Purification of an Almond Emulsin Fucosidase on Cibacron Blue ...

THEJOURNALOF BIOLOGICAL CHEMl6TRY Val. 257, No. 14. Issue of July 25. pp. 8205-8210, 1982 Printed in U.S.A.

Purification of an Almond Emulsin Fucosidase on Cibacron BlueSepharose and Demonstrationof Its Activity toward Fucose-containing Glycoproteins* (Received for publication, October 6, 1981)

Michael J. Imber$, LowrieR. Glasgow, and SalvatoreV. F%zo From the Departments of Pathology, Medicine and Biochemistry, Duke University Medical Center, Durham, North Carolina 2771 0

The almond emulsin fucosidase that specifically hy- this fucosidase 1250-fold which employsthe readily available drolyzes fucose in a(1-3) linkage to N-acetylglucosa- affiity medium Cibacron blue F3GA-Sepharose in the final step of the purification. Optimal conditions for enzymeactivity mine has been purified 1250-fold. The purification proon sul- are presented and its activity is demonstrated toward glycocedure includes ion exchange chromatography fopropyl-Sephadex (3-25, gel filtrationon Sephacryl S- protein substrates containing Fuc a(1-3)GlcNAc linkages. 200, andaffinitychromatography on CibacronblueEXPERIMENTAL PROCEDURES Sepharose 4B-CL. The molecular weight of thefucosidase was estimated by gel filtrationas approximately Materials 73,000. Enzyme activity wasmaximal at pH 5.3 in aceAlmond emulsin, bovine serum albumin, transfenin, cytochrome c, tate bufferand was dependent on ionic strength; at myoglobin, carbonic anhydrase, immunoglobulin G, andp-nitroleast 0.1 M NaCl was necessary for optimal activity. phenyl-glycosides were obtained from Sigma; Cibacron blue F3GA The purified enzyme was free of 8-galactosidase ac- from Polysciences,Inc., Warrington, P A GDP-[’4C]fucose and tivity toward the glycoprotein substrate [3H]galactosylNaBr3HI4from New England Nuclear; Sephadex and Sephacryl chroasialotransferrin and did not release fucose from sub- matographic media from Pharmacia, Piscataway, NJ; and Dowex AG strates containing fucose a(1-6) in linkage, (bovineIgG 50W-X8 resin, 100-200 mesh, from Bio-Rad. Human lactoferrin and glycopeptides) or in a(1-2) linkage, (2’-fucosyllactose). the a(1-3) fucosyltransferase from human milk were @ts of Dr. J.-P. Prieels, Universite Libre de Bruxelles, Belgium; orosomucoid was a The fucosidase displayed activity toward two glycoprotein substratesknown to contain fucose ina(1-3) link- gift of the American National Red Cross; and 2’-fucosyllactose wasa &t of Dr. V. Ginsberg, National Institutes of Health. All other age. Extensive incubations resulted in the release of reagents were the highest grade commercially available. 83%and 43% of the total fucose of asialoorosomucoid and lactoferrin, respectively. The fucosidase did not Preparation of Asialoorosomucoid release fucose from either the “slow” or the “fast” form ASORwas prepared from orosomucoid as previously described of az-macroglobulin, suggesting the absence of fucosyl (13). a(1-3) linkages on that glycoprotein. Preparation of fH]Galactosyl-asialotransferrin ~

~~

Fucosidases have been isolated from a number of prokaryotic and eukaryotic sources (1-6). Many of these enzymes display broad specificity or activity directed only to artificial substrates and this has limited their usefulness as structural probes for complex glycoconjugates.Kobata has recently described twofucosidase activities inalmondemulsin (7, 8). Type I is specific for Fuc a(1-3)GlcNAc‘ and Fuc a(1-4)GlcNAc linkages, and Type I1 is specific for Fuc a(1-2)Gal linkages present in milk oligosaccharides.The Type I activity has been useful in a number of carbohydrate structural studies (9-12). However, the procedure used to purify the fucosidase I achieves only a 98-fold purification and requires the complex milk oligosaccharide lacto-N-fucopentaose I1 coupled to Sepharose as an affinity medium (8). This publication describes a rapid procedure for purifying

Asialotransferrin was prepared from human transferrin as previously described (13). Tritiation of terminal galactosyl residues was performed according to Morell and Ashwell (14). Preparation of Bovine IgG Glycopeptides Glycopeptides were prepared from 10 g of bovine IgG by pronase digestion (15) and partial purification on Sephadex G-25 SF (3.5 x 150 cm) and Sephadex G-50(2 X 117 cm) equilibrated in0.1 M NH4HC03. Preparation of the Substrate of [i4C]Fucosyl-ASOR

