Skip Navigation


Glycobiology Advance Access originally published online on March 15, 2006
Glycobiology 2006 16(6):524-537; doi:10.1093/glycob/cwj102
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplemental Data
Right arrow All Versions of this Article:
16/6/524    most recent
cwj102v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (15)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Wu, A. M.
Right arrow Articles by Gabius, H.-J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, A. M.
Right arrow Articles by Gabius, H.-J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Interaction profile of galectin-5 with free saccharides and mammalian glycoproteins: probing its fine specificity and the effect of naturally clustered ligand presentation

Albert M. Wu1,2, Tanuja Singh2, June H. Wu3, Martin Lensch4, Sabine André4 and Hans-Joachim Gabius4

2 Glyco-Immunochemistry Research Laboratory, Institute of Molecular and Cellular Biology and 3 Department of Microbiology and Immunology, Chang-Gung University, Kwei-san, Tao-yuan, 333, Taiwan; and 4 Faculty of Veterinary Medicine, Institute of Physiological Chemistry, Ludwig-Maximilians-University, Veterinärstrasse 13, D-80539 Munich, Germany

1 To whom correspondence should be addressed; e-mail: amwu{at}mail.cgu.edu.tw

Received on December 25, 2005; revised on March 10, 2006; accepted on March 12, 2006


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Cell-surface glycans are functional docking sites for tissue lectins such as the members of the galectin family. This interaction triggers a wide variety of responses; hence, there is a keen interest in defining its structural features. Toward this aim, we have used enzyme-linked lectinosorbent (ELLSA) and inhibition assays with the prototype rat galectin-5 and panels of free saccharides and glycoconjugates. Among 45 natural glycans tested for lectin binding, galectin-5 reacted best with glycoproteins (gps) presenting a high density of Galß1-3/4GlcNAc (I/II) and multiantennary N-glycans with II termini. Their reactivities, on a nanogram basis, were up to 4.3 x 102, 3.2 x 102, 2.5 x 102, and 1.7 x 104 times higher than monomeric Galß1-3/4GlcNAc (I/II), triantennary-II (Tri-II), and Gal, respectively. Galectin-5 also bound well to several blood group type B (Gal{alpha}1-3Gal)- and A (GalNAc{alpha}1-3Gal)-containing gps. It reacted weakly or not at all with tumor-associated Tn (GalNAc{alpha}1-Ser/Thr) and sialylated gps. Among the mono-, di-, and oligosaccharides and mammalian glycoconjugates tested, blood group B-active II (Gal{alpha}1-3Gal ß1-4GlcNAc), B-active IIß1-3L (Gal{alpha}1-3Galß1-4GlcNAc ß1-3Galß1-4Glc), and Tri-II were the best. It is concluded that (1) Galß1-3/4GlcNAc and other Galß1-related oligosaccharides with {alpha}1-3 extensions are essential for binding, their polyvalent form in cellular glycoconjugates being a key recognition force for galectin-5; (2) the combining site of galectin-5 appears to be of a shallow-groove type sufficiently large to accommodate a substituted ß-galactoside, especially with {alpha}-anomeric extension at the non-reducing end (e.g., human blood group B-active II and B-active IIß1-3L); (3) the preference within ß-anomeric positioning is Galß1-4 ≥ Galß1-3 > Galß1-6; and (4) hydrophobic interactions in the vicinity of the core galactose unit can enhance binding. These results are important for the systematic comparison of ligand selection in this family of adhesion/growth-regulatory effectors with potential for medical applications.

Key words: glycoprotein / lectin / N-glycans / O-glycans / sialylation


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Cell-surface glycans act as versatile biochemical signals for cellular communication. Therefore, there is increasing interest in analyzing the interaction between carbohydrate epitopes and tissue lectins. This information can enable a functionally oriented interpretation of changes in glycosylation among cell types, in development or disease, currently often viewed phenomenologically (Haltiwanger and Lowe, 2004Go; Villalobo et al., 2006). Explicitly, these changes might modulate the reactivity for endogenous lectins. Owing to their prominent position at branch ends, the spatially accessible ß-galactosides serve as contact sites for various groups of endogenous lectins, among them the galectins (Hirabayashi and Kasai, 1993Go; Kaltner and Stierstorfer, 1998Go; Gabius et al., 2002Go, 2004Go; Gabius and Wu, 2006Go). Of note, this family comprises 14 members with three subgroups based on structural organization (i.e., monomeric and homodimeric prototype, tandem-repeat type, and chimera-type proteins) (Hirabayashi and Kasai, 1993Go; Gabius, 1997Go, 2001Go; Cooper, 2002Go; Leffler et al., 2004Go). As suggested by their names, they bind to ß-galactosides. Despite the abundance of this epitope, cell-surface reactivity of galectins can be very selective, for instance with strong preference of galectin-1 for glycans of ganglioside GM1 or the {alpha}5ß1-integrin in galectin-dependent growth regulation of neuroblastoma or colon/pancreatic carcinoma cells (Kopitz et al., 1998Go, 2001; Fischer et al., 2005Go). Equally intriguing, even closely related galectins can trigger distinct cellular activities, as exemplified by the different profiles of caspase involvement in apoptotic cell death of activated T cells (Sturm et al., 2004Go). Evidently, the lectin sites are capable of distinguishing between various natural ß-galactosides, and spatial parameters such as epitope clustering might also contribute to the selectivity in cell-surface binding. The importance of regulation of ligand affinity by spatial presentation and the genuine complexity of cell-surface glycan structures has been inferred for C-type lectins such as selectins or DC-SIGN or members of the I-type subfamily of siglecs, especially CD22 (siglec-2) (Varki, 1997Go; Angata and Brinkman-van der Linden, 2002; Pitarque et al., 2005Go; Gabius, 2006Go). Thus, to relate lectin structure to function, it is essential to thoroughly map a lectin’s profile of glycan binding. In addition to commonly used free carbohydrates, assays with natural glycoproteins (gps) and polysaccharides of known glycan composition afford the opportunity to probe the influence of spatial vicinity on reactivity. A recent study on galectin binding to the three mostly triantennary N-glycans presented by the gp asialofetuin has highlighted the enormous potential of clustering for producing high-affinity binding sites by showing negative cooperativity and a gradient of microscopic binding constants (Dam et al., 2005Go). To address both issues (i.e., fine specificity and influence of clustered presentation) in a systematic manner, we have performed a study on a mammalian prototype galectin (i.e., galectin-5) using a panel of natural gps for testing ligand capacity and an array of mono-, di-, and oligosaccharides for defining inhibitory capacity.

