Further Characterization and Immunochemical

0 downloads 0 Views 2MB Size Report
Sep 23, 1987 - ... York University, Ontario, and the disaccharide D-Gal-a-(l -+ 3)-~-Ara from gum arabic was pre- .... cathode lamp. In order to remove metal ions ... TABLE I. Purification of jackfruit lectin. Fraction. Protein. Titer". Specific activitv'.
Vol. 264, No. 16,Issue of June 5, pp. 9365-9372,1989 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc.

Further Characterization and Immunochemical Studies on the Carbohydrate Specificityof Jackfruit (Artocarpus integrifolia) Lectin* (Received for publication, September 23, 1987)

Hafiz Ahmed and Bishnu P. ChatterjeeS From the Department of Biological Chemistry, Indian Association for the Cultivation of Science, Jadavpur, Calcutta 700-032, India

The lectin from jackfruit (Artocarpus integrifolia) receptor for the peanut lectin, fits well to the combining site seeds has been purified by Rivanol (6,9-diamino-2- of the jackfruit lection too (3, 4), especially when this disacethoxyacridine lactate) treatment. The specific activ- charide is a-linked ( 5 ) .Recently, the purification and studies ity, molecular weights of parent lectin and its subunit, on the physicochemical, chemical, and biological properties its glycoprotein nature, and hemagglutination-inhibi- of this lectin have been reported by us (6). The lectin (Mr tion assays suggest that this preparation is identical to 42,000) is a tetramer havingfour apparently identical subunits that obtained by affinity chromatography on melibi- (Mr 11,400)joined noncovalently in the same molecule. It is ose-agarose adsorbent (Ahmed, H., and Chatterjee, B. aglycoprotein and contains 8% carbohydrate. Aminoacid P. (1986) in Lectins, Biology, Biochemistry, Clinical analysis revealed a high content of acidic amino acids and Biochemistry (B0g-Hansen, T. C., and van Driessche, relatively low amount of basic amino acids (6). CitraconylaE., eds) Vol. 5, pp. 125-133, Walter de Gruyter, New tion or maleylation as well as acetylationof the lectinmodified York).The lectin stronglyagglutinates human and several animal erythrocytes. The lectin contains five 75% of its lysyl and 70% of the tyrosyl residues, respectively, isolectins of PI values 7.1, 6.85, 5.5, 5.3, and 5.1. It is thus completely abolished the sugar binding affinity of the thermally stable and loses its activity above 75 “C. The lectin ( 7 ) .Jackfruit lectin, jacalinwas demonstrated to stimhemagglutinating activity remains unchanged in the ulate both human T and B lymphocytes (8) and precipitated presence of bivalent cationsviz., Ca2+,Mg2’, Mn2+,etc. among five classes of Igs only IgA and IgD (1). It could It is a metalloprotein. The lectin retains itsactivity by distinguish between two subclasses of IgA in human serum, dialysis with acetic acid followed by EDTA. It agglu- secretory IgA (SIgA) in human milk, and twoallotypes of tinates Ehrlich ascites cells. Equilibrium dialysis of IgA, and thus selectively separate IgA, from IgAz by precipilectin withmelibiose and quenching of fluorescence of tation or binding to immobilized jacalin (9, 10). Despite con4-methylumbelliferyl-cy-~-galactopyranoside by the siderable amount of study, the carbohydrate-binding requirelectin show that homotetrameric jackfruit lectin has ments of the lectin have not yet been established fully. two sugar-binding sites. The lectin precipitates well In the present paper, we describe a simpleand rapid method several galactomannans andglycoproteins having ter- for the isolation of this lectin and additional studies on its minal D-Gal-cy-(l + 6)- or D-Gal-@-(l+ 3)-~-GalNAc properties including the specificity and size of its combining residues. It hardly or does not precipitate polysaccha- site using quantitative precipitin assays. The comparison of rides having terminalD-Gal-a-(1+ 3) residues. Quan- its specificity to that of other D-Gal’/D-GalNAc-specific lectitative precipitin-inhibition studiesusing various hap- tins has also been reported herein. tens suggest that the -OCH2- group at C-1 and -OH groups at C-4 and partiallyat C-6 in the cy-glycoside of EXPERIMENTALPROCEDURES D-galactose configuration are importantforlectinMaterials-Jackfruit seeds were purchased from the local market. sugar interaction. Rivanol, 4-methylumbelliferyl-a-~-galactopyranoside, o-nitrophenyl

N-acetyl-a-D-galactosaminide, p-nitrophenyl N-aCetyl-a-D-galaCtOsaminide,p-nitrophenyl N-acetyl-P-D-galactosaminide, melibiose, raffinose, andstachyose were purchasedfrom Sigma. Methyl-NThe agglutinin present in the seeds of Artocarpus integri- acetyl-a-D-galactosaminide was prepared according to Sarkar and (11).Methyl-a-D-galactopyranoside was obtained from Behrfolia (jackfruit) has firstbeen reported by Chattejee et al. (1) Kabat ing Diagnostics (U.S.A.). D-Gal-P-(l- 3)-~-GalNAc was the product and has been found to react with alkali-labile asialo carbo- of Biocarb (Sweden). D-Gal-p-(l-+~)-D-G~~NAC-~-O-(CH~)~-N hydrate chains of various glycoproteins, which is represented (CHZ),-COOCH3, edible birdnest glycoprotein, desialylated glycoby the disaccharide, 3-0-~-D-galactopyranosyl-N-acetyl-D-gaphorin, and pig erythocyte mucoid were the kind gift of Prof. G. lactosamine linked to a protein core via serine or threonine. Uhlenbruck, Medical University Clinic, Cologne, Federal Republic of Itscarbohydrate specificityfora-D-galactosylresidue has Germany (F.R.G.). Ant egg glycoprotein was purified on immobilized lectin Sepharose4B column.Honeycombglycoprotein been demonstratedby precipitin reactionsof partially purified jackfruit (HCG) was isolated from its saline extract by 90% phenol followed lectin with various galactomannans by agar gel diffusion as by gel filtration on Sephadex G-100. Blood group B substance was reported from this laboratory earlier(2). Further experiments prepared from B erythrocyte stroma according to Nowotny (12). IgA

showed thatthiscarbohydratestructure,

awell-accepted

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ To whom correspondence should be addressed.