The radiolabeled substrate [’4C]fucosyl-ASORwas prepared by reacting GDP-[’4C]fucoseand ASOR with a purified a(1-3) fucosyltransferase from human milk (16). This transferase adds fucose in a(1-3) linkage with penultimate N-acetylglucosamine residues of desialated Asn-linked oligosaccharide chains (16). 3.2pCiof GDP[’4C]fucose in ethanol was dried under vacuum in a plastic tube (12 X 75 mm). Two mg of ASOR and 4 milliunits of a(1-3) fucosyltransferase were added in a fiial volume of 130 pl, 40 mM 4-morpholine* This work was supported by National Institutes of Health Re- propanesulfonic acid, 15 mM MnCh, pH 7.5. The solution was capped, search Grants HL-24066, CA-29589, AM-27199, and GM-25766. The vortexed, and incubated for 1 h a t 37 “C. The reaction mixture was costs of publication of this article were defrayed in part by the chromatographed on a Sephadex G-50 column (1.0 x 20 cm) equilipayment of page charges. This article must therefore be hereby brated in 0.15 M NaCl. Void volume fractions containing [’4C]fucosylmarked “aduertisement” in accordance with 18 U.S.C. Section 1734 ASOR were determined by scintillation counter. Greater than 90% of solely to indicate this fact. the radiolabeled fucosewas incorporated into protein. SubstratefPredoctoral Fellow, Medical Scientist Training Grant GM-07171. containing fractions were pooled and stored frozen at -20 “C until The abbreviations used are: Fuc, a-L-fucose; GlcNAc, ,8-D-N-ace- further use. tyl-glucosamine; Gal, /3-D-galatose; IgG, immunoglobulin G, ASOR, Descending paper chromatography in ethyl acetate/pyridine/water asialoorosomucoid; azM, as-macroglobulin, asM-MeNH2, methyla- (5:4:3)was performed on aliquots of the substrate to verify that mine-treated azM. incubation with fucosidase-containingsamples resulted in the release

8205

8206

Purification of an Almond Emulsin Fucosidase

of free fucose (17). Prior incubation of [14C]fucosyl-ASOR with almond emulsin resulted in the migration of "c radioactivity from the origin to a position identical with that of free fucose. Preparation of Human a d 4 and a&-MeNH2 ~ z M was purified from human plasma by a modification of the method of Kurecki et al. (18), using metal chelate chromatography on zinc-Sepharose 4B and gel filtration on ACA-22, as previously described (19). Purified a2M migrated as a single band in the native "slow" state during nondenaturing polyacrylamide gel electrophoresis (20) and the trypsin:azM binding stoichiometry was 1.9as determined by the method of Ganrot (21). (r2M-MeNH2was prepared from purified azM as previously described (19). Native polyacrylamide gel electrophoresis demonstrated the quantitative conversion from the native "slow" to theconformationally more compact "fast" state (20). a2M-MeNH2was inactive when assayed for trypsin-binding activity by the method of Ganrot (21). Preparation of Blue-Sepharose The affinity medium blue-Sepharose was prepared by coupling the dye Cibacron blue F3GA to Sepharose 4B-CL according to themethod of Travis et al. (22). Fucosidase Assay Using the Substratef4C]Fucosyl-ASOR