Galectin-5 was initially isolated from extracts of rat lung and referred to as RL-18 (Cerra et al., 1985Go). Sequence analysis and initial specificity assays were performed next, confirming its place in the galectin family and assigning it to the prototype group (Leffler and Barondes, 1986Go; Gitt et al., 1995Go). While behaving as a monomer in mass spectrometric and gel-filtration analyses, it also has the capacity to aggregate as shown in hemagglutination assays and binding studies with carbohydrate dendrimers, resembling chimera-type galectin-3 in this respect (Gitt et al., 1995Go; Vrasidas et al., 2003Go; Ahmad et al., 2004Go; André, Kaltner, et al., 2005; André, Kojima, et al., 2005). Of note, galectin-5 is unique for rat and is closely related to the C-domain of the tandem-repeat-type galectin-9 with 93% sequence identity on the DNA level, a proapoptotic effector with impact on TH1 immunity (Wada and Kanwar, 1997Go; Hirashima et al., 2004Go; Zhu et al., 2005Go), so that a specificity analysis also has relevance for this carbohydrate recognition domain. In other words, galectin-5 and this domain are attractive candidates for pinpointing correlations between few structural changes and any difference in ligand binding. In functional terms, the assumed role of galectin-5 in erythropoiesis with its special place in rat warrants to define binding properties, so far not included in a mapping by frontal affinity chromatography (Hirabayashi et al., 2002Go). Taking account of these issues, we have studied, in detail, the recognition profile of galectin-5 by both enzyme-linked lectinosorbent (ELLSA) and inhibition assays. The results provide a detailed view on the reactivity of galectin-5 with a panel of glycans, moving conspicuously beyond the available data (Leffler and Barondes, 1986Go), most notably by testing a wide array of gps.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Galectin-5/glycan interactions
Preliminary assays ascertained that galectin-5 maintained carbohydrate-specific binding activity after labeling with two to four biotin moieties. Albeit being monomeric in mass spectrometric analysis under mild conditions, which did not affect dimer status of galectins-1 and -7 (André, Kaltner, et al., 2005), the lectin showed hemagglutination activity with rabbit erythrocytes at a concentration as low as 15 µg/mL, indicating capacity for limited aggregate formation under these assay conditions. For comparison, homodimeric rat galectin-2 was active under identical conditions at 2–3 µg/mL, while human galectin-1 could be diluted to an active concentration of 0.3–0.5 µg/mL. To define binding properties of this galectin involved in erythropoiesis for natural gps, a model for the complexity of cell-surface glycans, we systematically determined the extent of reactivity under constant conditions. The avidity of galectin-5 for our test panel as studied by microtiter plate ELLSA is summarized in Table I (for a detailed documentation of the interaction profiles, please see Figure 1 in Supplementary Material). Among the gps tested, galectin-5 reacted best with four high-density Galß1-3/4GlcNAc (I/II)-containing glycoconjugates. These included human blood group B-active gps (cyst Beach phenol insoluble) and three blood group precursor (equivalent) determinants, cyst Beach P-1, cyst OG 10% 2x ppt, and cyst MSS 1st Smith degraded gp (Figure 1a in Supplementary Material). Less than 4 ng of the gp coated was required to yield an A405 value of 1.5 within 24 h when co-incubated with 250 ng of lectin. It also reacted strongly with other blood group B-active gp (cyst 19 in Figure 1a in Supplementary Material), high-density I/II-containing gps (cyst Tighe P-1 in Figure 1b, cyst MSS 2nd Smith degraded gp in Figure 1a and cyst Mcdon P-1 in Figure 1b presented in Supplementary Material), type II-containing N-glycans (human and bovine asialo {alpha}1-acid gp, sialylated and asialo Tamm–Horsfall gp (THGP), porcine thyroglobulin and its asialo derivative, asialo RSL, and asialofetuin in Figure 1d, f, and g in Supplementary Material), asialo bird nest gp (Figure 1f in Supplementary Material), blood group A- and H-active gps from hog gastric mucin and mild acid-hydrolyzed derivatives (Figure 1c in Supplementary Material), and blood group Lea-containing compounds from human ovarian cyst fluid (Figure 1e in Supplementary Material). T-containing human asialoglycophorin, Pneumococcus type 14 capsular polysaccharide which is composed of repeating poly II residues, blood group Ah-containing compounds from human ovarian cyst fluid, and native hog gastric mucin 4 (Ah, H) required 120–1100 ng to yield an A405 value of 1.5 within 24 h (Table I). In contrast, galectin-5 bound weakly or was completely inactive with most of the sialylated multiantennary II-containing N-glycans, Tn-containing salivary gps (sialylated and asialo OSM, BSM, and PSM in Table I), and blood group H-active cyst gp. To prove that the differences in ligand properties of these gps are not due to disparities in the extent of adsorption to the plate well surface and to substantiate the influence of ligand cluster presentation on binding, the inhibitory activity of each gp was also tested in solution using a standard ligand as matrix for galectin-5 binding.


View this table:
[in this window]
[in a new window]
  		Table I. The glycotope-binding specificity of galectin-5a

 

Figure 1
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1. Interspecies and inter-galectin sequence comparison of rat galectin-5 and the N-terminal domain of the rat tandem-repeat-type galectin-4 (G4-N) as well as homodimeric prototype chicken galectin CG-16. Amino acid sequences of the carbohydrate recognition domains of these three lectins studied previously (Wu et al., 2001Go, 2002Go, 2004) and herein were aligned using the program Multalin version 5.4.1. (http://prodes.toulouse.inra.fr/multalin/multalin.html). Identical residues between all three galectins are indicated as white letter on black background, whereas residues that are identical in two of the sequences are given as black letter on gray background. A consensus sequence calculated from the three galectin sequences is added to the alignment; consensus symbols represent: !, I or V; #, N, D, Q, or E.