The abbreviations used are: D-Gal, D-gahCtOSe; D-GalNAc, N acetyl-D-galactosamine; T, Thomsen-Friedenreich; JFL, jackfruit lectin; GS I, Griffonia simplicifolia I; Me, methyl; o-N02Ph-, o-nitrophenyl; L-Ara, L-arabinose; D-Gk, D-glucose; D-Man, D-mannose; DTal, D-talOSe; D-FUC,D-fucose; HPLC, high pressure liquid chromatography.

9365

9366

Carbohydrate Specificity

ofLectin Jackfruit

and dimeric IgA collected from a myeloma patient were kindly obIsoelectric Focusing-Isoelectric focusing was performed on a thin tained from Dr. C. K. Rao, Case Western Reserve University. layer polyacrylamide gel apparatus (FBE 300, Pharmacia Chemicals, Methyl-3-O-methyl-N-acetyl-a-~-galactosaminide was obtained Sweden) using 5% polyacrylamide gel containing 2% carrier amphofrom Dr. J. Palof this department. Galactomannansfrom fenugreek, lytes (pH 3-10) according to Rufo et al. (22). The marker proteins guar, Poinciana pulcherrima, and Crotalaria saltiana seeds were pu- used were myoglobin (PI 7.35 and 6.85), trypsin inhibitor (PI 4.55), rified as described elsewhere (13). Sesbania sesban seed galactoman- and a-chymotrypsinogen (PI 9.3). nan was generously supplied by Dr. N. Banerjee, Indian Institute of Metal Analyses-The purified JFL (50 mg) was dissolved in 4 ml Chemical Biology, Calcutta. Gum arabicwas the kind gift of Prof. G. of demineralized water and was refluxed for 30 min with a mixture 0. Aspinall, Department of Chemistry, York University, Ontario, and of concentrated HN03and 70% HClO, (1:l).Thematerial was the disaccharide D-Gal-a-(l -+ 3 ) - ~ - A r afrom gum arabic was pre- allowed to cool and was filtered through Whatman 40 filter paper. pared by the method as reported earlier (14). Hibiscus moscheutos Metal contents were measured by atomic absorption spectrophotommucilage polysaccharide was kindly donated by Prof. M. Tomoda, eter (Perkin-Elmer model 2380) using air acetylene flame and hollow Kyoritsu College of Pharmacy, Tokyo. Klebsiella Type 10 (K-10) cathode lamp. polysaccharide and D-Gal-a-(l -+ 3 ) - ~ - M a were n obtained from Dr. In order to remove metal ions from the lectin, J F L was dialyzed A. K. Sarkar of this laboratory. D-Gal-cu-(l+ 6)-D-Man was isolated exhaustively against 1 M acetic acid (23) followed by 0.1 M EDTA. from fenugreek galactomannan byhydrolyzing the polysaccharide Hemagglutination of native and dialyzed lectins was performed in with 0.3 N trifluoroacetic acid for 2.5 h and separated by HPLC on saline with or without additionof Caz+,M$+, and MnZ+(5-100 mM). p-Bondapakcarbohydrate column usingacetonitrile/water (70:30) Thermal Stability-JFL (60 pg/ml) was incubated separately for solvent. Methyl glycosides of D-fucose, D-talose, 2-deoxy-~-galactose, 10 and 30 min,respectively, in arange of temperatures 0-85 "C. and L-arabinose were obtained by refluxing the corresponding sugars Aliquots (0.1 ml) were withdrawn, cooled, and hemagglutination was with 1 N methanolic HCl for 6-8 h and the methyl a-glycosides of performed as described earlier (6). the sugars were separated by HPLC on p-Bondapak carbohydrate Effect ofpH-Hemagglutination of JFL was performed in presence column using the same solvent as described above. All other chemi- of different buffers (citrate-phosphate, pH 4.0-5.0; phosphate, pH cals, sugars, and enzymesused were of highest quality grade and 6.0-7.0; Tris-HC1, p H 7.4-9.0; borate, pH 10.0) of p H ranging from 4 purchased from various agencies as mentioned in our earlier paper to 10. (6). Fluorescence Quenching Study-The number of binding sites of Erythrocytes-Human blood samples were collected incitrateJ F L was determined by the quenchingof fluorescence of 4-methylumdextrose solutionfrom healthy donorsby vein puncture. Mouse blood belliferyl-a-D-galactopyranoside in a Fluorescence Spectrophotomewas obtained by cardiac puncture. The blood of duck and pigeon was ter (Perkin-Elmer, MPF 44A). In a cuvette containing 2 ml of 1 X obtained by killing the animals and that of buffalo and goat was M sugar in saline, increasing amounts (5-300 pl) of 1.25 X collected from a slaughterhouse. The erythrocyteswere treated with M J F L were added, and thefluorescence spectra of the residual sugar Pronase P (Streptomyces griseus) and neuraminidase (Vibrio chol- after each addition were recorded a t 27 "C. The excitations (Ex. slit erae) according to Chatterjee et al. (1). 16) was done at 318 nm, and emission spectra were recorded above Preparation of Jackfruit Lectin (JFL)-Ground seeds of jackfruit 330 nm with emission slit 0.9 and sensitivity3/5. As control, the saline (100 g) were suspended in 500 ml of 0.85% NaCl and stirredfor 1 h. and the lectin both were checked and found to have no fluorescence The slurry was filtered through cheese cloth and the filtrate was between 330 and 450 nm. Association constant was measured from centrifuged at 10,000 X g for 30 min. To the clear supernatant 400 the Scatchard plot (24). ml of 0.4% (w/v) Rivanol (6,9-diamino-2-ethoxyacridinelactate) was Eauilibrium Dialvsis-Eauilibrium dialysis of J F L (20 m d m l ) with added withvigorous stirring. The precipitate was removed bycentrifmelibiose (0.5-9.0 mg/ml) was performed in 1-ml Perspexcells sepugation a t 10,000 X g for 30 min. The excess Rivanol in the superarated by a semipermeable membrane. Dialysis was carried out for natant was precipitated by the addition of solid potassium bromide, 60 h a t 30 "C, and unbound melibiose in the sugar chamber was and the residue was removed by centrifugation a t 20,000 x g for 1 h. estimated by the method of Dubois et al. (20). Association constant The clear supernatant was exhaustively dialyzed against distilled was measured from the Scatchard plot (24). water, lyophilized, and storeda t -20 "C untilused. The yield was 720 Interaction with Ehrlich Ascites Carcinoma Cells-Ehrlich ascites mg. carcinoma cells were collected from the donor mouse (Swiss albino) Hemagglutination and Hemagglutination Inhibition Assays-These 18-20 g body weight of either sex and were suspended in sterile assays were performed in Takatsy microtiter plate in saline as described previously (6). The protein content of the lectinwas measured isotonic saline. A fixed number of the viable cells (usually 2 X lo6 according to Lowry et al. (15) using bovine serum albumin as standard. cells/20 g body weight) were implanted into the peritoneal cavityof Disc Gel Electrophoresis-Homogeneity of the purified lectin was each mouse. The tumorcells which multiplied relatively freely within judged by polyacrylamide discgel (7.5%) electrophoresisaccording to the peritoneal cavitywere withdrawn by a sterile needle and diluted Reisfeld et al. (16). Sodium dodecyl sulfate-polyacrylamide gel elec- withsaline (cell count 1.02 X lo6). The interaction of J F L with trophoresis was carried out in 10% polyacrylamide gels according to Ehrlich ascites carcinomawas tested by agglutination, and thespecificity of the reaction was determined by inhibition test in Takatsy Laemmli (17). Preparation ofdntisera-The rabbits were given injections of about microtiter plate as described earlier (6). Quantitative Precipitin Reaction-Quantitative precipitin reactions 2 mg of crude lectin in Freund's complete adjuvant (Difco Laboratowith 0.1 m l ( 1 mg/ml) J F L (14.0 gg of lectin N) were carried out with ries) intramuscularly once in 3 successive weeks followed by booster in Freund's incomplete adjuvant after a 2-week interval. The animals several galactomannans, polysaccharides, and glycoproteins (5-200 were bled by cardiac puncture in thefollowing week for 2 consecutive pg) in 5-ml conical centrifuge tubes. The final volume of the mixture days. After 1 month the immunization and bleeding schedule were was made up to 0.4 ml with 0.85% NaCl. The mixtures and blanks containing only lectin were set up in duplicate, thoroughly mixed, repeated and a second batch of antisera was obtained. and kept a t 4 "C for 72 h. The tubes were centrifuged for 40 min at Immunoelectrophoresis-Immunoelectrophoresis was carriedout was removed. The in Desaphor HL electrophoresis apparatus (Desaga, Heidelberg, F. R. 2000 X g at 4 "C, and the supernatant in each tube G . ) as described by Graber and Williams (18) using 1% agar in 0.05 precipitates were washed thrice with chilled saline, 1 ml each time, and then dissolved in 1 ml of 0.25 M acetic acid. The absorbance of M sodium barbital buffer, pH 8.2. The purified lectin was electrophoof nitrogen in resed a t 5 V/cm for 30 min. Migration was monitored by bromophenol the solution was measured at 280 nm, and the amount the precipitatewas calculated froma standard curve calibrated using blue dye as marker. Gels were allowed to develop overnight with rabbit anti-JFL (100 pl), fixed and stained with Coomassie Brilliant bovine serum albumin. Necessary correction was made in case of glycoprotein-induced JFL precipitation. Blue in acetic acid, and destained withacetic acid. Quantitative Precipitin Inhibition Assay-To 0.04 ml of JFL (5.6 Determination of Molecular Weight-The molecular weight of the purified lectin was estimated by gel filtration on a Sephadex G-100 pg of lectin N), in duplicate, increasing concentration of inhibitor (10"-104 p ~ was ) added and incubatedfor 2 h a t room temperature. column (90 X 1.6 cm) according to Andrews (19). An optimalamount of galactomannanor glycoprotein solution, Carbohydrate Analysis-The total neutral sugar in the purified J F L was estimated by the method of Dubois et al. (20). For identifi- needed to bring thesystem to equivalence, was then added to all the cation and quantitationof individual sugars, the lectin sample (2 mg) tubes. The final volume of the mixture was made up to 0.2 ml with saline, thoroughly mixed, and kept for 1 h at room temperature and was hydrolyzed with 4 N HC1 for 6h at 100 "C withinositolas standard and analyzed as their alditol acetates by gas-liquid chro- then at 4 "C for 72 h, and the amount of nitrogen in the precipitate was assayed as described above. The controls, one containing the matography (21).

