Isolation and Characterization of Jack Bean ,&Galactosidase* · Isolation and Characterization of Jack Bean ,&Galactosidase* (Received for publication, January 30, 1975) SU-CHEN LI,

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 250, No. 17, Issue of September 10. pp. 6786-6791, 1975

Printed in U.S.A.

Isolation and Characterization of Jack Bean ,&Galactosidase* (Received for publication, January 30, 1975)

SU-CHEN LI, MARY Y. MAZZOTTA, SU-FANG CHIEN, AND Yu-TEH LI$

From the Department of Biochemistry, Tulane University School of Medicine, New Orleans, Louisiana 70112 and Delta Regional Primate Research Center, Covington, Louisiana 70433

A simple procedure has been devised to isolate ,Sgalactosidase from jack bean meal. The final preparation gives one major protein band in disc gel electrophoresis. The substrate specificity of this enzyme toward some natural oligosaccharides, glycoproteins, and sphingoglycolipids has been examined in detail. Among three isomers of N-acetyllactosamine, GalPl+GGlcNAc is most readily hydrolyzed, followed by Galpl-4GlcNAc; while Galpl-XGlcNAc was hydrolyzed very slowly. This property can be used to distinguish the galactose linkage in asialo-GM, (Gal~l-3GalNAc~l-4Gal~l-4Glc-Cer) and that in lacto-N-neotetraosylceramide (Gal~l-4GlcNAc~lL3Gal~l+4Glc-Cer). For hydrolyzing glyco-

lipids, the effect of sodium taurodeoxycholate and sodium taurochenodeoxycholate on the rate of hydrolysis was carefully examined. This enzyme hydrolyzes lactosylceramide and asialo-GM1 faster than G,,. These results suggest that in addition to the type and linkage of the penultimate sugar unit, the sugar unit at the distal position of the saccharide chain also affects the hydrolysis rate. It also readily liberates 80% of n-galactosyl units from asialo al-acid glycoprotein. Escherichia coli /3galactosidase on the other hand cannot hydrolyze asialo-cu ,-acid glycoprotein, lactosylceramide, G, 1, asialo-G, 1, and lacto-N-neotetraosylceramide.

The molecular weight of this enzyme is about 75,000 and the isoelectric point is pH 8.0. With p-nitrophenyl P-n-galactopyranoside as substrate, optimal activity occurs at pH 2.8 with glycine-HCl buffer and at pH 3.5 with citrate-phosphate buffer. With lactose as substrate, the pH optimum in these two buffers are 2.8 and 4.0, respectively. K, values for p-nitrophenyl P-n-galactopyranoside, o-nitro- phenyl P-n-galactopyranoside and lactose are 0.51 mM, 0.63 mM, and 12.23 mM, respectively. Many inhibitors for this enzyme including inorganic ions, monosaccharides, and glycosides are investigated. In contrast to E. coli P-galactosidase, jack bean /3-galactosidase is not inhibited by p-aminophenyl thio-p- n-galactopyranoside.

p-n-Galactosyl units are common constituents of heterosac- charide chains in both glycoproteins and sphingoglycolipids. Although P-galactosidases have been isolated from a wide variety of sources (1, 21, very little is known about their specificity and biological significance. Among many /3-galac- tosidases isolated so far, the one isolated from jack bean has been used in studying the structure of a number of glycoconju- gates, including glycoproteins and sphingoglycolipids (3-22), even though it has never been extensively purified and characterized. This report describes a simple isolation proce- dure for jack bean P-galactosidase, its general properties and its specificity toward some natural substrates.

EXPERIMENTAL PROCEDURE

Materials-Jack bean meal was obtained from Nutritional Bio- chemical. p-Nitrophenyl p-n-galactopyranoside and neuraminidase from Clostridium perfringens were purchased from Sigma; isopropyl

*This work was supported by the National Science Foundation Grant GB 43571, the National Institutes of Health Grant NS 09626, and the National Foundation March of Dimes Grant l-356.

$ Recipient of Research Career Development Award l-K04-HD 50280 from the United States Public Health Service.

thio-@o-galactopyranoside from Schwa&Mann. G M 1 ’ and GM 2 gan- gliosides were isolated from normal human brain (23). The asialo derivatives of G,, and G,, gangliosides were prepared by hydrolyzing the ganglioside with 1 M HCOOH for 1 hour at 100’ as previously described (8). Globoside, trihexosylceramide, and lactosylceramide were isolated from human red cell stroma (24). al-Acid glycoprotein (orosomucoid) was isolated from human serum (25). The asialo derivative of cu,-acid glycoprotein was prepared by treating the glycoprotein with neuraminidase (26).