 

Inhibition of galectin-5/glycoform interaction by various polyvalent glycoconjugates
Table II documents the inhibitory potency of various gps when tested for interfering with binding of galectin-5 to the immobilized Galß1-3/4GlcNAc-containing gp (cyst Beach P-1). Details of the inhibition profile by individual gps are shown in Figure 2 in Supplementary Material. Among the glycans tested for inhibition of this interaction and expressed as nanogram amount, human blood group precursor gp (cyst Beach P-1), which contains a high density of Galß1-4GlcNAc and Galß1-3GlcNAc termini (II/I) at the non-reducing branch ends, was the best inhibitor, requiring only 8 ng to inhibit 50% of lectin/glycan binding. It was 2.5 x 102, 3.2 x 102, 4.3 x 102, and 1.7 x 104 times more potent than triantennary Galß1-4GlcNAc (Tri-II), monomeric Galß1-4/3GlcNAc (II/I), and Gal, respectively (Table II; curve 1 versus curves 28, 29, 31, and 33 in Figure 2a in Supplementary Material). Galectin-5 binding was also strongly inhibited by most of the other gps presenting high density of I/II epitopes. These included gps bearing human blood group precursor glycans (curves 2, 3, 8, 10, and 18), asialo bird nest gp (curve 4), blood group Lea-containing gp from human ovarian cyst fluid (asialo HOC 350 and cyst N-1 Lea 20% 2x, curves 9 and 11), blood group A- and H-active gps from hog gastric mucin 4 (curve 12), gps with multiantennary II-containing N-glycans (sialylated and asialo THGP, porcine thyroglobulin prior to and after desialylation, human and bovine asialo {alpha}1-acid gp, asialo RSL, and asialofetuin: curves 7, 13–17, 19, and 23 in Figure 2b in Supplementary Material), and also Pneumococcus type 14 capsular polysaccharide, which is composed of repeating poly-N-acetyllactosamine residues (curve 20). Their reactivities ranged from 180- to 3.5x103-fold higher than that of free Gal (Figure 2 in Supplementary Material, Table II, curves 2–4, 7–20 versus curve 33) and up to 65.0- and 87.5-fold higher than that of the free Galß1-4GlcNAc (II) and Galß1-3GlcNAc (I) disaccharides (curves 29 and 31; Table II), respectively. The extent of binding of galectin-5 was also strongly reduced by the presence of human blood group B-active cyst gps (cyst 19 and cyst Beach phenol insoluble, curves 5 and 6), blood group A- and H-active gps from hog gastric mucin and its mild acid-hydrolyzed derivatives (hog gastric mucins 9, 14, and 21; curves 21, 22, and 25), human asialoglycophorin which contains one N-glycan and 15 mucin-type O-glycans per molecule (curve 24; Figure 2), and blood group Ah-active gps (curves 26, 27, and 30). Sialylated human glycophorin (curve 34), Tn-glycophorin (curve 35), bird nest gp (curve 36), sialylated N-glycans with II-branch ends (curves 37–39), blood group H-active gp (curve 41), and Tn-containing salivary gps (curves 42–47) were tested from 278 to 556 ng, but did not reach 50% inhibition (Table II). Results of both the interaction and the inhibition assays showed consistency. These experiments thus defined the reactivity of galectin-5 to complex natural glycans as presented in gps or polysaccharides. As a means for mapping the selectivity toward mono-, di-, or oligosaccharides, a further series of measurements in the inhibition assay setting were required.


View this table:
[in this window]
[in a new window]
  		Table II. The inhibitory potency of various gps affecting the binding of galectin-5 (125 ng/50 µL) to a Galß1-3/4GlcNAc-containing gp (cyst Beach P-1, 5 ng/50 µL)a

 

Inhibition of galectin-5/glycoform interaction by mono-, di-, and oligosaccharides
Conceptually, galectin-5 binding to a gp presented as matrix was performed in the presence of free sugars. The amounts of ligands (expressed as nanomoles) required for 50% inhibition of the galectin-5/glycoform interaction with a Galß1-3/4GlcNAc-containing gp (cyst Beach P-1) are listed in Table III. The curves reflecting inhibitory capacity are illustrated in the Supplementary Material (Figure 3). Among the oligosaccharides and mammalian multiantennary II-containing glycotopes tested, blood group B-active II (linear B-trisaccharide; Gal{alpha}1-3Galß1-4GlcNAc) was the best inhibitor. It was slightly better than B-active IIß1-3L (linear B-pentasaccharide; Gal{alpha}1-3Galß1-4GlcNAcß1-3 Galß1-4Glc) and triantennary Galß1-4GlcNAc at the non-reducing end in N-linked glycopeptides (Tri-II) (curves 2 and 3). This blood group B-active II was 11.7- and 1.3 x 103-fold more active than free Galß1-4GlcNAc (II) and Gal, respectively (curve 1 versus curves 10 and 38 in Figure 3a in Supplementary Material). Galß1-4GlcNAc (II) was about 1.3 and 114 times more potent than Galß1-3GlcNAc (I) and Gal, respectively (curve 10 versus curves 11 and 38). Their reactivities could be enhanced by up to 3.5 times by increasing the length of carbohydrate chains. Linear B tri- and pentasaccharides were 133.3 and 80 times more effective than the blood group B-type disaccharide, revealing the importance of the lactose core. B-active II (curve 1) was about 7-fold more active than B-active L (Gal{alpha}1-3Galß1-4 Glc, curve 8), indicating that the N-acetyl group of the amino group at C-2 significantly enhanced the binding reactivity to the lectin. Galß1-4Glcß1-methyl (Lß) which is a building block in internal carbohydrate structures of human (brain) glycosphingolipids was as active as Galß1-4 Glc (L) (curve 13 versus curve 14), but about 1/4 as active as II-disaccharide, further proving that the subterminal N-acetyl group is important for binding (curve 13 versus curve 10). In this study, moderate inhibition of the galectin-5/cyst Beach P-1 interaction was also observed with Galß1-3GalNAc (T, curve 12), P (GalNAcß1-3Gal, curve 16), B (Gal{alpha}1-3Gal, curve 17), S (GalNAcß1-4Gal, curve 18), F (GalNAc{alpha}1-3GalNAc, curve 19), and A (GalNAc{alpha}1-3Gal, curve 22).


View this table:
[in this window]
[in a new window]
  		Table III. The inhibitory potency of various saccharides affecting the binding of galectin-5 (125 ng/50 µL) to a Galß1-3/4GlcNAc-containing gp (cyst Beach P-1, 5 ng/50 µL)a

 

As compiled in Table III, Galß1-4Gal (curve 32) was about 13.3 and 57 times less active than L (curve 14) and II (curve 10), implying that the configuration at C-4 and N-acetyl group at C-2 of the subterminal sugar unit are important for binding. Raffinose (Gal{alpha}1-6Glcß1-2DFruf) and stachyose (Gal{alpha}1-6Gal{alpha}1-6Glcß1-2DFruf) were equally active and about 1.6-fold better than melibiose (Gal{alpha}1-6Glc) (curves 36 and 37 versus curve 40), pointing to a certain degree of reactivity to {alpha}-anomers of Gal with potential for slight enhancement by chain extension. However, globally, {alpha}-anomers at the central Gal unit were rather weak inhibitors.