Carbohydrate Specificityof Jackfruit Lectin

9367

TABLE I Purification of jackfruit lectin Fraction

Protein

Titer"

Specific activitv'

Purification

mglml

-fold

Crude extract 3.1 128 41 1.0 (0.86)' 4096 (4096) 4096 (4763) Rivanol purified lectin a Hemagglutination titer was determined with human untreated B erythrocytes. * Specific activity is expressed as titer/mg of protein/ml. The figures in brackets of affinity chromatography-purified lectin adopted from Ref. 6.

1 100

(116)

1 0 i

10

30 40 50 Temperature ( Oc )-

20

60

70

80

FIG. 2. Thermal stability of jackfruit lectin. JFL (60pg/ml) and 30 min was heated a t temperature indicated for 10 min (-) (---), cooled, and hemagglutination was performed.

FIG.1. A, polyacrylamide (7.5%) disc gel electrophoresis of purified JFL (50pg, pH 4.3) at 3 mA for 4 h. B, immunoelectrophoresis of jackfruit lectin; upper well, Rivanol purified lectin; lowerwell, affinity purified lectin as described in Ref. 6; middle well, antiserum to jackfruit lectin. TABLE I1 Hemagglutination profiles of purified jackfruit lectin Concentration of lectin used in the assay was 0.25 mdml. Erythrocytes

Titer Untreated

Pronase treated

Human" 1,024 4,096 Mouse 128 512 Buffalo 256 1,024 Duck 256 4,096 Goat 0 32,768 Pigeon 16,384 65,536 a Irrespective of A, B, and 0 blood groups.