The following compounds were generous gifts: lacto-N-neotet- raosylceramide, Dr. S. Basu, University of Notre Dame, Indiana; Galpl-GMan, Dr. B. Lindberg, Stockholm University, Sweden; GalPl -4ManNAc, p-aminophenyl thio P-o-galactopyranoside (27), 2- [6-(6- aminohexanamido)hexanamido] ethyl - 1 - thio - fl - D - galacto- pyranoside, and its Sepharose 4A derivative (28), Dr. Y. C. Lee, Johns Hopkins University, Maryland; Galpl+3GlcNAc, Galpl- 4GlcNAc, Galpl-GGlcNAc, Dr. A. Gauhe, Max-Planck Institute,

‘The abbreviation and trivial names used are: G,,, Galpl-3GalNAcfll-4(NeuAcu2-3)

Galpl-4Glc@l-l’ceramide;

G ,,,GalNAc~l+4(NeuAc(u2-3)Galpl-4Glcpl-1’ceramide; globoside, Galpl-3Gal~l-4Gal~l-4Glc~l+l’ceramide; trihexosylceramide, Galol-4Gal&+4Glc@l-l’ceramide; lactosylceramide, Galpl+4Glcpl+l’ceramide; Cer, ceramide.

6786

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6787

elution profile is shown in Fig. 1. The fractions containing /?-galactosidase activity as indicated by the horizontal bar were pooled and precipitated by reverse dialysis against saturated (NH,),SO, solution. The precipitate was dissolved in 10 ml of 0.02 M sodium phosphate buffer, pH 8.0, and dialyzed exhaus- tively against this buffer.

Step 3. Chromatography on DEAE-Sephadex A-50-A 5-ml aliquot of the enzyme solution obtained from Step 2 was applied to a DEAE-Sephadex A-50 (2.5 x 30 cm) column previously equilibrated with 0.02 M sodium phosphate buffer, pH 8.0. The column was eluted with the same buffer and /3-galactosidase emerged at the void volume. There was a small protein peak devoid of enzyme activity eluted after the P-galactosidase peak. Under this condition, those glycosidases other than P-galactosidase remained firmly bound to the column. The fractions containing @-galactosidase activity were pooled and concentrated to 2 ml by ultrafiltration using an Amicon model 52 stirred cell with a UMlO membrane.

Step 4. Affinity Chromatography-One milliliter of the en- zyme solution obtained at Step 3 was exhaustively dialyzed against 0.01 M sodium citrate buffer, pH 4.0, and applied to a similarly equilibrated affinity column (1 x 12 cm) of Sepharose 4B coupled with 2- [6-(6-aminohexanamido)-hexanamido]- ethyl-1-P-thio-D-galactopyranoside. The column was subse- quently eluted with the same buffer at 5 ml/hour. The p-galactosidase activity was retarded relative to the main protein peak (Fig. 2). Those fractions containing P-galactosid- ase activity (tubes 11 to 19) were pooled and concentrated by ultrafiltration as described in Step 3. A summary of the specific acitivity and recovery during the purification proce- dure based on 200 g of jack bean meal is given in Table I.

General Properties of Jack Bean ,&Galactosidase

Purity-When examined by polyacrylamide gel electropho- resis at pH 9.0, the enzyme preparation obtained after affinity chromatography showed one major protein band (Fig. 3). In order to locate the enzyme activity, a parallel gel column was sliced into 2-mm sections and incubated with p-nitrophenyl P-D-galactopyranoside for 16 hours. The enzyme activity was found to coincide with the position of the major protein band on stained column. At pH 4.0, the enzyme migrated as one broad diffused band.2 For checking the cross-contamination of other glycosidases, 0.1 unit of the purified P-galactosidase was incubated separately with the following p-nitrophenylglyco- sides: a-n-galactopyranoside, (Y- and P-D-mannopyranosides, (Y- and @-D-glucopyranosides and LY- and P-2-acetamido-2- deoxy-n-glucopyranosides, according to the condition for assaying the @-galactosidase for 16 hours at 37”. No hydrolysis of the above mentioned substrates was observed. The prepara- tion is also devoid of protease activity as measured by pro- longed incubation with Azocoll.

Molecular Weight and Isoelectric Point-The molecular weight of @-galactosidase, estimated from its chromatographic mobility in Sephadex G-100 column, is about 75,000 (36). The isoelectric point of P-galactosidase was found to be pH 8.0 by isoelectrofocusing using ampholine of pH 7 to 10 according to the procedure described by Vesterberg and Svensson (37).

Optimum pH-With p-nitrophenyl P-n-galactopyranoside as substrate, optimal activity of this enzyme occurred at pH 2.8 with glycine-HCl buffer, and at 3.5 with citrate phosphate buffer (38). When lactose was used, the pH optimum was 2.8

Heidelberg, Germany; the freezing point-depression glycoprotein of antarctic fish, Z’rematomus borchgreuinki (29), Dr. A. L. DeVries, Scripps Institute of Oceanography, University of California, San Diego; fl-galactosidase isolated from Escherichia coli (k-12,3300), Dr. J. Tang, Oklahoma Medical Research Foundation. Oklahoma; agarose beads substituted with succinylaminoalkyl group attached to phenyl thio-P-n-galactopyranoside (40, 41), Dr. G. W. Jourdian, University of Michigan.