Monitoring the inhibitory activities of sugar derivatives with aglyconic substituents provides further information on the properties of the vicinity of the contact site for the central Gal unit. Among the respective monosaccharide variants studied, p-nitrophenyl {alpha}- and ß-derivatives of Gal were 7.7 and 4.4 times better than their methyl {alpha}- and ß-analogs, revealing that strong hydrophobicity at the anomeric position improves the inhibitory capacity (curve 29 versus curve 34 and curve 30 versus curve 39). Assays with monosaccharides underscored the requirement of contacts at least on the level of the disaccharide.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Galectins are effectors for regulating cell adhesion, migration, and proliferation by virtue of their binding and clustering of cell-surface ligands (Brewer, 2002Go; Villalobo et al., 2006). Because organs and even cells can co-express more than one family member, as documented for rat lung and various human tumor cells in vitro and in vivo (Cerra et al., 1985Go; Lahm et al., 2001Go, 2004Go), it is a clinically relevant challenge to define the functionality of each family member. Toward this long-term aim, mapping the carbohydrate specificity is a key step. The comparison of the resulting data sets will help answer the question of whether and to what extent structural changes translate into alterations of the ligand profile. Of note, even single-site mutations distant from the contact area with the ligand have been described to affect structural and thermodynamic aspects for a galectin (López-Lucendo et al., 2004Go). Besides measuring the reactivity with mono-, di-, and oligosaccharides, we included monitoring of the aspect of clustered presentation, which is a hallmark of cell-surface glycans, into our study by testing a panel of well-documented gps. Involvement of galectin-5 in erythropoiesis and its close similarity to the C-terminal domain of galectin-9, a key regulator of TH1 immunity, provided reasons to clarify its preferences for glycan binding.

To analyze the carbohydrate specificity of galectin-5, we employed the optimized ELLSA method (Duk et al., 1994Go; Lisowska et al., 1996Go), in which binding of the labeled lectin is detected with alkaline phosphatase-conjugated avidin. The notable advantage of this method is that the lectin amount required for the assay is about 1/10 to 1/3500 of that necessary for the quantitative precipitin and precipitin-inhibition assays (Wu et al., 1980Go, 1992Go, 1999Go). A wide range of defined gp and oligosaccharide preparations can thus be conveniently tested to characterize the binding profile of a lectin (Wu et al., 1980Go, 1992Go). Since the amounts of gps adsorbed onto the microwells are difficult to quantitate, the interactions of galectin-5 with the various test substances were examined using three parameters: (1) the amounts of gps added to wells that yielded an absorbance of 1.5 A405 units (Figure 1 in Supplementary Material and Table I), (2) the maximum absorbance obtained in assays with each gp after 24 h of incubation (Figure 1 in Supplementary Material and Table I), and (3) the amount of gp required to give 50% inhibition (Figure 2 in Supplementary Material and Table II).

A striking result of this study concerns the preference of galectin-5 for Galß1-3/4GlcNAc end groups. In addition to reacting strongly with human blood group precursor gps, this lectin also bound well to human blood group B-, Ah-, and Lea-active gps. Galectin-5 can thus recognize Galß1-3/4GlcNAc cores, harboring the abovementioned substitutions and extensions. Because these determinants are abundantly found on erythrocytes, the preferential binding of galectin-5 to histo-blood group epitopes supports the concept of intimate protein/carbohydrate interactions in these cells or their erythropoietic precursors (Gitt et al., 1995Go). To figure out whether other family members such as the cross-linking galectin-9 may compete for the same ligands and monomeric galectin-5 may thus block galectin-9 binding and effector functionality, our analysis needs to be extended accordingly under the same conditions.

From these sets of results, we conclude that (1) a high density of polyvalent Galß1-3/4GlcNAc (I/II) (curves 1–4 in Figure 2 in Supplementary Material) and substituted derivatives thereof is a very potent factor promoting galectin-5/glycoform binding; (2) when a blood group B (Gal{alpha}1-) determinant is added to terminal Galß1-4GlcNAc (II) and IIß1-3L core oligosaccharides, the reactivities are increased (curves 1 and 2 in Figure 3 in Supplementary Material); (3) the combining site of galectin-5 appears to be of the shallow-groove type sufficiently large to accommodate a pentasaccharide with {alpha}-anomeric Gal at the non-reducing end and most complementary to the tri- (B-active II) and pentasaccharides (B-active IIß1-3L); (4) galectin-5 has the following order of preference for ß-anomeric galactosides—Galß1-4 ≥ Galß1-3 > Galß1-6; and (5) a hydrophobic interaction in the vicinity of the binding site for the central Gal moiety can increase affinity, a result relevant for design of synthetic inhibitors. Using a recently proposed system to classify specificity by increasing the length of the ligand in a stepwise manner (Wu, 2003Go), binding properties of galectin-5 are defined as follows: (1) the monosaccharide specificity is GalNAc/Gal > GlcNAc, Glc, and LFuc (inactive); (2) the specificities for mammalian disaccharide structural unit is II > I > T > L > P > B > S/F/A > Tn and E (inactive); (3) the most active ligand is B-active II; (4) single multiantennary chains or small clusters of glycotopes such as Di-II, Tri-II, and biantennary N-glycans with terminal I/II determinants have limited impact, as also observed by direct interaction study with the LEC14-type N-glycan (André, Kojima, et al., 2005), and Tn-containing glycopeptides (MW < 3.0 x 103) are inactive (this result sets galectin-5 apart from the macrophage C-type lectin which can avidly home in on the Tn epitope; Iida et al., 1999Go); and (5) high-density multivalent or complex clustering of glycotopes plays a prominent role in enhancing galectin-5 reactivity. Having defined the specificity profile of galectin-5, the hypothesis can next be tested that particular changes in the sequence of galectins have an impact on ligand profiles, whereby crucial sites relevant for further scrutiny might be inferred.