Neuraminidase treated

4,096 512 512 512 512 32,768

same amount of galactomannan and JFL as in the other tubes and the other containingJFL only, were also set up. RESULTS

Purification of Lectin-Rivanol precipitated nonagglutinable contaminating proteins from the crude extract settingthe lectin free in the supernatant. The excess Rivanol from the supernatant was removed by precipitation with solid potassium bromide as insoluble bromide salt. Since contamination of lectin with Rivanol hampers its serological property, the solution was assayed spectrophotometrically in the range of 300-400 nm in order to determine the presence of Rivanol, (characteristic absorption at 360 nm) and was found to be completely free from it. The purification fold achieved was 100 (Table I). Homogeneity-The lectin was proved to be homogeneous electrophoretically and immunoelectrophoretically as it

showed a single band in polyacrylamide gel electrophoresis (pH 4.3) (Fig. lA) and produced a single arc with its rabbit antiserum by immunoelectrophoresis (Fig. 1B). Molecular Muss-The lectin showed a molecular mass of 41,000 dalton as determined bygel filtration. Onsodium dodecyl sulfate-polyacrylamide gel electrophoresis it dissociated into single species ofM, 10,000 suggesting it to be a tetramer. Carbohydrate Analysis-The neutral carbohydrate in JFL was estimated to be 7%. The carbohydrate analysis of the lectin revealed the presence of xylose (4.1%),mannose (0.9%), galactose (l.l%), glucosamine (1.6%), and trace amounts of glucose (0.2%) and mannosamine (0.4%). Hemagglutination and Hemagglutination Inhibition Assays-The purified lectin agglutinated human 0, A, B, and AB erythrocytes equally well. It strongly agglutinated untreated and enzyme treated mouse, buffalo, duck, and pigeon erythrocytes (Table 11). Untreated erythrocytes of goat were not agglutinated by JFL but could be whenthey were treated with the enzyme Pronase or neuraminidase. The hemagglutination of the lectin was strongly inhibited by D-Gal-@-(l --., ~ ) - D - G ~ ~ N A ~ - ~ - O - ( C H ~ ) ~ N H C O COOCH3, Me-a-D-Gal and Me-a-D-GalNAc (requiring 0.78, 3.1, and 6.2 mM, respectively, to inhibit two hemagglutination doses of lectin) while Me-B-D-Gal wasinactive even up to 400 mM concentration. Isoelectric Focusing-Isolectric focusing of JFL on 5%polyacrylamide gel produced fivebands of PI values 7.1, 6.85, 5.5, 5.3, and 5.1, respectively. Metal Analyses-Metal analyses of JFL by atomic absorption spectroscopy showed that iron was present in highest amount (1.0 mol/mol of lectin) followedby calcium and magnesium (0.55mol and 0.46 mol/mol of lectin, respectively). Chromium and zinc were present in smaller amounts (0.28 and 0.1 mol, respectively) while manganese was negligible (0.01 mol). The treatmentof JFL by dialysis in presence of acetic acid

Carbohydrate Specificity of Jackfruit Lectin

9368

0

0.4

0.8

1.2

r

1.6

2.0

SI Micrograms of polysaccharide

,

I

r FIG. 3. A, Scatchard plot for the binding of 4-methylumbelliferyla-D-galactopyranoside to jackfruit lectin at 30 "C. The number of the binding sites (valency) was obtained from the intercept on the x axis, and the association constant was determined from the Scatchard l/n; where n = number of binding equation l / r = l / n K , X l / c sites; K. = association constant; c = concentration of free sugar; r = ratio of mol of sugar bound/mol of lectin. B, Scatchard plot for the binding of melibiose to jackfruit lectin. Equilibrium was attained in 0.85% NaCl at 30 'C for 60 h. The number of binding sites and association constant were determined as described in A.

+

V

5 2 0

0

40 80 120 160 200 Micrograms of glycoprotein

FIG. 4. Quantitative precipitin curves of the purified jackfruit lectin (14.0 fig N) with polysaccharides ( A ) and glycoproteins ( B ) .Total volume was 400 pl. Symbols of polysaccharides and glycoproteins are shown in Table 111.

followed by EDTA and NaC1, respectively, neither affected the hemagglutinating activity of the lectin, nor did incorporation of Ca2+and/or other bivalent ions in saline show any enhancement in the agglutination titer. Metal contents of two sugar-binding sites/mol of lectin with an association acetic acid-EDTA-dialyzed JFL measured by atomic absorp- constant 0.21 X lo3M" (Fig, 3B). tion spectrophotometer revealed the presence of calcium in Interaction with Mouse Ascites Cell-JFL strongly agglutinhighest amount followed by iron and magnesium. ated ascites cell developedin Swiss albino mice. The minimum Thermal Stability-JFL was significantly stable for 10-15 amount of purified JFL required to aggiutinate ascites cell days at room temperature. Fig. 2 shows its thermal stability. (1.02 X lo6 cells/ml) was36 gg/ml. The agglutination of The hemagglutinating activity decreased gradually with in- ascites cell with JFL was strongly inhibited by T-disaccharide, crease of temperature and was completely lost either at 75 "C D-Gal-@-(l -+ 3)-~-GalNAc,methyl-a-D-galactoside, and (30 min of incubation) or at 85 "C (10 min of incubation). melibiose suggesting that theinteraction was specific. Effect of pH-JFL agglutinated erythrocytes almost with Quantitative Precipitin Assay-The quantitative precipitin equal potency in the pH range 5-10 although in the presence reaction curves of the purified JFL with various galactomanof phosphate buffer, pH 6-7, its activity was enhanced slightly. However, it was very labile at pH below 5 or above nans, glycoproteins, and polysaccharides are shown in Fig. 4, A and B , and the quantity of them giving 50% precipitation 10. and the maximum amount of the lectin N precipitated by FluorescenceQuenching and Equilibrium Dialysis-The maximum fluorescence of 4-methylumbe~liferyl-a-~-galacto-these substances are given in Table 111. The lectin was well pyranoside was observed at 374 nm. The observed fluores- precipitated by galactomannans from fenugreek, guar, S. sescence was quenched with successive addition of the lectin at ban, and P. pulcherrima of which guar required significantly 27 "C. The number of moles of free and bound sugar were less amount, 5 gg only.A glucogalactomannan from C. saltiana also precipitated JFL obviously requiring a higher amount, 98 calculated from the difference of quenching, and the Scatchard plot was drawn. The Scatchard plot showed that the pg, whereas polysaccharides from Klebsiella Type 10 (K10) number of binding sites for JFL is two (Fig. 3A). The associ- and gum arabic did poorly. Amongthe glycoproteins, desialyation constant (KO) calculated from the Scatchard plot was 6 lated glycophorin, birdnest glycoprotein, honey comb glycoX lo5 M-'. The Scatchard plot based on the results of equilib- protein, ant egg glycoprotein, and IgA precipitated 50-100% rium dialysis using melibiose as binding sugar also showed of the lectin added and required 51, 71, 86, 94, and 90 gg,

