/ . Biochem., 80, 209-215 (1976)
Lactosylceramide /3-Galactosidase Kenji NISHIMURA and Reiko AMANO Department of Biology, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka 431-31 Received for publication, January 28, 1976
Porcine thymus lactosylceramide ^-galactosidase was purified by a simple procedure. In the final step of isoelectric focusing the enzyme was separated into two peaks of p / 6.3 (peak I) and 7.0 (peak II), which showed 3,600- and 4,000-fold enhancement of lactosylceramide-hydrolysing activity, respectively. The two peaks had identical mobility on polyacrylamide gel electrophoresis. The apparent molecular weight was 34,000. Neither monosialoganglioside (GMi) nor galactosylceramide was hydrolysed by the purified enzyme fractions. The optimal pH was at 4.6, and sodium taurocholate was essential for the reaction. The apparent Km was 2.3 xlO~5 M. The reaction was stimulated by sodium chloride and linoleic acid, while it was strongly inhibited by Triton X-100 and bovine serum albumin. Galactosylceramide, />-nitrophenyl /3galactoside, and />-nitrophenol were weak inhibitors. No effects of GM! and galactose were observed on the hydrolysis of lactosylceramide.
A glycolipid storage disease of GMrgangliosidosis is attributed to congenital deficiency of /9-galactosidase, which hydrolyses the nonreducing galactose end of GMi (1). Lactosylceramidosis, found by Dawson and Stein (2), is another type of disorder caused by a deficiency of the galactosidase which is responsible for the catabolism of lactosylceramide. These findings suggest the presence of galactosidases at each catabolic process of glycolipids. Many reports on the properties of glycolipid-^-galactosidases have been published recently {3-15), but several investigators (6, 13—15) have been unable to separate lactosylceramide-hydrolysing activity from activity toward GM 1 F galactosyl-
ceramide, or other glycolipids. This paper reports the partial purification and properties of porcine thymus ^-galactosidase, which is highly specific for lactosylceramide. METHODS Purification of the Enzyme — Porcine thymus tissue (4-6 months) was obtained from a slaughter house, and stored at —20°. All the procedures were carried out at 0—4°. The tissue was thawed before use and homogenized in 4 volumes of distilled water. The homogenate was centrifuged at 25,000X0 for 30 min, and the supernatant was collected. The supernatant fraction was concentrated by ammonium sulfate fractionation between 20
Abbreviations : GMi, monosialoganglioside ; TCA, trichloroacetic acid. Vol. 80, No. 2, 1976
209
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
Partial Purification and Properties of Porcine Thymus
210
ness in a test tube and the dry residue was suspended uniformly in 0.3 ml of distilled water using a conventional sonicator bath. The incubation system consisted of 0.3 ml of the substrate mixture, 0.3 ml of Mcllvaine buffer (23) and 0.1 ml of enzyme. Incubation was carried out at 37° for 30 min. [8H]Galactose liberated from lactosylceramide was measured by the method of Radin and Arora (24) except that the scale was reduced to 7/10. Galactosylceramide /3-galactosidase [EC 3.2.1.46] was assayed by the same procedures as lactosylceramide /J-galactosidase, using substrate of similar specific activity. GMi /3-galactosidase was measured by the method of Miyatake and Suzuki (9). These assays were undertaken at the optimal pH of 5.4 in Mcllvaine buffer. p-Nitrophenyl-fi-galactosidase—The incubation system contained 0.1 ml of enzyme, 0.5 ml of Mcllvaine buffer (pH 4.6) and 0.2 ml of 10 mM />-nitrophenyl /3-galactosidc. The mixture was incubated at 37° for 1 hr and the reaction was stopped by adding 1 ml of 5% trichloroacetic acid (TCA). After brief centrifugation, 1 ml of clear supernatant was made alkaline with 0.2 ml of 1 N KOH and 2 ml of borate buffer (pH 9.5), and liberated p-n\trophenol was measured at 410 nm. Assays of other glycosidases were carried out with 0.1 ml of enzyme, 0.2 ml of 5 mM substrate and 0.5 ml of Mcllvaine buffer at the appropriate optimal pH as follows; phenyl-agalactosidase [EC 3. 2.1. 22]: pH 4.6, phenyla-glucosidase [EC 3. 2.1. 20]: pH 4.6, phenyl^-glucosidase [EC 3. 2.1. 21]: pH 4.6, phenyla-N-acetylgalactosaminidase [EC 3. 2.1.49]: pH 3.0, phenyl-/J-N-acetylgalactosaminidase [EC 3.2.1.53]: pH 5.0 and phenyl-^-N-acetylglucosaminidase [EC 3.2.1.30]: pH 5.0. Incubation was stopped by adding 1 ml of 5% TCA after 1 hr at 37°, and liberated phenol was measured by the method of Lowry et al. (17). The value obtained in substrate-free runs was subtracted from each experimental series, fiGlucuronidase [EC 3. 2.1. 31] was assayed according to the method of Barrett (25). Disc Gel Electrophoresis — Samples were dialyzed exhaustively against distilled water for several days to remove carrier ampholytes. / . Biochem.