We have previously reported the carbohydrate specificity of the prototype chicken galectin CG-16 and of the N-terminal domain of the tandem-repeat-type rat galectin-4 G4-N using this type of analysis (Wu et al., 2001Go, 2002Go, 2004). Homodimeric CG-16 is a developmentally regulated galectin from chicken liver with capacity for efficient cross-linking (Gupta et al., 1996Go; Wu et al., 2001Go). Owing to the presence of defined amino acid differences between the carbohydrate recognition domains of the three galectins (i.e., CG-16, G4-N, and galectin-5, shown in Figure 1) and the differences in quaternary structure, our data render it possible to answer the question as to whether and to what extent these disparities translate into detectable changes on the level of ligand selection. As summarized in Table IV, the reactivities of binding for high-density I/II-containing gps (such as I/II in human blood group precursors and multiantennary II in human asialo {alpha}1-acid gp and asialofetuin) can be ranked as G4-N > galectin-5 > CG-16. The degree of enhancement in Gal-5 and CG-16 was less when compared with G4-N (Wu et al., 2004Go), indicating that the extent of polyvalent effects on carbohydrate/galectin interactions must be individually evaluated to establish its rules. Effects of the presence of ABH- and Lea-active sugars and sialic acids on binding can be summarized as CG-16 > galectin-5 > G4-N (Table IV). Next, T/Tn-containing gps were ligands for the gastrointestinal G4-N and to some extent for galectin-5, while CG-16 showed negligible binding. G4-N and galectin-5, when used as histochemical tools as currently tested for human galectins in histopathology (Plzák et al., 2004Go), can therefore detect the presence of both ABH- and Lea-containing glycotopes and their precursors, while CG-16 can distinguish between ABH-active glycotopes and their precursors. CG-16 and also human galectin-1 (Lee et al., 1990Go) clearly prefer Gal to GalNAc on the level of the monosaccharide in contrast to G4-N and galectin-5. By computationally drawing the carbohydrate contact sites for CG-16/galectin-5, Glu/His and Lys/Gln substitutions in the carbohydrate recognition domains (please see also Figure 1) have been delineated (André, Kaltner, et al., 2005). They can contribute to explain the described changes. However, the current paucity of structural data on galectins, when interacting with complex ligands, noted long-range effects of single amino acid substitutions (López-Lucendo et al., 2004Go), and the potential for an effect of the ligand on a galectin’s gyration radius, shown for galectin-1 in solution (He et al., 2003Go), render the reliability of computational predictions not free of ambiguity. Broadening of the database by determining reactivity profiles will be instrumental to substantiate the concept that subtle sequence changes in a lectin family can bring about disparate binding profiles (Wu et al., 1999Go, 2000, 2004; Wu, 2003Go). In parallel, the spatial way glycans of cellular glycoconjugates are presented, which is for instance modulated by branching or substitutions (Reuter and Gabius, 1999Go), will also have a bearing on their affinity toward lectins. This notion has recently been validated for the presence of the bisecting GlcNAc unit and core fucosylation in complex-type biantennary N-glycans and galectin-1 (Unverzagt et al., 2002Go; André et al., 2004Go) and further extended by defining branching as regulatory element (Dam et al., 2005Go; André et al., forthcoming). Based on our reported data, the strong binding to the B-type substitution and the tolerance for Le-type additions provide attractive targets for structural research to define the nature of the extended binding site of galectin-5, as performed recently for human galectin-1 and its interaction with ganglioside GM1’s pentasaccharide in solution (Siebert et al., 2003Go). Such results will be a valuable resource for the optimization of computational calculations of the complete binding sites in galectins beyond the Gal core to predict structures of high-affinity and discriminatory ligands for medical applications. These presented data also point to the possibility for distinct selectivity of galectin-5 in its involvement in erythropoiesis.


View this table:
[in this window]
[in a new window]
  		Table IV. Comparison of binding specificity of galectin-5 with those of two previously studied ABH- and I/II-specific galectins (Wu et al., 2001Go, 2002Go, 2004)

 


   Materials and Methods
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Lectin production and biotinylation
The cDNA for rat galectin-5 was cloned from the RNA population of adult kidney and inserted into the NcoI/BamHI sites of the commercial pQE-60 expression vector (Qiagen, Hilden, Germany). Recombinant production in E. coli strain M15 (Qiagen) and TB medium (Roth, Karlsruhe, Germany) at 30°C followed using an isopropyl-ß-D-thiogalactoside concentration of 0.5 mM and lectin purification with affinity chromatography as crucial step was performed using lactose-bearing Sepharose 4B, obtained after divinyl sulfone activation of the resin, as described (Gabius, 1990Go; André et al., 1999Go, André, Kaltner, et al., 2005). Quality controls for purity and activity were performed by one- and two-dimensional gel electrophoresis, gel filtration on a Superose 12 HR 10/30 column (Pharmacia, Freiburg, Germany), and mass spectrometry as well as hemagglutination with glutaraldehyde-fixed, trypsin-treated rabbit erythrocytes (André et al., 2001Go; André, Kaltner, et al., 2005). For biotinylation, the lectin stabilized by iodoacetamide treatment was labeled with the N-hydroxysuccinimide ester derivative of biotin (Sigma, Munich, Germany) under activity-preserving conditions, as described in detail previously (Gabius et al., 1991Go), and checked for label incorporation by a proteomics protocol (Purkrábková et al., 2003Go).

Glycoproteins and polysaccharides
The blood group A-, B-, H-, Lea-, and Leb-active substances were purified from human ovarian cyst fluid by digestion with pepsin and precipitation with increasing concentrations of ethanol (Kabat, 1956Go; Lloyd and Kabat, 1968Go; Maisonrouge-McAuliffe and Kabat, 1976Go); the dried ethanol precipitates were extracted with 90% phenol. Codes such as OG, MSS, Tighe, Mcdon, Tij, and JS represent different patients from which the samples were taken. Supernatants were fractionally precipitated by the addition of 50% ethanol in 90% phenol to the indicated concentrations (Kabat, 1956Go). The designation 10% or 20% (ppt) refers to a fraction precipitated from phenol at an ethanol concentration of 10 or 20%; 2x signifies that a second round of phenol extraction and ethanol precipitation was carried out (e.g., Cyst OG 20% 2x). HOC 350, a sialic-acid-rich gp isolated from human ovarian cyst fluid (Pusztai and Morgan, 1961Go), was kindly provided by the late Dr W.M. Watkins (Faculty of Medicine, Imperial College of Science, Technology and Medicine, London, UK). Regardless of their A-, B-, H-, or Lewis- (a, b, x, and y) activity, the purified water-soluble blood group substances from human ovarian cyst fluids have a similar precursor structure. They are polydispersed macromolecules (MW > 200,000 Da) of similar composition (75–85% carbohydrate, 15–25% protein), presenting multiple mucin-type O-glycans which can be branched (Lloyd and Kabat, 1968Go; Maisonrouge-McAuliffe and Kabat, 1976Go; Wu, 1988Go).

