Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 (34)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Henrick, K.
Right arrow Articles by Hughes, R. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henrick, K.
Right arrow Articles by Hughes, R. C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Glycobiology Pages 45-57  

Evidence for subsites in the galectins involved in sugar binding at the nonreducing end of the central galactose of oligosaccharide ligands: sequence analysis, homology modeling and mutagenesis studies of hamster galectin-3
Introduction
Results and discussion
Materials and methods
References

Evidence for subsites in the galectins involved in sugar binding at the nonreducing end of the central galactose of oligosaccharide ligands: sequence analysis, homology modeling and mutagenesis studies of hamster galectin-3

Evidence for subsites in the galectins involved in sugar binding at the nonreducing end of the central galactose of oligosaccharide ligands: sequence analysis, homology modeling and mutagenesis studies of hamster galectin-3

Kim Henrick1, Sulemana Bawumia, Erminia A.M. Barboni2, Bruno Mehul3, R.Colin Hughes4

National Institute for Medical Research, Mill Hill, London NW7 1AA, United Kingdom

Received on April 17, 1997; revised on July 4, 1997; accepted on July 15, 1997

A model of the carbohydrate recognition domain CRD, residues 111-245, of hamster galectin-3 has been made using homology modeling and dynamics minimization methods. The model is based on the known x-ray structures of bovine galectin-1 and human galectin-2. The oligosaccharides NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc and GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc, known to be specific high-affinity ligands for galectin-3, as well as lactose recognized by all galectins were docked in the galectin-3 CRD model structure and a minimized binding conformation found in each case. These studies indicate a putative extended carbohydrate-binding subsite in the hamster galectin-3 involving Arg139, Glu230, and Ser232 for NeuNAc-[alpha]2,3-; Arg139 and Glu160 for fucose-[alpha]1,2-; and Arg139 and Ile141 for GalNAc-[alpha]1,3- substituents on the primary galactose. Each of these positions is variable within the whole galectin family. Two of these residues, Arg139 and Ser232, were selected for mutagenesis to probe their importance in this newly identified putative subsite. Residue 139 adopts main-chain dihedral angles characteristic of an isolated bridge structural feature, while residue 232 is the C-terminal residue of [beta]-strand-11, and is followed immediately by an inverse [gamma]-turn. A systematic series of mutant proteins have been prepared to represent the residue variation present in the aligned sequences of galectins-1, -2, and -3. Minimized docked models were generated for each mutant in complex with NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc, GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4- Glc, and Gal-[beta]1,4-Glc. Correlation of the computed protein-carbohydrate interaction energies for each lectin-oligosaccharide pair with the experimentally determined binding affinities for fetuin and asialofetuin or the relative potencies of lactose and sialyllactose in inhibiting binding to asiolofetuin is consistent with the postulated key importance of Arg139 in recognition of the extended sialylated ligand.

Key words: galectins/sugar binding sites/mutagenesis

Introduction

Galectins are a family of metal ion independent [beta]-galactoside binding lectins of unknown function (Barondes et al.). Previous site directed mutagenesis studies on human and bovine galectin-1 (Abbott and Feizi, 1991; Hirabayashi and Kasai, 1992, 1994), and x-ray crystallographic studies on bovine galectin-1 complexed with N-actetyllactosamine (Liao et al., 1994), human galectin-2 complexed with lactose (Lobsanov et al., 1993), and of bovine galectin-1 complexes with biantennary octa- and nonasaccharides (Bourne et al., 1994) have clearly established the lectin residues responsible for recognition of the galactose sugar. These residues are highly conserved for all the galectins that have been sequenced to date (for review see Hirabayashi and Kasai, 1993; Barondes et al., 1994; Hughes, 1994).

Several biochemical studies have demonstrated that there is a higher level of complexity involved for carbohydrate recognition by the galectins, than just the primary requirement of a central galactose sugar. Bovine (Abbott et al., 1988; Solomon et al., 1991; Ahmed et al., 1996), human (Ahmed et al., 1990; Lee et al., 1990), and rat (Leffler and Barondes, 1986) galectin-1, and galectin-3 homologs of rat (Leffler and Barondes, 1986), human (Sparrow et al., 1987; Feizi et al., 1994), and hamster (Sato and Hughes, 1992) have been examined in detail for carbohydrate binding specificity (Table I). There is a broadly similar pattern of affinities among the galectin-3 members, albeit with some species specific differences, but very significant differences to galectin-1 members. Both lectins bind to Type I Gal-[beta]1,3-GlcNAc and II Gal-[beta]1,4-GlcNAc chains. However, extension at the nonreducing end of the central disaccharide with GalNAc-[alpha]1,3- and Fuc-[alpha]1,2- units greatly enhances affinity to galectin-3 but has smaller effect or significantly reduces galectin-1 binding. Hence, human blood group A carbohydrate epitope GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc is a high affinity ligand for galectin-3 but is relatively poorly recognized by galectin-1. Similarly, NeuNAc-[alpha]2,3- or Fuc-[alpha]1,2- derivatives of lactose generally show increased binding to galectin-3 but reduced binding to galectin-1 (Table I). Overall, these data, together with other studies (Knibbs et al., 1993; Ramkumar et al., 1995), have suggested an extended binding site in galectin-3 that may accommodate at least a tri-or tetrasaccharide.