A substrate mixture was prepared by diluting stock [14C]fucosylASOR 25-fold with 0.1 M NaOAc, 0.1 M NaCl, pH 5.3. Fifty p1 of this mixture contained approximately 12,000 cpm of ['4C]fucose. Five- to fifteen-pl aliquots of fractions to be assayed for fucosidase activity were incubated with 50-pl aliquots of substrate mixture a t 37 "C for times ranging from 10 min to several hours. The reaction was then quenched by the addition of 1.0 ml of 50 m~ citric acid. Ion exchange chromatography of the reaction mixture was performed in Pasteur pipette columns containing 0.5 ml of Dowex AG 50W-X8 resin and equilibrated in 50 m~ citric acid. The sulfonate groups of the resin retain protein-associated radioactivity at low pH but allow free ["C] fucose to elute freely. The diluted reaction mixture was applied to the column and the effluent was collected directly in a minivial for scintillation counting. The column was then washed with 1.5 ml of 50 m~ citric acid. 1.5 ml of Aquasol I1was added to the combined column effluents and the I4C radioactivity of the mixture was determined. Controls were determined by substituting buffer for the fucosidase-containing sample in the incubation. The assay was linear with respect to time and enzyme concentration when less than 30% of the total counts were released. Fucosidase Assay Using the SubstrateLactoferrin The fucose-containing glycoprotein lactoferrin was used as a substrate to quantitate fucosidase activity following each step of the purification. Assay mixtures contained 5 mg of lactoferrin (131 nmol of fucose), 0.05 M NaOAc, 0.1 M NaCl, pH 5.3, and were incubated with aliquots of pooled fucosidase activity in a final volume of 200 pl at 37 "C for 1-6 h. Background incubations were performed without added fucosidase. Release of free fucose was determined directly on each reaction mixture by the fucose dehydrogenase-linked assay of Tsay and Dawson (23).A standard plot for this assay was determined using 5-50 nmol of L-fucose.The assay was linear with respect to time and enzyme concentration up to a release of 25 nmol of L-fucose per incubation. Determination of Other Glycosidase Activities Activity toward the p-nitrophenyl-glycoside substratesp-nitrophenyl-,B-galactoside, -,B-N-acetylglucosaminide,-a-galactoside, and -a-mannoside was determined on the solution of almond emulsin and final fraction of purified fucosidase. Assay mixtures contained 3 mM p-nitrophenyl-glycoside substrates, 0.05 M citrate, 0.1 M NaC1, pH 5.0, and were incubated with aliquots of enzyme activity in a final volume of 70 pl at 37 "C for 1 h. The reaction was quenched by the addition of 1.0 ml of0.2 M Na2C03 and the absorbance at 400 nm was determined. Background incubations were performed on assay buffers alone. P-Galactosidase activity toward an asialoglycoprotein substrate was also determined. The assay mixtures contained [3H]galactosylasialotransferrin (4300 cpm of [3H]galactose), 0.05 M NaOAc, 0.1 M NaC1, pH 5.3, and were incubated with aliquots of enzyme activity in a final volume of70 pl at 37 "C for 1 h. The reaction was quenched and the release of [3H]galactosewas determined by the Dowex column assay described above for [14C]fucosyl-ASOR.

Molecular Weight Estimation by Gel Filtration on Sephacryl S-200 The method of Andrews (24) was used to estimate the molecular weight of the native, affinity-purified fucosidase. Gel filtration was performed on a column of Sephacryl S-200 (1.5 X 149 cm) equilibrated in 0.05 M NaOAc, 0.15 M NaCI, pH 5.0, a t 4 "C. Blue dextran was used to determine VO,and the following globular proteins were used to calibrate the column: cytochrome c (&Ir = 11,700), myoglobin (M, = 17,800),carbonic anhydrase (Mr = 29,000), ovalbumin (Mr = 43,000), and bovine serum albumin (Mr= 68,000). Each of these proteins was dissolved in equilibration buffer (5-10 mg/ml) and applied to the column. Two-ml fractions were collected at a flow rate of20 ml/h. Elution positions were determined spectrophotometrically. The purified fucosidase (200 pl) was likewise applied to the Sephacryl column. The elution position of the fucosidase activity was determined by measuring [14C]fucoserelease from ["C]fucosyl-ASOR, as described above. Analytical Methods Protein was determined by the method of Lowry et al. (25). Bovine serum albumin was used as the standard. Total fucose was estimated by the method of Dische and Shettles(26; see also Ref. 27). Protease activity was determined by the caseinolytic method of Highsmith and Rosenberg (28). Native polyacrylamide gel electrophoresis was performed at pH 3.8 and 9.5 according to Gabriel (29). Fucosidase Purification Ion Exchange Chromatography on Sulfopropyl (SP)-Sephadex C25-All steps were performed in a cold room a t 4 "C unless otherwise described. Almond emulsin (2.2 g) was dissolved in 200 ml of 0.05 M NaOAc, 0.1 M NaCl, pH 5.0. The solution was applied to a column of SP-Sephadex (2.3 X 18 cm) equilibrated in 0.05 M NaOAc, 0.1 M NaCl, pH 5.0. Fractions (12.2 m l ) were collected at a flow rate of 15 ml/h. The column was washed with 1.0 liter of equilibration buffer. A linear gradient of increasing ionic strength was applied to elute the fucosidase activity from the ion exchange column. The buffer reservoirs of the gradient apparatus contained 400 ml of0.05 M NaOAc, 0.1 M NaC1, pH 5.0, and 400 ml of0.05 M NaOAc, 0.4 M NaCl, pH 5.0, respectively. Fractions (5 m l ) were collected at a flow rate of 15 ml/h. Absorbance at 280 nm and conductivity determinations were made on fractions equilibrated a t room temperature. Fucosidase activity was determined by measuring [14C]fucoserelease from [14C]fucosylASOR, as described above. Fractions 66-83 were pooled (86 m l ) and concentrated to 7 ml with a Millipore CX-30 concentrator. Gel Filtration on Sephacryl S-200-The concentrated SP-Sephadex pool was applied to a column of Sephacryl S-200 (5 X 85 cm) equilibrated in 0.05 M NaOAc, 0.1 M NaCl, pH 5.0. The column was eluted with equilibration buffer at a flow rate of 124 ml/h and 13-ml fractions were collected. Fractions 57-64(104 ml) were pooled and dialyzed for 24 h against several changes of 0.05 M NaOAc, pH 5.0, until a conductivity less than 2 mmho was obtained. Affmity Chromatography on Blue-Sepharose-The dialyzed Sephacryl S-200 pool was applied to a column of blue-Sepharose (0.7 X 2.0 cm) equilibrated in 0.05 M NaOAc, pH 5.0. The column was washed with 100 ml of equilibration buffer a t a flow rate of 20 ml/h and 10.4-ml fractions were collected. Bound fucosidase activity was pulsed from the column by elution with 0.05 M NaOAc, 0.25 M NaC1, pH 5.0, and 0.5-ml fractions were collected. Fractions 2-9 (3.6 ml) were pooled and stored at 4 "C as stock fucosidase. RESULTS

