Purification and Characterization of a Thermotolerant -Galactosidase ...

2 downloads 0 Views 215KB Size Report
LUTZ FISCHER,* CHRISTIAN SCHECKERMANN, AND FRITZ WAGNER. Institute .... acrylamide gel in a Mini Protean II system (12 by 7 cm; Bio-Rad, Richmond,.
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1995, p. 1497–1501 0099-2240/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 61, No. 4

Purification and Characterization of a Thermotolerant b-Galactosidase from Thermomyces lanuginosus LUTZ FISCHER,* CHRISTIAN SCHECKERMANN,

AND

FRITZ WAGNER

Institute of Biochemistry and Biotechnology, Technical University of Braunschweig, D-38106 Braunschweig, Germany Received 8 September 1994/Accepted 2 February 1995

A new inducible intracellular b-galactosidase (EC 3.2.1.23) of the thermophilic fungus Thermomyces lanuginosus was purified by fractional salt precipitation, hydrophobic interaction, and anion exchange chromatography. The first 22 amino acid residues were determined by N-terminal sequencing. Electrophoretic investigations revealed a dimeric enzyme with a molecular mass of 75 to 80 kDa per identical subunit and an isoelectric point of 4.4 to 4.5. The native b-galactosidase was identified as a glycoprotein by the enzyme-linked immunosorbent assay technique. The b-galactosidase activity was optimal at pH 6.7 to 7.2, and the enzyme displayed stability between pH 6 and 9. It was completely stable at pH 6.8 and 47&C for 2 h. After 191 h at 50&C, the remaining b-galactosidase activity of an enzyme fraction after salt precipitation was 58%. The b-galactosidase hydrolyzed p- and o-NO2-phenyl-b-D-galactopyranoside, lactose, lactulose, MeOH-b-D-galactopyranoside, phenyl-b-D-galactopyranoside, and p-NO2-phenyl-a-L-arabinopyranoside. The kinetic constants (Km) measured for p- and o-NO2-phenyl-b-D-galactopyranoside and b-lactose were 4.8, 11.3, and 18.2 mM, respectively.

b-Galactosidases (b-D-galactoside galactohydrolase or lactase; EC 3.2.1.23) are widely distributed in nature. With regard to reaction mechanism (2, 16, 18) and three-dimensional structure (9), the most thoroughly investigated b-galactosidase is obtained from E. coli (pH optimum, pH 7), whereas the most significant industrial b-galactosidases originate from yeasts and several Aspergillus species (pH optima, between pH 4 and 5). These fungal b-galactosidases are applied in food technology for hydrolyzing the b-(1,4) linkage between galactose and glucose in lactose, the worldwide production of which is approximately 5.75 million metric tons per year (6). This enzymatic cleavage overcomes some disadvantages, like the moderate solubility and lower sweetness of lactose compared with galactose and glucose and its partial indigestibility (21), normally encountered in the food industry with the use of lactose-containing products. Therefore, depending upon which conditions are required (with regard to temperature, pH, substrate, ion strength, etc.), a variety of lactases with suitable properties are desirable. Besides the established industrial applications of b-galactosidases in cleaving O-b-D-galactosidic bonds, a great deal of interest has been directed towards their potential use in regio- and stereospecific synthesis of glycoconjugates in recent years (1, 4, 7, 11, 17, 20, 22). Whereas many available b-galactosidases are relatively thermolabile, enzymes derived from thermophilic microorganisms demonstrate a higher degree of thermostability than b-galactosidases derived from mesophilic ones. A better thermostability would be a significant advantage (13) for prospective biotechnological applications either in the food industry or in glycoconjugate synthesis. This article describes the purification, characterization, and partial N-terminal amino acid sequence of a new intracellular b-galactosidase from the ther-