To investigate further the nature of this putative extended binding site, and in the absence of available crystal structures of galectin-3 with or without bound sugars, we have constructed a model of the carbohydrate recognition domain (CRD) residues 111-243 (Figure 1) of hamster galectin-3. The model was used in docking experiments with various extended sugars to identify amino acid residues likely to be involved in ligand binding additional to those well established for binding of simple [beta]-galactosides. Some key predictions have been tested by site-directed mutagenesis and binding assays using wild-type and mutant galectin-3.

Table I . Binding specificities of galectins; selected data for inhibition of binding to asialofetuin
  Galectin 1 Galectin 3
  Bovinec Humand Rate Hamsterf Humang Rate
Gal-[beta]1,4-Glc 1 1 1 1 1 1
NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc 0.4 0.6 0.3 (1.8)a 1.2 0.3
GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc 0 0.3 0.3 (4.5)b 32 25
Gal-[beta]1,4-GlcNAc 3.7 7.8 5 2 11.3 7
Fuc-[alpha]1,2-Gal-[beta]1,4-Glc 0.8 0.3 0.4 1.4 1.8 1.5
nd, Not determined. Unless specified otherwise inhibition is expressed as I50 of each compound relative to lactose I50.
aRelative value for (NeuNAc-[alpha]2,3)0 or 1 R.
bRelative value for (GalNAc-[alpha]1,3[Fuc-[alpha]1,2])0 or 1 R. R = Gal-[beta]1,3-GlcNAc-[beta]1,3-Gal-[beta]1,4-Glc.
cSolomon et al. (1991).
dAhmed, et al. (1990, 1996).
eLeffler and Barondes (1986).
fSato and Hughes (1992).
gSparrow et al. (1987).

Results and discussion

The hamster galectin 3 CRD model

As a starting point for the present work homology modeling of the CRD (residues 111-243) of hamster galectin-3 (Mehul et al., 1994) was carried out. The x-ray structures of bovine galectin-1 (Liao et al., 1994) and human galectin-2 (Lobsanov et al., 1993) consist of 12 [beta]-strands making an extended anti-parallel [beta]-sheet, which associates in a [beta]-sandwich motif to form symmetrical homodimers. For 79 residues in [beta]-strands 1-12 the structures superpose with RMS deviations of less than 0.92 and an overall RMS of 0.53 using the program Profit(Martin, 1995). From a homology modeling point of view the monomer is not complicated by the presence of disulfide linkages, cis-prolines, nor regions of extended random coil. Secondary structure for the CRD of hamster galectin-3 predicted from the sequence between residues 111 to 243 (Mehul et al., 1994) was generated using the programs Sopma (Geourjon and Deleage, 1994), Predator(Frishman and Argos, 1996), Nnpredict(Kneller et al., 1990), and PHDsec (Rost and Sander, 1994). A consensus prediction (Figure 1), prepared using Alscript(Barton, 1993), was in close agreement to the secondary structure assignments found for the x-ray structures of galectins-1 and -2 by use of the program Stride (Frishman and Argos, 1995). Following the structural comparison, the amino acid sequences of 19 galectin CRDs were aligned (Figure 2) using the program Clustalw(Thompson et al., 1994). Minor visual adjustments were then carried out to correlate the assigned x-ray and predicted topologies (Figure 1).


Figure 1 Alignment of the sequences of bovine galectin-1 and human galectin-2 with the carbohydrate recognition domain (CRD) of hamster galectin-3. Major secondary structure elements (t, turns; broad arrows, [beta]-strands 1-12) are shown as deduced from crystal structural analysis for galectins-1 and -2 and modeling for galectin-3 CRD as discussed in the text. Conserved residues are shown in boldface type.


Figure 2 Multiple sequence alignment of galectins. The complete sequences of galectin-1 and -2 and the CRD sequences of galectin-3, galectin-4, galectin-5, galectin-7, and galectin-8 are shown. Here we have defined chicken galectin-1 as the CL16 protein (Sakakura et al., 1990) and used galectin-1m (for monomer) to describe the CL14 protein (Ohyama and Kasai, 1988). This classification derives from their sequence identity at the residue 161 insertion-deletion region as described in the text. Residues in yellow are highly conserved; other partially conserved regions are depicted in pale blue, and the variable regions are shown in red. Residues 139 and 232 of the hamster galectin-3 sequence chosen for mutagenesis are shown in green.