Fucosidase Purification-The elution profiles for each of the three chromatographic steps in the purification are presented in Figs. 1-3, and Table I summarizes the purification at each state. An overall 1250-fold purification and 23% yield of fucosidase was obtained from commercially available almond emulsin.The major loss of activity occurred during the first step, ionexchange chromatography on SP-Sephadex. The 10% loss determined after gel filtration on Sephacryl s200 occurred during the concentration step prior to chromatography. The blue-Sepharose column elution was initially attempted with a linear salt gradient, but major activity losses resulted from enzyme dilution. Pulse gradient elution, however, provided a high yield and a concentrated stock of purified enzyme.

Purification Almond Emulsin of an

Fucosidase

8207

Almond emulsin contains a glycopeptidase which cleaves paspartylglycosamine linkages (30). This enzyme is alsoabsent I from the final fucosidase preparation by virtue of the same I I reasoning. The final preparation of purified fucosidase was I not free of contaminating proteins. Native gel electrophoresis 0 at pH 3.8 and 9.5 revealed multiple discrete protein bands still X present. E Fucosidase activity was remarkably stable in the final pu6 s rified and concentrated state. Greater than 90% of the activity U in freshly purified fucosidaseremained following refrigeration 0 4 s at 4 "C for 6 months with added 0.02% NaN3. However, t freezing and thawing resulted in a one-third loss of initial activity. 2 : Molecular Weight Estimation by Gel Filtration-The elu0.4 V 2 tion position of affiity-purified fucosidase wasdetermined on 0.2 a column of Sephacryl S-200 calibrated by well characterized I020304050 10203040!506070eo90l100 globular proteins (Fig. 4). Assuming that the fucosidase is a Flow -Through Elution Gradient globular protein, the observed elution position is consistent Fraction Number with M, = 73,000. FIG. 1. Step 1: ion exchange chromatography of almond Effects of Assay Medium on Fucosidase Actiuity-Fig. 5 emulsin on SP-Sephadex (2-26. Detailsareprovidedunder "Experimental Procedures." Fifteen-p1 aliquots were incubated for 60 presents the results of activity assays using an acetate/NaCl rnin at 37 "C with the substrate ['4C]fucosyl-ASOR. The shaded area buffer. The most interesting observation was the ionic indicates fractions pooled containingfucosidase activity. strength dependence of fucosidase activity (Fig. 5B). Aliquots of purified enzyme dialyzed against acetate buffer without NaC1, and assayed in the absence of added NaC1, displayed almost no activity. The activity was completely recovered, however, followingthe addition of NaCl to the assay medium. Maximal activity was observed at pH 5.3 (Fig. 5A) and variations in acetate concentration between 25 and 100 m~ did not greatly affect activity (Fig. 5 0 . Other buffer systems were decidedlyless desirable as assay media. Fucosidase activity increased with increasing ionic strength until reaching a maximum level at 80 mM NaCl in the presence of malonate buffer (Fig. 6A). Further salt addi5 .06 tion actually inhibited activity. Increasing concentrations of .04 malonate itself above 20 m~ markedly inhibited activity (Fig. .o 2 6B). This same pattern of salt and buffer concentration dependence was also observed at pH 5.0 in cacodylate/NaCl. The standard buffer composition used in fucosidaseassays Fraction Number or digests was 0.05 M NaOAc, 0.1 M NaCl, pH 5.3. Table I1 FIG.2. Step 2: gel filtration of pooled activity from Step 1, on Sephacryl S-200. Detailsareprovidedunder"Experimental I 1 1 Procedures." Fifteen-pl aliquots were incubated for 60 rnin at 37 "C with the substrate ['4C]fucosyl-ASOR. The shaded urea indicates fractions pooled containing fucosidase activity. Apply Ewllibratlng Almond Buffer Emulsln

-In Gradlent

. I

;

