and the soluble sugars extracted from 20 mg of dried powder with organic solvents according to Saravitz et al. (1987). The solvent was evaporated at 40 C and the sugars resuspended in 1 ml water. The sugars were analyzed by HPLC on a Shimadzu series 10A chromatograph using an analytical column [aminopropil (-NH2)] eluted with an acetonitrile–water isocratic mixture (80:20 v/v). Individual sugars were automatically identified and quantified by comparison with retention times and known concentrations of standard sugars. Gentiobiose was used as an internal standard as it does not interfere with the other sugars and it is not found in soybean seeds. 3.7. Protein determination Protein was quantified by the Coomassie Blue binding method (Bradford, 1976) with bovine serum albumin as standard.
Avigad, G., Dey, P.M., 1997. Carbohydrate metabolism: storage carbohydrates, In: Dey, P.M., Harborne, J.B. (Ed.), Plant Biochemistry, Academic Press, San Diego, pp. 143–204. Bergmeyer, H.U., Bernt, E., 1974. Determination of glucose with oxidase and peroxidase. In: Bergmeyer, H.U. (Eds.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, pp. 1205–1215. Blum, H., Beier, H., Gross, H., 1987. Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8, 93–99. Bradford, M.M., 1976. A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle dye binding. Anal. Biochem. 72, 680–685. Campillo, H., Shannon, L.M., 1982. An a-galactosidase with hemagglutinin properties from soybean seeds. Plant Physiol. 69, 628– 631. Cruz, R., Park, Y.K., 1982. Production of a-galactosidase and its application to the hydrolysis of galactoligosaccharides in soybean milk. J. Food Sci. 47, 1973–1975. Cruz, R., Silva, A.L., 1986. Soybean (Glycine max), endogenous alpha-galactosidase and invertase during the germinative process. Arq. Biol. Technol. 29, 435–443. Davis, M.O., Hata, D.J., Johnson, S.A., Walker, J.C., Smith, D.S., 1996. Cloning, expression and characterization of a blood group B active recombinant a-d-galactosidase from soybean (Glycine max). Biochem. Mol. Biol. Int. 39, 471–485. De Lumen, O.B., 1992. Molecular strategies to improve protein quality and reduce flatulence in legumes: a review. Food Structure 11, 33–46. De Rezende, S.T., Felix, C.R., 1997. Raffinose-hydrolyzing activity of Aspergillus fumigatus. Biotech. Letters 19, 217–220. De Rezende, S.T., Felix, C.R., 1999. Production and characterization of raffinose-hydrolysing and invertase activities of Aspergillus fumigatus. Folia Microbiol. 44, 191–195. Dey, P.M., Pridham, J.B., 1972. Biochemistry of a-galactosidases. Adv. Enzymol. 36, 91–130. Dhar, M., Mitra, M., Hata, J., Butnariu, O., Smith, D., 1994. Purification and characterization of Phaseolus vulgaris a-d-galactosidase isozymes. Biochem. Mol. Biol. Int. 34, 1055–1062. Gitzelmann, R., Auricchio, S., 1965. The handling of soy a-galactoside by a normal and galactosemic child. Pediatrics 36, 231–232. Hustedt, H., Kroner, K.K., Kula, M.R., 1985. Applications of phase partitioning in biotechnology. In: Walter, H., Brooks, D.E., Fisher, D. (Eds.), Partitioning in Aqueous Two-Phase Systems, Academic Press, Orlando. Laemmli, U.K., 1970. Cleavage of structural of structural protein during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Miller, G.L., 1956. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31, 426–428. Mulimani, V.H., Thippeswamy, S., Ramalingam, 1997. Effect of soaking, cooking and crude a-galactosidase treatment on the oligosaccharide content of soybean flour. Food Chem. 59, 279–282. Oliveira-Neto, O.B., Gomes-Filho, E., Prisco, J.T., Ene´as-Filho, J., 1998. Partial purification and properties of cotyledonary a-galacto
V.M. Guimara˜es et al. / Phytochemistry 58 (2001) 67–73 sidase from three cultivars of Vigna unguiculata. R. Bras. Fisiol. Veg. 10, 91–96. Porter, J.E., Herrmann, K.M., Ladish, M.R., 1990. Integral kinetics of a-galactosidase purified from Glycine max for simultaneous hydrolysis of stachyose and raffinose. Biotechnol. Bioeng. 35, 15–22. Sanni, A.I., Onilude, A.A., Ogundoye, O.R., 1997. Effect of bacterial galactosidase treatment on the nutritional status of soybean seeds and its milk derivative. Nahrung 41, 18–21. Saravitz, D.M., Pharr, D.M., Carter Jr., T.E., 1987. Galactinol syn-
73
thase activity and soluble sugars developing seeds of four soybean genotypes. Plant Physiol. 83, 185–189. Scalabrini, P., Rossi, M., Spettoli, P., Matteuzzi, D., 1998. Characterization of Bifidobacterium strains for use in soymilk fermentation. Int. J. Food Microbiol. 39, 213–219. Zeilinger, S., Kristufek, D., Arisan-Atac, I., Hodits, R., Kubicek, C.P., 1993. Conditions of formation, purification, and characterization of an a-galactosidase of Trichoderma reesei RUT C-30. Appl. Environ. Microbiol. 59, 1347–1353.