In general, the P-1 fractions (e.g., Cyst Mcdon P-1 or Tighe P-1) represent the non-dialyzable portion of the blood group substances after mild hydrolysis at pH 1.5–2.0 for 2 h which removes most of the L-fucopyranosyl end groups as well as some blood group A- and B-active oligosaccharide side chains (Kabat et al., 1948Go; Leskowitz and Kabat, 1954Go; Allen and Kabat, 1959Go). The 1st Smith-degraded products of blood group A-active substances (MSS 10% 2x), in which almost all of the terminal sugar groups of the branches are removed, were prepared as described earlier (Wu et al., 1982Go, 1984). Hog gastric mucin 4, a blood group A- and H-substance, was derived from crude hog stomach mucin as described previously (van Halbeek et al., 1982Go). Its acid hydrolysis with HCl at pH 2, 100°C, for 90 min yields hog gastric mucin 9, while at pH 1.5, 100°C, for 2 and 5 h, hog gastric mucins 14 and 21 (respectively) are produced. Extensive hydrolysis leads to impairment of blood group activities (Wu et al., 1992Go). Human and bovine {alpha}1-acid gp and porcine thyroglobulin were purchased from Sigma (St. Louis, MO). Human {alpha}1-acid gp contains tetra-, tri-, and diantennary complex-type glycans in the ratio of 2 : 2 : 1 (Fournet et al., 1978Go; Fournier et al., 2000Go). Fetuin (Gibco Laboratories, Grand Island, NY), which is the major gp in fetal calf serum, harbors six glycan chains per molecule, three mucin-type O-glycans, and three mostly triantennary N-glycans (Nilsson et al., 1979Go; Dam et al., 2005Go). Mucus gp (or native bird-nest gp), the so-called nest-cementing substance (Wieruszeski et al., 1987Go) from the salivary gland of Chinese swiftlets (genus Collocalia), was extracted with distilled water at 60°C for 20 min from commercial bird nest substance (Kim Hing, Singapore). Tamm–Horsfall gp (THGP), which was kindly provided by the late Dr W.M. Watkins, was isolated with 0.58 M NaCl from the urine of a single donor (W.T.J.M.) with the Sd (a+) blood group by the method of Tamm and Horsfall with slight modifications (Donald et al., 1982Go; Wu et al., 1995Go). The rat-sublingual gp (RSL) was prepared by the method of Moschera and Pigman (1975)Go. Ovine salivary gp-major (OSM-major), bovine submandibular gp-major (BSM), and porcine salivary mucin (PSM) were purified according to the method of Tettamanti and Pigman (1968)Go with some modifications (Herp et al., 1979Go, 1988Go).

Production of asialoglycoproteins was performed by mild acid hydrolysis in 0.01 N HCl at 80°C for 90 min; thereafter, the solution was dialyzed against distilled water for 2 days to remove reagents (Wu and Pigman, 1977Go). Glycophorin was prepared by the method described previously (Duk et al., 1994Go). Desialylation of glycophorin was performed by mild acid hydrolysis (Tettamanti and Pigman, 1968Go). The Tn-type glycophorin (Tn-glycophorin) was obtained by removing galactose residues from asialoglycophorin using periodate oxidation and mild acid hydrolysis (Smith degradation) (Wu et al., 1982Go; Duk et al., 1994Go). The Pneumococcus type 14 polysaccharide was a generous gift from the late Dr E.A. Kabat (Department of Microbiology, Columbia University, New York, NY) (Lindberg et al., 1977Go). Capsular polysaccharide from E. coli (colominic acid) consisting of poly-{alpha}2,8-N-acetylneuraminic acid was purchased from Sigma.

Inhibiting sugars
Mono-, di-, and oligosaccharides used were obtained from Dextra (Reading, Berkshire, UK) or Sigma. Triantennary glycopeptides with mostly type II termini and the 2,4,2,-branching pattern were prepared from asialofetuin by pronase digestion and repeatedly fractionated by BioGel P-4 400-mesh column chromatography (Wu et al., 1998Go). The Tn clusters used for this study were mixtures of Tn-containing glycopeptides from asialo OSM in the filterable fraction (molecular mass cut-off <3000 Da) (Wu et al., 1997Go).

Enzyme-linked lectinosorbent assay
The assay was performed according to the procedures described by Duk et al. (1994)Go. The volume of each reagent solution applied to wells of the plate was 50 µL/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). TBS-T was used for washing plates between incubation steps.

The 96-well microtiter plates (Nunc-Immuno plate, Kamstrupvej, Denmark) were coated with gps in 0.05 M sodium carbonate buffer (0.05 M NaHCO3/0.05 M Na2CO3), pH 9.6, and kept overnight at 4°C. After washing the plate, solution with biotinylated galectin-5 was added, and the plate was incubated for 30 min. The plates were next carefully washed to remove free lectin, and the ExtrAvidin/alkaline phosphatase solution (diluted 1 : 10,000; Sigma) was then added to detect the specifically bound probes. After 1 h, the plates were washed at least four times to remove the conjugate and then incubated with p-nitrophenyl phosphate (Sigma phosphatase substrate 5 mg tablets) in 0.05 M carbonate buffer, pH 9.6, containing 1 mM MgCl2 (1 tablet/5 mL). The resulting absorbance was read at 405 nm in a microtiter plate reader after 24 h incubation at 20°C in the dark with the substrate-containing solution. For inhibition studies, serially diluted inhibitor samples were mixed with an equal volume of lectin solution containing a fixed amount of galectin-5 (250 ng/well). The control lectin sample was diluted 2-fold with TBS-T. After 30 min at 20°C, samples were tested by the binding assay, as described above. The inhibitory activity was estimated from the inhibition curve and is expressed as the amount of inhibitor (ng or nmol per well) giving 50% inhibition of binding of the control galectin binding.


   Supplementary Material
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Supplementary Material
Conflict of interest statement
 Acknowledgments
 References
 
Supplementary data are available at Glycobiology online (http://glycob.oxfordjournals.org/).


   Conflict of interest statement
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
Acknowledgments
 References
 
None declared.


   Acknowledgments
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
References
 
This study was supported by grants from the Chang-Gung Medical Research Project (CMRPD no. 33022), Kwei-san, Tao-yuan, Taiwan, the National Science Council, Taiwan (NSC 94-2320-B-182-044, NSC 94-2320-B-182-053), and the Mizutani Foundation for Glycoscience (Tokyo, Japan) as well as a travel grant of the Deutscher Akademischer Austauschdienst (DAAD; Bonn, Germany).