Enzyme Assays-During the enzyme isolation, P-galactosidase ac- tivity was routinely assayed at 37” by using p-nitrophenyl fl-n-galac- topyranoside as substrate. An enzyme solution (1 to 50 ~1) was added to 1 ml of 2 mM p-nitrophenyl P-D-galactopyranoside dissolved in 0.05 M glvcine-HCl buffer, DH 3.5. After incubation for meset time. 3.0 ml of b.; M sodium borate buffer, pH 9.8, were added-to stop the’ reaction; absorbance of the resultant solution was read at 400 nm. A continu- ously monitored spectrophotometric assay method (30) was used to determine the K,. One unit of fl-galactosidase is defined as the amount of enzyme which hydrolyzes 1 Fmol of p-nitrophenyl P-o-galac- topyranoside/min under the conditions described above. The specific activity of the enzyme was expressed as units per mg of protein. Protein was determined by the method of Lowry et al. (31) with crystalline bovine serum albumin as standard. When lactose was used as substrate the liberated free n-glucose was determined by glucose oxidase as previously described (32).

Analytical Methods-Analytical thin layer chromatography of gan- gliosides and neutral sphingoglycolipids was performed according to the procedures previously described (8, 33). For quantitative estima- tion of glycolipids on thin layer plates, the intensities of the bands were compared visually with that of the known quantities of the glycolipid (5). The conversion rate of a given glycolipid to its derivative with one less sugar unit by glycosidases was determined by following both the rates of disappearance of the original glycolipid and the appearance of the new glycolipid (5). When glycoprotein, oligosaccharide, or sphingo- glycolipid was used as substrate, free o-galactose in the enzyme digests was quantitatively determined by automated anion exchange chroma- tography (34). Isoelectrofocusing was performed on the LKB 110.ml column using ampholine of pH 6 to 10 according to the procedure described by the manufacturer. Polyacrylamide disc gel electrophore- sis was performed according to the procedure described by Davis (35).

RESULTS

Purification of fi-Galactosidase from Jack Bean Meal

Unless otherwise indicated, all operations for the enzyme isolation were carried out at O-4”. Centrifugation was carried out in a Sorvall RCB-B refrigerated centrifuge. Samples were routinely centrifuged for 20 min at the following speeds: 7,000 rpm for GS-3 large capacity rotor; 9,000 rpm for GSA high speed rotor; 12,000 rpm for SS-34 superspeed rotor.

Step 1. Extraction and Ammonium Sulfate Precipitation- A 200-g portion of jack bean meal was suspended in 1.2 liters of water and stirred for 2 hours at room temperature. The suspension was strained through cheesecloth. The turbid filtrate was adjusted to pH 5.5 at room temperature with 1.5 M sodium citrate, pH 2.7, and centrifuged to obtain 900 ml of extract. Solid (NH,),SO, was added to this extract to obtain 30% saturation. After standing for 2 hours, the precipitate was removed by centrifugation and was discarded. In some batches of jack bean meal, little or no precipitate occurred at this saturation. p-Galactosidase in the supernatant was precipitated by adding solid (NH,),SO, to 60% saturation. After standing overnight, the precipitated protein containing P-galactosidase was collected by centrifugation and dissolved in 100 ml of 0.1 M sodium phosphate buffer, pH 7.0, to obtain an opaque solution.

Step 2. Gel filtration on Sephadex G-200-A 25.ml portion of the enzyme solution obtained at Step 1 was applied to a Sephadex G-200 column (5 x 90 cm) which had been equili- brated with 0.1 M sodium phosphate buffer, pH 7.0. The column was eluted with the same buffer at 25 ml/hour. The ?This may indicate that enzyme is still not pure.

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6788

Fraction Number Froctlon Number

FIG. 1 (left). Sephadex G-200 filtration of @-galactosidase prepara- tion obtained at Step 1. The enzyme solution (25 ml) containing 1.8 g of protein was applied to a Sephadex G-200 column (5 x 90 cm) which had been previously equilibrated with 0.1 M sodium phosphate buffer, pH 7.0. The column was eluted with the same buffer at a flow rate of 25 ml/hour. 0, absorption at 280 nm; 0, P-galactosidase activity ex- pressed as the absorption at 400 nm produced by incubating 50-~1 aliquots of each fraction with substrate for 10 min according to the assay conditions described in the text; 20-ml fractions were collected.

