- Email: [email protected]
Multiple recognition systems adopting four different glycotopes at the same domain for the Agaricus bisporus agglutinin–glycan interactions
FEBS Letters 584 (2010) 3561–3566
journal homepage: www.FEBSLetters.org
Multiple recognition systems adopting four different glycotopes at the same domain for the Agaricus bisporus agglutinin–glycan interactions Albert M. Wu *, Jia-Hau Liu, Yu-Ping Gong, Chia-Chen Li, En-Tzu Chang Glyco-immunochemistry Research Laboratory, Institute of Molecular and Cellular Biology, College of Medicine, Chang-Gung University, Kwei-san, Tao-yuan 333, Taiwan
a r t i c l e
i n f o
Article history: Received 13 May 2010 Revised 13 July 2010 Accepted 13 July 2010 Available online 17 July 2010 Edited by Sandro Sonnino Keywords: Carbohydrate specificity Combining site Complex N-glycan Lectin Agaricus bisporus agglutinin
a b s t r a c t For the GalNAca1? specific Agaricus bisporus agglutinin (ABA) from an edible mushroom, the mechanism of polyvalent Galb1?3/4GlcNAcb1? complex in ABA–carbohydrate recognition has not been well defined since Gal and GlcNAc are weak ligands. By enzyme-linked lectinosorbent and inhibition assays, we show that the polyvalent Galb1?3/4GlcNAcb1? in natural glycans also play vital roles in binding and we propose that four different intensities of glycotopes (Galb1-3GalNAca1-, GalNAca1Ser/Thr and Galb1-3/4GlcNAcb1-) construct three recognition systems at the same domain. This peculiar concept provides the most comprehensive mechanism for the attachment of ABA to target glycans and malignant cells at the molecular level. Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction The Agaricus bisporus agglutinin (ABA) in edible mushroom has been found to have a potent anti-proliferative effect on malignant colon cells with considerable therapeutic potential as an antineoplastic agent [1–3]. This lectin has a molecular mass of 6.4 104 Da and consists of four apparently identical subunits (Mr. 16 000) [4]. Previous studies on the structural requirements for glyco-recognition of ABA were focused on GalNAca1-related ligands at reducing ends of O-glycans [5]. It was also found that multi-complex N-glycans may be involved in glyco-recognition, while Gal, Man and GlcNAc were weak or inactive binders. To further clarify the mechanisms of multi-complex N-glycans in ABA– carbohydrate recognition process, the inhibition of ABA–complex N-glycan (asialo human a1-acid glycoprotein) binding was evaluated. In contrast to the previous reports, the results revealed that the inactive Galb1-related ligands and its cluster features of multi-antennary IIb (Galb1-4GlcNAcb1-) in N-glycans also play a vital role in binding. Thus, three recognition systems of four different intensities of mammalian structural units (glycotopes) involved in ABA–glycan interactions at the same combining site or two nearby sites within the same domain are defined. The four different structural units are: Galb1-3GalNAca1- glycotope (Ta, most active one), GalNAca1-Ser/Thr (Tn), Galb1-3GlcNAcb1- (Ib) and * Corresponding author. Fax: +886 3 211 8456. E-mail address: [email protected] (A.M. Wu).
Galb1-4GlcNAcb1- (IIb); the three recognition systems are – Ta for O-glycan, Tn for O-glycans and Ib/IIb for O/N-glycans and Pneumococcus type 14 capsular polysaccharides; the order of potency of glycotopes is Ta > Tn, Ib IIb; polyvalency is essential for all interactions. This hypothesis provides the most comprehensive concept to explain the ability of ABA to recognize target glycans and malignant cells at the molecular level [5–7]. Furthermore, through this study, the current structural concept of the combining sites of lectins has also been expanded. It is expected that this approach can also be applied to many other lectins. 2. Materials and methods 2.1. Lectin preparation and biotinylation Agaricus bisporus agglutinin (ABA) was purchased from Sigma Chemical Co., (St. Louis, MO, USA). Lectin biotinylation was performed as described previously [8,9]. 2.2. Glycoproteins and polysaccharides The blood group precursors and Smith-degraded products (e.g., cyst Mcdon P-1, Tighe P-1 and cyst JS 1st Smith degraded) were purified and prepared from human ovarian cyst fluid [10–12]. Human/bovine a1-acid gps, and fetuin, which are multi-antennary IIb containing N-glycans [13], were purchased from Sigma Chemical Co. (St. Louis).
0014-5793/$36.00 Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2010.07.021
3562
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Tn and sialyl Tn-containing bovine, porcine, and ovine submandibular mucins (BSM, PSM and OSM), were purified according to the procedure of Tettamanti and Pigman [14] and its modifications [14–16]. Desialylation of sialoglycoproteins was performed by mild acid hydrolysis in 0.01 N HCl at 80 °C for 90 min and followed by dialysis against distilled H2O for 2 days to remove small fragments [14,15]. Pneumococcus type XIV and Gala1?4Gal containing Okra polysaccharides were prepared as described previously [17,18]. Mannan isolated from yeast, Saccharomyces cerevisiae, which consists of complex DMan with a1,6 linkage on backbone and a1,2/a1,3 linkages on side chains [19,20]. Poly-2,8-N-acetylneuraminic acid capsular polysaccharide (colominic acid) from Escherichia coli [21] were purchased from Sigma Chemical Co. (St. Louis). 2.3. Monosaccharides and oligosaccharides Mono-, di- and oligosaccharides were either purchased from Sigma Chem. Co. (USA) or prepared by Dextra (Reading, Berkshire, UK).
assay [8,9]. The inhibitory activity was estimated from the inhibition curve and is expressed as the amount of inhibitor (ng per well) giving 50% inhibition of the control ABA binding. 3. Results 3.1. Interaction of ABA with diverse glycoproteins and polysaccharides The avidities of ABA to various glycoproteins and polysaccharides were analyzed by a microtiter plate enzyme-linked lectinosorbent assay (ELLSA). The interaction intensities are summarized in Table 1. Among the polyvalent glycotopes tested, ABA reacted strongly with Ta/Tn-containing O-glycans (#12 to #16), Ib or IIbcontaining O-glycans (#7 to #9, #11), IIb-containing N-glycans (#1 to #6) and with Pneumococcus type 14 ps of repeating unit of IIb (#17), but not other polysaccharides (#18 to #20). These results indicate that polyvalent IIb structural units in complex N-glycans and in high density polyforms (Pneumococcus type 14 capsular ps, #17) are ascertainably one of the recognition factors in ABA recognition process.
2.4. Lectin–enzyme binding assays (ELLSA)
3.2. Inhibition of ABA – IIb-containing N-glycan (asialo human a1-acid gp) recognition by N-glycans, O-glycans and other polysaccharides
The assays were performed according to the procedures described by Duk et al. [8]. The volume of each reagent solution applied to wells of the plate was 50 ll/well, and all incubations, except for coating, were performed at 20 °C. The reagents, if not indicated otherwise, were diluted with TBS containing 0.05% Tween 20 (TBS-T) and TBS-T was used to wash plates between incubations. For inhibition studies, serially diluted inhibitor samples were mixed with fixed amount of biotinylated ABA. The control lectin sample was diluted twofold with TBS-T. After 30 min incubation at 20 °C, the samples were tested by using the binding
In order to confirm that ABA binding to complex N-glycans is glyco-specific, the ABA interaction with asialo human a1-acid gp (N-glycan) was inhibited by various N-gycans, O-glycans and other glycans and expressed as 50% mass (nanogram) inhibition. As shown in Table 2, all O-glycans (#7 to #14) were very active ligands and up to 4.1 105-fold as active as GalNAc (#15) and all N-glycans (#1 to #6) were also active ligands and up to 2.0 104-fold as active as GalNAc. These results suggest that polyvalent Galb1?3/4GlcNAc (Ib/IIb) can be assigned as one of the ABA recognition systems.