mophilic fungus Thermomyces lanuginosus (formerly Humicola lanuginosa). MATERIALS AND METHODS Microorganism and culture conditions. T. lanuginosus ATCC 16455 was grown aerobically in suspension culture for 24 h. The medium contained, per liter, 20 g of soluble starch, 10 g of yeast extract, 6 g of lactose, 1 g of MgSO4, and 1 g of K2HPO4, adjusted to pH 6.8. The cultivations were carried out in Erlenmeyer flasks (liquid medium, 20% of total volume) on a rotary shaker at 508C and 100 rpm. Assays. b-Galactosidase activity was determined at 378C with p-nitrophenolb-D-galactopyranoside (pNO2PheGal) as the substrate dissolved in 100 mM potassium phosphate buffer, pH 6.8. The hydrolysis was monitored spectrophotometrically at 405 nm. One nanokatal of activity was defined as 1 nmol released s21 under the described conditions. The absorbance values were converted into molar concentrations on the basis of a molar absorption coefficient of 9,600 liters mol21 cm21 at 405 nm. Tests on other chromogenic substrates like p-nitrophenyl-a-D-galactopyranoside, p-nitrophenyl-a-D-glucopyranoside, p-nitrophenyl-b-D-glucopyranoside, p-nitrophenyl-a-L-fucopyranoside, p-nitrophenyl-a-Larabinopyranoside, and p-nitrophenyl-b-L-arabinopyranoside were performed under the same conditions. The hydrolysis of phenyl-b-D-galactopyranoside was monitored by high-performance liquid chromatography (HPLC). The enzymatic reaction (under the conditions described above) was stopped with 10% (vol/vol) trichloroacetic acid after 3 min. The sample was centrifuged, and the clear supernatant was subjected to quantitative analysis by HPLC on an RP 18 column (125 by 4.0 mm; Merck, Darmstadt, Germany) with MeOH and H2O at a 40/60 (vol/vol) ratio as the eluent and UV detection at 220 nm. The activities of the b-galactosidase against lactose, lactulose, and MeOH-bgalactopyranoside were estimated spectrophotometrically in a coupled enzyme test at 378C and 340 nm (reduction of NAD1) with b-galactose dehydrogenase (Sigma, Deisenhofen, Germany). Since some of the substrates used (Sigma) contained small amounts of b-galactose, the assay solution was preincubated with 50 mg of b-galactose dehydrogenase (0.26 U) in 0.1 M potassium phosphate buffer (pH 6.8) and 0.6 mM NAD1 until the A340 was constant. The enzyme assay was started by addition of b-galactosidase (final volume, 1 ml). Enzyme purification. T. lanuginosus cells were grown for 24 h and harvested at the beginning of the stationary growth phase. The wet biomass (150 g) obtained after centrifugation (22,000 3 g, 20 min, 48C) of the culture broth was resuspended in 0.1 M potassium phosphate buffer (total suspension volume, 500 ml), pH 6.8, and the cells were disrupted with a glass bead mill (Dyno Mill, type KDL; W. Bachofen AG, Maschinenfabrik, Basel, Switzerland) at 48C. The suspension was centrifuged (22,000 3 g, 20 min, 48C), and the supernatant was taken as crude extract for further purification. The protein concentration was determined as described by Bradford (3). Crude extract (190 ml) was subjected to ammonium sulfate fractionation at

* Corresponding author. Mailing address: Institute of Biochemistry and Biotechnology, Technical University of Braunschweig, Spielmannstr. 7, D-38106 Braunschweig, Germany. Phone: 49-531-3915730. Fax: 49-531-3915763. 1497

1498

FISCHER ET AL.

APPL. ENVIRON. MICROBIOL. TABLE 1. Purification of T. lanuginosus b-galactosidase Characteristic of b-galactosidase at indicated step

Step

Crude extract (NH4)2SO4 precipitation Hydrophobic interaction chromatography Anion exchange chromatography Mono Q 10/10 (pH shift) Mono Q 10/10 (pH 8.0) Mono Q 5/5 (pH 6.5)

Total protein (mg)

Total activity (nkat)