   Abbreviations
 
DAra, D-Arabinose; BSM, bovine submandibular glycoprotein-major; ELLSA, enzyme-linked lectinosorbent assay; DFuc, D-fucopyranose; LFuc, L-fucopyranose; Gal, D-galactopyranose; GalNAc, 2-acetamido-2-deoxy-D-galactopyranose; Glc, D-glucopyranose; GlcNAc, 2-acetamido-2-deoxy-D-glucopyranose; gp, glycoprotein; HOC, human ovarian cyst fluid; Man, D-mannopyranose; OSM, ovine salivary glycoprotein-major; PSM, porcine salivary mucin; TBS-T, Tris/HCl-buffered saline containing 0.05% Tween 20;
The structural units in mammalian glycans used to define binding properties of galectin-5 are: A, GalNAc{alpha}1-3Gal, human blood group A-specific disaccharide; Ah, GalNAc{alpha}1-3(LFuc{alpha}1-2)Gal, human blood group A-specific trisaccharide containing the crypto H determinant; B, Gal{alpha}1-3Gal, human blood group B-specific disaccharide; Bh, Gal{alpha}1-3(LFuc{alpha}1-2)Gal, human blood group B-specific trisaccharide containing the crypto H-determinant; H, LFuc{alpha}1-2Gal, human blood group H-specific disaccharide; I, Galß1-3GlcNAc, human blood group type I precursor sequence; II, Galß1-4GlcNAc, human blood group type II precursor sequence; E, galabiose, Gal{alpha}1-4Gal sequence, a receptor of the uropathogenic E. coli ligand; L, Galß1-4Glc, lactose; T, Galß1-3GalNAc, Thomsen–Friedenreich antigen; Tn, GalNAc{alpha}1-Ser/Thr; F, GalNAc{alpha}1-3GalNAc, Forssman antigen; P, GalNAcß1-3Gal; S, GalNAcß1-4Gal


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Supplementary Material
 Conflict of interest statement
 Acknowledgments
 References
 
Ahmad, N., Gabius, H.-J., André, S., Kaltner, H., Sabesan, S., Roy, R., Liu, B., Macaluso, F., and Brewer, C.F. (2004) Galectin-3 precipitates as a pentamer with synthetic multivalent carbohydrates and forms heterogeneous cross-linked complexes. J. Biol. Chem., 279, 10841–10847.[Abstract/Free Full Text]

Allen, P.Z. and Kabat, E.A. (1959) Immunochemical studies on blood groups XXII. Immunochemical studies on the nondialyzable residue from partially hydrolyzed blood group A, B, and O(H) substances (P1 fraction). J. Immunol., 82, 340–357.[Abstract/Free Full Text]

André, S., Kaltner, H., Lensch, M., Russwurm, R., Siebert, H.-C., Fallsehr, C., Tajkhorshid, E., Heck, A.J.R., von Knebel Doeberitz, M., Gabius, H.-J., and Kopitz, J. (2005) Determination of structural and functional overlap/divergence of five proto-type galectins by analysis of the growth-regulatory interaction with ganglioside GM1 in silico and in vitro on human neuroblastoma cells. Int. J. Cancer, 114, 46–57.[CrossRef][ISI][Medline]

André, S., Kojima, S., Gundel, G., Russwurm, R., Schratt, X., Unverzagt, C., and Gabius, H.-J. (Forthcoming) Branching mode in complex-type triantennary N-glycans as regulatory element of their ligand properties. Biochim. Biophys. Acta.

André, S., Kojima, S., Prahl, I., Lensch, M., Unverzagt, C., and Gabius, H.-J. (2005) Introduction of extended LEC14-type branching into core-fucosylated biantennary N-glycan. Glycoengineering for enhanced cell binding and serum clearance of the neoglycoprotein. FEBS J., 272, 1986–1998.[CrossRef][Medline]

André, S., Kojima, S., Yamazaki, N., Fink, C., Kaltner, H., Kayser, K., and Gabius, H.-J. (1999) Galectins-1 and -3 and their ligands in tumor biology. J. Cancer Res. Clin. Oncol., 125, 461–474.[CrossRef][ISI][Medline]

André, S., Pieters, R.J., Vrasidas, I., Kaltner, H., Kuwabara, I., Liu, F.-T., Liskamp, R.M.J., and Gabius, H.-J. (2001) Wedgelike glycodendrimers as inhibitors of binding of mammalian galectins to glycoproteins, lactose maxiclusters, and cell surface glycoconjugates. Chembiochem, 2, 822–830.[CrossRef][ISI][Medline]

André, S., Unverzagt, C., Kojima, S., Frank, M., Seifert, J., Fink, C., Kayser, K., von der Lieth, C.-W., and Gabius, H.-J. (2004) Determination of modulation of ligand properties of synthetic complex-type biantennary N-glycans by introduction of bisecting GlcNAc in silico, in vitro and in vivo. Eur. J. Biochem., 271, 118–134.[ISI][Medline]

Angata, T. and Brinkman-van der Linden, E.C.M. (2002) I-type lectins. Biochim. Biophys. Acta, 1572, 294–316.[Medline]

Brewer, C.F. (2002) Binding and cross-linking properties of galectins. Biochim. Biophys. Acta, 1572, 255–262.[Medline]

Cerra, R.F., Gitt, M.A., and Barondes, S.H. (1985) Three soluble rat ß-galactoside-binding lectins. J. Biol. Chem., 260, 10470–10477.

Cooper, D.N.W. (2002) Galectinomics: finding themes in complexity. Biochim. Biophys. Acta, 1572, 209–231.[Medline]

Dam, T.K., Gabius, H.-J., André, S., Kaltner, H., Lensch, M., and Brewer, C.F. (2005) Galectins bind to the multivalent glycoprotein asialofetuin with enhanced affinities and a gradient of decreasing binding constants. Biochemistry, 44, 12564–12571.[CrossRef][Medline]

Donald, A.S.R., Soh, C.P.C., Watkins, W.M., and Morgan, W.T.J. (1982) N-Acetyl-D-galactosaminyl-ß-(1->4)D-galactose: a terminal non-reducing structure in human blood group Sda-active Tamm-Horsfall urinary glycoprotein. Biochem. Biophys. Res. Commun., 93, 1132–1139.[CrossRef]

Duk, M., Lisowska, E., Wu, J.H., and Wu, A.M. (1994) The biotin/avidin-mediated microtiter plate lectin assay with the use of chemically modified glycoprotein ligand. Anal. Biochem., 221, 266–272.[CrossRef][ISI][Medline]

Fischer, C., Sanchez-Ruderisch, H., Welzel, M., Wiedenmann, B., Sakai, T., André, S., Gabius, H.-J., Khachigian, L., Detjen, K.M., and Rosewicz, S. (2005) Galectin-1 interacts with the {alpha}5ß1 fibronectin receptor to restrict carcinoma cell growth via induction of p21 and p27. J. Biol. Chem., 280, 37266–37277.[Abstract/Free Full Text]