Table 1 Binding of ABA to human blood group active, sialo-, asialo-glycoproteins and polysaccharides by ELLSA.a Sample No. (#)
N-Glycans 1 2 3 4 5 6 O-Glycans 7 8 9 10 11 12 13 14 15 16 Polysaccharides 17 18 19 20 a
Glycoprotein (lectin determinantsb; blood group specificity)
1.5 (A405) unit (ng)
Maximum A405 absorbance Absorbance reading
Binding intensityc
Asialo bovine a1-acid gp (IIb) Asialo fetuin (IIb > Ib, Ta) Fetuin (sialyl IIb, Ta) Asialo human a1-acid gp (IIb) Bovine a1-acid gp (sialyl IIb) Human a1-acid gp (sialyl IIb)
2.5 3.0 4.5 7.0 26.0 200.0
4.4 2.7d 3.0d 3.0d 2.3 2.3d
5+ 5+ 5+ 5+ 4+ 4+
Cyst Mcdon P-1 (Ib, IIb, Ta, Tn) Cyst Tighe P-1 (Ib, IIb, Ta, Tn) Cyst Beach P-1 (Ib, IIb, Ta, Tn) Asialo BSM (Tn, GlcNAcb1?3Tn) Cyst JS 1st Smith degraded (Ib, IIb, Ta, Tn) BSM (sialyl Tn, GlcNAcb1?3Tn) Asialo PSM (Ta, Tn, Ah, H) Asialo OSM (Tn) PSM (sialy Ta, Tn) OSM (sialyl Tn)
0.3 0.4 0.7 0.8 1.0 10.0 11.0 17.0 20.0 50.0
4.4 3.0d 3.0d 4.0 3.0d 3.6 3.0d 3.0d 1.7d 2.0d
5+ 5+ 5+ 5+ 5+ 5+ 5+ 5+ 3+ 4+
Pneumococcus type 14 ps (repeating units of IIb) Mannane Okrae (poly Gala1-4Gal) Colominic acide (poly Neu5Aca2-8Neu5Ac)
110.0
2.2d 0.1d 0.1
4+
d
10 ng of biotinylated lectin were added to various gps, ranging from 0.016 ng to 5 lg. The bolded symbols in parentheses indicates the human blood group activity and/or lectin determinants [23–26]: A (GalNAca1-3Gal); Ah (GalNAca1-3[LFuca1-2]Gal); H (LFuca1-2Gal); Ta (Galb1-3GalNAca1-); Tn (GalNAca1-Ser/Thr); Ib/IIb (Galb1-3/4GlcNAcb1-). c The results were interpreted according to the measured A405 after 2 h incubation as follows: 5+ (O.D. P 2.5), 4+ (2.5 > O.D. P 2.0), 3+ (2.0 > O.D. P 1.5), 2+ (1.5 > O.D. P 1.0), + (1.0 > O.D. P 0.5), ± (0.5 > O.D. P 0.2), and (O.D. < 0.2). d According to [5]. e Negative control. b
3563
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Table 2 Amount of various glycoproteins and saccharides giving 50% inhibition of ABA (10 ng/50 ll) binding to a IIb-containing complex N-glycan (asialo human a1-acid gp, 20 ng/50 ll).a Sample No. (#)
Inhibitorb
N-Glycans 1 Asialo bovine a1-acid gp (IIb) 2 Asialo fetuin (IIb > Ib, Ta) 3 Asialo human a1-acid gp (IIb) 4 Bovine a1-acid gp (sialy IIb) 5 Fetuin (sialy IIb, Ta) 6 Human a1-acid gp (sialy IIb) O-Glycans 7 Asialo BSM (sialyl Tn, GlcNAcb1?3Tn) 8 Cyst Mcdon P-1 (Ta, Tn, Ib, IIb) 9 Asialo PSM (Ta, Tn, Ah, H) 10 Cyst JS 1st Smith degraded (Ib, IIb, Ta, Tn) 11 Asialo OSM (Tn) 12 BSM (sialyl Tn, GlcNAcb1?3Tn) 13 PSM (sialyl Ta, Tn) 14 OSM (sialyl Tn) Monosaccharides and polysaccharides 15 GalNAc 16 Gal 17 GlcNAc 18 Man 19 Mannan
Quantity giving 50% inhibition (ng)
Relative potencyc
42.0 46.5 137.0 340.0 440.0 840.0
2.0 104 1.7 104 5.9 103 2.3 103 1.8 103 9.6 102
2.0 2.0 2.5 5.0 10.0 16.5 36.0 175.0
4.1 105 4.1 105 3.2 105 1.6 105 8.0 104 4.8 104 2.3 104 4.6 103
8.0 105 >5.0 106 >3.8 106 >5.0 106 >2.8 103
(4.4% inhibition) (8.4% inhibition) (13.0% inhibition) (3.5% inhibition)
1.0 – – – –
a
The inhibitory activity is expressed as the amount of inhibitor (nanogram) giving 50% inhibition. Total volume is 50 ll. The symbol in parentheses indicates the human blood group activity and/or lectin determinants [23–26]. Expressed in bold are: A (GalNAca1-3Gal); Ah (GalNAca13[LFuca1-2]Gal); H (LFuca1-2Gal); Ta (Galb1-3GalNAca1-); Tn (GalNAca1-Ser/Thr); Ib/IIb (Galb1-3/4GlcNAcb1-). c Relative potency (RP) = Quantity of GalNAc required for 50% inhibition is taken as 1.0/Quantity of sample required for 50% inhibition. The concentration was 2.8 102 ng for non-reactive polysaccharides (Okra and Colominic acid). b
Table 3 Amount of glycoproteins giving 50% inhibition of ABA (2.5 ng/50 ll) by asialo PSM (5 ng/50 ll).a,b
a b c d e
Sample No.
Inhibitor (multivalent structural units)
Quantity giving 50% inhibition (ng)
Relative potencyc
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Asialo PSM (Ta, Tn, Ah, H) Cyst Tighe P-1 (Ib, IIb, Ta, Tn) Asialo OSM (Tn) PSM (sialyl Ta, Tn) OSM (sialyl Tn) Asialo human a1-acid gp (mIIb) Pneumococcus type 14 ps (IIb) Human a1-acid gp (sialyl mIIb) Galb1-3GalNAc (T) Galb1-3GlcNAc (I) GalNAc Galb1-4Glc (L) GlcNAc Man Glc Gal Mannan Galb1-4GlcNAc (II)
0.07 0.4 3.0 10.0 20.0 22.0 110.0 200.0 3.8 104 1.1 105 3.3 105 5.1 105d 4.0 107d >2.2 106e >3.8 106e >4.0 106e >277.8 (18.3% inhibition) >2.5 105 (5.1%inhibition)
4.7 106 8.3 105 1.1 105 3.3 104 1.7 104 1.5 104 3.0 103 1.7 103 8.7 3.0 1.0 0.6 0.008 <0.15 <0.09 <0.08 – –
Adopted from part of Tables 2 and 3 [5]. The inhibitory activity is expressed as the amount of inhibitor giving 50% inhibition of the control lectin binding. Total volume is 50 ll. Relative potency (RP) = Quantity of GalNAc required for 50% inhibition is taken as 1.0/Quantity of sample required for 50% inhibition. Extrapolation. Maximum amount of inhibitor tested.