The initial model of hamster galectin-3 CRD was then constructed using the Looksuite of programs, a set of tools that includes knowledge-based homology modeling based on the programs Segmod (Levitt, 1992), Cara(Lee and Subbiah, 1991; Lee, 1994), and Encad (Levitt, 1983). Input of the sequence alignment as discussed above, together with the bovine galectin-1 monomer coordinates (Liao et al., 1994) generated a minimized homology model with no manual intervention. In the case of loop regions, where galectin-3 differs from galectin-1, appropriate library structural fragments were found that allowed modeling of galectin-3 to the galectin-1 core [beta]-strand framework to be carried out without difficulty. In particular the structure of galectin-1 has several turns containing glycines with main chain dihedral angles that prohibit substitution with any other amino acid. However, the modeling selected suitable replacement fragments from the library of known structures. Lookuses a molecular mechanics procedure to generate a relaxed low energy structure and all polar and nonpolar hydrogen atoms are included in the minimization procedure (Powell, 1977). Inspection of the initial model using the program O (Jones et al., 1991) and checking the overall geometry using Procheck (Laskowski, et al., 1993) showed good stereochemistry with no residues with disallowed main chain torsion angles and no short intramolecular contacts. The model has fewer residues in ideal conformations compared to the x-ray structures. However, the parameters that have been used as indicators of chemical stereochemistry for the model are well within acceptable limits, i.e., the 'G-factors" (Morris et al., 1992) which should be above -0.5 are: dihedrals = -0.32, covalent = 0.16, and overall = -0.29.

Comparison of the galectin-3 model with the galectin-1 and galectin-2 monomers found in the x-ray structures showed only two significant conformational differences. Superposition shows strand-1 differs with an RMS of [sim]2.5 , but the rest of the [beta]-sheet structure is essentially equivalent. The insertion loop at position 161 (the numbering here and elsewhere is taken from the hamster galectin-3 sequence; see Figure 1) is different in all three structures and, as will be discussed later, this region does contribute to specificity of carbohydrate recognition. The second insertion/deletion region involving an Asn residue at position 222 is distant from the carbohydrate binding site and contains little information specific to sugar binding. The other differences are at position 124 and 211. For hamster galectin-3 the residues Arg124 (close to an insertion/deletion region) and Ala211 (within the turn from strand-9 to strand-10), are both glycine in galectin-1 and galectin-2 with dihedral angles ca. [Phi] 100 and [psi] -20°. The RMS deviation of 59 equivalent C-[alpha] atoms from strands 2 to 12 between galectin-1 and modeled galectin-3 CRD is 0.59 .

Table II . Synthetic oligonucleotides used for site-directed mutagenesis and sequencing
(i) Primers for PCR
Mutation Primer name Primer sequence Template
RR (S) MB06 5[prime]TGGCCATGGCAGACGGTTTTTCGC3[prime]  
  (AS) MB21 5[prime]GCGAATTCTTGAATCATGGTGGGTGCAGCACTGGTAAGGGTTATATCACCACGAATTTCC3[prime] L1
GR (S) MB06    
  (AS) MB21   GS
RG (S) MB06    
  (AS) MB46 5[prime]TAGCGAATTCTTAGATCATGGTGGGTGCAGCACTGGTAAGGGTTATATCACCACCAATTTCC3[prime] L1
GG (S) MB06    
  (AS) MB46   GS
SG (S) MB06    
  (AS) MB46   SS
SS (S) MB43 Phos-5[prime]CCCAATGCAAACAGCATTATTCTAAAT3'  
  (AS) MB44 5[prime]CTTCACTGTGCCCAT3' L1
SR (S) MB43    
  (AS) MB44   RR
GS (S) MB48 Phos-5[prime]CCCAATGCAAACGGGATTATTCTAAAT3'  
  (AS) MB44   L1
(ii) Partial sequences
  139 232
Wild type RS - - N A N (R) I I L - - - - E I (S) G D I- -
Single mutants GS G S
  SS S S
  RR R R
  RG R G
Double mutants GR G R
  GG G G
  SR S S
  SG S G
In (i) the wild type hamster galectin 3 cDNA plasmid template L1 containing Arg139 and Ser232 (named RS) was initially used for PCR using sense (S) MB06 and antisense (AS) MB21 oligonucleotide primers. The resulting PCR product RR contains the mutation Ser232->Arg (underlined). Other mutants were obtained and named similarly using the primers and templates as indicated.

