Carbohydrate binding specificity of the lectin from the PEA (Pisum sativum)

Biochimica et Biophysica Acta, 379 (1975) 456-461

© Elsevier Scientific Publishing Company, Amsterdam --- Printed in The Netherlands BBA 36939 C A R B O H Y D R A T E B I N D I N G SPECIFICITY OF T H E LECTIN F R O M T H E PEA ( P I S U M S A T I V U M )

J. P. VAN WAUWE, F. G. LOONTIENS and C. K. DE BRUYNE Laboratorium voor Algemene en Biologische Scheikunde, Rijksuniversiteit van Gent, K. L. Ledeganckstraat 35, B-9000 Gent (Belgium)

(Received July 19th, 1974)

SUMMARY Hapten inhibition measurements on the precipitin reaction beteeen Pisum sativum lectin and Pichiapinus phosphomannan showed the lectin to bind D-mannose, D-glucose, D-fructose and L-sorbose. Unmodified hydroxyl groups at the C-4 and the C-6 positions of the D-glucopyranose ring were essential for binding to the protein. Modification of the C-2 hydroxyl group was allowed in the D-glucopyranose ring, but not in the D-mannopyranose configuration. Substitution of the hydroxyl hydrogen atom at the C-3 position of D-glucose increased the binding efficiency. With the exception of gentiobiose, the fl-linked glycobioses tested were not bound to the lectin, whereas the a-linked glycobioses were potent inhibitors. In general, the P. sativum lectin was found to be less sensitive to structural variation of inhibiting carbohydrates than concanavalin A, the lectin from Canavalia ensiformis.

INTRODUCTION Lectins or phytohemagglutinins are a group of proteins, mainly of plant origin, with interesting biological features and applications [1], governed by their binding with carbohydrates. Their interaction with polysaccharides resembles the antibodyantigen reaction [2, 3] and can be specifically inhibited by simple sugars, allowing detailed studies of the carbohydrate binding sites of lectins. Earlier studies, performed with inhibition of agglutination of human erythrocytes, indicated that the lectin from the pea (Pisum sativum) binds sugars which fall into M/ikel/i's group III [4, 5]. The binding of p-substituted phenyl a-D-gluco- and mannopyranosides showed Hammett relationships with negative Q values [6], identical with the value [7] for concanavalin A, suggesting a common dependence of the electronic properties of the substituents for both lectins. However, differences in specificity of the pea lectin and concanavalin A have been observed. The pea lectin more readily agglutinated tumor cells [8] and erythrocytes [9]; about 10 ~ of 12SI-labeled pea lectin bound to spleen cells and displaceable by unlabeled pea lectin, could not be displaced by concanavalin A [10]. We therefore examined the sugar binding specificity of the pea lectin, using a