3.3. Inhibition of ABA – Ta/Tn-containing O-glycan (asialo PSM) recognition by N-glycans, O-glycans and other polysaccharides ABA – Ta/Tn-containing O-glycan (asialo PSM) inhibition assay was repeated to establish the location of the combining site of ABA for O-glycans and complex N-glycans (polyvalent IIb glycotopes). As summarized in Table 3, Ta/Tn-containing O-glycans (#1, #3 to #5); IIb-containing complex N- and/or O-glycans (#2, #6 and #8) and Pneumococcus type 14 polysaccharides (#7) strongly inhibit the interaction, demonstrating that four active glycotopes – GalNAca1-Ser/Thr (Tn), Galb1-3GalNAca1- (Ta) and Galb1-3/4GlcNAcb1- (Ib/IIb), share the same combining site (Structure 1) or nearby sites.
4. Discussion In human diet, there are many lectins that resist heat and digestion [1]. In colon and faeces, these lectins remain in active form with notable effects on the function of gastrointestinal cells [1,3,4,22]. Agaricus bisporus agglutinin (ABA) isolated from edible mushroom is one of the dietary lectins that possesses considerable therapeutic potential with potent anti-proliferative effects and is an anti-neoplastic agent toward malignant colon cancer cells [1,3,4]. Previous studies on the structural requirements for ABA glyco-recognition were focused on GalNAc related ligands (Ta, Galb1-3GalNAca1-; Tn, GalNAca1-Ser/Thr) in O-glycans and analyzed by enzyme-linked lectinosorbent/inhibition assays [5].
3564
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Structure 1. Four different intensities of mammalian glycotope units (i–iv) constructing three recognition systems (i–iv) for Agaricus bisporus agglutinin–glycan interactions.
It was also found that many complex N-glycans may also be involved in glyco-recognition, while Gal, Man and GlcNAc were weak or inactive ligands. Later, roles of GlcNAc-exposed N-glycans were studied by lectin microarray and frontal affinity chromatography [7] and the existence of another independent carbohydrate-binding site for GlcNAc in glycans was reported [6]. In order to further clarify these observations, natural glycoproteins were used to elu-
cidate the polyvalent effects of weak Galb1?4GlcNAcb (IIb) and weak Galb1?3GlcNAcb (Ib) in N-glycan on the recognition process. To eliminate the possibility of non-specific interactions, the roles of complex N-glycans in ABA–glycan recognition process were further studied by the inhibition approach, in which ABA – complex N-glycan (asialo human a1-acid glycoprotein) rather than O-glycan (asialo PSM) interaction system [5] was used in this study. The
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
results indicate that both O-glycans and many IIb-containing N-glycans tested can inhibit not only ABA–O-glycan interactions, but also ABA–N-glycan interactions. Thus, it is proposed that the combining site of ABA consists of one major site (a potent –OH configuration of GalNAc for Tn and Ta or –OH configuration of a weak GlcNAcb1?for Ib and IIb) and two subsites (Galb1-3/4 at nonreducing end) for accommodating four different intensities of mammalian glycotopes and three recognition systems (Tn, Ta and Galb1-3/4GlcNAcb1-). When the inactive Gal is b1-3 linked to GalNAca1- (Tn) at non-reducing end, the Galb1-3GalNAca1(Ta, the most active glycotope) sequence is formed. GlcNAc is a very poor monosaccharide inhibitor (#13 in Table 3) and different from GalNAc at the –OH configuration at carbon-4. Combination of these two weak or inactive monosaccharides (Gal and GlcNAc) by b1-3/4 linkages as Ib/IIb (Galb1-3GlcNAcb1-/Galb1-4GlcNAcb1-) resulted in a greater than 360 times enhancement of binding (#10, Galb1-3GlcNAcb1- vs. #13, GlcNAc in Table 3). The glycotopes of Ib and IIb (Galb1-3/4GlcNAcb1-) existing in both O- and N-glycans are constructed as the third recognition system. Since the results of both inhibition profiles (Tables 2 and 3) were similar, in which intensity of polyvalent Ta/Tn were more potent than polyvalent Ib/IIb (#7, #9, #11 to #14 vs. #1 to #6 in Table 2; #1 and #3 vs. #6 and #7 in Table 3). These active glycotopes – GalNAca1-Ser/ Thr (Tn), Galb1-3GalNAca1- (Ta) and Galb1-3/4GlcNAcb1- (Ib/IIb) are proposed to share the same combining site through three different recognition systems (Structure 1). Its size should be as large as disaccharide and most complementary to Galb1-3GalNAca1(Ta). If ‘‘two intrinsic sugar-binding sites in one polypeptide; one for Ta and the other for GlcNAc” suggested by Nakamura-Tsuruta [6,7] exist, the inhibitory profiles, as shown in the Tables 2 and 3, would be expected to be different. Therefore, it is proposed that the GalNAc and GlcNAc should share the same major site and also configuration of carbons-2, -3 and -5. Otherwise, they must be located at nearby sites. During the past three decades, many synthetic glycans or ligands have been used to study the binding properties of lectins or antibodies, especially in the worldwide consortium systems. However, these interactions are not in existence in nature. In this study, all the polyvalent glycoforms used in this study were from natural sources, which clearly demonstrated the theme of our study (Tables 2 and 3). The combining site of antibodies has been proposed to be divided into only two shapes: the cavity and the groove type by Kabat [27]. Based on our three decades of study, the combining sites of most lectins appear to be more complicated than Kabat proposed. Recently, we found that different intensities of four mammalian structural units constructing four recognition systems at two different combining sites for the process of Ralstonia solanacearum lectin–glycan interactions (unpublished data) [29] in which the inhibitory profiles are very different. In addition to our current report on ‘‘duality of the carbohydrate-recognition system of Pseudomonas aeruginosa-II lectin (PA-IIL)” sharing the same configurations of epimeric –OH at carbon -2, and -4 of LFuc for both LFuc-containing glycans and mannan recognition [28]. Now, at least three types – cavity, groove and hybrid types (cavity as major site and groove as subsite) are frequently found in plant lectins. Better systems are being explored and organized in our laboratory for plant, microbial and animal lectins. From the results of this study, the Kabat’s structural concept of the combining site of lectins has been expanded and upgraded [28]. It is expected that this approach and concept can be applied to other lectin recognition systems. Combining previous and current results [5–7], it is proposed that: (a) the combining site of ABA is composed of one major site a potent –OH configuration of GalNAc and a weak –OH configuration of GlcNAc i.e. GalNAc GlcNAc and two subsites (Galb1-3/4) at non-reducing end for fitting four different intensities of glyco-
3565