FIG. 2 (center). Affinity column chromatography of jack bean @galactosidase. One milliliter of the enzyme solution obtained after Step 3, containing 2.8 mg of protein, was exhaustively dialyzed against 0.05 M sodium citrate buffer, pH 4.0, and applied to a column (1 x 8 cm)

TABLE I Purification of @-galactosidase from 200 g of jack bean meal

steps Total Total units proteins

Specific activity Recovery

Step 1. Extraction and

WL),SO, ppt Step 2. Sephadex G-200

filtration Step 3. DEAE-Sephadex

chromatography

Step 4. Affinity chroma- tography

m&T unltlmg 9%

314 1010 0.031 100

260 630 0.41 83

140 18 7.7 45

32 1.7 18.6 10

with glycine-HCl buffer and was 4.0 with citrate-phosphate buffer. The apparent differences in the pH optimum might be attributed to difference in ionic strength of the buffers.

pH Stability and Heat Stability-The stability of P-galac- tosidase at various pH values was studied by placing the enzyme in 0.05 M citrate phosphate, sodium phosphate, and sodium borate buffers ranging in pH from 2 to 10 at room temperature for 15 hours. Then, the enzyme activity was assayed at 37” as described under “Experimental Procedures.” The enzyme retained 80 to 100% activity from pH 3 to 10 but rapidly lost the activity below pH 3. At pH 2.2, it retained only 3% activity. The enzyme was stable at 60” for 1 hour in 0.05 M

sodium phosphate buffer, pH 7.0; however at 65” for 10 min, it lost more than 50% of activity.

Inhibition Studies-The effect of metal ions on P-galactosid- ase activity was investigated by preincubation of the enzyme with the inhibitor for 30 min at room temperature before the

of 2- [6-(6-amino-hexanamido)hexanamido]-ethyl-l-thio-P-n-galactopy- ranoside-linked Sepharose (3.3 wmol of ligands/ml of gel). The column was eluted with the same buffer. 0, absorption at 280 nm; 0, fl-galac- tosidasc activity expressed as the absorption at 400 nm produced by incubating 25-~1 aliquots of each fraction substrate for 30 min accord- ing to the assay conditions described in the text. One milliliter per fraction was collected.

FIG. 3 (right). Disc gel electrophoresis of jack bean /3-galactosidase at pH 9.0. 1, crude extract (400 pg); 2, preparation obtained after Sephadex G-200 filtration (200 fig); 3, preparation obtained after DEAE-Sephadex A-50 chromatography (200 pg); 4, preparation ob- tained after affinity column chromatography (100 L(R). The gels were stained with Amido black stain.

addition of p-nitrophenyl fl-n-galactopyranoside. The results are summarized in Table II. Of the various metal ions tested,

Ag+, Hgz+, and Zn2+ to a lesser extent were potent inhibitors.

P-Galactosidase was not inhibited by ethylenediaminetet- raacetate. This table also includes the inhibition of /3-galacto- sidase activity by several sugar derivatives. It is of interest to note that isopropyl thio-/3-n-galactopyranoside but not p- aminophenyl thio-P-n-galactopyranoside exert significant in- hibitory effect on the enzyme. As shown in the same table,

2 - [6 - (6 - aminohexanamido)hexanamido] - ethyl 1 thio - @-n-galactoside is also a potent inhibitor for jack bean /?- galactosidase. This inhibitor was, therefore, chosen to prepare an affinity column for this enzyme (see Fig. 2). The following K, values were obtained by Dixon plots (39); p-aminophenyl thio-/3-n-galactoside, 46 mM; isopropyl thio-P-n-galactopy- ranoside, 16 mM; 2- [6-(6-aminohexanamido)hexanamido]- ethyl-l-thio-/I-n-galactopyranoside, 1.3 mM; n-galactono-1,4- lactone, 0.1 mM. All these compounds are competitive inhib- itors.

Effect of Substrate Concentration-The effects of varying substrate concentration on the reaction rate for o- and p-

nitrophenyl P-n-galactopyranosides and lactose was measured at 37” using 0.05 M sodium citrate buffer, pH 4.0. The apparent Michaelis constant (K,) for each substrate was determined from the Lineweaver-Burk plots to be: p-nitrophenyl P-D-

galactopyranoside, 0.51 mM; o-nitrophenyl P-n-galctopyrano- side, 0.63 mM; lactose, 12.3 mM.

Specificity of Jack Bean P-Galactosidase-The action of jack bean P-galactosidase on various saccharides, glycoproteins and glycolipids are summarized in Table III. As shown in this table, under the condition described Galpl+GGlcNAc was com-

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6789

TABLE II

Effect of various inhibitors on activity of P-galactosidase

0.Galactosidase, 0.4 ml (contained 0.03 unit), was preincubated at 25” with 0.1 ml of inhibitor for 30 min. Then the enzyme activity was determined at 37” by adding 0.5 ml of 4 mM p-nitrophenyl

@-o-galactopyranoside. Enzyme, inhibitors and substrates were dis- solved in 0.05 M sodium citrate buffer, pH 4.0.

their specificities toward Gal/3+3GlcNAc are very different. In contrast to jack bean p-galactosidase, the enzyme from E. coli hydrolyzes l+3-linked isomer faster than I-4-linked isomer. It also hydrolyzes lactose faster than jack bean P-galactosidase.