Modeling of carbohydrate to protein interactions

In follow-up studies, various oligosaccharides were docked into the galectin-3 CRD modeled structure and a minimized binding conformation was found. For each binding model, a nonreducing end galactose was initially fixed at the position found in the crystal structures of galectins-1 and -2. For lactose binding to galectins, the conserved residues His153, Asn155, Arg157, Asn169, Trp176, and Glu179 are well known from the x-ray structures of galectins-1 and -2 to be critical (see Ahmed and Vesta, 1994, for review), and these make major contributions to the nominal total protein-carbohydrate interactions energies for hamster galectin-3CRD as computed from the modeling (Table III). The predicted roles in sugar-binding of these and additional significant residues are shown in Table IV. Previously, Arg181 has been suggested to be a factor in distinguishing between Glc and GlcNAc and also between Gal-[beta]1,4-GlcNAc and Gal-[beta]1,3-GlcNAc (Ahmed and Vasta,1994). However, we find for the modeled galectin-3 CRD that Arg181 is only weakly involved in direct sugar binding to Glc- O3, although it does play a crucial role in positioning Asn169 and Glu179: Lys or His at the equivalent position in other galectins (see Figure 2) accommodates the same role. Additional residues Gly177, Tyr216, and Arg219 also play significant roles in positioning the Asn169 side-chain and Arg 178 contributes to the conformation of residue Glu179 that is involved in direct glucose binding. In particular, the main chain carbonyl O-atom of Glu179 is influenced by the side chain of residue 178. Position 178, therefore, may exert indirect effects on the space available to a glucose residue and hence discriminate Gal-[beta]-1,3 or [beta]-1,4-Glc linkages, and possibly Type I and II Gal-[beta]1,3 or [beta]1,4-GlcNAc linkages. Galectin-1 sequences have much shorter side-chains, Ala or Thr, in this position that are unable to influence the main chain of Glu179.