topes and three recognition systems (Tn, Ta and Galb1-3/4GlcNAc); (b) the order of potency of glycotope units is Ta > Tn, Ib > IIb; (c) polyvalency is essential for all interactions; (d) the order of the intensities of polyvalency enhancement, opposing to that of (b), is approximately IIb > Ib > Tn Ta; (e) the combining size of ABA should be as large as disaccharide and most complementary to Ta; (f) the sites for all glycotope units are located at the same site or nearby sites as an independent GlcNAc subsite has been reported [6,7]. Thus, an additional recognition factor responsible for the polyvalent effect of a poor Galb1?4GlcNAcb (IIb) and a weak Galb1?3GlcNAcb (Ib) on N-glycan–ABA interactions was established. The peculiar concept of four different intensities of glycotope units constructing three recognition systems provides an advanced knowledge about the combining sites of ABA, which can be used to explain the ABA attachment process to target glycans or malignant cells at the molecular level. Acknowledgements This work was supported by Grants from the Chang-Gung Medical Research Project (CMRP No. 180481 and 170442), Kwei-san, Tao-yuan, Taiwan, the National Science Council (NSC 97-2628-B182-002-MY3, and 97-2320-B-182-020-MY3), Taipei, Taiwan. References [1] Milton, J.D. and Rhodes, J.M. (1995) Dietary galactose-binding lectins and their effects on human colonic epithelial cells in: Lectin: Biomedical Perspectives (Pusztai, A. and Bardocz, S., Eds.), pp. 225–234, Taylor & Francis, London. [2] Yu, L.G., Fernig, D.G., Smith, J.A., Milton, J.D. and Rhodes, J.M. (1993) Reversible inhibition of proliferation of epithelial cell lines by Agaricus bisporus (edible mushroom) lectin. Cancer Res. 53, 4627–4632. [3] Yu, L.G., Fernig, D.G., White, M.R., Spiller, D.G., Appleton, P., Evans, R.C., Grierson, I., Smith, J.A., Davies, H., Gerasimenko, O.V., Petersen, O.H., Milton, J.D. and Rhodes, J.M. (1999) Edible mushroom (Agaricus bisporus) lectin which reversibly inhibits epithelial cell proliferation blocks nuclear localization sequence-dependent nuclear protein import. J. Biol. Chem. 274, 4890–4899. [4] Sueyoshi, S., Tsuji, T. and Osawa, T. (1985) Purification and characterization of four isolectins of mushroom (Agaricus bisporus). Biol. Chem. Hoppe-Seyler 366, 213–221. [5] Wu, A.M., Wu, J.H., Herp, A. and Liu, J.H. (2003) Effect of polyvalencies of glycotopes on the binding of a lectin from the edible mushroom, Agaricus bisporus. Biochem. J. 371, 311–320. [6] Carrizo, M.E., Capaldi, S., Perduca, M., Irazoqui, F.J., Nores, G.A. and Monaco, H.L. (2005) The antineoplastic lectin of the common edible mushroom (Agaricus bisporus) has two binding sites, each specific for a different configuration at a single epimeric hydroxyl. J. Biol. Chem. 280, 10614–10623. [7] Nakamura-Tsuruta, S., Kominami, J., Kuno, A. and Hirabayashi, J. (2006) Evidence that Agaricus bisporus agglutinin (ABA) has dual sugar-binding specificity. Biochem. Biophys. Res. Commun. 347, 215–220. [8] Duk, M., Lisowska, E., Wu, J.H. and Wu, A.M. (1994) The biotin/avidinmediated microtiter plate lectin assay with the use of chemically modified glycoprotein ligand. Anal. Biochem. 221, 266–272. [9] Lisowska, E., Duk, M. and Wu, A.M. (1996) Preparation of biotinylated lectins and application in microtiter plate assays and western blotting in the biomethod-series of birkhäuser publishers. Basel, Boston, Berlin 7, 115–128. [10] Kabat, E.A. (1956) Blood Group Substances, Their Chemistry and Immunochemistry, Academic Press, New York. pp. 135–139. [11] Wu, A.M., Kabat, E.A., Nilsson, B., Zopf, D.A., Gruezo, F.G. and Liao, J. (1984) Immunochemical studies on blood groups. Purification and characterization of radioactive 3H-reduced di- to hexasaccharides produced by alkaline betaelimination- borohydride 3H reduction of Smith degraded blood group A active glycoproteins. J. Biol. Chem. 259, 7178–7186. [12] Wu, A.M. (1988) Structural concepts of the human blood group A, B, H, Lea, Leb, I, and I active glycoprotein purified from human ovarian cyst fluid. Adv. Exp. Med. Biol. 228, 351–394. [13] Wu, A.M. and Sugii, S.J. (1988) Differential binding properties of Ga1NAc and/ or Ga1 specific lectins. Adv. Exp. Med. Biol. 228, 205–263. [14] Tettamanti, G. and Pigman, W. (1968) Purification and characterization of bovine and ovine submaxillary mucins. Arch. Biochem. Biophys. 124, 41–50. [15] Wu, A.M. and Pigman, W. (1977) Preparation and characterization of armadillo submandibular glycoproteins. Biochem. J. 161, 37–47. [16] Herp, A., Borelli, C. and Wu, A.M. (1988) Biochemistry and lectin binding properties of mammalian salivary mucous glycoproteins. Adv. Exp. Med. Biol. 228, 395–435. [17] Lindberg, B., Lonngren, J. and Powell, D.A. (1977) Structural studies on the specific type-14 pneumococcal polysaccharide. Carbohydr. Res. 58, 177–186.
3566
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
[18] Whistler, R.L. and Conrad, H.E. (1954) Crystalline galactobiose from acid hydrolysis of okra mucilage; 2-O-(D-galactopyranosyluronic acid)-L-rhamnose from okra mucilage. J. Am. Chem. Soc. 76, 1673–1674. [19] Yoshida, Y., Naito, E., Mizukoshi, H., Watanabe, Y., Kimura, K., Yokoi, W., Sato, T., Okumura, T., Ito, M. and Sawada, H.J. (2009) Side-chain structure of cell surface polysaccharide, mannan, affects hypocholesterolemic activity of yeast. J. Agric. Food Chem. 57, 8003–8009. [20] Gemmill, T.R. and Trimble, R.B. (1999) Review: overview of N- and O-linked oligosaccharide structures found in various yeast species. Biochim. Biophys. Acta 1426, 227–237. [21] Kwiatowski, B. and Stirm, S. (1987) Polysialic acid depolymerase. Meth. Enzymol. 138, 786–792. [22] Ryder, S.D., Smith, J.A., Rhodes, E.G., Parker, N. and Rhodes, J.M. (1994) Proliferative responses of HT29 and Caco2 human colorectal cancer cells to a panel of lectins. Gastroenterology 106, 85–93. [23] Wu, A.M. (2001) Expression of binding properties of Gal/GalNAc reactive lectins by mammalian glycotopes. Adv. Exp. Med. Biol. 491, 55–64. [24] Wu, A.M. (2002) Carbohydrate structural units in glycosphingolipids as receptors for Gal and GalNAc reactive lectins. Neurochem. Res. 27, 593–600.
[25] Wu, A.M. (2003) Carbohydrate structural units in glycoproteins and polysaccharides as important ligands for Gal and GalNAc reactive lectins. J. Biomed. Sci. 10, 676–688. [26] Yang, Z., Tsai, M.S., Wu, J.H., Herp, A. and Wu, A.M. (2008) Defining the carbohydrate specificities of Erythrina corallodendron lectin (ECorL) as polyvalent Galb1–4GlcNAc (II) > monomeric II > monomeric Gal and GalNAc. Chang Gung Med. J. 31, 26–43. [27] Kabat, E.A. (1988) Antibody combining sites: how much of the antibody repertoire are we seeing? How does it influence our understanding of the structural and genetic basis of antibody complementarity? Adv. Exp. Med. Biol. 228, 1–45. [28] Wu, A.M., Gong, Y.P., Li, C.C. and Gilboa-Garber, N. (2010) Duality of the carbohydrate-recognition system of Pseudomonas aeruginosa-II lectin (PA-IIL). FEBS Lett. 584, 2371–2375. [29] Kostlánová, N., Mitchell, E.P., Lortat-Jacob, H., Oscarson, S., Lahmann, M., Gilboa-Garber, N., Chambat, G., Wimmerová, M. and Imberty, A. (2005) The fucose-binding lectin from Ralstonia solanacearum: A new type of b-properller architecture formed by oligomerization and interacting with fucoside, fuccosyllactose, and plant xyloglucan. J. Biol. Chem. 280, 27839–27849.