457 series of monosaccharides, synthetic glycosides and oligosaccharides as inhibitors of the precipitin reaction between the pea lectin and Pichia pinus phosphomannan, in order to decide if the distinction in biological activity of both lectins could be explained in terms of differences in sugar binding specificity. MATERIALS AND METHODS The lectin from Pisum sativum was prepared by affinity chromatography using Sephadex G-75 [11]. The protein was stored at 4 °C in 1 M NaCI. Protein concentrations were determined at 280 nm (Zeiss PMQII-M4QII) in 1 M NaCI, using P 10/o ~l cm equal to 12.0. Pichia pinus phosphomannan was isolated as described [12]. All carbohydrates used in this study were examined for their purity by thinlayer chromatography. Phenyl 3-O-methyl-fl-D-glucopyranoside triacetate was obtained by a modified Michael condensation [13], m.p. 145-145.5 °C, [a] o zz = --66.3 (c 1, CHCI3). Catalytic deacetylation [14] (sodium methoxide) yielded the corresponding phenyl 3-0methyl-fl-D-glucopyranoside, m.p. 152-153 °C, [a]o z2 = --66.2 (c 1, methanol). 2,3,4-Tri-O-methyl-D-glucopyranose was synthetized [15] by methylation cf 1,6-anhydro-fl-D-glucopyranose and subsequent hydrolysis. The fluoro sugars used were a gift from Dr A. B. Foster and Dr E. Bessell. 3-O-fl-D-Galactopyranosyl-D-glucopyranose, 2-acetamido-2-deoxy-3-O(fl-D-galactopyranosyl)-D-glucopyranose and nigerose were kindly provided by Dr A. Gauhe and Dr E. T. Reese. Other carbohydrates used were obtained commercially. Hapten inhibition of the lectin-P, pinus phosphomannan turbidity formation was carried out as reported [6]. RESULTS The inhibition of the P. pinus phosphomannan-lectin reaction by several monosaccharides and their derivatives is shown in Table I. We have classed as noninhibitors those carbohydrates which show less than 3 ~ inhibition at a final concentration of 15 mM. Among the monosaccharides tested, o-mannose was found to be a more potent inhibitor than D-glucose or D-fructose. Except L-sorbose (about 8.8 times less active than o-mannose), other monosaccharides failed to inhibit the lectin-polysaccharide precipitation reaction. Thus, monosaccharides having equatorial hydroxyl groups at C-3 and C-4 in the CI(D) conformation, possess the capacity to inhibit. Compared with D-glucose, methyl a-D-glucopyranoside is about two times as active, whereas methyl fl-D-glucopyranoside shows a 5-fold decrease in activity. The same effect can be noticed in the case of phenyi D-glucopyranosides, the a-anomer being approximately 8 times more effective than the fl-anomer. The effect of modification at C-2 in o-glucopyranose and o-mannopyranose is entirely comparable to the results of the inhibition studies with concanavalin A [16]: the similar degree of inhibition of 2-acetamido-2-deoxy-D-glucose, 2-O-methyl-Dglucose, 2-deoxy-D-glucose and 2-deoxy-2-fluoro-D-glucose as compared to D-glucose

458 TABLE I INHIBITION OF THE P. S A T 1 V U M LECTIN-P. P I N U S PHOSPHOMANNAN PRECIPITATION BY MONOSACCHARIDES AND THEIR DERIVATIVES The reaction mixture (3 ml, 25 °C) contained P. sativum lectin (1.50 rag), 150/~gphosphomannan and different amounts of inhibitor in 1 M NaC14).017 M Tris-HC1 (final pH 8.1). The reaction was started by addition of the lectin. After exactly 15 rain, the resulting turbidity was measured photometrically at 405 nm. o-Galactose, L-fucose, D-xylose, o-ribose, L-arabinose are all noninhibitors at a final concentration of 15 raM. Inhibitor

Monosaccharides o-mannose D-glucose D-fructose L-sorbose Modification at C-I methyl a-D-mannopyranoside methyl a-D-glucopyranoside methyl fl-D-glucopyranoside phenyl a-D-mannopyranoside phenyl a-D-glucopyranoside phenyl/3-D-glucopyranoside Modification at C-2 2-deoxy-2-fluoro-D-mannose 2-acetamido-2-deoxy-D-mannose 2-deoxy-2-fluoro-D-glucose 2-acetamido-2-deoxy-D-glucose 2-O-methyl-D-glucose 2-deoxy-D-glucose Modification at C-3 and C-4 3-O-methyl-D-glucose 3-O-benzyl-D-glucose 2,3 di-O-methyl-D-glucose methyl 2,3-di°O-methyl-a-D-glucopyranoside phenyl 3-O-methyl-fl-glucopyranoside 3-deoxy-3-fluoro-D-glucose 4-deoxy-4-fluoro-D-glucose 2,3,4-tri-O-methyl-o-glucose Modification at C-6 6-deoxy-D-glucose 6-deoxy-6-fluoro-D-glucose methyl a-D-xylopyranoside methyl ~-D-xylopyranoside phenyl a-D-xylopyranoside phenyl fl-o-xylopyranoside