Carbohydrate Specificity

9369

ofLectin Jackfruit

TABLE I11 Comparative precipitating activitiesof various polysaccharides and glycoproteins with jackfruit lectin Substances Symbol

0

A 0 0

X

0 V

A 0

Substances

Giving 50% precipitation

Fenugreek Guar

S. sesban P. pulcherrima C. saltiana K-10 Gum arabic 4.3 AEG BNG HCG Desialylated glycophorin PEM IgA Dimeric IgA “B” substance 28.6

27 5 30 26 98 >200 0.6>120 94 71 86 51 200 >200 4.0 >200

Lectin

Giving maximum precipitation

100 20 57.1 80 180 70.0 180 80 40 100 150 150 150 200 78.6 150 180 200

Maximum precipitated

Recovery

&

%

9.8 8.2 8.0 11.8 9.8 0.8

70.0 58.6 84.3 5.7

7.1 12.5 9.9 14.0 7.0 11.0 4.1

50.7 89.3 10.7 100.0 50.0 29.3

26

2o

/-...i.i

t I

lCi’

1oo

Id Micromoles Inhibitor added

2

10

o3

1

FIG. 5. Inhibition by mono- and oligosaccharides of precipitation of fenugreek galactomannan with purified jackfruitlectin (6.6 pg N). Total volume was 200 pl. Symbols of inhibitors are shown in Table IV.

respectively. Pig erythrocyte mucoid reacted relatively weakly times) than methyl-a-D-Gal (4.7 p M ) . Melibiose (D-Gal-a-(1 2-deoxy-methyl-a-~-Gal, and D-Gal-a-(l + 6)and required 200 pg to cause 50% of precipitation of the lectin + 6)-~-Glc), added. Dimeric IgA and blood group B substance were signif- &Man were good inhibitors requiring 14.5, 18 and 22 pM, (174 p M ) was poor. icantly less active in precipitating the lectin and gave maxi- respectively, while D-Gal-a-(l+ 3)-~-Ara Methyl-a-D-Tal (69 p M ) , raffinose (73 p M ) , methyl-a-D-Fuc mum precipitation of 29.3 and 28.6%, respectively. (112 p M ) inhibited Quantitative Precipitin Inhibition Assay-The abilities of (83 PM) and methyl-2-O-methyl-a-~-Gal various sugars to inhibit the precipitation of jackfruit lectin the lectin-galactomannan precipitation in almost equal po, (200 with fenugreek galactomannan are shown in Fig. 5 and the tency. The free sugars uiz., D-Gal (220 p ~ ) D-GalNAc (263 pM), 2-deoxy-~-Gal(280 p M ) , and D-Tal(312 amounts required for 50% inhibition are listed in Table IV. p M ) , D-FUC ) very poor inhibitors. Methyl-B-D-Gal, lactose, Of the monosaccharides and their glycosides tested, the a- p ~ were methyl-a-D-Man, L-Ara, and methyl-a-Llinked glycosides of D-Gal and D-GalNAc were found to be methyl-a”&, most active while their corresponding P-anomers were inac- Ara, either inhibited precipitation very slightly or not at all, even at a very high concentration. tive. D-Gal-@-(l+ ~)-D-G~~NAC-~-O-(CH~)~-NHCO(CH~)~COOCH, was found to be the best inhibitor requiring only DISCUSSION 0.63 p~ for 50% inhibition while D-Gal-P-(l + 3)-~-GalNAc requiring 3.0 p M was almost five times less active than the The purification procedure described herein consists of former. 0- and p-nitrophenyl a-D-GalNAc requiring 1.1 and precipitation of nonagglutinable proteins from jackfruit seed 1.4 p ~ respectively, , were seven to eight times better than extract by Rivanol. The molecular weight of the native lectin ) abouttwice less active than methyl a-D-GalNAc (9.2 p ~ and and its subunit, itscarbohydrate composition, hemagglutinathe most potent (0.63 p ~ ) whereas , 0- and p-nitrophenyl ation/agglutination-inhibition, ability to precipitate galactoD-Gal (3.7 and 3.2 p M ) were slightly more potent (1.3-1.5 mannans and glycoproteins, etc. suggested that the lectin

9370

Carbohydrate Specificity of Jackfruit Lectin TABLE IV Amount of different mono- and oligosaccharides giving 50% inhibition of precipitation of jackfruit lectin by fenugreek gukctomannan Line no.

Symbol

20 26 7

0

6 5 19 8 10

0

@

3

0

2 25 18 21 9

0 Q

Inhibitors

Quantity giving 50% inhibition

Relative potency"

PM

12 4 1

I

0

e xt

v

8 0 H

*

14

@

17

0

16 23 11 24 13 22 15

A X

0

* V

0

D-Gal Galactitol Me-a-D-Gal Me-P-D-Gal o-NOzPh-a-D-Gal p-NOzPh-a-D-Gal D-GalNAc Me-a-D-GalNAc Me-3-O-Me-a-~-GalNAc o-NOZPh-a-n-GalNAc p-NOZPh-a-D-GalNAc p-NO,Ph-p-D-GalNAc ~-Gal-a-(l+3)-L-Ara ~-Gal-a-(1-.3)-~-Man D-Gal-a-(l-&)-D-Glc (melibiose) D-Gal-a-(l-&)-D"an ~-Gal-p-(1-+3)-~-GalNAc D-Gal-p-(1-+3)-DGalNAc-a-o-(CH&NHCO-(CHz)?COOCH3 D-Gal-P-(1-4)-D-Glc (lactose) D-Ga1-d l"rG)-D-Glc-P(1-~2)-~-Fru(raffinose) D-Gal-a-(l-rG)-D-Gal-a(l-*s)-D-GlC-P-(1-~2)D - F (stachyose) ~ Me-2-O-Me-a-D-Gal 2-Deoxy-~-Gal P-Deoxy-Me-a-~-Gal D-Tal Me-a-D-Tal