journal homepage: www.FEBSLetters.org
Multiple recognition systems adopting four different glycotopes at the same domain for the Agaricus bisporus agglutinin–glycan interactions Albert M. Wu *, Jia-Hau Liu, Yu-Ping Gong, Chia-Chen Li, En-Tzu Chang Glyco-immunochemistry Research Laboratory, Institute of Molecular and Cellular Biology, College of Medicine, Chang-Gung University, Kwei-san, Tao-yuan 333, Taiwan
a r t i c l e
i n f o
Article history: Received 13 May 2010 Revised 13 July 2010 Accepted 13 July 2010 Available online 17 July 2010 Edited by Sandro Sonnino Keywords: Carbohydrate specificity Combining site Complex N-glycan Lectin Agaricus bisporus agglutinin
a b s t r a c t For the GalNAca1? specific Agaricus bisporus agglutinin (ABA) from an edible mushroom, the mechanism of polyvalent Galb1?3/4GlcNAcb1? complex in ABA–carbohydrate recognition has not been well defined since Gal and GlcNAc are weak ligands. By enzyme-linked lectinosorbent and inhibition assays, we show that the polyvalent Galb1?3/4GlcNAcb1? in natural glycans also play vital roles in binding and we propose that four different intensities of glycotopes (Galb1-3GalNAca1-, GalNAca1Ser/Thr and Galb1-3/4GlcNAcb1-) construct three recognition systems at the same domain. This peculiar concept provides the most comprehensive mechanism for the attachment of ABA to target glycans and malignant cells at the molecular level. Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction The Agaricus bisporus agglutinin (ABA) in edible mushroom has been found to have a potent anti-proliferative effect on malignant colon cells with considerable therapeutic potential as an antineoplastic agent [1–3]. This lectin has a molecular mass of 6.4 104 Da and consists of four apparently identical subunits (Mr. 16 000) [4]. Previous studies on the structural requirements for glyco-recognition of ABA were focused on GalNAca1-related ligands at reducing ends of O-glycans [5]. It was also found that multi-complex N-glycans may be involved in glyco-recognition, while Gal, Man and GlcNAc were weak or inactive binders. To further clarify the mechanisms of multi-complex N-glycans in ABA– carbohydrate recognition process, the inhibition of ABA–complex N-glycan (asialo human a1-acid glycoprotein) binding was evaluated. In contrast to the previous reports, the results revealed that the inactive Galb1-related ligands and its cluster features of multi-antennary IIb (Galb1-4GlcNAcb1-) in N-glycans also play a vital role in binding. Thus, three recognition systems of four different intensities of mammalian structural units (glycotopes) involved in ABA–glycan interactions at the same combining site or two nearby sites within the same domain are defined. The four different structural units are: Galb1-3GalNAca1- glycotope (Ta, most active one), GalNAca1-Ser/Thr (Tn), Galb1-3GlcNAcb1- (Ib) and * Corresponding author. Fax: +886 3 211 8456. E-mail address: [email protected] (A.M. Wu).
Galb1-4GlcNAcb1- (IIb); the three recognition systems are – Ta for O-glycan, Tn for O-glycans and Ib/IIb for O/N-glycans and Pneumococcus type 14 capsular polysaccharides; the order of potency of glycotopes is Ta > Tn, Ib IIb; polyvalency is essential for all interactions. This hypothesis provides the most comprehensive concept to explain the ability of ABA to recognize target glycans and malignant cells at the molecular level [5–7]. Furthermore, through this study, the current structural concept of the combining sites of lectins has also been expanded. It is expected that this approach can also be applied to many other lectins. 2. Materials and methods 2.1. Lectin preparation and biotinylation Agaricus bisporus agglutinin (ABA) was purchased from Sigma Chemical Co., (St. Louis, MO, USA). Lectin biotinylation was performed as described previously [8,9]. 2.2. Glycoproteins and polysaccharides The blood group precursors and Smith-degraded products (e.g., cyst Mcdon P-1, Tighe P-1 and cyst JS 1st Smith degraded) were purified and prepared from human ovarian cyst fluid [10–12]. Human/bovine a1-acid gps, and fetuin, which are multi-antennary IIb containing N-glycans [13], were purchased from Sigma Chemical Co. (St. Louis).
0014-5793/$36.00 Ó 2010 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2010.07.021
3562
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Tn and sialyl Tn-containing bovine, porcine, and ovine submandibular mucins (BSM, PSM and OSM), were purified according to the procedure of Tettamanti and Pigman [14] and its modifications [14–16]. Desialylation of sialoglycoproteins was performed by mild acid hydrolysis in 0.01 N HCl at 80 °C for 90 min and followed by dialysis against distilled H2O for 2 days to remove small fragments [14,15]. Pneumococcus type XIV and Gala1?4Gal containing Okra polysaccharides were prepared as described previously [17,18]. Mannan isolated from yeast, Saccharomyces cerevisiae, which consists of complex DMan with a1,6 linkage on backbone and a1,2/a1,3 linkages on side chains [19,20]. Poly-2,8-N-acetylneuraminic acid capsular polysaccharide (colominic acid) from Escherichia coli [21] were purchased from Sigma Chemical Co. (St. Louis). 2.3. Monosaccharides and oligosaccharides Mono-, di- and oligosaccharides were either purchased from Sigma Chem. Co. (USA) or prepared by Dextra (Reading, Berkshire, UK).
assay [8,9]. The inhibitory activity was estimated from the inhibition curve and is expressed as the amount of inhibitor (ng per well) giving 50% inhibition of the control ABA binding. 3. Results 3.1. Interaction of ABA with diverse glycoproteins and polysaccharides The avidities of ABA to various glycoproteins and polysaccharides were analyzed by a microtiter plate enzyme-linked lectinosorbent assay (ELLSA). The interaction intensities are summarized in Table 1. Among the polyvalent glycotopes tested, ABA reacted strongly with Ta/Tn-containing O-glycans (#12 to #16), Ib or IIbcontaining O-glycans (#7 to #9, #11), IIb-containing N-glycans (#1 to #6) and with Pneumococcus type 14 ps of repeating unit of IIb (#17), but not other polysaccharides (#18 to #20). These results indicate that polyvalent IIb structural units in complex N-glycans and in high density polyforms (Pneumococcus type 14 capsular ps, #17) are ascertainably one of the recognition factors in ABA recognition process.
2.4. Lectin–enzyme binding assays (ELLSA)
3.2. Inhibition of ABA – IIb-containing N-glycan (asialo human a1-acid gp) recognition by N-glycans, O-glycans and other polysaccharides
The assays were performed according to the procedures described by Duk et al. [8]. The volume of each reagent solution applied to wells of the plate was 50 ll/well, and all incubations, except for coating, were performed at 20 °C. The reagents, if not indicated otherwise, were diluted with TBS containing 0.05% Tween 20 (TBS-T) and TBS-T was used to wash plates between incubations. For inhibition studies, serially diluted inhibitor samples were mixed with fixed amount of biotinylated ABA. The control lectin sample was diluted twofold with TBS-T. After 30 min incubation at 20 °C, the samples were tested by using the binding
In order to confirm that ABA binding to complex N-glycans is glyco-specific, the ABA interaction with asialo human a1-acid gp (N-glycan) was inhibited by various N-gycans, O-glycans and other glycans and expressed as 50% mass (nanogram) inhibition. As shown in Table 2, all O-glycans (#7 to #14) were very active ligands and up to 4.1 105-fold as active as GalNAc (#15) and all N-glycans (#1 to #6) were also active ligands and up to 2.0 104-fold as active as GalNAc. These results suggest that polyvalent Galb1?3/4GlcNAc (Ib/IIb) can be assigned as one of the ABA recognition systems.