Concn giving 50 % inhibition (mM) 0.70 1.65 3.00 6.20 0.30 0.84 7.42 0.20 0.44 3.50 0.60 4.30 (1%)* 1.50 1.40 1.50 1.80 0.13 0.098 0.15 0.048 0.21 5.10 4.30 (2 ~) 3.50 (2 ~) 5.70 (2 ~) 4.10 (5 ~) 101.0 (25 ~) 43.5 16.8 (0 ~) 36.0 (5 ~)

* Number in parentheses represent the percentage inhibition given by the concentration of carbohydrate noted. indicates a n o n p a r t i c i p a t i o n of the C-2 hydroxyl g r o u p o f the D-glucopyranose structure. A l t h o u g h D-mannose a n d its 2-fluoro a n a l o g u e are equally effective as inhibitors, 2 - a c e t a m i d o - 2 - d e o x y - o - m a n n o s e completely failed to inhibit, suggesting a specific interaction o f the axial C-2 hydroxyl g r o u p of the m a n n o p y r a n o s e configura-

459 tion with the protein. This is further substantiated by the fact that methyl a-D-mannopyranoside is about three times more potent as inhibitor than methyl a-D-glucopyranoside. The effect of modification at the C-3 hydroxyl group shows the following pecularities: 3-deoxy-3-fluoro-D-glucose is about three times less active as an inhibitor than D-glucose. On the contrary, 3-O-methyl of 3-O-benzyl substitution of D-glucose greatly enhances the binding power of these derivatives. Thus, 3-O-methyl-D-glucose (0.13 mM for 50 ~ inhibition) and 3-O-benzyl-D-glucose (0.098 mM for 50 ~o inhibition) are considerably better inhibitors than D-glucose. Both 3-O-substituted D-glucose derivatives are absolutely noninhibitors in the concanavalin A-dextran system [16]. Phenyl 3-O-methyl-fl-D-glucopyranoside is also an excellent inhibitor, binding about 17 times better than phenyl fl-D-glucopyranoside. 2,3-Di-O-methyl-D-glucose and 3-0methyl-D-glucose are equally potent, whereas methyl 2,3-di-O-methyl-a-D-glucopyranoside is the best inhibitor tested. 4-Deoxy-4-fluoro-D-glucose and 2,3,4-tri-O-methyl-D-glucose are both noninhibitors, demonstrating the implication of the C-4 hydroxylgroup in the binding. Modification of the C-6 hydroxyl group of the o-glucopyranose ring leads to a drastic loss in ability of the resulting carbohydrates to bind to the lectin. Compared with the methyl D-glucopyranosides, methyl a-D-xylopyranoside binds about 240 times less effective than methyl a-D-glucopyranoside, whereas methyl fl-D-xylopyranoside shows only a 5.8 fold loss in binding as compared with methyl-fl-D-glucopyranoside. On the other hand, both anomers of phenyl D-xylopyranoside were found to be noninhibitors. The binding potency of a- and fl-linked dissacharides for the P. sativum lectin and concanavalin A is shown in Table II, expressed with respect to the inhibition by methyl a-D-mannopyranoside. Disaccharides containing the a-glycosidic bond are active in both systems. a,a-Trehalose (0.30 mM for 50 ~ inhibition), the most effective disaccharide tested, is approximately 5.4 times more active than D-glucose, whereas maltose (4-O-a-DTABLE II INHIBITION OF THE PEA LECTIN AND CONCANAVALIN A SYSTEMS BY OLIGOSACCHARIDES, AS COMPARED WITH THE INHIBITION BY METHYL a-D-MANNOSIDE The same experimental procedure is employed as described in Table I. The concentration yielding 50 ~ inhibition by the oligosaccharides is expressed relative to the inhibition of methyl a-D-mannoside. Cellobiose, laminaribiose, 3-O-fl-D-galactopyranosyl-D-glucose,2-acetamido-2-deoxy-3-O-(fl-Dgalactopyranosyl)-D-glucose,lactose, melibiose, raffinose were all noninhibitors at a concentration of 15 mM. Inhibitor