Table III . Predicted protein-carbohydrate interaction energies (kcal/mol)
(i) Gal-[beta]1,4-Glc model complexes
  RS RR RG SR SG GG GR GS SS
Ser139 -0.1                
Arg139 -3.6 -1.1 -3.4            
Ile141 -0.3 -0.4 -0.2 -0.3 -0.3 -0.3 -0.4 -0.3 -0.3
Asn143 -0.5 -0.4 -0.3 -0.6 -0.4 -0.4 -0.5 -0.4 -0.4
His153 -3.7 -5.7 -2.2 -5.7 -2.1 -2.1 -5.6 -2.3 -2.1
Asn155 -1.0   -0.8 -0.6 -0.9 -0.9   -0.9 -0.9
Arg157 -2.8 -5.3 -2.4 -5.2 -2.4 -2.4 -5.4 -2.5 -2.3
Asn162 -0.3 -0.5 -0.3 -0.6 -0.3 -0.3 -0.5 -0.3 -0.3
Val167 -0.7 -1.2 -0.5 -1.1 -0.7 -0.7 -1.2 -0.7 -0.6
Asn169 -1.6 -1.5 -1.3 -1.7 -1.3 -1.3 -1.6 -1.4 -1.3
Trp176 -4.7 -4.2 -4.5 -4.4 -4.7 -4.7 -4.3 -4.7 -4.7
Glu179 -6.1 -5.6 -6.2 -5.6 -6.3 -6.2 -5.5 -6.2 -6.2
Arg181 -0.3 -0.8 -0.3 -0.7 -0.3 -0.3 -0.8 -0.2 -0.3
Arg232   -0.1         -0.1    
Totals -25.6 -26.9 -22.4 -26.8 -19.7 -19.6 -25.9 -19.9 -19.4
(ii) GalNAc-[alpha]1,3-[[alpha]1,2-Fuc]-Gal-[beta]1,4-Glc model complexes
  RS RR RG SR SG GS GG GR SS
Tyr113 -0.6               -0.5
Gly139               -0.2  
Ser139       -0.8 -0.5     -0.4  
Arg139 -10.3 -10.2 -10.7            
Ile141 -1.7 -1.4 -1.6 -1.4 -1.5 -1.5 -1.5 -1.4 -1.6
Asn143 -1.8 -2.4 -1.5 -2.5 -1.6 -1.8 -1.6 -2.5 -1.9
Ile150 -0.4 -0.4 -0.4 -0.4 -0.4 -0.4 -0.4 -0.4 -0.4
His153 -5.9 -6.1 -5.5 -6.5 -6.0 -6.4 -6.1 -6.1 -6.4
Asn155 -1.1 -1.0 -1.2 -0.8 -0.9 -0.7 -0.9 -1.0 -0.9
Arg157 -6.9 -7.6 -7.1 -7.5 -7.1 -8.6 -7.1 -7.9 -6.7
Glu160 -6.9 -7.1 -7.0 -6.3 -5.9 -7.1 -6.0 -7.1 -5.9
Asn161 -0.3 -0.3 -0.4 -0.3 -0.2 -0.4 -0.2 -0.4 -0.2
Asn162 -0.8 -1.2 -1.0 -1.7 -1.4 -1.5 -1.4 -1.2 -1.4
Val167 -0.9 -1.0 -0.9 -1.0 -0.9 -1.1 -0.9 -1.0 -0.9
Asn169 -1.6 -1.5 -1.4 -1.6 -1.7 -1.6 -1.7 -1.5 -1.7
Lys171   -0.1 -0.4 -0.2   -0.5 -0.4 -0.1  
Trp176 -5.1 -5.2 -5.1 -5.2 -5.3 -5.2 -5.3 -5.2 -5.3
Glu179 -5.6 -5.6 -5.4 -5.4 -5.5 -5.2 -5.1 -5.2 -5.3
Arg181 -0.8 -1.1 -0.8 -1.0 -0.8 -1.1 -0.8 -1.1 -0.8
Arg232   -3.5   -3.2       -3.3  
Ser232 -0.8               -0.6
Totals -51.5 -55.7 -50.4 -45.8 -39.7 -43.1 -39.6 -45.4 -40.3
(iii) NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc model complexes
  RS RRa RRa RG SR SG GG GR GS SS
Tyr113 -5.0 -0.4 -0.5 -2.5 -0.4 -4.8 -4.8 -0.7 -0.9 -0.9
Arg139 -2.1 -10.6 -5.1 -2.5            
Ser139         -0.4 -0.2        
Ile141 -2.9 -1.7 -1.7 -3.1 -1.7 -2.7 -2.7 -1.8 -1.3 -1.4
Asn143 -1.7 -0.9 -0.5 -1.9 -0.5 -1.7 -1.7 -1.0 -1.4 -1.2
Ile150 -0.3     -0.4   -0.3 -0.3 -0.1 -0.6 -0.6
His153 -5.2 -6.3 -2.1 -5.4 -1.9 -5.3 -5.3 -5.4 -1.3 -1.3
Asn155 -1.5 -0.3 -3.3 -1.2 -4.0 -1.4 -1.4 -0.1 -1.2 -1.2
Arg157 -4.5 -6.1 -4.5 -4.9 -5.0 -4.9 -4.9 -5.7 -2.2 -2.2
Glu160 -0.5 -0.3 -0.6 -0.5 -0.4 -0.5 -0.5      
Asn162 -0.3 -0.5 -0.3 -0.4 -0.3 -0.3 -0.3 -1.5 -0.3 -0.3
Val167 -1.0 -1.2 -0.8 -1.0 -0.8 -1.0 -1.0 -0.2 -0.4 -0.4
Asn169 -1.5 -1.7 -0.9 -1.5 -0.9 -1.5 -1.5 -1.6 -0.6 -0.6
Lys171 -1.2 -1.1 -0.1 -1.3 -0.1 -1.2 -1.2 -1.1 -1.2 -0.7
Trp176 -5.9 -5.2 -4.8 -6.1 -4.6 -5.9 -5.9 -5.1 -4.9 -4.9
Glu179 -6.1 -5.0 -5.9 -6.0 -6.0 -5.9 -6.0 -5.7 -6.5 -6.5
Arg181 -0.5 -0.8 -0.5 -0.6 -0.5 -0.5 -0.5 -0.8 -0.3 -0.3
Glu230 -1.3     -0.8   -0.9 -0.9   -2.3 -2.2
Ser232 -3.6               -1.4 -1.3
Arg232   -2.8 -10.1   -10.1     -2.8    
Gly233 -0.4 -0.2 -0.5   -0.5 -0.2 -0.2 -0.3    
Asp234                    
Totalc -45.5 -45.3 -42.3 -40.1 -38.1 -39.2 -39.1 -33.9 -26.8 -26.0
aFor the RR mutant two near equal conformations (R1 and R2) were found.

The Trp176 in strand-6 essential for galactose binding (Hirabayashi and Kasai, 1992, 1994) follows the glycine dihedral-specific turn in the x-ray structures between strands 5 and 6. The turn for residues 172-175 is different in galectins-1, -2, and -3 and is likely to position the Trp176 side chain with a slightly different orientation towards the carbohydrate. This tight turn contains a glycine with specific dihedrals in an area populated mainly by glycine in galectin-1 (sequence DAGA) and galectin-2 (sequence DGSN). The galectin-3 sequence is QDNN. In galectin-1 this is a Type I[prime] turn and for galectin-2 it is a Type IV turn as found using the program Promotif (Wilmot and Thornton, 1990; Hutchinson and Thornton, 1996). Therefore, this turn sequence may also influence the selection of either a Gal-[beta]-1,3- or [beta]-1,4-GlcNAc structure. For the superimposed structures derived as above, the Trp176 C-[alpha] is moved 0.75 in comparing galectin-1 to galectin-2, and 0.67 between galectin-1 and -3.