1,540.00

3,326.4

2.2

1.0

100.0

674.00

2,089.4

3.1

1.4

62.8

42.60

2,112.1

49.6

22.5

63.5

4.10 0.35 0.15

1,124.9 737.4 400.0

274.4 2,106.9 2,666.7

124.7 957.7 1,212.1

33.8 22.2 12.0

48C. The pellet, obtained at between 45 and 60% ammonium sulfate saturation, was resuspended in a buffer containing 50 mM Tris (pH 7.0), 0.5 M (NH4)2SO4, 3 mM EDTA, and 0.2 M saccharose (buffer A). All subsequent chromatographic steps were performed with a fast protein liquid chromatography (FPLC) system (Pharmacia, Freiburg, Germany) at room temperature. Hydrophobic interaction chromatography. The enzyme solution obtained after ammonium sulfate fractionation was loaded onto a phenyl-Sepharose column (26 mm by 10 cm, highly substituted, 6% agarose fast flow; Pharmacia, Uppsala, Sweden) equilibrated with buffer A. The bound proteins were eluted in step gradients at 250 and 50 mM (NH4)2SO4 in the same buffer system. The b-galactosidase eluted in the presence of 250 mM (NH4)2SO4, and the active fractions were then pooled (60 ml) and lyophilized. Anion exchange chromatography on Mono Q 10/10 with pH shift. The lyophilized sample, after hydrophobic interaction chromatography, was dissolved in a buffer containing 50 mM KH2PO4-NaHPO4 (pH 6.0), 3 mM EDTA, and 0.2 M saccharose (buffer B) and passed through PD-10 columns (Pharmacia, Uppsala, Sweden) to discard low-molecular-weight compounds. The active enzyme solution was loaded onto a Mono Q 10/10 column (Pharmacia, Uppsala, Sweden) equilibrated with buffer B and washed with 6 column volumes. The column was developed with a pH gradient of 6.0 to 4.6 in the same buffer. The b-galactosidase eluted at approximately pH 5.3. Anion exchange chromatography on Mono Q 10/10 at pH 8.0. A sample containing the pooled active fractions of the former step was adjusted to pH 8.0 and loaded onto the same column, which was then equilibrated with a buffer containing 20 mM N-methyl-diethanolamine (pH 8.0), 0.3 M KCl, 3 mM EDTA, and 0.2 M saccharose. The column was washed with 6 column volumes of the buffer and then developed with an increasing salt gradient. The b-galactosidase eluted at approximately 440 mM KCl in the buffer system. Anion exchange chromatography on Mono Q 5/5 at pH 6.5. Active fractions were passed through PD-10 columns equilibrated with a buffer containing 20 mM bis-trispropane (pH 6.5), 3 mM EDTA, and 0.2 M saccharose and loaded onto an equilibrated Mono Q 5/5 column. The column was washed with 6 column volumes in the presence of 0.2 M KCl. The column was developed with a salt gradient, and the b-galactosidase eluted at approximately 230 mM KCl in the buffer system. Electrophoretic analysis. Sodium dodecyl sulfate (SDS)-denatured and native proteins were separated by electrophoresis (Phast-System; Pharmacia, Uppsala, Sweden) on commercially available polyacrylamide gels (Pharmacia, Uppsala, Sweden) and were stained with a 0.4% silver nitrate solution (silver stain kit; Pharmacia, Sweden). The protocols for the separation methods and gels are described by the manufacturer’s instructions (Phast-System; Pharmacia, Uppsala, Sweden). For activity staining, a gel obtained after native polyacrylamide electrophoresis was immediately incubated in assay buffer containing pNO2PheGal for 10 min at room temperature. For molecular mass estimation, the following markers were used: myosin (212 kDa), a2-macroglobin (170 kDa), b-galactosidase (116 kDa), transferrin (76 kDa), and glutamic dehydrogenase (53 kDa). Size exclusion chromatography for molecular mass studies. Gel filtrations (with active fractions obtained after the last-but-one purification step) were carried out with both a prepacked Superose 12 column (1-cm inside diameter; 30-cm gel bed height; 200-ml sample volume) and a self-packed Sephacryl S-300 column (1.6-cm inside diameter; 82-cm gel bed height; 1-ml sample volume) (both from Pharmacia, Uppsala, Sweden) equilibrated with 0.5 M Tris-HCl (pH 7.0), 3 mM EDTA, 0.1 M KCl, and 0.2 M saccharose. For molecular mass estimation, the following markers were used: thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), albumin (67 kDa), and ovalbumin (43 kDa). N-terminal amino acid sequence. A pure b-galactosidase sample was filtered through a polyvinylidene difluoride membrane (3-kDa size limit; Prospin; Applied Biosystems) by centrifugation. Continuous automated Edman degradation in an automated sequencer equipped with an on-line HPLC unit (Model 473A; Applied Biosystems) was performed.