I

The final pool of purified fucosidase was assayed for the presence of contaminating exoglycosidase and protease activities. The purified fucosidase was assayed for ,&galactosidase, @-hexosaminidase,p-galactosidase, and a-mannosidase activity toward synthetic p-nitrophenyl-glycoside substrates. The color yield from these experiments was essentially at background levels which indicates that these activities are present at less than a few thousandths of a per cent of the activity originally present in the almond emulson. More importantly, there was no remaining p-galactosidase activity toward the radiolabeled asialoglycoprotein,[3H]galactosyl-asialotransferrin, nor was there detectable proteolytic activity toward radiolabeled casein. These data also demonstrate that GlcNAc and mannose were not released by the fucosidase preparation since these residues cannot be released without concomitant loss of the terminal [3H]galactoseresidues from the glycopeptide which has the carbohydrate structure: [3H]Galpl,4GlcNAcp1,2Mana1,3

\ Man~l,4GlcNAc~1,4GlcNAc-Asn

/ ["H]Gal/31,4GlcNAc/31,2Mam1,6

I 1

!

j14t: 13

Ft

Iit

(li I

I I

4 8 12 16 Flow -Through Pulse

\

\

'.4

!

1

4

e

12

V

4

Fraction Number

FIG. 3. Step 3: affinity chromatography of pooled activity from Step 2, on blue-Sepharose. Detailsare provided under "Experimental Procedures." Five-$ aliquots were incubated 15 min at 37 "C with the substrate ['4C]fucosyl-ASOR. Values of ['4C]fucose counts/rnin released were multiplied by 12 before plotting for cornparison with activity levels shown in Figs. 1 and 2.

Purification Almond Emulsin an of

8208

Fucosidase

TABLEI Summary of steps leading to thepurification of almond emulsin fucosidase Details regarding each step areprovided under “Experimental Procedures.” Fraction

Volume

Milligramunits“ Total

Specific step ity”Total

Yield

Purification

%

-fold

Step

ml

Total

1. Almond emulsin 2. SP-Sephadex pool 4. Sephacryl S-200 pool 4. Blue-Sepharose pulse

124 200 2400 0.052 100 1 1 100 86 148 47.8 39 39 0.32 6.3 6.3 9.7 35.8 104 3.68 75 29 11.4 71.5 0.44 23 64.50 79 3.6 28.4 1250 17.5 a Aliquots from each step were assayed for their ability to release fucose from human lactoferrin as described under “Experimental Procedures.” Total units aregiven as nanomoles of fucose released/min; specific activity is given as nanomoles of fucose released/min/mg of protein.

100

3 \ 3

-

80 -

1.6 4..->

c

1.4

U

t

L

\

1 44 . 4 . 4 8 . 4 6 . 4 . 20

I

I

I

I

I 5.0

a - 600

EX 0

E

--

40-

20

fl Malonic Acid

0.I E NoCl

pH 5.5

log Molecular Wefght

FIG. 4. Plot of V,/ Voagainst log molecular weight for protein standards and fucosidase on a Sephacryl S-200 column. Details are provided under “Experimental Procedures.’’ Protein standards (0)used were: I , cytochrome c (Mr = 11,700);2,myoglobin (M, = 17,800); 3, carbonic anhydrase (Mr = 29,000); 4, ovalbumin (M, = 43,000); and 5, bovine serum albumin (BSA) (M, = 68,000). The V,/ VOvalue obtained for affinity-purified fucosidase activity is indicated by the arrow.

2o

t

40 80 120 160 NoCl

(mM)

40 80 120 I60 MalonicAcid

(e)

FIG. 6. Effects of NaCl and malonic acid buffer concentrations on fucosidase activity. Assay buffers were prepared a t various

combinations of [NaCI] and [malonic acid] at pH5.5. For each series of assays, either [NaCI] or [malonic acid] was held constant (as shown in eachpanel), while the other was varied: A, [NaCl] and B, [malonic acid]. Fucosidase activity assays were performed as described in the legend to Fig. 5 and under “Experimental Procedures.”