A putative extended binding site in modeled galectin-3 CRD

It was evident from an examination of the galectin-3 CRD model that there was an extension to the galactose-binding site that lies further inside the folded protein at the nonreducing end of the bound lactose. In docking studies using the tri-saccharide NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc and tetrasaccharide GalNac-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc, models were found to show very strong binding to each nonreducing terminal sugar group. Figure 3 shows the modeled galectin-3 CRD complexed with GalNac-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc (Figure 3A) and NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc (Figure 3B), and highlights the known lactose binding residues (Hirabayashi and Kasai, 1992, 1994) and the extra sugar-binding amino acid residues proposed from this work. Table IV gives a summary of the predicted role for each amino acid in the new putative subsite involved in sugar binding by the modeled galectin-3 CRD as deduced from this work.


Figure 3 Space filling models of A blood group tetrasaccharide GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal[beta]1,4-Glc (A) and NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc (B) to wild type (RS) galectin-3 CRD. Oligosaccharides are in red. Conserved binding residues are shown in brown and mutant positions 139 and 232 in yellow. Other important binding residues are shown in green (160 and 230) and blue (residues 161, 162, 171, and 216). See text and Table IV for further discussion.

The proposed fucose subsite in binding of GalNac-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc is flanked by Glu160 and Arg139. Glu160 is conserved in all the galectins-3 and makes major contributions to binding energies in docking of fucosylated ligand: galectin-1 has Ala or Cys at this position; and although galectin-2 also has Glu160 this is followed by a deletion after residue 161, and therefore Glu160 does not have the same position in the subsite as that modeled for galectin-3. In the galectin-3 CRD model Asn162 has a structural role in positioning of Glu160 and makes weaker interactions with Arg181. Hence the variable insertion-deletion region of the galectins, delineated by Asn residues 161 and 162 in galectin-3 CRD (Figure 1), is likely to influence binding specificity. Major contacts with the GalNAc are predicted to involve Arg139 and Lys171; the latter may have a water-mediated interaction with the GalNAc-N2,O7 atoms. This Lys is conserved for both galectin-1 and -3 but is a Leu in galectin-2. In addition, there is a van der Waals contact between the side chain of Ile141 and the C3 of GalNAc that imposes stereochemical constraints in the binding pocket.

Two important residues of the modeled galectin-3 CRD directly interacting with a NeuNAc-[alpha]2,3- unit are predicted to be Arg139 and Ser 232 (Table IV). Arg139 is well-conserved in all galectins-3 and is either Ser or Gly in galectins-1 and -2, respectively. Ser 232 is also well- conserved in galectins-3 but is variously Gly or Asp in galectins-1 and Arg in human galectin-2. Additionally, Glu230 could contribute significantly to the binding energies for NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc and have a positioning role for the side-chain of Ser 232.

Modeling of oligosaccharide interactions with mutant proteins

In order to test the quality of the predictions deduced from oligosaccharide docking into the galectin-3 CRD model, the wild-type modeled CRD was modified at specific sites. We chose to replace Arg139 with Ser or Gly and Ser 232 with Gly and Arg, residues that are found in equivalent positions in galectins-1 or -2 (Figure 1). Inspection of the crystal structures of the solved galectin molecules showed that these residues are outside insertion-deletion regions within the aligned set of known galectins (Figure 2) and hence could be investigated by mutagenesis without additional complication, to determine their possible effects upon carbohydrate binding within the galectin series. Single and double mutant hamster galectin-3 models were generated with Lookfrom the wild-type model and minimized (see Table II for nomenclature) and oligosaccharide-docking was performed. Table III gives the computed individual energy contributions between mutant proteins and carbohydrate for each of the oligosaccharides examined.

Certain of the mutations substituted well in the modeling for the wild-type residues in forming the sugar-binding fold or in sugar binding. For example, a Ser139 positions Glu160 in similar fashion as an Arg 139, through Ser-OG -Glu160-OE2 interactions. Similarly, an Arg 232 residue as in mutant RR adopts a conformation that is hydrogen-bonded to Asn 143, adjusting Asn143 with respect to His 153. In addition, it interacts with Glu 230 through Arg 232-NE-Glu 230-OE1 contacts and forms hydrogen-bonding to GalNAc-O3,O7, NeuNAC-O8,O10,O1A and weakly with Gal-O3 contributing significantly to sugar binding. These and other interactions involving mutated residues are included in Table IV. Figure 4A, drawn with Molscript(Kraulis 1991) and Raster3d(Merritt and Murphy 1994), illustrates aspects of the interactions between the RR mutant and GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc. For the RR-galectin-3 complex with NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc, two equally stable conformations termed R1 and R2 were found. In the first the NeuNAc group is directed toward Arg139 whereas in the second Arg232 contributes more to the binding (Figure 4B).