Inhibitors Inhibitor Relative concentration activity

H&l,

Y”

AgNO, ZnSO, CuSO,, MgSO,, MnCl,, BaCl,, CaCl,

EDTA o-Galactono-l,4-1actone

o-Galactose D-Glucose p-Aminophenyl thio-fi-o-galactopyrano.

side Isopropyl thio-fl-D-galactopyranoside

AHA-AHA-AES-@Gal”

TllM

1.0

1.0 1.0 1.0 9

1.0 2.0

5.0 5.0

27

38 14-98

103 27

34 100

5.0 5.0

5.0

100 70

70

a 2. [6-(6-Aminohexanamido)hexanamido]-ethyl-l-thio-~-o-galacto-

pyranoside.

TABLE III

Specificity of jack bean and Escherichia coli P-galactosidase

For the hydrolysis of N-acetyllactosamine isomers and various disaccharides, incubation mixture contained the following components in 100 ~1: substrate, 0.125 /*mol; 4 rmol of sodium citrate buffer, pH

4.0 (for jack bean P-galactosidase), or sodium phosphate buffer, pH 7.0

(for E. co2i /3-galactosidase), 0.3 unit of enzyme. For the liberation of o-galactose from asialo-cu,-acid glycoprotein, the reaction mixture

contains following components in 150 ~1: asialo-a ,-acid glycoprotein, 1 mg; 4 Kmol of sodium citrate buffer, pH 4.0 or sodium phosphate buffer, pH 7.0; 0.3 unit of enzyme. For the hydrolysis of sphingo-

glycolipids, incubation mixture contains the following components in 400 ~1: substrate, 0.02 pmol; 15 pmol of sodium citrate buffer, pH 4.0 or

sodium phosphate buffer, pH 7.0; sodium taurodeoxycholate, 300 pg; enzyme, 0.75 unit.

Jack bean P-galactosidase efficiently cleaves @-n-galactosyl residues from asialo-oc ,-acid glycoprotein. Approximately 80% of the total bound n-galactose was liberated within 2 hours of incubation. Prolonged incubation did not liberate more D-

galactose from the glycoprotein suggesting that about 20% of the total n-galactose in this glycoprotein is resistant to jack bean P-galactosidase. This result is analogous to that of Hughes and Jeanloz (26), who found that P-galactosidase of Diplococcus pneumoniae released 80% of n-galactose from the asialo+,- acid glycoprotein. Distler and Jourdian (40) also found that P-galactosidase of bovine testis liberated 87% of n-galactose from asialo-cu ,-acid glycoprotein. In agreement with Hughes and Jeanloz (26), we also found that E. coli P-galactosidase was not able to hydrolyze n-galactosyl units from asialo-oc,-acid glycoprotein. n-Galactosyl units in freezing point depression glycoprotein were found to be resistant to jack bean P-galac- tosidase. Although this glycoprotein was reported to contain terminal n-galactosyl units linked l-4 to N-acetyl-galactosa- mine (29), only 3% of the n-galactose was liberated after 24 hours of incubation.

R Hydrolysis

Substrate

Gal&+GGlcNAc Gal&+4GlcNAc

Gal&+3GlcNAc

Gal&+4Glc Galpl-6Man Galfll-6ManNAc

Asialo ol,-acid gly- coprotein

Lactosylceramide

Aziilo-G,, G

Incubation time Jack bean E. coli

0.galactosidase 8.galactosidase

min YO YO 10 100 100 10 75 55 10 1 92 10 42 100 10 19 N.D.” 10 10 N.D.

120 79 0 120 100 0

120 120 82 9 0 0

a Not determined

pletely hydrolyzed in 10 min, while 75% hydrolysis of Gal- /3-4GlcNAc and virtually no hydrolysis of Galpl-3GlcNAc were detected. The effect of aglycon moiety on the rate of hydrolysis was examined by replacing the GlcNAc residue with Glc, Man, or ManNAc. The results show a reduction of the hydrolysis because of different aglycons. For comparison, specificity of Escherichia coli P-galactosidase towards N- acetyllactosamine isomers and lactose was also studied:

We also have examined the liberation of n-galactosyl units from a number of sphingoglycolipids by jack bean P-galactosid- ase. This enzyme requires the addition of a detergent such as the crude sodium taurocholate isolated from canine bile for the hydrolysis of lipid substrates. Although crude sodium tauro- cholate has been widely used to stimulate the enzyme hydrol- ysis of sphingoglycolipids, this preparation is an ill defined mixture of bile salts containing more than 10 components which often interfere with the analysis of thin layer chromatog- raphy. The stimulating effect of the crude bile salts is also inconsistent from lot to lot. We, therefore, examined various pure bile salts obtained from Calbiochem for their ability to stimulate the hydrolysis of n-galactosyl units from sphingo- glycolipids. Among the sodium salts of taurochenodeoxycho- late, taurodeoxycholate, taurocholate, taurodehydrocholate, and taurolithocholate, we found that sodium taurochenodeoxy- cholate and sodium taurodeoxycholate were most active in stimulating jack bean P-galactosidase to hydrolyze GM1-gangli- oside and lactosylceramide. Fig. 4 shows the stimulation of the enzymic hydrolysis of G,,-ganglioside and lactosylceramide as a function of the concentration of sodium taurodeoxycholate and sodium taurochenodeoxycholate. In general, sodium tauro- deoxycholate is better than sodium taurochenodeoxycholate. With 0.8 unit of the enzyme and 50 WM substrate concentration, the stimulation of G,, hydrolysis by sodium taurodeoxycholate reached the maximum at the concentration of 300 &0.4 ml (1.43 mM). Rapid decrease in hydrolysis rate occurred when the detergent concentration was greater than 400 &0.4 ml (1.91 mM). Under the same condition, stimulation of G,, hydrolysis by sodium taurochenodeoxycholate reached the maximum at the concentrations of 500 kg/O.4 ml (2.39 mM). Greater concen- tration of taurochenodeoxycholate did not cause significant decrease in hydrolysis rate. For the hydrolysis of lactosylcera- mide (50 FM), less detergent is required for maximal rate of hydrolysis: 200 kg/O.4 ml (0.95 mM) for sodium taurodeoxycho- late and 400 pg/O.4 ml (1.91 mM) for sodium taurochenodeoxy- cholate. Therefore, at the substrate concentration of 50 FM, we

chose 1.43 mM (300 pg/O.4 ml) sodium taurodeoxycholate for

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6790

acetyllactosamine at different rates. The l+6-linked N-acetyl- lactosamine is most readily hydrolyzed, followed by l-4- linked isomer. The enzyme hydrolyzes l-3-linked N-acetyllac- tosamine very slowly (Table III). This property can be used to distinguish the nature of n-galactosyl linkage in asialo-Gh,, (Gal~l-3GalNAc~l-4Gal~l-4Glc-Cer), which is slowly hy- drolyzed, and that in lacto N-neotetraosylceramide (Gal~l-4GlcNAc/I-3Galfll-4Glc-Cer), which is rapidly hy- drolyzed. Jack bean @alactosidase is also able to distinguish the linkage of terminal n-galactosyl units in lacto-N-tetraose (Gal/31-3GlcNAc~l+3Gal/31-4Glc) and lacto-2\i-neotetraose (Gal/?l-4GlcNAc~l-3Gal/3l+4Glc) (10, 19). The nature of the aglycon moiety also affected the rate of hydrolysis. As shown in Table III, Galfil-4GlcNAc was hydrolyzed faster than Gal- Pl-4Glc or Gal/3+4ManNAc. In contrast to jack bean ,8- galactosidase, bovine testes P-galactosidase hydrolyzes l-3- linked Gal-GlcNAc faster than the l-4 isomer, and the l-6-linked isomer is hydrolyzed at the lowest rate. By using two P-galactosidases isolated from bovine testes and jack bean, it should be possible to distinguish the Gal/3-XGlcNAc, Galpl+4GlcNAc, and Galpl-6GlcNAc linkages. The specific- ity of E. coli @galactosidase toward N-acetyllactosamines on the other hand, is quite different from the /I-galactosidase of jack bean or bovine testes.

Jack bean P-galactosidase can readily hydrolyze n-galactosyl units from asialo-cu,-acid glycoprotein or asialo glycopeptide prepared from human (Y,-protease inhibitor.3 However, it hydrolyzes the freezing point depression glycoprotein very slowly. The freezing point depression glycoprotein has been reported to contain Galpl-4GalNAc-Thr sequence (29). The explanation for the resistance may lie in the n-galactosyl unit being linked to GalNAc instead of GlcNAc. Alternatively, the proximity of this disaccharide unit to the polypeptide moiety may result in a steric hindrance, unfavorable for the approarh of the P-galactosidase.

When a glycosidase is used to hydrolyze the saccharide units in glycolipids, careful consideration should be given to the amount and the nature of the detergent used. As shown in Fig. 4, different detergents activate at different optimal concentra- tion. It seems reasonable to postulate that the situation is even more complex, namely the optimal micelle organization and architecture for presentation of substrate to enzyme may depend on the concentration of substrate and detergent and furthermore that the optimal ratio may vary with different substrate.