DFUC

Me-a-D-Fuc L-Ara 27 U Me-a-L-Ara Me-a-D-Man Me-a-D-Gle a Calculated in relation to Me-a-D-Gal. A

220.0 620.0 4.7 Tested up to10,000 p~ 3.7 3.2 200.0 9.2 16.0 1.4 1.1 400.0 174.0 230.0 14.5 22.0 3.0 0.63

Tested up to 10,000 p M 73.0 130.0 112.0 280.0 18.0 312.0 69.0 263.0 83.0 Tested up to 10.000 p M 6600.0 Tested up to 10,000 p M Tested up to 10,000 p M

0.024 0.007 1 34% inhibition 1.3 1.4 0.02 0.5 0.29 3.6 4.3 0.012 0.03 0.020 0.32 0.21 1.57 7.5

27% inhibition 0.06

0.036 0.04 0.017 0.26 0.015 0.07 0.018 0.06 14% inhibition 0.0007 42% inhibition 46% inhibition

purified by this method is identical in physical and chemical use commercially because of its large production, minimum properties to that purified over melibiose-agarose affinity cost, and less time consumption. absorbent (6). Besides human, JFL strongly agglutinated various animal The purification of peanut agglutinin, a true anti-Tlectin, erythrocytes. With goat erythrocytes it behaves like an "inby Rivanol precipitation (25) led us to purify this anti-T-like complete" nonagglutinating lectin as the cells were agglutinlectin by the same procedure. Horejsi andSmetana (26) ated after treatment with Pronase or neuraminidase. This reported that IgG could be separated from the mixtures of observation was also noted with chicken red cells (1). immunoglobulins in human serum by Rivanol treatment JFL is a metalloprotein containing iron in the highest whereby except IgG all other immunoglobulins were removed amount. This might have absorbed to the surface of the by precipitation leaving IgG in solution. Antiserum produced protein since on dialysis against acetic acid followed by in rabbits against JFL cross-precipitated human IgG in agar EDTA, its iron contentwas lost in appreciable amount. geldiffusion: thereby facilitating the purification of this JFL is thermally stable like many other plant lectins uiz., lectin. Another anti-T-like lectin, Artocarpin, from the seeds A . lakoocha (28) and Erythrina indica (29). of Artocarpus lakoocha having specificity for D-Gal-P-(l+ 3)The fluorescence quenching and equilibrium dialysis studD-GalNAc and bearing identical physical, chemical, and immunochemical properties to JFLwas also purified by Rivanol ies show that homotetrameric JFL possesses two sugar-bindtreatment (27). Although the purification fold (100) is slightly ing sites. The two sugar-binding sites were also present in less than that (116) achieved by affinity chromatography (6), homotetrameric lima bean (component 111) (30) and soybean the method adopted in the present study is advantageous to (31) lectins although recent studies were reported to possess four binding sites/tetramer (32, 33). JFL prepared by Suresh Kumar et al. (34) on immobilized guar galactomannan also B. P. Chattejee, manuscript in preparation.