Sp act (nkat/mg)

Purification (fold)

Yield (%)

Evidence for protein glycosylation. An immunoassay (DIG glycan/protein double labeling kit; Boehringer, Mannheim, Germany) was used for the qualitative estimation of possible sugar moieties covalently bound to the b-glycosidase. In this test, glycoproteins and proteins are visibly distinguished through two simultaneous, but specific, staining reactions with antibodies. The protein samples, containing pure b-galactosidase and glycosylated and nonglycosylated marker proteins, were incubated first with periodate and subsequently with digoxigenin-3-O-succinyl-ε-aminocaproic acid-hydrazide (DIG-hydrazide) to oxidize and label possible sugar residues. Then, the free amino groups of the proteins were labeled by reaction with fluorescein and the samples were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) with a 10% polyacrylamide gel in a Mini Protean II system (12 by 7 cm; Bio-Rad, Richmond, Calif.). The proteins were blotted onto a nitrocellulose membrane (0.45-mm pore size; BA 85; Schleicher & Schuell, Dassel, Germany). The filters were incubated as described in the kit protocol. Metal dependency. An enzyme solution (containing 5 mM EDTA) obtained after ammonium sulfate precipitation was desalted and transferred to 25 mM MOPS (3-morpholine propanesulfonic acid) buffer, pH 6.8, containing 14.2 mM metal salt (see below) by PD-10 gel filtration (Pharmacia, Freiburg, Germany). The solution (final volume, 700 ml) was incubated for 30 min at 378C. The assay was then started by addition of 300 ml of 15 mM pNO2PheGal in 25 mM MOPS, pH 6.8. The metal salt concentration during the assay was 10 mM. The metal salts tested were Co(NO3)2, MnSO4, FeSO4, MgSO4, ZnSO4, CaCl2, CuSO4, MnCl2, FeCl2, Na2MoO4, CoCl2, Ni(NO3)2, and SnCl2. The measured activities were compared with the activity of an enzyme solution without metal salt under the same conditions. pH stability. In each case, 2.5 ml of crude extract was passed through a PD-10 gel filtration column (Pharmacia, Freiburg, Germany) to remove low-molecularmass compounds and to change the buffer system to (i) 0.2 M potassium phosphate buffer (pH 6, 7, and 8), (ii) 0.2 M Tris/HCl (pH 7, 8.5, and 9), or (iii) 0.2 M glycine (pH 9 and 10). A 1,500-ml sample of the obtained enzyme solution was covered with 200 ml of oil and incubated at 4 and at 378C. After 20 h, the b-galactosidase activities of the samples were measured. Temperature stability. (i) Crude extract (500 ml) was covered with oil (100 ml) and incubated at 0, 30, 37, 42, 47, 50, 53, and 568C. After an incubation time of 1 or 2 h, samples were withdrawn and immediately placed on ice before measurement of the b-galactosidase activity. (ii) Crude extract (1,000 ml) was covered with oil (100 ml) and incubated at 0, 50, and 538C. After 1, 3, 5, 7, 10, 23, 30, and 48 h, samples were withdrawn and the b-galactosidase activities were measured. (iii) Diluted enzyme solution (1,000 ml) obtained after ammonium sulfate precipitation (the resuspended pellet of the 45 to 60% satisfied salt fraction in 0.1 M potassium phosphate buffer, pH 6.8) was covered with oil (100 ml) and incubated at 0 and 508C for 191 h. Samples were withdrawn after intervals, and the b-galactosidase activities were measured.

RESULTS Purification of the b-galactosidase. T. lanuginosus cells were harvested at the beginning of the stationary growth phase and disrupted with a glass bead mill. The procedure for purification of the b-galactosidase is summarized in Table 1. Starting with 190 ml of crude extract containing a b-galactosidase activity of 17.5 nkat/ml, the first purification step applied was a fractional ammonium sulfate precipitation. All following steps were carried out by column chromatography with FPLC equipment. The hydrophobic interaction chromatography resulted in a 22.5-fold purification without any loss of total activity. In the subsequent three anion exchange chromatography steps, the

VOL. 61, 1995

THERMOTOLERANT b-GALACTOSIDASE FROM T. LANUGINOSUS

1499

TABLE 2. Characterization of T. lanuginosus b-galactosidase Parameter

Result

Molecular mass (kDa) as determined by: Electrophoretic investigations (SDS) Polyacrylamide gradient gel 8–25%..................... Polyacrylamide homogenous gel 12.5% .............. Size exclusion chromatography Sephacryl S-300 gel ................................................ Superose 12 gel ......................................................