Fournet, B., Montreuil, J., Strecker, G., Dorland, L., Haverkamp, J., Vliegenthart, J.F.G., Binette, J.P., and Schmid, K. (1978) Determination of the primary structures of 16 asialo-carbohydrate units derived from human plasma {alpha}1-acid glycoprotein by 360-MHz 1H-NMR spectroscopy and permethylation analysis. Biochemistry, 17, 5206–5214.[CrossRef][Medline]

Fournier, T., Medjoubi, N.N., and Porquet, D. (2000) {alpha}1-acid glycoprotein. Biochim. Biophys. Acta, 1482, 157–171.[CrossRef][Medline]

Gabius, H.-J. (1990) Influence of type of linkage and spacer on the interaction of ß-galactoside-binding proteins with immobilized affinity ligands. Anal. Biochem., 189, 91–94.[CrossRef][ISI][Medline]

Gabius, H.-J. (1997) Animal lectins. Eur. J. Biochem., 243, 543–576.[ISI][Medline]

Gabius, H.-J. (2001) Glycohistochemistry: the why and how of detection and localization of endogenous lectins. Anat. Histol. Embryol., 30, 3–31.[CrossRef][ISI][Medline]

Gabius, H.-J. (2006) Cell surface glycans: the why and how of their functionality as biochemical signals in lectin-mediated information transfer. Crit. Rev. Immunol., 26, 43–79.[ISI][Medline]

Gabius, H.-J., André, S., Kaltner, H., and Siebert, H.-C. (2002) The sugar code: functional lectinomics. Biochim. Biophys. Acta, 1572, 165–177.[Medline]

Gabius, H.-J., Siebert, H.-C., André, S., Jiménez-Barbero, J., and Rüdiger, H. (2004) Chemical biology of the sugar code. Chembiochem, 5, 740–764.[CrossRef][Medline]

Gabius, H.-J., Wosgien, B., Hendrys, M., and Bardosi, A. (1991) Lectin localization in human nerve by biochemically defined lectin-binding glycoproteins, neoglycoprotein and lectin-specific antibody. Histochemistry, 95, 269–277.[ISI][Medline]

Gabius, H.-J. and Wu, A.M. (2006) The emerging functionality of endogenous lectins: a primer to the concept and a case study on galectins including medical implications. Chang Gung Med. J., 29, 37–62.[Medline]

Gitt, M.A., Wisers, M.F., Leffler, H., Herrmann, J., Xia, Y.-R., Massa, S.M., Cooper, D.N.W., Luis, A.J., and Barondes, S.H. (1995) Sequence and mapping of galectin-5, a ß-galactoside-binding lectin, found in rat erythrocytes. J. Biol. Chem., 270, 5032–5038.[Abstract/Free Full Text]

Gupta, D., Kaltner, H., Dong, X., Gabius, H.-J., and Brewer, C.F. (1996) Comparing cross-linking activities of lactose-specific plant and animal lectins and a natural lactose-binding immunoglobulin G fraction from human serum with asialofetuin. Glycobiology, 6, 843–849.[Abstract/Free Full Text]

Haltiwanger, R.S. and Lowe, J.B. (2004) Role of glycosylation in development. Annu. Rev. Biochem., 73, 491–537.[CrossRef][ISI][Medline]

He, L., André, S., Siebert, H.-C., Helmholz, H., Niemeyer, B., and Gabius, H.-J. (2003) Detection of ligand- and solvent-induced shape alterations of cell-growth-regulatory human lectin galectin-1 in solution by small angle neutron and X-ray scattering. Biophys. J., 85, 511–524.[Medline]

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.[Medline]

Herp, A., Wu, A.M., and Moschera, J. (1979) Current concepts of the structure and nature of mammalian salivary mucous glycoproteins. Mol. Cell. Biochem., 23, 27–44.[CrossRef][ISI][Medline]

Hirabayashi, J., Hashidate, T., Arata, Y., Nishi, N., Nakamura, T., Hirashima, M., Urashima, T., Oka, T., Futai, M., Müller, W.E.G., and others. (2002) Oligosaccharide specificity of galectins: a search by frontal affinity chromatography. Biochim. Biophys. Acta, 1572, 232–254.[Medline]

Hirabayashi, J. and Kasai, K.-i. (1993) The family of metazoan metal-independent ß-galactoside binding lectins: structure, function and molecular evolution. Glycobiology, 3, 297–304.[Abstract/Free Full Text]

Hirashima, M., Kashio, Y., Nishi, N., Yamauchi, A., Imaizumi, T.-a., Kageshita, T., Saita, N., and Nakamura, T. (2004) Galectin-9 in physiological and pathological conditions. Glycoconj. J., 19, 593–600.[CrossRef][ISI][Medline]

Iida, S.-i., Yamamoto, K., and Irimura, T. (1999) Interaction of human macrophage C-type lectin with O-linked N-acetylgalactosamine residues on mucin glycopeptides. J. Biol. Chem., 274, 10697–10705.[Abstract/Free Full Text]

Kabat, E.A. (1956) Blood Group Substances: Their Chemistry and Immunochemistry. Academic Press, NY, pp. 135–139.

Kabat, E.A., Baer, H., Bezer, A.E., and Knaub, V. (1948) Immunochemical studies on blood groups. VII. Chemical changes associated with destruction of blood group activity and enhancement of the type XIV cross reactivity by partial hydrolysis of hog and human blood group A, B, and O substances. J. Exp. Med., 88, 43–57.[Abstract]

Kaltner, H. and Stierstorfer, B. (1998) Animal lectins as cell adhesion molecules. Acta Anat., 161, 162–179.[CrossRef][ISI][Medline]

Kopitz, J., von Reitzenstein, C., André, S., Kaltner, H., Uhl, J., Ehemann, V., Cantz, M., and Gabius, H.-J. (2001) Negative regulation of neuroblastoma cell growth by carbohydrate-dependent surface binding of galectin-1 and functional divergence from galectin-3. J. Biol. Chem., 276, 35917–35923.[Abstract/Free Full Text]

Kopitz, J., von Reitzenstein, C., Burchert, M., Cantz, M., and Gabius, H.-J. (1998) Galectin-1 is a major receptor for ganglioside GM1, a product of the growth-controlling activity of a cell surface ganglioside sialidase, on human neuroblastoma cells in culture. J. Biol. Chem., 273, 11205–11211.[Abstract/Free Full Text]

Lahm, H., André, S., Hoeflich, A., Fischer, J.R., Sordat, B., Kaltner, H., Wolf, E., and Gabius, H.-J. (2001) Comprehensive galectin fingerprinting in a panel of 61 human tumor cell lines by RT-PCR and its implications for diagnostic and therapeutic procedures. J. Cancer Res. Clin. Oncol., 127, 375–386.[CrossRef][ISI][Medline]<