Methyl a-o-mannopyranoside D-Glucose Trehalose Maltose Isomaltose Nigerose Sucrose Gentiobiose

Pisum lectinP. pinus phosphomannan

system 1 5.4 1 2.5 3.0 5.8 3.3 35

Concanavalin A dextran system (ref. 17) 1 47 3.8 7.2 3.8 8.2 38 158

460 glucopyranosyl-D-glucose), isomaltose (6-O-a-D-glucopyranosyl-D-glucose) and sucrose (a-D-glucopyranosyl fl-o-fructofuranoside) are somewhat less active, but still superior to D-glucose. The fl-linked glucobioses tested failed to exhibit any significant inhibition except for gentiobiose (6-O-fl-o-glucopyranosyl-D-glucose) which was a relatively good inhibitor (10.5 mM for 50 ~ inhibition). Oligosaccharides containing a non-reducing o-galactopyranosyl residue, such as the bloodgroup disaccharide 2-acetamido-2-deoxy-3-O-(fl-D-galactopyranosyl)-Dglucose, 3-O-fl-D-galactopyranosyl-D-glucose, lactose, raffinose and melibiose are ineffective as inhibitors. Other 3-O-linked disaccharides containing a reducing glucose residue are either as effective as D-glucose (nigerose) or totally ineffective (laminaribiose). DISCUSSION

In many respects, the specificity of the P. sativum lectin is similar to concanavalin A, both proteins binding sugars of M~ikel/i.s Group HI with the D-arabinohexopyranosyl configuration at C-3, C-4 and C-5. Distinction between the two lectins can be made when comparing their degree of specificity toward variation in structure of the inhibiting carbohydrates. Thus, the pea lectin binds D-mannose and 2-acetamido-2-deoxy-D-glucose, both among the most common constituents of glycoproteins, almost with the same combining strength (D-mannose is only 2.4 times more active than 2-acetamido-2-deoxy-D-glucose). On the other hand, concanavalin A shows a more marked difference in binding capacity of these two carbohydrates, D-mannose being about 6 times more potent than 2-acetamido-2-deoxy-D-glucose [17]. The pea lectin is also less sensitive to the anomeric configuration of the inhibiting saccharides. For the/3- and a-methyl D-glucopyranosides, the ratio of molarities producing 50~o inhibition is only 8. Once again, this value is lower than for the concanavalin A-dextran system, where the presence of a C-1 /%methoxyl group is believed to force this anomer into an unfavourable binding orientation [18]. In the latter system, methyl a-D-glucopyranoside is about 27 times more potent as an inhibitor than the corresponding fl-anomer. Although 3-O-benzyl-D-glucose, 3-O-methyl-D-glucose and its derivatives fail to interact with concanavalin A, they are found to be excellent inhibitors of the pea lectin reaction, 3-O-methyl substitution enhancing the inhibitor potency in both aand/~-D-glucopyranosides as demonstrated by the powerful inhibition of methyl 2,3di-O-methyl-a-D-glucopyranoside and phenyl 3-O-methyl-fl-D-glucopyranoside. Interconversion of the anomeric substituents and the 3-O-substituent seems feasable by rotation over 180° around the C6-Cs-C2-axis. However, the identical binding of the aromatic aglycon of phenyl fl-D-glucopyranoside and phenyl 3-O-methyl-fl-D-glucopyranoside is shown by an almost equal binding ratio of phenyl/3-o-glucopyranoside to D-glucose and of phenyl 3-O-methyl-fl-D-glucopyranoside to 3-O-methyl D-glucose, being 2.1 and 1.6 respectively. These values suggest an identical binding location of the aglycon in both phenyl glucopyranosides. It should be noted that hapten inhibition of the precipitin reaction of the Hi agglutinogen with Pisum arvense lectin [19] and of P. pinus phosphomannan with Lens culinaris lectin (our unpublished results) also demonstrated 3-O-methyl-D-glucose to be a potent inhibitor.