Jack bean P-galactosidase hydrolyzes about 40% of the intact G,,-ganglioside in 24 hours, while under the same condition, it completely hydrolyzes the asialo-GM, in about 3 hours. The result suggests that, in addition to the type and linkage of the penultimate sugar unit, the sugar unit at the distal position of the saccharide chain also affect the hydrolysis rate.

From the study presented in this report, it is clear that specific glycosidases are useful for the determination of both the anomeric configuration and the sequential arrangement of saccharide units in glycoproteins and glycolipids. It should be emphasized, however, that the specificities of glycosidases are rather complex. The same categories of glycosidases isolated from different sources often vary considerably in their sub- strate specificities. One should, therefore, interpret the results of enzyme hydrolysis with extreme caution. When used with

3 S. K. Chan, D. C. Rees, S.-C. Li, and Y.-T. Li, manuscript in preparation.

200 400 600 800 1000 -

Ng Detergent Added

FIG. 4. Effect of sodium taurodeoxycholate and sodium tauro- chenodeoxycholate concentration on the hydrolysis of lactosylceramide and GMnl-ganglioside catalyzed by jack bean P-galactosidase. Incuba- tion mixture contains the following components in 400 ~1; lactosylcera- mide or G,,-ganglioside, 0.02 pmol; sodium citrate buffer (15 pmol of citrate), pH 4.0; 0.75 unit of P-galactosidase and different quantity of the detergent. After incubation at 37” for 6 hours, the reaction mixture contained lactosylceramide as substrate was partitioned with 4 vol- umes of chloroform/methanol (2/l), while G ,,-ganglioside, with 4 volumes of chloroform/methanol (l/l). Chloroform-methanol layer was evaporated to dryness and analyzed by thin layer chromatography (3, 33). -, lactosylceramide; mm-1 G,,-ganglioside; 0, sodium tauro- deoxycholate; 0, sodium taurochenodeoxycholate.

the hydrolysis of lactosylceramide and G M ,-gangalioside. As shown in Table III, jack bean fl-galactosidase hydrolyzed lactosylceramide much faster than G M ,-ganglioside. Lactosyl- ceramide was almost completely hydrolyzed in 2 hours, while only 40% of G,,-ganglioside was hydrolyzed after 24 hours of incubation. In contrast to G ,,-ganglioside, asialo-G ,+,, was readily hydrolyzed by jack bean P-galactosidase. E. coli /3-galactosidase was not able to hydrolyze either G,,, asialo- G M 1, or lactosylceramide under a variety of conditions.

DISCUSSION

In order for a glycosidase to be useful for the structural analysis of complex saccharide chains, it is imperative that the enzyme free from other glycosidases can be easily isolated and that detailed information about its substrate specificity is available. By using the simple procedure described above, P-galactosidase is completely separated from other glycosi- dases. For practical purposes, P-galactosidase isolated by DEAE-Sephadex can be used to study the structure of complex carbohydrate chains.

It is surprising that p-aminophenyl thio-P-n-galactopyrano- side, an effective competitive inhibitor of E. coli P-galactosid- ase (41), does not inhibit the hydrolysis of p-nitrophenyl P-n-galactopyranoside by jack bean P-galactosidase even at 20 mM concentration (Table II). This lack of inhibition by aryl thio-@-n-galactopyranoside is reflected in the lack of affinity of this /3-galactosidase to the affinity column prepared by suc- cinylaminoalkyl group attached to phenyl thio-@-n-galac- topyranoside (42). However, we found that thio-P-n-galac- topyranoside attached to an alkyl aglycon exhibited competi- tive inhibition to this P-galactosidase.

Jack bean P-galactosidase hydrolyzes three isomers of N-

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6791

304663051 16. Gahmberg, C. G., and Hakomori, S. (1973) J. Biol. Chem. 248,

4311-4317 17. Kornfeld, R., and Kornfeld, S. (1970) J. Biol. Chem. 245,

2535-2545 18. Baenziger, J., Kornfeld, S., and Kochwa, S. (1974) J. Biol. Chem.

care, combination of glycosidases with methylation analysis

probably offers the most powerful tool available for the

elucidation of the complete structure of heterosaccharide

chains.

Acknowledgments-The authors are grateful to Dr. Y. C.

Lee, Johns Hopkins University, for providing affinity beads,

and for his careful reading of this manuscript and valuable

suggestions. Thanks are also due to Dr. J. E. Muldrey, Tulane

University, for his advice and comments during the prepara-

tion of the manuscript.