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
and 55% saturation. The fraction collected by brief centrifugation was suspended in a small volume of 1 M acetate buffer dropwise to the suspension, mixing continuously with a magnetic stirrer. The suspension was mixed for at least 15 min, then centrifuged at 25,000 xg for 30 min. The clear supernatant was dialyzed overnight against 0.1 M phosphate buffer (pH 7.4) containing 0.2 M sodium chloride, and concentrated to a small volume by ultrafiltration using PM-10 membranes (Amicon) under nitrogen (2 kg/cm J ). The sample was loaded on a column (2.6x45 cm) packed with Sephadex G-200 and eluted with the same phosphate buffer. Fractions of 4 ml were collected and aliquots were subjected to enzyme assay. Fractions having /3-galactosidase activity • were pooled and dialysed against distilled water overnight. Isoelectric focusing was performed at 4° on an LKB8101 column (110 ml) with pH 3.5—10 Ampholine (at 2%, v/v) in a sucrose gradient (16). The column was unloaded under gravity and the effluent was monitored at 280 nm. An aliquot of each 2 ml fraction was subjected to enzyme assay. Protein concentration was determined by the method of Lowry et al. (17). Preparation of Labeled Substrates—Lactosylceramide was isolated from a partial hydrolysate of bovine brain gangliosides (0.1 M HC1 for 1 hr at 80°) and purified by Florisil column chromatography (18). GMt was prepared by treatment with neuraminidase [EC 3.2.1.18] (19), and purified by silicic acid column chromatography (20). Galactosylceramide was obtained from bovine brain (21). ['HJLactosylceramide, [*H]galactosylcer amide, and [3H]GMi were prepared by the method of Radin (22), and diluted to appropriate specific activities with carrier glycolipids. They were all stored in chloroform: methanol ( 2 : 1 , v/v) at —20°. Lactosylceramide fi-Galactosidase—A substrate mixture for the standard assay procedures was prepared as follows: a mixture of solutions of 8H-labeled lactosylceramide (0.1 mg, 50,000 dpm, per test tube), sodium taurocholate (0.1 mg, Wako Chemical Co.), and linoleic acid (0.2 mg) was evaporated to dry-
K. NISHIMURA and R. AMANO
LACTOSYLCERAMIDE /3-GALACTOSIDASE FROM PORCINE THYMUS
RESULTS The results of a purification run with porcine thymus /3-galactosidase are summarized in Table I. An artificial substrate, />-nitrophenyl /3-galactoside, was also used besides lactosylceramide. Such artificial substrates have been used by a number of investigators (3—15), mainly for convenience in the assay procedure. The extent of purification was 27 times with ^-nitrophenyl /3-galactoside, while it was 543 times with lactosylceramide. In all the experimental runs, including that presented in the table, the yield of the step after ammonium sulfate fractionation exceeded 100%. In small-scale experiments,
when the tissue was homogenized in 9 volumes of water, the yield of enzyme activity in the 25,000X0 supernatant was about 80%, but when the enzyme was prepared on a larger scale, as shown in the table, the tissue was homogenized in 4 volumes of water, and the loss of enzyme in the 25,000 Xg sediment was disregarded. Thus, the yield could be much more although it might be suggestive of the presence of an inhibitory factor in the crude homogenate, or the presence of an endogenous substrate. Similar circumstances have been observed in several cases with other glyco-
sidases (13, 27). Galactosylceramide was separated in a supernatant at pH 5.0 by Bowen and Radin (13). Norden et al. (14) obtained GMi galactosidase as a supernatant at pH 4.6. Distler and Jourdian (6) extracted testicular /3-galactosidase at pH 4.0. In this experiment, lactosylceramidase was separated as a supernatant from the bulk of the proteins at pH 3.0. When citrate buffer at an identical pH was used in place of acetate buffer, considerable loss of the activity was observed. The specific activity of this fraction was increased to 534 nmoles/
TABLE I. Purification of porcine thymus /3-galactosidase.