461 O u r results obtained with the inhibitory power of methyl D-xylopyranosides are comparable to those o f concanavalin A [18], where it was postulated that the 6hydroxymethyl g r o u p o f methyl fl-o-glucopyranoside binds in the position occupied by the C-2 hydroxyl group o f methyl a-o-glucopyranoside. Thus, methyl fl-o-xylopyranoside (43.5 m M for 50~o inhibition) is about 4.6 times more potent than the corresponding a - a n o m e r (101 m M for 25~o inhibition). Phenyl o-xylopyranosides, tested at high concentrations fail completely to inhibit. The inability of the aromatic xylopyranosides to bind to concanavalin A (our unpublished results) gives additional support to the different binding mechanism o f phenyl- and methylglycosides to these lectins [18]. Like concanavalin A [17], the pea lectin does not exhibit a p r o n o u n c e d sensitivity to variation in structure of the a-disaccharides. With the exception of gentiobiose, which is only 11 times less potent than isomaltose, fl-linked glucobioses are p r o b a b l y sterically hindered when binding to both lectins. The present results suggest that the differences in agglutinating activities between P. sativum lectin and concanavalin A are due to the lower degree of specificity of the pea iectin. The most striking difference is the enhanced binding of 3-O-substituted glucose molecules to the pea lectin, although this property is not reflected in the binding of a- or fl-(1-3)-linked disaccharides. REFERENCES 1 Sharon, N. and Lis, H. (1972) Science 177, 949-959 2 Goldstein, I. J., Hollerman, C. E. and Merrick, J. M. (1965) Biochim. Biophys. Acta 97, 68-76 3 Van Wauwe, J. P., Loontiens, F. G. and De Bruyne, C. K. (1973) Biochim. Biophys. Acta 313, 99-105 4 Boyd, W. (1963) Vox Sang. 8, 1-32 5 Betail, G., Genaud, L. and Coulet, M. (1972) Ann. Inst. Pasteur 123, 831-840 6 Van Wauwe, J. P., Loontiens, F. G., Carchon, H. A. and De Bruyne, C. K. (1973) Carbohydr. Res. 30, 249-256 7 Loontiens, F. G., Van Wauwe, J. P., De Gussem, R. and De Bruyne, C. K. (1973) Carbohydr. Res. 30, 51-62 8 Vesely, P., Entlicher, G. and Kocourek, J. (1972) Experientia 28, 1085-1086 9 Bures, L., Entlicher, G., Ticha, M. and Koucourek, J. (1973) Experientia 29, 1546-1547 10 Trowbridge, I. S. (1973) Proc. Natl. Acad. Sci. U.S. 70, 3650--3654 11 Entlicher, G., Kostir, J. V. and Kocourek, J. V. (1970) Biochim. Biophys. Acta 221,272-281 12 Jeanes, A., Pittsley, J. E., Watson, P. R. and Dimler, R. J. (1961) Arch. Biochem. Biophys. 92, 343-350 13 De Bruyne, C. K. and Van Wijnendaele, F. (1967) Carbohydr. Res. 4, 102-104 14 Thompson, A. and Wolfrom, M. L. (1963) Methods Carbohydr. Res. 2, 215 15 Bourne, E. J. and Peat, S. (1950) Adv. Carbohydr. Chem. 5, 145-190 16 Poretz, R. D. and Goldstein, I. J. (1970) Biochemistry 9, 1890-2896 17 Goldstein, I. J., Hollerman, C. E. and Smith, E. E. (1965) Biochemistry 4, 876-883 18 Brewer, C. F., Sternlicht, H., Marcus, D. M. and Grollman, A. P. (1973) Biochemistry 12, 44484457 19 Scheinberg, S. L. and Reckel, R. P. (1965) Nature 205, 250-254