REFERENCES

1. Wallenfels, K., and Malhotra, 0. P. (1961) Adv. Carbohydr. Chem. 16, 239-298

2. Li, S.-C., and Li, Y.-T. (1975) in Handbook of Biochemistry and Molecular Biology (Fasman, G. D., ed) 3rd Ed. Chemical Rubber Company Press, Cleveland, Ohio, in press

3. Li, Y.-T., and Li, S.-C. (1972) Methods Enzymol. 28, 702-713 4. Li, Y.-T., Li, S.-C., and Shetlar, M. R. (1968) J. Biol. Chem. 243,

656-665 5. Hakomori, S., Siddiqui, B., Li, Y.-T., Li, S.-C., and Hellerqvist, C.

G. (1971) J. Biol. Chem. 246, 2271-2277 6. Li, Y.-T., and Li, S.-C. (1971) J. Biol. Chem. 246, 376993771 7. Laine, R., Sweeley, C. C., Li, Y.-T., Kisie, A., and Rapport, M. M.

(1972) J. Lipid Res. 13, 519-524 8. Svennerholm, L., MBnsson, J.-E., and Li, Y.-T. (1973) J. Biol.

Chem. 248, 740-742 9. Li, Y.-T., Mdnsson, J.-E., Vanier, M.-T., and Svennerholm, L.

(1973) J. Biol. Chem. 248, 2634-2636 10. Arakawa, M., Ogata, S., Muramatsu, T., and Kobata, A. (1974) J.

Biochem. (Tokyo) 75, 707-714 11. Krantz, M. J., Lee, Y. C., and Hung, P. P. (1974) Nature 248,

684-686 12. Siddiqui, B., Kawanami, J., Li, Y.-T., and Hakomori, S. (1972) J.

Lipid Res. 13, 657-662 13. Stellner, K., and Hakomori, S. (1974) J. Biol. Chem. 249,

1022-1025 14. Baenziger, J., and Kornfeld, S. (1974) J. Biol. Chem. 249,

7260-7269 15. Wherrett, J. R., and Hakomori, S. (1973) J. Biol. Chem. 248,

249, i889-1896 19. Kobata. A.. and Ginsburg. V. (1972) J. Biol. Chem. 247.1525-1529 20. M&tensson, E., ehrnan; R.; Graves, M., and Svennerholm, L.

(1974) J. Biol. Chem. 249, 4132-4137 21. Koscielak, J., Prasek, A., Gorniak, H., Gardas, A., and Gregor, A.

(1973) Eur. J. Biochem. 37, 214-225 22. Sung, S. J., Esselman, W. J., and Sweeley, C. C. (1973) J. Biol.

Chem. 248, 6528-6533 23. Svennerholm, L. (1972) Methods Carbohyd. Chem. 6, 464-474 24. Hakomori. S.. and Strvcharz. G. D. (1968) Biochemistry 7,

1279-1286 _

25. Li, Y.-T., and Li, S.-C. (1970) J. Biol. Chem. 245, 825-832 26. Hughes, R. C., and Jeanloz, R. W. (1964) Biochemistry 3,

1535-1543 27. Chipowsky, S., and Lee, Y. C. (1973) Carbohyd. Res. 31,339-346 28. Lee, R. T., and Lee, Y. C. (1974) Carbohyd. Res. 34, 151-160 29. Shier, W. T., Lin, Y., and De Vries, A. L. (1972) Biochim. Biophys.

Acta 263, 406-413 30. Ford, J. R., Nunley, J. A., Li, Y.-T., Chambers, R. T., and Cohen,

W. C. (1973) A&. Biochem. 54, 120-128 31. Lowrv. 0. H.. Rosebrouah. N. J.. Farr. A. L.. and Randall. R. J.

(1951) J. Biol. Chem.193, 265-274 32. Suzuki, H., Li, S.-C., and Li, Y.-T. (1970) J. Biol. Chem. 245,

781-786 33. Li, Y.-T., Mazzotta, M. Y., Wan, C. C., Orth, R., and Li, S.-C.

(1973) J. Biol. Chem. 248, 7512-7515 34. Lee, Y. C. (1972) Methods Enzymol. 28, 63-73 35. Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121, 404-427 36. Andrews, P. (1964) Biochem. J. 91,222-233 37. Vesterberg, O., and Svensson, H. (1966) Acta Chem. Stand. 20,

820-834 38. McIlvaine, T. C. (1921) J. Biol. Chem. 49, 183-186 39. Dixon, M. (1953) Biochem. J. 55, 170-171 40. Distler, J. J., and Jourdian, G. W. (1973) J. Biol. Chem. 248,

6772-6780 41. Steers, E., Cuatrecasas, P., and Pollard, H. B. (1971) J. Biol.

Chem. 246, 196-200 42. Cuatrecasas, P. (1970) J. Biol. Chem. 245, 3059-3065

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

S C Li, M Y Mazzotta, S F Chien and Y T LiIsolation and characterization of jack bean beta-galactosidase.

1975, 250:6786-6791.J. Biol. Chem. 

  http://www.jbc.org/content/250/17/6786Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/250/17/6786.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from