The presence of 10 I“Hg2+reduced activity by almost 80%, perhaps due to its binding surface sulfhydryl groups. pH 5.3 Activity of the Purified Fucosidase toward Fucose-containing Glycoconjugates-The purified enzyme was specific for fucose ina(1-3) linkage with N-acetylglucosamine residues of Asn-linked oligosaccharide chain branches. Enzyme incuba4.5 5.0 5.5 6.0 O.! 0.2 0.3 0.4 0.5 5 0 100 1 5 0 2 0 0 tions wereperformed with glycoprotein, glycopeptide, and pH NaCl (y1 NaOAc (a) oligosaccharide substrates which contain fucose in a number FIG. 5. Effects of pH and concentrations of NaCl and NaOAc of different glycosyl linkages, in order to verify this specificity on fucosidase activity. Assay buffers were prepared at various (Table 111). combinations of pH, [NaCl], and [NaOAc]. For each series of assays, The glycoprotein asialoorosomucoid (Mr= 40,000)contains 2 of these variables were held constant (asshown in eachpanel) and 0.7% fucose or 1.7 mol of fucose/mol (31, 32). Structural 1 was varied: A , pH; B , [NaCl]; and C, [NaOAc]. Two-pl aliquots of purified fucosidase were incubated with 50 p1 of the appropriate assay studies indicate that all fucose in ASOR is present in the buffer and 2 plof stock [“C]fucosyl-ASOR for 90 min a t 37 “C. glycosyl linkage Fuc a(1-3)GlcNAc (33, 34). Extensive incuRelease of [‘4C]fucose was determined as described under “Experi- bations with increasing amounts of fucosidase resulted in the release of at least 83% of the fucose calculated to be present mental Procedures.” on ASOR (Table 111) and this result was consistent with the summarizes the effects on activity of various additional re- presumed specificity of the enzyme. agents. Almost 50% of control activity was inhibited in the The iron-binding protein lactofemin (Mr= 76,500) contains presence of 10 mM L-fucose. The divalent cations Ca2+,Mn2+ 2 mol of fucose/mol (35). The fucose, however, is present in and Mg2+,and EDTA had little effect on fucosidase activity. (~(1-6)linkage to the core N-acetylglucosamine as well as in 0.1

E NaCl

202


8209

DISCUSSION TABLEI1 Effects of fucose, divalent metal ions, and EDTA on fucosidase Cibacron blue-Sepharose has been used successfully to puactivity rify many enzymes requiring NAD or ATP as a cofactor (43Assay buffers were prepared containing 0.05 M NaOAc, 0.1 M NaCl, 46). It was therefore surprising to find that blue-Sepharose pH 5.3, and additional substances listed in the table. Fifteen plof was an excellent affinity medium for the almond emulsin purified fucosidase, diluted IO-fold in the appropriate assay buffer, fucosidase I, especially since a number of fucose-containing was incubated with5 pl of stock [’4C]fucosyl-ASOR and70 pl of assay buffer for 60 min at 37 “C. Release of [“C]fucose was determined as media have not proved very useful. Yoshima et al. (8), reported only an additional %fold purification of fucosidase I described under “Experimental Procedures.”