Purification steps Homogenate 25,000xg supernatant (NH4),SO4 fractionation Supernatant at pH 3 Sephadex G-200
Purification steps
(rag)
Total activity (^moles/hr)
9,100 2,977 1,881 395.7 33.4
88.0 54.1 106.3 211.4 176.0
"Vol. 80, No. 2, 1976
Specific activity (nmoles/mg protein/hr) 9.7
18.2 56.5 534.4 5,270
Purification (fold) 1 1.9
5.8 55.1 543
/>-Nitrophenyl ^-galactoside
Protein (rag)
Homogenate 25,000xg supernatant (NH4),SO4 fractionation Supernatant at pH 3 Sephadex G-200
Lactosylceramide
Protein
Total activity (/imoles/hr)
Specific activity (nmoles/mg protein/hr)
Purification (fold) 1
9,100 2,977
731.2 337.1
80.1 113.2
1.4
1,881 395.7
235.4 204.7 71.1
125.6 523.0
1.6 6.5
33.4
2,129
26.6
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
Polyacrylamide gel electrophoresis was carried out with 7.5% gels at pH 2.3 according to the method described by Davis (26). Proteins on the gels were detected by staining with Brilliant cresyl blue. Bovine serum albumin
Lactosylceramide /3-Galactosidase Kenji NISHIMURA and Reiko AMANO Department of Biology, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka 431-31 Received for publication, January 28, 1976
Porcine thymus lactosylceramide ^-galactosidase was purified by a simple procedure. In the final step of isoelectric focusing the enzyme was separated into two peaks of p / 6.3 (peak I) and 7.0 (peak II), which showed 3,600- and 4,000-fold enhancement of lactosylceramide-hydrolysing activity, respectively. The two peaks had identical mobility on polyacrylamide gel electrophoresis. The apparent molecular weight was 34,000. Neither monosialoganglioside (GMi) nor galactosylceramide was hydrolysed by the purified enzyme fractions. The optimal pH was at 4.6, and sodium taurocholate was essential for the reaction. The apparent Km was 2.3 xlO~5 M. The reaction was stimulated by sodium chloride and linoleic acid, while it was strongly inhibited by Triton X-100 and bovine serum albumin. Galactosylceramide, />-nitrophenyl /3galactoside, and />-nitrophenol were weak inhibitors. No effects of GM! and galactose were observed on the hydrolysis of lactosylceramide.