200 200–220

Glycosylated protein ......................................................

Yesa

Isoelectric point ..............................................................

4.4–4.5a

Metal dependency ..........................................................

No

75–80a 75–80a

pH (optimum)................................................................. 6.7–7.2 pH (stability)................................................................... 6–9 (378C, 20 h) Temperature stability at 508C (% remaining activity) Crude extract .............................................................. 47 after 48 h After (NH4)2SO4 precipitation................................. 58 after 191 h a

FIG. 1. SDS-PAGE of the pure b-galactosidase on an 8 to 25% polyacrylamide gradient gel after staining with silver nitrate. Lane 1, marker proteins (A, myosin [212 kDa]; B, a2-macroglobulin [170 kDa]; C, b-galactosidase [116 kDa]; D, transferrin [76 kDa]; E, glutamic dehydrogenase [53 kDa]). Lane 2, pure b-galactosidase. (For details, see Materials and Methods).

elution conditions were altered each time (pH shift, 6.0 to 4.6; salt gradients at pH 8.0 and 6.5) and led to 1,212-fold purification. The enzyme fraction after the third and last Mono Q step had a specific activity of 2,667 nkat/mg of protein (160 U/mg), and the yield was 12.0%. The SDS-PAGE of this fraction on an 8 to 25% polyacrylamide gradient gel showed a single protein band after being stained with silver nitrate (Fig. 1, lane 2). The purified enzyme was subject to N-terminal amino acid sequencing by Edman degradation. The first 22 residues of the purified b-galactosidase are depicted in Fig. 2. In addition, a sample of the pure enzyme fraction was also run by native PAGE. One gel obtained after native electrophoresis was stained with silver nitrate as usual, but another gel was incubated in assay buffer containing pNO2PheGal. Both gels demonstrated a single stained band at the same place on the gels. The silver nitrate staining led to a dark band, whereas the activity staining resulted in a yellow band caused by the pNO2-phenol ion released because of b-galactosidase activity (not shown). Properties of the b-galactosidase. The properties of the b-galactosidase from T. lanuginosus are summarized in Table 2. Electrophoretic investigations of the pure enzyme in the presence of SDS indicated a molecular mass of 75 to 80 kDa for the unfolded, denatured b-galactosidase subunit. As only a single protein band occurred in the SDS gel (Fig. 1, lane 2) and the protein sequencing indicated homogeneity (Fig. 2), it was concluded that the b-galactosidase consists of two identical subunits. The molecular mass of the native b-galactosidase was

FIG. 2. Partial N-terminal amino acid sequence of T. lanuginosus b-galactosidase.

Estimated with pure enzyme.

estimated by size exclusion chromatography at an apparent 200 to 220 kDa during gel filtration. To ascertain whether the purified b-galactosidase contained carbohydrate moieties, the homogeneous enzyme fraction was investigated by immunoassay. After glycoprotein-specific staining, a resulting blue band supplied evidence for the existence of carbohydrate structures covalently bound to the protein. The isoelectric point was estimated by isoelectric focusing on a polyacrylamide gel at 4.4 to 4.5. No metal dependency of the b-galactosidase was observed. All tested metal salts (see Materials and Methods) led to a decrease in enzyme activity. The pH optimum for the hydrolysis of pNO2PheGal was between pH 6.7 and 7.2. pH and temperature stability. The b-galactosidase was incubated for 20 h at 378C at various pH values. The enzyme was stable over a broad pH range, since no loss of enzyme activity was observed between pH 6 and 9 during this time. The thermal stability of the b-galactosidase is demonstrated in Fig. 3. At temperatures up to 478C, no loss in catalytic activity occurred after 2 h, whereas at 568C nearly all enzyme was dena-

FIG. 3. Temperature stability of the b-galactosidase from T. lanuginosus (for details, see Materials and Methods).

1500

FISCHER ET AL.

APPL. ENVIRON. MICROBIOL.