9371

Carbohydrate Specificityof Jachfruit Lectin

contained two sugar-binding sites (35). Like Ricinus communis,Phaseolus vulgaris (36,37), and GS I (38) lectins J F L also agglutinated mouse ascites cells. This is possibly due to its attachment with the carbohydratereceptor,a-D-Gal residues presentintheterminal position of Ehrlichascites cell plasmamembrane (39). However, the agglutinating activityof J F L for ascites cells (36 pg/ml; minimum amount requiredfor agglutination) was36 and 750 times less compared with rat lymphocytes (1 pg/m12) and human erythrocytes (48 ng/ml) (6),respectively. The quantitative precipitin and precipitin inhibition studies of jackfruit lectin give insight into thespecificity and size of its combining site. From the previous report it is known that FIG. 6. The sites of lectin-sugar interactions are repreJFL has binding affinityfor terminal nonreducing a-D-galac- sented by shaded area. (3tosyl units (2,6) as well as for D-Gal-@-(l+ 3)-~-GalNAc 5). The present study confirms the specificity of JFL with precipitation although the amount requiredwasveryhigh some additional information. From Table IV it is evident that (Table IV).However, absence of hydroxymethyl group at C-6 the precipitation of J F L with fenugreek galactomannan was as in L-Ara (only 14% inhibn. at 10 mM) and even in Me-astrongly inhibited by a-glycosides of D-Gal and D-GalNAc as L-Ara, (6.6 mM) did not cause them to bind with the lectin. disaccharideshaving well asD-Gal-a-(l + 6)units.The Galactitol was a very poor inhibitor indicating that the terminal galactosylresidue linked glycosidically 1,2, 1,3, or combining site of JFL, unlike peanut (46), did notrecognize 1,4 ( a l p ) t o any subterminal sugar, except D-Gal-P-(l+ 3)- the open chain structure of galactose. This behavior was also D-GalNAc, hardly inhibited the lectin-galactomannan precip- observed in GS I lectin (42). itation. Thus, D-Gal-a-(l+ 3)-~-Ara/~-Man was a very poor Among the monosaccharides and oligosaccharides tested inhibitor. The very weak precipitation of K 10 (40) and gum D-Gal-P-(l + ~ ) - D - G ~ ~ N A C - ~ - O - ( C H ~ ) ~ - N H C O ( arabic (14) polysaccharides with JFL (Table 111) supported COOCH, was the most potent inhibitor. This is supported by the above results. Blood group B substance (41) precipitated the fact that birdnest glycoprotein (47), desialylated glycothe lectinweakly and the amountrequired to precipitate50% phorin (48), pig erythrocyte mucoid (49), and IgA (50) all of the lectinwas very high (>200 pg), although the inhibitory having D-Gal-P-(l+ 3)-~-GalNAcresidue a-0-glycosidically effect of D-Gal-a-(l -+ 3 ) - ~ - G a is l yet to be tested and this linked to Ser or Thr, precipitatedwell. JFLHowever, a higher may arrive at a definite conclusion of the above statement. In amount of pig erythrocyte mucoid is required because D-Galthis respect JFL differs from GS I (42) lectin as the latter (3-(1 + 3)-~-GalNAc in themucoid is substituted by sialic precipitated well gum arabic and blood group B substance. acid. It is logical to mention here that JFL was precipitated J F L did not precipitatecarboxyl-reduced mucilage polysac- by [D-Gal-P-(l + ~)-D-G~~NAC-~-O-(CH~)~-CONH] 3oBSA charide from H.moscheutos in which galactosyl residue is a- butneitherwasitprecipitated by [D-Gal-P-(l 3)-Dglycosidically 1,2-linked although this is not present in the G~~NAC-P-O-(CH~)~-CONH]~~HSA nor its hemagglutinanonreducing terminal position (43). tioninhibited by D-Gal-@-(l ~)-D-G~~NAC-P-O-(CH~)~In the caseof di- or oligosaccharides having D-Gal-a-(l COOCH3 (5). Thisfinding clearly demonstrates that JFL 6) linkage, there may be great opportunity for rotation about recognizes extended linkage a t C-1 of T-disaccharide and this 1,6-glycosidic bonds which causes the greater reactivity of is obviously a in nature. From the inhibition results of Me-aoligosaccharide with this linkage on the basis of the -CH2 D-Gal (4.7 p ~ )Me-a-D-GalNAc , (9.2 pM), and D-Gal-P-(l group. Plant galactomannans such as those from fenugreek 3)-~-GalNAc (3 pM), it could presumably be said that the8S. sesban(aegyptiaca)(45) containing D-Gal unit of T-disaccharide either had a very little effect or (2),guar(44),and terminal D-Gal-a-(l + 6) residues were well precipitated by not and thus substitution at C-3 of a-D-Gal/a-D-GalNAc was J F L (Table 111). not important. This fact was confirmed as Me-3-O-Me-a-~The better inhibition causedby 0- and p-nitrophenyl a-D- GalNAc inhibitedtheprecipitationalmost like Me-a-DGalNAc over Me-a-D-GalNAc (Table IV) was possibly due to GalNAc. However, the effect of P-D-Gal unit could be ascernonspecific hydrophobic interaction between aromatic agly- tained by comparing the inhibition results of D-Gal-P-(l + con of the sugar moiety and the combining siteof the lectin. ~)-D-G~~NA~-~-O-(CH~)~-NHCO-(CH~)~-COOCH with Dp-Nitrophenyl-P-D-GalNAc inhibited the precipitation very G~~NAC-(Y-O-(CH~)~-NHCO(CH~)~-COOCH~ and D-Gal-P-(l weakly (400 p M ) presumably for the same reason. + 3)-D-GalNAc-a-O-Me with Me-a-D-GalNAc. The signifiThe substitution or absence of -OH a t C-2 of D-Gal had cant inhibition of P-D-Gal-(l + 3)-~-GalNAc containing no little influence on binding to JFL because the binding affinity extended a-aglycon moiety is due to the presence of a-OH at of Me-a-D-GalNAc and 2-deoxy-Me-a-~-Gal was not affected C-1 of D-GalNAc in solution (51). much in comparison to Me-a-D-Gal. The inversion of -OH at The present quantitative inhibition studies suggest that the C-2 is not favorablefor binding since Me-a-D-Tal was 15 combining siteof J F L is recognized by the -OH groups mainly times less potent than Me-a-D-Gal (Table IV). The inhibitory at C-4 and partiallya t C-6 of D-Gal/D-GalNAc and extended potency of Me-P-O-Me-cy-~-Galwas reduced much probably up to a-CH2-aglyconic part asshown in Fig. 6. The other Ddue to steric impediment. galactosyl-binding lectin from the tunicate Didemnum candidum (DCL-I) (52) interacts with hydroxyls on carbons 2, 3, The -OH at C-4 of hapten inhibitor is important since and 4, while lectins from peanut (53) and GS I (42) interact inversion as in Me-a-D-Glc, made it inactive (Table IV). The -OH at C-5 cannotbeinteractingwiththelectin at C-2, C-4, and C-6hydroxylgroups. M . pomifera lectin combining site since it forms the hemiacetalic ring with C-1. recognizes C-4 and C-6-OH groups of a-glycosides (54), and The -OH at C-6 is not very critical as D-FUC inhibited the thus closely resembles JFL. lectin slightly less than D-Gal did. This result is contrary to that reportedby Sastry etal. (4) as Me-a-D-Fuc inhibited the Acknowledgments-We wish to express sincere thanks to Dr. J.

-

-

-

-

9372

Carbohydrate Specificity of Jackfruit Lectin

Paland Dr. S. Chowdhury of this department for their helpful discussions in preparing the manuscript and toDr. M. Sarkar of the Indian Institute of Chemical Biology, Calcutta, for her kind help in performing the isoelectric focusing experiment. "