Table IV . Binding role for amino acid residues in the carbohydrate binding pockets of galectins
(i) Conserved residues for mammalian galectins (see Figure 2 for sequences)
Residue Role Interactions Residue variations
Gly 131 Structural    
Asn 143 Positioning of sugar residues NeuNAc-O4,O10 wea Asn143-OD1...His153-ND1 Sometimes Asp
His 153 Gal-O4 H-bond to Asn143  
Asn 155 Gal-O4 Asn155-OD1...Arg157-NE
Asn155-ND2...Ile140-O		  
Pro 156 Structural    
Arg 157 Gal-O4,O5 secondary role to Glc-O3,O4 weak hydrophobic CZ packs to Fuc-C6 H-bond to Asn155
Arg157-NE...Glu160-OE2
Arg157-NH1...Asn162-OD1
Arg157-NH2...Glu179-OE2		  
Val 167 Hydrophobic contribution    
Asn 169 Gal-O6 Asn169-OD1...Arg219-NH2
Tyr216-OH...Asn169-O
Asn169-OD1...Gly177-N
Asn169-ND2...Gly177-O		  
Trp 176 Hydrophobic packing to Gal-C4 controls acceptable stereochemistry of sugar    
Gly 177 Structural with main chain positioning of Asn169    
Glu 179 GlcNAc-N2
Glc/GlcNAc-O3
Glc-O2
Gal-O6		 Held in place by H-bond to
Arg157 and Arg181
Glu179-OE1...Arg181-NH1,NH2
Glu179-O...Arg178-NE,NH2		 Position 178 is variable
can be A,T,S,P,Q,R,K,E
Arg 181 Glc-O3 weak positioning role H-bonds to Asn169 and Glu179 Sometimes Lys
Phe 187 Structural packing role within binding cavity    
Gly 190 Structural can only be a Gly from dihedrals    
Tyr 216 Structural Tyr216-OH...Asn169-O Sometimes Phe
Arg 219 Structural Holds Asn 169 side chain in place  
Gly 233 Structural can only be a Gly from dihedrals    
(ii) Residues binding to complex sugars at proposed new subsite
Residue Role Interactions Residue variations
Arg 139 Arg - GalNAc-O6
Fuc-O4
NeuNAc-O6,O7,O9,O1B		 Arg-NH2/NE interaction with Glu160 and
Asp234
Arg139-NE,NH2...Glu160-OE2
Arg139-NH2...Asp234-OD2		 S, G, R, N
Ile 141 Hydrophobic packing
to GalNAc-C3		   L,V,I,A,H,Q,S,N
Always hydrophobic in galectin-3
Ile 150 Small contributions to the binding   L,I,V,A but always hydrophobic
Glu 160 Fuc-O4 Interacts with Arg 139 weak with Arg157 and
Asn162
Glu160-OE1...Asn162-ND2		 Variable in an insert/deletion region
 
 
Always Glu in galectin-3
Asn 162 Structural Positioning of Glu160
 
Weak interaction with Arg181 and Arg157
Asn162-OD1...Arg181-NH2		 together with Asn161 this is an insert-deletion region
Asn161 may be H,S,N,D,K.
Asn162 may be D,N.
Lys 171 May contribute to the potential field of the binding site No direct role in binding K,L,M,Y,Q,Falways Lys in galectin- 1 & galectin-3, Leu in galectin-2
Glu 230 NeuNAc-O10 Positioning role for Ser232 side chain, Variable S,A,E,G,D,Q usually Gly in galectin-3
Ser 232 Ser->NeuNAc-O8
Arg->GalNAc-O3,O7
->NeuNAc-O8,
     O10,O1A
->Gal-O3 weak		 Ser positioned by Glu230
Ser232-OG...Glu230-OE1
Arg232-NE...Glu230-OE1		 G D R S I H D K A E

Carbohydrate geometry in the modeled galectin-3 CRDs


Figure 4 Outlines of potential ligand binding sites in hamster galectin-3 CRD mutants. (A) The complete binding pocket for mutant-RR complexed with GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc. The lectin-galactose interactions shown are essentially the same for all combinations of mutant lectin and carbohydrate that were modeled. (B) A partial composite schematic of the interactions found for lectin mutants to the sialyl group of NeuNAc-[alpha]2,3- Gal-[beta]1,4-Glc. The binding contributions for an Arg at positions 139 and 232 are shown together with the potential hydrogen bonding between a Ser at position 139. Hamster galectin-3 modeling also showed that for sialyllactose, Asn143 and Glu230 potentially play a significant role in binding.