TABLE 3. Substrate specificity and kinetic constants of pure T. lanuginosus b-galactosidase Substrate

o-NO2-phenyl-b-D-galactopyranoside p-NO2-phenyl-b-D-galactopyranoside p-NO2-phenyl-a-L-arabinopyranoside Phenyl-b-D-galactopyranoside Lactulose Lactose MeOH-b-D-galactopyranoside p-NO2-phenyl-b-D-glucopyranoside p-NO2-phenyl-b-L-arabinopyranoside p-NO2-phenyl-a-D-mannopyranoside p-NO2-phenyl-a-L-fucopyranoside p-NO2-phenyl-a-D-galactopyranoside p-NO2-phenyl-a-D-glucopyranoside

Characteristic of b-galactosidase Hydrolytic activity (%)a

Km (mM)

kcat (s21)b

kcat/Km (s21 z mM21)

100.0 85.8 11.9c 8.4 4.8 1.8 1.6 0 0 0 0 0 0

11.3 4.8 75.4d 35.9 5.0 18.2 6.4

941 1,035 501d 65 33 18 16

83.3 215.6 6.6d 1.8 6.6 1.0 2.5

100% enzyme activity 5 2,740.4 nkat z mg21. Calculated from theoretical Vmax values obtained from Hanes plots. c Measured at maximal solubility of p-NO2-phenyl-b-L-arabinopyranoside (21.8 mM). d Calculated on activity values obtained in the range of 0 to 21.8 mM substrate. a b

tured after 1 h. In consequence, the enzyme stability was tested at 508C with both a crude extract and an enzyme solution after ammonium sulfate fractionation. The crude extract retained 47% of its activity after 48 h, whereas the further-purified enzyme solution (after ammonium sulfate precipitation) was more stable and retained 58% activity after 191 h. Substrate specificity and kinetic constants. All compounds tested for hydrolysis by the purified b-galactosidase are summarized in Table 3. The hydrolytic activities, except for those of p-NO2-phenyl-a-L-arabinopyranoside, whose maximal solubility at pH 6.8 and 378C is 21.8 mM, were measured under substrate saturation conditions. The highest hydrolytic activities were observed with o- and p-NO2-phenyl-b-D-galactopyranoside as substrates. The a-configured p-NO2-phenyl-D-galactopyranoside was not hydrolyzed. Other substrates which underwent hydrolysis included the disaccharides lactose and lactulose, the alkyl glycoside MeOH-b-D-galactopyranoside, the aryl-glycoside phenyl-b-D-galactopyranoside, and p-NO2phenyl-a-L-arabinopyranoside. All other compounds tested as substrates failed to be hydrolyzed by the purified b-galactosidase. The Km, kcat, and kcat/Km values were determined (Table 3). The kcat values were calculated on the basis of the theoretical Vmax values obtained from Hanes plots and under the assumption of a dimer and one active site per subunit. The b-galactosidase’s highest affinities, as indicated by the Km values, were to p-NO2-phenyl-b-D-galactopyranoside, lactulose, and MeOH-b-D-galactopyranoside, with Km values of 4.8, 5.0, and 6.4 mM, respectively. As mentioned before, the nitro-arylgalactosides p- and o-NO2-phenyl-b-D-galactopyranoside were the most favorable substrates, with kcat/Km values of 215.6 and 83.3 s21 z mM21, respectively. The km, kcat, and kcat/Km values for p-NO2-phenyl-a-L-arabinopyranoside were calculated by measurement of the activities of the b-galactosidase over the range of 0 to 21.8 mM. DISCUSSION The intracellular b-galactosidase described in this paper is, as with the previously described a-amylase (10) and lipase (8), another enzyme from the thermophilic fungus T. lanuginosus which has demonstrated a reasonable thermostability. The enzyme retains some 58% of its activity at 508C after 191 h of incubation and suffers no loss in activity at 478C and 2 h of incubation (Fig. 3). The higher thermostability (608C, 20 h) of