activity on lacto-N-fucopentaose I1 coupled to Sepharose, with a 70% loss of activity. Initial attempts in this laboratory mM to purify the enzyme on agarose-eaminocaproylfucosamine cPm 1. No addition 220 1.00 (47),and on fucosyloxirane-Sepharose(48),were unsuccessful. 212 0.96 2. L-Fucose 2.5 The fucosidase bound very tightly to blue-Sepharose at low 0.54 118 10 ionic strength, but eluted readily in the presence of 0.1 M 0.31 50 69 NaC1. Since the enzyme activity in acetate buffer displays a 1.13 249 3. Ca2+ 10 parallel dependence on ionic strength (Fig. 5), it is possible 225 1.02 4. Mn2’ 10 that the fucosidase undergoes ionic strength-dependent con245 1.11 5. Mg2’ 10 50 0.23 6. Hg2’ formational changes. At very low ionic strengths (g < 2.0 10 237 1.08 7. EDTA 10 mmho), the enzyme may exist in an inactive state which has a high affinity for the Cibacron blue dye. In the presence of increasing salt, the fucosidase may assume an active conformation which does not bind to Cibacron blue. TABLEI11 The fiial preparation of purified enzyme is of sufficient Fucosidase activity toward fucose-containing glycoconjugates Substrates were incubated at 37 “C with purified fucosidase in a enzymatic homogeneity to be useful as a carbohydrate strucfinal volume of 200 pl, 0.05 M NaOAc, 0.1 M NaCl, pH 5.3. Amounts tural tool. Only trace levels of glycosidase activities toward of substrates,fucosidase, and lengths of incubations arelisted in table. synthetic p-nitrophenyl-glycoside substrates were detected Release ofL-fucosewasdetermined by the method of Tsay and and there was no P-galactosidase activity toward asialoglycofor protein substrates in the purified fucosidase. The final prepDawson (23). Duplicatesamplesandcontrolswereperformed each digestion. Other details are provided under “ExperimentalPro- aration was also free of contaminating protease activity. The cedures.” specificity of the final preparation for Fuc a(1-3)GlcNAc Milli- Total Fucosi- A! (37 L-Fucose reSubstrate linkages was supported by its lack of activity toward subleased grams fucose dase C) strates containing Fuc a(1-2)Gal and Fuc a(1-6)GlcNAc linknmol pl h nmol YOtotal ages (Table 111). 1.8 77 5 35 23 18 1. ASOR The purifiedfucosidasewas active toward glycoprotein 37 52 10 35 substrates containing Fuc a(1-3)GlcNAc linkages (Table 111). 64 25 3583 32 34 3.6 94 5 35 2. Lactofenin The present results are consistent with recent structural stud10 3541 39 ies of oligosaccharides obtained from orosomucoid (33, 34) 25 35 40 43 and lactoferrin (36, 37). Almost all of the fucose present on 50 3543 40 ASOR was removed following extensive incubations, in agree0 0 3. (x~M 50 24 85 2.0 ment with reports that all fucose present in orosomucoid 1.8 77 24 50 0 0 4. mM-MeNH2 exists in a(1-3) linkage (33,34). Almost half of the lactoferrin 0 0 90 35 15 5. IgG glycopeptides 170 20 18 0 0 6. 2”Fucosyllactose fucose was removed, in agreement with the anticipated content of a(1-3)-linked fucose (36,37). Finally, the fucosidase was used to investigate the fucoserich (31/mol), plasma protease inhibitor apM (49).The native a(1-3) linkage to GlcNAc, as in ASOR (36,37). The glycopepform of this glycoprotein is a tetramer of four identical ( M , tides of lactoferrin are heterogeneous with regard to their = 185,000) subunits which undergo a conformational rearcontent of Fuc a(1-3)GlcNAc (36). The quantity of Fuc rerangement when azM binds a protease or is exposed to methleased from lactoferrin by the purified fucosidase (Table 111) ylamine (38, 40). Studies were directed toward whether the is in good agreement with the estimated Fuc in a(1-3) linkage exposure of Fuc a(1-3)GlcNAc linkages on azM, if present, to GlcNAc based on the proposed glycopeptide structures might be dependent on the conformational state. The fucosi(36). dase, however, released no fucose from either form of the The plasma protease inhibitor, azmacroglobulin (Mr = glycoprotein (Table 111) suggesting that Fuc a(1-3) linkages 726,000), contains 31 mol of fucose/mol (38,39). No report of are either missing altogether or are buried within the quaterthe nature and/or distribution of all of the fucosyl linkages nary structure of azM. present in azM has yet been made. The fucosidase was incubated with the native, electrophoretic “slow” form of azM, Acknowledgment-We wish to thank Dr. Robert L. Hill for his and with the conformationally compact, electrophoretic “fast” review and helpful criticismof this manuscript. form, generated by exposure to methylamine (40). However, REFERENCES no release of fucose was observed with either form of azM (Table 111). Fuc a(1-3)GlcNAc linkages are presumably ab1. Bahl,0. P. (1970) J. Biol. Chem. 245,299-304 2. Iijima, Y.,Muramatsu, T., and Egami, F. (1971) Arch. Biochem. sent or enzymatically inaccessible on this glycoprotein. Biophys. 145.50-54 Finally, no fucose release occurred from bovine IgG glyco3. Iijami, Y.,and Egami, F. (1971) J.Biochem. (Tokyo) 70, 75-78 peptides, known to contain Fuc a(1-6)GlcNAc linkages in the 4. Wiederschain, G. Y.,and Rosenfeld, E. L. (1971) Biochem. Biocore region (41, 42), or from the milk oligosaccharide 2‘phys. Res. Commun. 44, 1008-1014 fucosyllactose, which contains fucose in a(1-2) linkage with 5. Aminoff, D., and Furukawa, K. (1970) J.Biol. Chem. 245, 1659galactose (42). 1669 Additions