REFERENCES 1. Chatterjee, B. P., Vaith, P., Chatterjee, S., Karduck, D., and Uhlenbruck, G. (1979) Int. J. Biochem. 10, 321-327 2. Chatterjee, B. P., Sarkar, N., and Rao, A. S. (1982) Carbohydr. Res. 104, 348-353 3. Chatterjee, B. P., and Uhlenbruck, G. (1982) Experientia 3 8 , 1225-1226 4. Sastry, M. V. K., Banarjee, P., Patanjali, S. R., Swamy, M. J., Swarnalata. G. V.. and Surolia.. A. (1986) . . J. Biol. Chem. 2 6 1 . 11726-11733 5. Chatteriee. B. P.. Ahmed. H..Uhlenbruck. G.. Janssen. E.. Kolar. C., an-d Seiler, F . R. (1985) Behring I n s t . Mitt. 78, 1481158 ' 6. Ahmed, H., andChatterjee, B. P. (1986) in Lectins, Biology, Biochemistry, Clinical Biochemistry (Bcig-Hansen, T. C., and van Driessche, E., eds) Vol. 5, pp. 125-133, Walter de Gruyter, New York I . Appukuttan, P. S., and Basu, D. K. (1985) J. Biosci. 7 , 7-14 8. Bunn-Moreno, M. M., and Campos-Neto, A. (1981) J. Zmmunol. 127,427-429 9. Kondoh, H., Kobayashi, K., Hagiwara, K., and Kajii, T. (1986) J. Immunol. Methods 88,171-173 10. Kobayashi, K., Hagiwara, K., Kondoh, H., and Vaerman, J. P. (1987) Adu. Ewp. Med. Biol. 2 1 6 B , 1193-1197 11. Sarkar, M., and Kabat, E. A. (1979) Carbohydr. Res. 69,143-149 12. Nowotny, A. (1955) Acta Physiol. Acad. Sci. Hung. 8,25-31 13. Chowdhury, S., Ray, S., and Chatterjee, B. P. (1988) Glycoconjugate J. 5, 27-34 14. Aspinall, G . O., and Kneble, M. C. (1986) Carbohydr. Res. 1 5 7 , 257-260 15. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 16. Reisfeld, R. A., Lewis, U. J., and Williams, D. E. (1962) Nature 195, 281-283 17. Laemmli, U. K. (1970) Nature 2 2 7 , 680-685 18. Graber, P., and Williams, C.A. (1953) Biochim. Biophys. Acta 10, 193-194 19. Andrews, P. (1965) Biochem. J. 96,595-606 20. Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. (1956) Anal. Chem. 2 8 , 350-356 21. Pal, R., Ahmed, H., and Chatterjee, B. P. (1987) Biochem. Arch. 3,399-412 22. Ruffo, G. A., Jr., Singh, J . P., Babcock, D. F., and Lardy, H. A. (1982) J. Bid. Chem.257,4627-4632 23. Paulova, M., Ticha, M., Entlicher, G., Kostir, J. V., and Kocourek, J. (1971) Biochim. Biophys. Acta 2 5 2 , 388-395 N. Y. Acad. Sci. 51,660-672 24. Scatchard, G. (1949) Ann. 25. Merry, A. H., Rawlinson, V. I., and Stratton, F. (1981) Immunological Commun. 10, 265-273 26. Horejsi, J., and Smetana, R. (1956) Acta Med. Scand. 1 5 5 , 6570

27. Chatterjee, B. P., Ahmed, H., and Chowdhury, S. (1987) Abstract in the 4th European Carbohydrate Symposium July 12-17, p. B-8, Darmstadt, F. R. G. 28. Chowdhurv. S., Ahmed, H.. and Chatteriee. . . Carboh" . B. P. (1987) ydr. Res.- 1 5 9 , 137-148 29. Bhattacharyya, L., Das, P. K., and Sen, A. (1981)Arch. Biochem. Biophys. 2 1 1,459-470 30. Bessler, W., and Goldstein, I. J. (1974) Arch. Biochem. Biophys. 165,444-445 31. Lotan, R., Siegelman, H. W., Lis, H., and Sharon, N. (1974) J. Bwl. Chem. 249, 1219-1224 32. Roberts, D. D., and Goldstein, I. J. (1984)Arch. Biochem. Biophys. 230,316-320 33. DeBoeck, H., Lis, H.,Van Tilbeurgh, H., Sharon, N., and Loontiens, F. G. (1984) J. Biol. Chem. 2 5 9 , 7067-7074 34. Suresh Kumar, G., Appukuttan, P. S., and Basu, D. K. (1982) J. Biosci. 4, 257-261 35. Appukuttan, P. S., and Basu, D. K. (1985) FEBS Lett. 180,331334 36. Tomita, M., Oswa, T., Sakurai, Y., and Ukita, T. (1970) Int. J. Cancer 6,283-289 37. Nachbar, M. S., Oppenheim, J. D., and Aull, F. (1976) Biochim. Biophys. Acta 419,512-529 38. Eckhardt, A. E., Barbara, N.M., and Goldstein, I. J. (1982) Cancer Res. 42,2977-2979 39. Eckhardt, A. E., and Goldstein, I. J. (1983) Biochemistry 2 2 , 5290-5297 40. Sarkar, A. K., and Roy, N. (1986) Carbohydr. Res. 152, 205-215 41. Watkins, W. M. (1972) in Glycoproteins Part B (Gottschalk, A., ed) pp. 830-891, Elsevier Scientific Publishing Co., Amsterdam 42. Hayes, C. E., and Goldstein, I. J. (1974) J. Biol. Chem. 2 4 9 , 1904-1914 43. Tomoda, M., Shimada, K., Shimizu, N., Kanari, M., and Kaneko, E. (1986) Carbohydr. Res. 151,29-35 44. Smith, F., and Montgomery, R. (1959) The Chemistry of Plant Gums and Mucilages, pp. 324-338, Reinhold, New York 45. Bhattacharyya, S. B., Das, A. K., Banerji, N., and Farooqi, M. I. H. (1983) Phytochemistry 22,161-164 46. Pereira, M. E. A., Kabat, E. A., Lotan, R., and Sharon, N. (1976) Carbohydr. Res. 51,107-118 47. Farrar, G. H., Uhlenbruck, G., and Karduck, D. (1980) Hoppe Seyler's 2. Physiol. Chem. 361,473-476 48. Thomas, D.B., and Winzler, R. J . (1969) J. Biol. Chem. 2 4 4 , 5943-5946 49. Newman, R. A., Gloeckner, W. M., and Uhlenbruck, G . (1976) Eur. J. Biochem. 64,373-380 50. Vaith, P., and Uhlenbruck, G. (1978) 2. Zmmunforsch. 154, 114 51. Flowers, H. M., and Shapiro, D. (1965) J. Og. Chem. 30, 20412043 52. Vasta. G.R.. and Marchalonis. J. J. (1986) . . J. Biol. Chem. 261. 9182-9186 53. Lotan. R.. Skutelskv. E.. Danon., D.., and Sharon., N. (1975) J. . Biol: Chem. 250,85168523 54. Sarkar, M., Wu, A. M., and Kabat, E. A. (1981) Arch. Biochem. Biophys. 209,204-218