the lipase and quite similar thermostability of the a-amylase (508C, 4 h) could be due to the fact that both are extracellular enzymes of the same microorganism, and extracellular enzymes are known to be more stable than intracellular ones. Another intracellular and quite thermostable b-galactosidase from a Thermoanaerobacter sp. (12) with a half-life of 2.58 h at 758C was unfortunately produced only under anaerobic conditions and in low quantities. Typical cell yields after fermentation of this microorganism were 3.3 g (wet biomass) liter21, resulting in 1.5 nkat of b-galactosidase activity g21 (wet biomass). In comparison, T. lanuginosus was harvested with a yield of 7 g (wet biomass) liter21, resulting in 31.4 nkat g21 (wet biomass) at 378C. This leads to approximately 45 times more b-galactosidase liter21. The advantageous pH stability of the b-galactosidase from T. lanuginosus between pH 6 and 9 (Table 2) is perhaps not unprecedented for these enzymes. The Aspergillus niger and Aspergillus oryzae b-galactosidases (5) were described as stable between pH 2.5 and 8.0 and pH 2.5 and 7.0, respectively, and Bacillus macerans b-galactosidase (14) was also stable between pH 6 and 9. Here, we utilized this property for varying the conditions in anion exchange chromatography on Mono Q (Table 1). First, the ion concentration of the elution buffer was kept constant and the protein elution was effected by pH shift, and then the ion concentrations of the elution buffers were increased at different, but constant, pH values. The pure b-galactosidase fraction obtained after the last purification step had a specific activity of 160 U mg21. Other purified b-galactosidases, for instance, those of Streptococcus thermophilus (19) and Thermoanaerobacter sp. (12), had similar specific activities after the last purification step (466 and 99 U mg21, respectively). The size of the native b-galactosidase from T. lanuginosus was estimated by electrophoretic and gel chromatography investigations, but unfortunately the methods were not exactly in accord with each other (Table 2). This could be due to the fact that the enzyme is glycosylated and not globular in its dimeric configuration. After denaturation of the protein by SDS, an accurate determination of a molecular mass of 75 to 80 kDa was made, both by gradient (Fig. 1) and by homogenous PAGE (data not shown). As sequencing of the pure b-galactosidase by Edman degradation yielded a definite amino acid order (Fig. 2), it must consist of homogenous subunits. Therefore, and in

THERMOTOLERANT b-GALACTOSIDASE FROM T. LANUGINOSUS

VOL. 61, 1995

consideration of the size estimation experiments with the native enzyme, two subunits for the native configuration are assumed, with a total theoretical molecular mass of 150 to 160 kDa. Other dimeric b-galactosidases described in the literature were those of Penicillium citrinum (two 60-kDa subunits) and Corynebacterium murisepticum (two 100-kDa subunits) (15). The preferred substrates for the b-galactosidase were nitroaryl-b-D-galactosides (Table 3). Apart from that, the substrates had to contain a b-D-galactopyranoside residue as the glycone moiety, since other nitro-aryl-b-D-monosaccharides were not hydrolyzed. The pure enzyme cleaved O-glycosidic bonds of the D-configured galactopyranosides at C-1 stereospecifically in the b-anomeric form. One single exception was the acceptance of p-NO2-phenyl-a-L-arabinopyranoside as a substrate, which was also described for the Thermoanaerobacter b-galactosidase (12). One explanation for this phenomenon could be the homologous three-dimensional ring structure of the pentose arabinopyranoside and the hexose galactopyranoside, whereby the difference between the two monosaccharides exists in the additional H2(OH)C group at C-5 in the case of the galactopyranoside. The fact that this group is missing as a possible fixing point when the pyranose moiety is bound to the enzyme’s active center, as is the case with p-NO2-phenyl-a-L-arabinopyranoside, could be the reason for its acceptance as a substrate in its a- but not b-anomeric configuration. However, more experiments have to be done to provide proofs for this hypothesis. The aglycone moiety of the substrates of the b-galactosidase from T. lanuginosus can vary from glycosyl, alkyl, or aryl residues (Table 3), and the enzyme is catalytically active for days at 508C (Table 2). These are the major reasons why the enzyme should be an interesting tool in the synthesis of glycoconjugates, a possibility which is presently under investigation in our laboratory. Furthermore, the information concerning the Nterminal sequence could be used to find the proper gene and to overexpress the b-galactosidase from T. lanuginosus in a suitable host system.

3.

4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

14. 15.

16.

17.

ACKNOWLEDGMENTS

18.

We thank M. Kies (GBF, Braunschweig, Germany) for performing the N-terminal amino acid sequencing and Ian Nicholls for linguistic advice.

19.

20.

REFERENCES 1. Attal, S., S. Bay, and D. Cantacuzene. 1992. Enzymatic synthesis of b-galactosylpeptides and of b-(1,3)-digalactosyl serine derivatives using b-galactosidase. Tetrahedron 48:9251–9260. 2. Bader, D. E., M. Ring, and R. E. Huber. 1988. Site-directed mutagenic

21. 22.