~

Concentration

[“CIFucose released

Relative activitv

8210


6. Reglero, A., and Cabezas, J. A. (1976) Eur. J. Biochem. 66, 379387 7. Ogata-Arakawa, M., Muramatsu, T., and Kobata, A. (1977)Arch. Biochem. Biophys. 181,353-358 8. Yoshima, H., Takasaki, S., Ito-Mega, S., and Kobata, A. (1979) Arch. Biochem. Biophys. 194,394-398 9. Yamashita, K., Tachibana, Y., Nakayama, T., Kitamura, M., Endo, Y., and Kobata, A. (1980) J . Biol. Chem. 255,5635-5642 10. Lamblin, G., Lhermitte, M., Boersma, A,, Roussel, P., and Reinhold, V. (1980) J . Biol. Chem. 255,4595-4598 11. Santer, U. V., and Glick, M.C. (1980) Biochem. Biophys. Res. Commun. 96, 219-226 12. DiCiocci,R.A,, Piskorz, C., Salamida, G., Barlow, J. J., and Matta, K. J . (1981) Anal. Biochem. 111, 176-183 13. Prieels, J.-P., Pizzo, S. V., Glasgow, L. R., Paulson, J . C., and Hill, R. L. (1978) Proc. Natl. Acad. Sci.U. S. A . 75, 2215-2219 14. Morell, A. G., and Ashwell, G. (1972) Methods Enzymol. 28,205208 15. Koide, N., and Muramatsu, T. (1974) J . Biol. Chem. 249, 48974904 16. Prieels, J.-P., Monnom, D., Dolmans, M., Beyer, T. A., and Hill, R. L. (1981) J. Biol. Chem. 256, 10456-10463 17. Trevelan, W. E., Procter, D. P., and Harrison, J. S. (1950) Nature (Lond.)166,444-446 18. Kurecki, T., Kress, L.F., and Laskowski,M., Sr. (1979) Anal. Biochem. 99,415-420 19. Imber, M. J., and Pizzo, S.V. (1981) J. Biol. Chem. 256, 81348139 20. Nelles, L. P., Hall, P. K., and Roberts, R. C. (1980) Biochim. Biophys. Acta 623,46-56 21. Ganrot, P. 0. (1966) Clin. Chim. Acta14,493-501 22. Travis, J., Bowen, J., Tewksbury, D., Johnson, D., and Pannell, R. (1976) Biochem. J . 157,301-306 23. Tsay, G. C., and Dawson, G (1977) Anal. Biochem. 78,423-427 24. Andrews, P. (1965) Biochem. J. 96, 595-606 25. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 26. Dische, Z., and Shettles, L. B. (1948) J . Biol. Chem. 175,595-603 27. Spiro, R. G. (1966) Methods Enzymol. 8,3-26 28. Highsmith, R. F., and Rosenberg, R. D. (1977) Thromb. Res. 11, 131-140 29. Gabriel, 0. (1971) Methods Enzymol. 22, 565-578

30. Takahashi, N., and Nishibe, N. (1981) Biochim. Biophys. Acta 657,457-467 31. Jeanloz, R. W. (1972) in The Glycoproteins (Gottschalk, A., ed) pp. 565-611, Elsevier, New York 32. Schmid, K. (1975) in The Plasma Proteins (Putnam, F. W., ed) Vol. 1 pp. 184-228, Academic Press, New York 33. Fournet, B., Montreuil, J., Strecker, G., Dorland, L., Haverkamp, J., Vliegenthart, J. F. G., Binette, J. P., and Schmid, K. (1978) Biochemistry 17, 5206-5214 34. Schmid, K., Binette, J. P., Dorland, L., Vliegenthart, J. F. G., Fournet, B., and Montreuil, J . (1979) Bwchim. Biophys. Acta 581,356-359 35. Montreuil, J., and Spik, G. (1975) in Proteins of Zron Storage and Transport in Biochemistry and Medicine (Crichton, R. R., ed) pp. 27-38, North-Holland Publishing Co., Amsterdam 36. Matsumoto, A,, Yoshima, H., Takasaki, S., and Kobata, A. (1982) J. Biochem. 92, 143-145 37. Spik, G., and Mazurier, J . (1977) in Proteins of Zron Metabolism (Brown, E. B., Aisen, P., Fielding, J., and Crichton, R., eds) pp. 143-151, Grune and Stratton,New York 38. Hall, P. K., and Roberts, R. C. (1978) Biochem. J. 171.27-38 39. Bourillon, R., and Razafamaleo, E. (1972) in The Glycoproteins (Gottschalk, A., ed) pp. 699-715,, Elsevier, New York 40. Barrett, A. J., Brown, M. A,, and Sayers, C. A. (1979) Biochem. J. 181,401-418 41. Tai, T.,Ito, S., Yamashita, K., Muramatsu, T., and Kobata, A. (1975) Biochem. Biophys. Res. Commun. 65,968-974 42. Kornfeld, R., and Kornfeld, S. (1976) Annu. Rev. Biochem. 45, 217-237 43. Thompson, S.T., Cass, K. H., and Stellwagen, E. (1975) Proc. Natl. Acad. Sci.U. S. A . 72,669-672 44. Thompson, S. T., and Stellwagen, E. (1976) Proc. Natl. Acad. Sci. U. S. A . 73, 361-365 45. Wilson, J. E. (1976) Biochem. Biophys. Res. Commun. 72, 816823 46. Cheng, Y.-C., and Domin, B. (1978) Anal. Biochem. 85,425-429 47. Robinson, D., and Thorpe, R. (1974) FEBS Lett. 45, 191-193 48. Vertblad, P. (1976) Biochim. Biophys. Acta 434, 169-176 49. Starkey, P.M., and Barrett, A. J. (1977) in Research Monographs

in Cell and Tissue Physiology. Proteinases in Mammalian Cells and Tissues (Barrett, A .J., ed) Volume 2, pp. 663-696, North-Holland Publishing Co., Amsterdam