A glycolipid storage disease of GMrgangliosidosis is attributed to congenital deficiency of /9-galactosidase, which hydrolyses the nonreducing galactose end of GMi (1). Lactosylceramidosis, found by Dawson and Stein (2), is another type of disorder caused by a deficiency of the galactosidase which is responsible for the catabolism of lactosylceramide. These findings suggest the presence of galactosidases at each catabolic process of glycolipids. Many reports on the properties of glycolipid-^-galactosidases have been published recently {3-15), but several investigators (6, 13—15) have been unable to separate lactosylceramide-hydrolysing activity from activity toward GM 1 F galactosyl-
ceramide, or other glycolipids. This paper reports the partial purification and properties of porcine thymus ^-galactosidase, which is highly specific for lactosylceramide. METHODS Purification of the Enzyme — Porcine thymus tissue (4-6 months) was obtained from a slaughter house, and stored at —20°. All the procedures were carried out at 0—4°. The tissue was thawed before use and homogenized in 4 volumes of distilled water. The homogenate was centrifuged at 25,000X0 for 30 min, and the supernatant was collected. The supernatant fraction was concentrated by ammonium sulfate fractionation between 20
Abbreviations : GMi, monosialoganglioside ; TCA, trichloroacetic acid. Vol. 80, No. 2, 1976
209
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
Partial Purification and Properties of Porcine Thymus
210
ness in a test tube and the dry residue was suspended uniformly in 0.3 ml of distilled water using a conventional sonicator bath. The incubation system consisted of 0.3 ml of the substrate mixture, 0.3 ml of Mcllvaine buffer (23) and 0.1 ml of enzyme. Incubation was carried out at 37° for 30 min. [8H]Galactose liberated from lactosylceramide was measured by the method of Radin and Arora (24) except that the scale was reduced to 7/10. Galactosylceramide /3-galactosidase [EC 3.2.1.46] was assayed by the same procedures as lactosylceramide /J-galactosidase, using substrate of similar specific activity. GMi /3-galactosidase was measured by the method of Miyatake and Suzuki (9). These assays were undertaken at the optimal pH of 5.4 in Mcllvaine buffer. p-Nitrophenyl-fi-galactosidase—The incubation system contained 0.1 ml of enzyme, 0.5 ml of Mcllvaine buffer (pH 4.6) and 0.2 ml of 10 mM />-nitrophenyl /3-galactosidc. The mixture was incubated at 37° for 1 hr and the reaction was stopped by adding 1 ml of 5% trichloroacetic acid (TCA). After brief centrifugation, 1 ml of clear supernatant was made alkaline with 0.2 ml of 1 N KOH and 2 ml of borate buffer (pH 9.5), and liberated p-n\trophenol was measured at 410 nm. Assays of other glycosidases were carried out with 0.1 ml of enzyme, 0.2 ml of 5 mM substrate and 0.5 ml of Mcllvaine buffer at the appropriate optimal pH as follows; phenyl-agalactosidase [EC 3. 2.1. 22]: pH 4.6, phenyla-glucosidase [EC 3. 2.1. 20]: pH 4.6, phenyl^-glucosidase [EC 3. 2.1. 21]: pH 4.6, phenyla-N-acetylgalactosaminidase [EC 3. 2.1.49]: pH 3.0, phenyl-/J-N-acetylgalactosaminidase [EC 3.2.1.53]: pH 5.0 and phenyl-^-N-acetylglucosaminidase [EC 3.2.1.30]: pH 5.0. Incubation was stopped by adding 1 ml of 5% TCA after 1 hr at 37°, and liberated phenol was measured by the method of Lowry et al. (17). The value obtained in substrate-free runs was subtracted from each experimental series, fiGlucuronidase [EC 3. 2.1. 31] was assayed according to the method of Barrett (25). Disc Gel Electrophoresis — Samples were dialyzed exhaustively against distilled water for several days to remove carrier ampholytes. / . Biochem.