Table 1 Binding of ABA to human blood group active, sialo-, asialo-glycoproteins and polysaccharides by ELLSA.a Sample No. (#)
N-Glycans 1 2 3 4 5 6 O-Glycans 7 8 9 10 11 12 13 14 15 16 Polysaccharides 17 18 19 20 a
Glycoprotein (lectin determinantsb; blood group specificity)
1.5 (A405) unit (ng)
Maximum A405 absorbance Absorbance reading
Binding intensityc
Asialo bovine a1-acid gp (IIb) Asialo fetuin (IIb > Ib, Ta) Fetuin (sialyl IIb, Ta) Asialo human a1-acid gp (IIb) Bovine a1-acid gp (sialyl IIb) Human a1-acid gp (sialyl IIb)
2.5 3.0 4.5 7.0 26.0 200.0
4.4 2.7d 3.0d 3.0d 2.3 2.3d
5+ 5+ 5+ 5+ 4+ 4+
Cyst Mcdon P-1 (Ib, IIb, Ta, Tn) Cyst Tighe P-1 (Ib, IIb, Ta, Tn) Cyst Beach P-1 (Ib, IIb, Ta, Tn) Asialo BSM (Tn, GlcNAcb1?3Tn) Cyst JS 1st Smith degraded (Ib, IIb, Ta, Tn) BSM (sialyl Tn, GlcNAcb1?3Tn) Asialo PSM (Ta, Tn, Ah, H) Asialo OSM (Tn) PSM (sialy Ta, Tn) OSM (sialyl Tn)
0.3 0.4 0.7 0.8 1.0 10.0 11.0 17.0 20.0 50.0
4.4 3.0d 3.0d 4.0 3.0d 3.6 3.0d 3.0d 1.7d 2.0d
5+ 5+ 5+ 5+ 5+ 5+ 5+ 5+ 3+ 4+
Pneumococcus type 14 ps (repeating units of IIb) Mannane Okrae (poly Gala1-4Gal) Colominic acide (poly Neu5Aca2-8Neu5Ac)
110.0
2.2d 0.1d 0.1
4+
d
10 ng of biotinylated lectin were added to various gps, ranging from 0.016 ng to 5 lg. The bolded symbols in parentheses indicates the human blood group activity and/or lectin determinants [23–26]: A (GalNAca1-3Gal); Ah (GalNAca1-3[LFuca1-2]Gal); H (LFuca1-2Gal); Ta (Galb1-3GalNAca1-); Tn (GalNAca1-Ser/Thr); Ib/IIb (Galb1-3/4GlcNAcb1-). c The results were interpreted according to the measured A405 after 2 h incubation as follows: 5+ (O.D. P 2.5), 4+ (2.5 > O.D. P 2.0), 3+ (2.0 > O.D. P 1.5), 2+ (1.5 > O.D. P 1.0), + (1.0 > O.D. P 0.5), ± (0.5 > O.D. P 0.2), and (O.D. < 0.2). d According to [5]. e Negative control. b
3563
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Table 2 Amount of various glycoproteins and saccharides giving 50% inhibition of ABA (10 ng/50 ll) binding to a IIb-containing complex N-glycan (asialo human a1-acid gp, 20 ng/50 ll).a Sample No. (#)
Inhibitorb
N-Glycans 1 Asialo bovine a1-acid gp (IIb) 2 Asialo fetuin (IIb > Ib, Ta) 3 Asialo human a1-acid gp (IIb) 4 Bovine a1-acid gp (sialy IIb) 5 Fetuin (sialy IIb, Ta) 6 Human a1-acid gp (sialy IIb) O-Glycans 7 Asialo BSM (sialyl Tn, GlcNAcb1?3Tn) 8 Cyst Mcdon P-1 (Ta, Tn, Ib, IIb) 9 Asialo PSM (Ta, Tn, Ah, H) 10 Cyst JS 1st Smith degraded (Ib, IIb, Ta, Tn) 11 Asialo OSM (Tn) 12 BSM (sialyl Tn, GlcNAcb1?3Tn) 13 PSM (sialyl Ta, Tn) 14 OSM (sialyl Tn) Monosaccharides and polysaccharides 15 GalNAc 16 Gal 17 GlcNAc 18 Man 19 Mannan
Quantity giving 50% inhibition (ng)
Relative potencyc
42.0 46.5 137.0 340.0 440.0 840.0
2.0 104 1.7 104 5.9 103 2.3 103 1.8 103 9.6 102
2.0 2.0 2.5 5.0 10.0 16.5 36.0 175.0
4.1 105 4.1 105 3.2 105 1.6 105 8.0 104 4.8 104 2.3 104 4.6 103
8.0 105 >5.0 106 >3.8 106 >5.0 106 >2.8 103
(4.4% inhibition) (8.4% inhibition) (13.0% inhibition) (3.5% inhibition)
1.0 – – – –
a
The inhibitory activity is expressed as the amount of inhibitor (nanogram) giving 50% inhibition. Total volume is 50 ll. The symbol in parentheses indicates the human blood group activity and/or lectin determinants [23–26]. Expressed in bold are: A (GalNAca1-3Gal); Ah (GalNAca13[LFuca1-2]Gal); H (LFuca1-2Gal); Ta (Galb1-3GalNAca1-); Tn (GalNAca1-Ser/Thr); Ib/IIb (Galb1-3/4GlcNAcb1-). c Relative potency (RP) = Quantity of GalNAc required for 50% inhibition is taken as 1.0/Quantity of sample required for 50% inhibition. The concentration was 2.8 102 ng for non-reactive polysaccharides (Okra and Colominic acid). b
Table 3 Amount of glycoproteins giving 50% inhibition of ABA (2.5 ng/50 ll) by asialo PSM (5 ng/50 ll).a,b
a b c d e
Sample No.
Inhibitor (multivalent structural units)
Quantity giving 50% inhibition (ng)
Relative potencyc
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Asialo PSM (Ta, Tn, Ah, H) Cyst Tighe P-1 (Ib, IIb, Ta, Tn) Asialo OSM (Tn) PSM (sialyl Ta, Tn) OSM (sialyl Tn) Asialo human a1-acid gp (mIIb) Pneumococcus type 14 ps (IIb) Human a1-acid gp (sialyl mIIb) Galb1-3GalNAc (T) Galb1-3GlcNAc (I) GalNAc Galb1-4Glc (L) GlcNAc Man Glc Gal Mannan Galb1-4GlcNAc (II)
0.07 0.4 3.0 10.0 20.0 22.0 110.0 200.0 3.8 104 1.1 105 3.3 105 5.1 105d 4.0 107d >2.2 106e >3.8 106e >4.0 106e >277.8 (18.3% inhibition) >2.5 105 (5.1%inhibition)
4.7 106 8.3 105 1.1 105 3.3 104 1.7 104 1.5 104 3.0 103 1.7 103 8.7 3.0 1.0 0.6 0.008 <0.15 <0.09 <0.08 – –
Adopted from part of Tables 2 and 3 [5]. The inhibitory activity is expressed as the amount of inhibitor giving 50% inhibition of the control lectin binding. Total volume is 50 ll. Relative potency (RP) = Quantity of GalNAc required for 50% inhibition is taken as 1.0/Quantity of sample required for 50% inhibition. Extrapolation. Maximum amount of inhibitor tested.