1501

replacement of Glu-461 with Gln in b-galactosidase (E. coli): evidence that Glu-461 is important for activity. Biochem. Biophys. Res. Commun. 153: 301–305. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248–254. Crout, D. H. G., D. A. MacManus, and P. Critchley. 1990. Enzymatic synthesis of glycosides using b-galactosidase of E. coli: regio- and stereo-chemical studies. J. Chem. Soc. Perkin Trans. I 7:1865–1868. Crueger, A., and W. Crueger. 1984. Carbohydrates, p. 421–457. In K. Kieslich (ed.), Biotechnology, vol. 6a. Verlag Chemie, Weinheim, Germany. Gonza ´lez, M. I., and S. Sua ´rez Doval. 1994. Klyveromyces lactis immobilisation on corn grits for milk whey lactose hydrolysis. Enzyme Microb. Technol. 16:303–310. Herrmann, G. F., Y. Ichigawa, C. Wandrey, F. C. A. Gaeta, J. C. Paulson, and C. H. Wong. 1993. A new multi-enzyme system for a one pot synthesis of sialyl oligosaccharides: combined use of b-galactosidase and a-(2,6)-sialyl transferase coupled with regeneration in situ of CMP-sialic acid. Tetrahedron Lett. 34:3091–3094. Ibrahim, C. O., N. Nishio, and S. Nagai. 1987. Production of a thermostable lipase by Humicola lanuginosa grown on sorbitol-corn steep liquor medium. Agric. Biol. Chem. 51:2145–2151. Jacobson, R. H., X.-J. Zhang, F. DuBose, and B. W. Matthews. 1994. Threedimensional structure of b-galactosidase from E. coli. Nature (London) 369:761–766. Jensen, B., and J. Olsen. 1992. Physicochemical properties of a purified a-amylase from the thermophilic fungus Thermomyces lanuginosus. Enzyme Microb. Technol. 14:112–116. Johansson, E., L. Hedbys, and P.-O. Larsson. 1991. Enzymatic synthesis of monosaccharide-amino acid conjugates. Enzyme Microb. Technol. 13:781– 787. Lind, D. L., R. M. Daniel, D. A. Cowan, and H. W. Morgan. 1989. b-Galactosidase from a strain of the anaerobic thermophile, Thermoanaerobacter. Enzyme Microb. Technol. 11:180–186. Martinek, K., and V. V. Mozhaev. 1993. Need for thermostability, its benefits and main strategies for thermostabilization, p. 76–83. In M. N. Gupta (ed.), Thermostability of enzymes. Springer-Verlag, Berlin. Miyazaki, Y. 1988. Purification, crystallization and properties of b-galactosidase from Bacillus macerans. Agric. Biol. Chem. 52:625–631. Priyolkar, M. R., C. K. K. Nair, and D. S. Pradhan. 1989. Purification and characterization of an inducible b-galactosidase from Corynebacterium murisepticum. Arch. Microbiol. 151:49–53. Ring, M., D. E. Bader, and R. E. Huber. 1988. Site-directed mutagenesis of b-galactosidase (E. coli) reveals that Tyr-503 is essential for activity. Biochem. Biophys. Res. Commun. 152:1050–1055. Sauerbrei, B., and J. Thiem. 1992. Galactosylation and glucosidation by use of b-galactosidase. Tetrahedron Lett. 33:201–204. Sinnot, M. L. 1978. Ions, ion-pairs and catalysis by the lacZ b-galactosidase of E. coli. FEBS Lett. 94:1–9. Smart, J., and B. Richardson. 1987. Molecular properties and sensitivity to cations of b-galactosidase from Streptococcus thermophilus with four enzyme substrates. Appl. Microbiol. Biotechnol. 26:177–185. Stevenson, D. E., R. A. Stanley, and R. H. Furneaux. 1993. Optimization of alkyl b-D-galactopyranoside synthesis from lactose using commercially available b-galactosidases. Biotechnol. Bioeng. 42:657–666. Uhlig, H. 1991. Enzyme arbeiten fu ¨r uns: technische Enzyme und ihre Anwendungen, p. 77–78. Hanser, Munich. Vulfson, E. N., and B. A. Law. 1993. Enzymatic synthesis and modification of carbohydrates. J. Chem. Technol. Biotechnol. 56:409.