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
and 55% saturation. The fraction collected by brief centrifugation was suspended in a small volume of 1 M acetate buffer dropwise to the suspension, mixing continuously with a magnetic stirrer. The suspension was mixed for at least 15 min, then centrifuged at 25,000 xg for 30 min. The clear supernatant was dialyzed overnight against 0.1 M phosphate buffer (pH 7.4) containing 0.2 M sodium chloride, and concentrated to a small volume by ultrafiltration using PM-10 membranes (Amicon) under nitrogen (2 kg/cm J ). The sample was loaded on a column (2.6x45 cm) packed with Sephadex G-200 and eluted with the same phosphate buffer. Fractions of 4 ml were collected and aliquots were subjected to enzyme assay. Fractions having /3-galactosidase activity • were pooled and dialysed against distilled water overnight. Isoelectric focusing was performed at 4° on an LKB8101 column (110 ml) with pH 3.5—10 Ampholine (at 2%, v/v) in a sucrose gradient (16). The column was unloaded under gravity and the effluent was monitored at 280 nm. An aliquot of each 2 ml fraction was subjected to enzyme assay. Protein concentration was determined by the method of Lowry et al. (17). Preparation of Labeled Substrates—Lactosylceramide was isolated from a partial hydrolysate of bovine brain gangliosides (0.1 M HC1 for 1 hr at 80°) and purified by Florisil column chromatography (18). GMt was prepared by treatment with neuraminidase [EC 3.2.1.18] (19), and purified by silicic acid column chromatography (20). Galactosylceramide was obtained from bovine brain (21). ['HJLactosylceramide, [*H]galactosylcer amide, and [3H]GMi were prepared by the method of Radin (22), and diluted to appropriate specific activities with carrier glycolipids. They were all stored in chloroform: methanol ( 2 : 1 , v/v) at —20°. Lactosylceramide fi-Galactosidase—A substrate mixture for the standard assay procedures was prepared as follows: a mixture of solutions of 8H-labeled lactosylceramide (0.1 mg, 50,000 dpm, per test tube), sodium taurocholate (0.1 mg, Wako Chemical Co.), and linoleic acid (0.2 mg) was evaporated to dry-
K. NISHIMURA and R. AMANO
LACTOSYLCERAMIDE /3-GALACTOSIDASE FROM PORCINE THYMUS
RESULTS The results of a purification run with porcine thymus /3-galactosidase are summarized in Table I. An artificial substrate, />-nitrophenyl /3-galactoside, was also used besides lactosylceramide. Such artificial substrates have been used by a number of investigators (3—15), mainly for convenience in the assay procedure. The extent of purification was 27 times with ^-nitrophenyl /3-galactoside, while it was 543 times with lactosylceramide. In all the experimental runs, including that presented in the table, the yield of the step after ammonium sulfate fractionation exceeded 100%. In small-scale experiments,
when the tissue was homogenized in 9 volumes of water, the yield of enzyme activity in the 25,000X0 supernatant was about 80%, but when the enzyme was prepared on a larger scale, as shown in the table, the tissue was homogenized in 4 volumes of water, and the loss of enzyme in the 25,000 Xg sediment was disregarded. Thus, the yield could be much more although it might be suggestive of the presence of an inhibitory factor in the crude homogenate, or the presence of an endogenous substrate. Similar circumstances have been observed in several cases with other glyco-
sidases (13, 27). Galactosylceramide was separated in a supernatant at pH 5.0 by Bowen and Radin (13). Norden et al. (14) obtained GMi galactosidase as a supernatant at pH 4.6. Distler and Jourdian (6) extracted testicular /3-galactosidase at pH 4.0. In this experiment, lactosylceramidase was separated as a supernatant from the bulk of the proteins at pH 3.0. When citrate buffer at an identical pH was used in place of acetate buffer, considerable loss of the activity was observed. The specific activity of this fraction was increased to 534 nmoles/
TABLE I. Purification of porcine thymus /3-galactosidase.
Purification steps Homogenate 25,000xg supernatant (NH4),SO4 fractionation Supernatant at pH 3 Sephadex G-200
Purification steps
(rag)
Total activity (^moles/hr)
9,100 2,977 1,881 395.7 33.4
88.0 54.1 106.3 211.4 176.0
"Vol. 80, No. 2, 1976
Specific activity (nmoles/mg protein/hr) 9.7
18.2 56.5 534.4 5,270
Purification (fold) 1 1.9
5.8 55.1 543
/>-Nitrophenyl ^-galactoside
Protein (rag)
Homogenate 25,000xg supernatant (NH4),SO4 fractionation Supernatant at pH 3 Sephadex G-200
Lactosylceramide
Protein
Total activity (/imoles/hr)
Specific activity (nmoles/mg protein/hr)
Purification (fold) 1
9,100 2,977
731.2 337.1
80.1 113.2
1.4
1,881 395.7
235.4 204.7 71.1
125.6 523.0
1.6 6.5
33.4
2,129
26.6
Downloaded from https://academic.oup.com/jb/article-abstract/80/2/209/801859 by University of Edinburgh user on 24 January 2019
Polyacrylamide gel electrophoresis was carried out with 7.5% gels at pH 2.3 according to the method described by Davis (26). Proteins on the gels were detected by staining with Brilliant cresyl blue. Bovine serum albumin