3.3. Inhibition of ABA – Ta/Tn-containing O-glycan (asialo PSM) recognition by N-glycans, O-glycans and other polysaccharides ABA – Ta/Tn-containing O-glycan (asialo PSM) inhibition assay was repeated to establish the location of the combining site of ABA for O-glycans and complex N-glycans (polyvalent IIb glycotopes). As summarized in Table 3, Ta/Tn-containing O-glycans (#1, #3 to #5); IIb-containing complex N- and/or O-glycans (#2, #6 and #8) and Pneumococcus type 14 polysaccharides (#7) strongly inhibit the interaction, demonstrating that four active glycotopes – GalNAca1-Ser/Thr (Tn), Galb1-3GalNAca1- (Ta) and Galb1-3/4GlcNAcb1- (Ib/IIb), share the same combining site (Structure 1) or nearby sites.
4. Discussion In human diet, there are many lectins that resist heat and digestion [1]. In colon and faeces, these lectins remain in active form with notable effects on the function of gastrointestinal cells [1,3,4,22]. Agaricus bisporus agglutinin (ABA) isolated from edible mushroom is one of the dietary lectins that possesses considerable therapeutic potential with potent anti-proliferative effects and is an anti-neoplastic agent toward malignant colon cancer cells [1,3,4]. Previous studies on the structural requirements for ABA glyco-recognition were focused on GalNAc related ligands (Ta, Galb1-3GalNAca1-; Tn, GalNAca1-Ser/Thr) in O-glycans and analyzed by enzyme-linked lectinosorbent/inhibition assays [5].
3564
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
Structure 1. Four different intensities of mammalian glycotope units (i–iv) constructing three recognition systems (i–iv) for Agaricus bisporus agglutinin–glycan interactions.
It was also found that many complex N-glycans may also be involved in glyco-recognition, while Gal, Man and GlcNAc were weak or inactive ligands. Later, roles of GlcNAc-exposed N-glycans were studied by lectin microarray and frontal affinity chromatography [7] and the existence of another independent carbohydrate-binding site for GlcNAc in glycans was reported [6]. In order to further clarify these observations, natural glycoproteins were used to elu-
cidate the polyvalent effects of weak Galb1?4GlcNAcb (IIb) and weak Galb1?3GlcNAcb (Ib) in N-glycan on the recognition process. To eliminate the possibility of non-specific interactions, the roles of complex N-glycans in ABA–glycan recognition process were further studied by the inhibition approach, in which ABA – complex N-glycan (asialo human a1-acid glycoprotein) rather than O-glycan (asialo PSM) interaction system [5] was used in this study. The
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
results indicate that both O-glycans and many IIb-containing N-glycans tested can inhibit not only ABA–O-glycan interactions, but also ABA–N-glycan interactions. Thus, it is proposed that the combining site of ABA consists of one major site (a potent –OH configuration of GalNAc for Tn and Ta or –OH configuration of a weak GlcNAcb1?for Ib and IIb) and two subsites (Galb1-3/4 at nonreducing end) for accommodating four different intensities of mammalian glycotopes and three recognition systems (Tn, Ta and Galb1-3/4GlcNAcb1-). When the inactive Gal is b1-3 linked to GalNAca1- (Tn) at non-reducing end, the Galb1-3GalNAca1(Ta, the most active glycotope) sequence is formed. GlcNAc is a very poor monosaccharide inhibitor (#13 in Table 3) and different from GalNAc at the –OH configuration at carbon-4. Combination of these two weak or inactive monosaccharides (Gal and GlcNAc) by b1-3/4 linkages as Ib/IIb (Galb1-3GlcNAcb1-/Galb1-4GlcNAcb1-) resulted in a greater than 360 times enhancement of binding (#10, Galb1-3GlcNAcb1- vs. #13, GlcNAc in Table 3). The glycotopes of Ib and IIb (Galb1-3/4GlcNAcb1-) existing in both O- and N-glycans are constructed as the third recognition system. Since the results of both inhibition profiles (Tables 2 and 3) were similar, in which intensity of polyvalent Ta/Tn were more potent than polyvalent Ib/IIb (#7, #9, #11 to #14 vs. #1 to #6 in Table 2; #1 and #3 vs. #6 and #7 in Table 3). These active glycotopes – GalNAca1-Ser/ Thr (Tn), Galb1-3GalNAca1- (Ta) and Galb1-3/4GlcNAcb1- (Ib/IIb) are proposed to share the same combining site through three different recognition systems (Structure 1). Its size should be as large as disaccharide and most complementary to Galb1-3GalNAca1(Ta). If ‘‘two intrinsic sugar-binding sites in one polypeptide; one for Ta and the other for GlcNAc” suggested by Nakamura-Tsuruta [6,7] exist, the inhibitory profiles, as shown in the Tables 2 and 3, would be expected to be different. Therefore, it is proposed that the GalNAc and GlcNAc should share the same major site and also configuration of carbons-2, -3 and -5. Otherwise, they must be located at nearby sites. During the past three decades, many synthetic glycans or ligands have been used to study the binding properties of lectins or antibodies, especially in the worldwide consortium systems. However, these interactions are not in existence in nature. In this study, all the polyvalent glycoforms used in this study were from natural sources, which clearly demonstrated the theme of our study (Tables 2 and 3). The combining site of antibodies has been proposed to be divided into only two shapes: the cavity and the groove type by Kabat [27]. Based on our three decades of study, the combining sites of most lectins appear to be more complicated than Kabat proposed. Recently, we found that different intensities of four mammalian structural units constructing four recognition systems at two different combining sites for the process of Ralstonia solanacearum lectin–glycan interactions (unpublished data) [29] in which the inhibitory profiles are very different. In addition to our current report on ‘‘duality of the carbohydrate-recognition system of Pseudomonas aeruginosa-II lectin (PA-IIL)” sharing the same configurations of epimeric –OH at carbon -2, and -4 of LFuc for both LFuc-containing glycans and mannan recognition [28]. Now, at least three types – cavity, groove and hybrid types (cavity as major site and groove as subsite) are frequently found in plant lectins. Better systems are being explored and organized in our laboratory for plant, microbial and animal lectins. From the results of this study, the Kabat’s structural concept of the combining site of lectins has been expanded and upgraded [28]. It is expected that this approach and concept can be applied to other lectin recognition systems. Combining previous and current results [5–7], it is proposed that: (a) the combining site of ABA is composed of one major site a potent –OH configuration of GalNAc and a weak –OH configuration of GlcNAc i.e. GalNAc GlcNAc and two subsites (Galb1-3/4) at non-reducing end for fitting four different intensities of glyco-
3565
topes and three recognition systems (Tn, Ta and Galb1-3/4GlcNAc); (b) the order of potency of glycotope units is Ta > Tn, Ib > IIb; (c) polyvalency is essential for all interactions; (d) the order of the intensities of polyvalency enhancement, opposing to that of (b), is approximately IIb > Ib > Tn Ta; (e) the combining size of ABA should be as large as disaccharide and most complementary to Ta; (f) the sites for all glycotope units are located at the same site or nearby sites as an independent GlcNAc subsite has been reported [6,7]. Thus, an additional recognition factor responsible for the polyvalent effect of a poor Galb1?4GlcNAcb (IIb) and a weak Galb1?3GlcNAcb (Ib) on N-glycan–ABA interactions was established. The peculiar concept of four different intensities of glycotope units constructing three recognition systems provides an advanced knowledge about the combining sites of ABA, which can be used to explain the ABA attachment process to target glycans or malignant cells at the molecular level. Acknowledgements This work was supported by Grants from the Chang-Gung Medical Research Project (CMRP No. 180481 and 170442), Kwei-san, Tao-yuan, Taiwan, the National Science Council (NSC 97-2628-B182-002-MY3, and 97-2320-B-182-020-MY3), Taipei, Taiwan. References [1] Milton, J.D. and Rhodes, J.M. (1995) Dietary galactose-binding lectins and their effects on human colonic epithelial cells in: Lectin: Biomedical Perspectives (Pusztai, A. and Bardocz, S., Eds.), pp. 225–234, Taylor & Francis, London. [2] Yu, L.G., Fernig, D.G., Smith, J.A., Milton, J.D. and Rhodes, J.M. (1993) Reversible inhibition of proliferation of epithelial cell lines by Agaricus bisporus (edible mushroom) lectin. Cancer Res. 53, 4627–4632. [3] Yu, L.G., Fernig, D.G., White, M.R., Spiller, D.G., Appleton, P., Evans, R.C., Grierson, I., Smith, J.A., Davies, H., Gerasimenko, O.V., Petersen, O.H., Milton, J.D. and Rhodes, J.M. (1999) Edible mushroom (Agaricus bisporus) lectin which reversibly inhibits epithelial cell proliferation blocks nuclear localization sequence-dependent nuclear protein import. J. Biol. Chem. 274, 4890–4899. [4] Sueyoshi, S., Tsuji, T. and Osawa, T. (1985) Purification and characterization of four isolectins of mushroom (Agaricus bisporus). Biol. Chem. Hoppe-Seyler 366, 213–221. [5] Wu, A.M., Wu, J.H., Herp, A. and Liu, J.H. (2003) Effect of polyvalencies of glycotopes on the binding of a lectin from the edible mushroom, Agaricus bisporus. Biochem. J. 371, 311–320. [6] Carrizo, M.E., Capaldi, S., Perduca, M., Irazoqui, F.J., Nores, G.A. and Monaco, H.L. (2005) The antineoplastic lectin of the common edible mushroom (Agaricus bisporus) has two binding sites, each specific for a different configuration at a single epimeric hydroxyl. J. Biol. Chem. 280, 10614–10623. [7] Nakamura-Tsuruta, S., Kominami, J., Kuno, A. and Hirabayashi, J. (2006) Evidence that Agaricus bisporus agglutinin (ABA) has dual sugar-binding specificity. Biochem. Biophys. Res. Commun. 347, 215–220. [8] Duk, M., Lisowska, E., Wu, J.H. and Wu, A.M. (1994) The biotin/avidinmediated microtiter plate lectin assay with the use of chemically modified glycoprotein ligand. Anal. Biochem. 221, 266–272. [9] Lisowska, E., Duk, M. and Wu, A.M. (1996) Preparation of biotinylated lectins and application in microtiter plate assays and western blotting in the biomethod-series of birkhäuser publishers. Basel, Boston, Berlin 7, 115–128. [10] Kabat, E.A. (1956) Blood Group Substances, Their Chemistry and Immunochemistry, Academic Press, New York. pp. 135–139. [11] Wu, A.M., Kabat, E.A., Nilsson, B., Zopf, D.A., Gruezo, F.G. and Liao, J. (1984) Immunochemical studies on blood groups. Purification and characterization of radioactive 3H-reduced di- to hexasaccharides produced by alkaline betaelimination- borohydride 3H reduction of Smith degraded blood group A active glycoproteins. J. Biol. Chem. 259, 7178–7186. [12] Wu, A.M. (1988) Structural concepts of the human blood group A, B, H, Lea, Leb, I, and I active glycoprotein purified from human ovarian cyst fluid. Adv. Exp. Med. Biol. 228, 351–394. [13] Wu, A.M. and Sugii, S.J. (1988) Differential binding properties of Ga1NAc and/ or Ga1 specific lectins. Adv. Exp. Med. Biol. 228, 205–263. [14] Tettamanti, G. and Pigman, W. (1968) Purification and characterization of bovine and ovine submaxillary mucins. Arch. Biochem. Biophys. 124, 41–50. [15] Wu, A.M. and Pigman, W. (1977) Preparation and characterization of armadillo submandibular glycoproteins. Biochem. J. 161, 37–47. [16] Herp, A., Borelli, C. and Wu, A.M. (1988) Biochemistry and lectin binding properties of mammalian salivary mucous glycoproteins. Adv. Exp. Med. Biol. 228, 395–435. [17] Lindberg, B., Lonngren, J. and Powell, D.A. (1977) Structural studies on the specific type-14 pneumococcal polysaccharide. Carbohydr. Res. 58, 177–186.
3566
A.M. Wu et al. / FEBS Letters 584 (2010) 3561–3566
[18] Whistler, R.L. and Conrad, H.E. (1954) Crystalline galactobiose from acid hydrolysis of okra mucilage; 2-O-(D-galactopyranosyluronic acid)-L-rhamnose from okra mucilage. J. Am. Chem. Soc. 76, 1673–1674. [19] Yoshida, Y., Naito, E., Mizukoshi, H., Watanabe, Y., Kimura, K., Yokoi, W., Sato, T., Okumura, T., Ito, M. and Sawada, H.J. (2009) Side-chain structure of cell surface polysaccharide, mannan, affects hypocholesterolemic activity of yeast. J. Agric. Food Chem. 57, 8003–8009. [20] Gemmill, T.R. and Trimble, R.B. (1999) Review: overview of N- and O-linked oligosaccharide structures found in various yeast species. Biochim. Biophys. Acta 1426, 227–237. [21] Kwiatowski, B. and Stirm, S. (1987) Polysialic acid depolymerase. Meth. Enzymol. 138, 786–792. [22] Ryder, S.D., Smith, J.A., Rhodes, E.G., Parker, N. and Rhodes, J.M. (1994) Proliferative responses of HT29 and Caco2 human colorectal cancer cells to a panel of lectins. Gastroenterology 106, 85–93. [23] Wu, A.M. (2001) Expression of binding properties of Gal/GalNAc reactive lectins by mammalian glycotopes. Adv. Exp. Med. Biol. 491, 55–64. [24] Wu, A.M. (2002) Carbohydrate structural units in glycosphingolipids as receptors for Gal and GalNAc reactive lectins. Neurochem. Res. 27, 593–600.
[25] Wu, A.M. (2003) Carbohydrate structural units in glycoproteins and polysaccharides as important ligands for Gal and GalNAc reactive lectins. J. Biomed. Sci. 10, 676–688. [26] Yang, Z., Tsai, M.S., Wu, J.H., Herp, A. and Wu, A.M. (2008) Defining the carbohydrate specificities of Erythrina corallodendron lectin (ECorL) as polyvalent Galb1–4GlcNAc (II) > monomeric II > monomeric Gal and GalNAc. Chang Gung Med. J. 31, 26–43. [27] Kabat, E.A. (1988) Antibody combining sites: how much of the antibody repertoire are we seeing? How does it influence our understanding of the structural and genetic basis of antibody complementarity? Adv. Exp. Med. Biol. 228, 1–45. [28] Wu, A.M., Gong, Y.P., Li, C.C. and Gilboa-Garber, N. (2010) Duality of the carbohydrate-recognition system of Pseudomonas aeruginosa-II lectin (PA-IIL). FEBS Lett. 584, 2371–2375. [29] Kostlánová, N., Mitchell, E.P., Lortat-Jacob, H., Oscarson, S., Lahmann, M., Gilboa-Garber, N., Chambat, G., Wimmerová, M. and Imberty, A. (2005) The fucose-binding lectin from Ralstonia solanacearum: A new type of b-properller architecture formed by oligomerization and interacting with fucoside, fuccosyllactose, and plant xyloglucan. J. Biol. Chem. 280, 27839–27849.