The orientation of the Gal-[beta]1,4-Glc remained almost unchanged during optimization for all combinations of mutant protein and carbohydrate tested. The glycosidic linkages were allowed free rotation during minimization. Apart from the RR-mutant complexed with NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc,the conformations of bound oligosaccharides generated in the modeling were close to the starting values. The Gal-[beta]1,4-Glc linkage has torsion angles marginally different to that observed in the x-ray galectin structures but well within the minima described for this linkage type (Imberty et al., 1991). In all cases the minimized complexes contain sugar linkages that show [Phi] and [Psi] angles which do not correspond to previously calculated lowest energy conformations in isolated sugars. Calculations exploring the conformational space available for isolated polysaccharides frequently give lowest energy conformations that have contributions from intramolecular H-bonds (Imberty et al., 1995). Protein/sugar complexes are expected to replace these hydrogen bonds by intermolecular contacts. The sugar geometries found in our present study were within allowed conformational states and generated no steric conflicts. The Gal-[beta]1,4-Glc [Phi], O(5)-C(1)-O(1)-C(4[prime]) and [Psi], C(1)-O(1)-C(4[prime])-C(5[prime]) averaged angles were -70.7° and -129.9°, respectively. These values lie in an allowed region close to one set of calculated minima, at [Phi] and [Psi] values of -75.9° and -115.4°, found for the A type 2 antigen (Imberty et al., 1995). The Fuc-[alpha]1,2-Gal linkage was found to have average values of 81.8° and 132.2° for [Phi], O(5)-C(1)-O(1)-C(2[prime]) and [Psi], C(1)-O(1)-C(2[prime])-C(3[prime]), respectively, which differs from other reports. For the isolated blood group A trisaccharide, GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]-OMe, a molecular mechanics study gave low energy conformations for the Fuc-Gal linkage at either [Phi], [Psi] of -69.5°, -93.5° or -94.9°, -169.5° (Imberty et al., 1995). The first calculated minima is close to that observed for the complex of Lewis-B blood group determinant and GS-IV lectin (L. Delbaere et al., PDB entry ILED) where a conformation of [Phi] -65° and [Psi] -100° was found for Fuc-[alpha]1,2-Gal. The blood group A trisaccharide has also been docked into a model of the Dolichos biflorus seed lectin and here, the calculated minima conformations gave stable complexes (Imberty et al., 1994). The GalNAc-[alpha]1,3-Gal linkage in the galectin-3 model complexes averaged values of [Phi] 101.9° and [Psi] 161.9°, close to the reported calculated low-energy of type-B structures (Imberty et al., 1995). The combined linkage angles for the docked GalNAc-[alpha]1,3-[Fuc-[alpha]1,2]-Gal-[beta]1,4-Glc sugar do not match any of the three calculated conformational families for this A type 2 tetrasaccharide, calculated for the [beta]-OMe glycoside (Imberty et al., 1995). The differences could arise from the interactions with the protein surface of the tetrasaccharide, compared with the conformational space available to the molecule in solution.

For both the Fuc and GalNAc substituted sugars essentially one conformation was found. By contrast the NeuNAc-[alpha]2,3-Gal-[beta]1,4-Glc model complexes fell into three conformational groups with different values for [Phi], O(6)-C(2)-O(3[prime])-C(3[prime]) and [Psi], C(2)-O(3[prime])-C(3[prime])-C(4[prime]). For the RS, RG, SG, GG, GS, and SS mutants the respective values are 171.5° and 120.5°; for the RR-2 and SR mutants, -167.0° and 164.5°, and for the RR-1 and GR mutants, 67.0° and 172.0°. Here RR-1 and RR-2 refer to the alternative conformations of the sialyllactose in the RR mutant. In each case the differences are generated by competing interactions between the two possible (Arg139, 232) side chains.

Mutant binding to sialylated and nonsialylated ligands

We determined next experimentally the binding properties of wild type and eight mutants (Table II) of hamster galectin-3. We used two separate binding assays: direct binding of tritiated fetuin or asialofetuin to recombinant wild-type and mutant proteins, and inhibition by lactose or NeuNAc-[alpha]2,3-lactose of wild-type or mutant galectin-3 binding to asialofetuin. Fetuin contains three N-linked oligosaccharides, 83% of which are triantennary glycans containing Gal-[beta]1,4-GlcNAc termini bearing variable amounts of NeuNAc-[alpha]2,3- linked residues (Green et al., 1988).

Binding isotherms of wild type RS galectin-3 to fetuin gave a biphasic curve representing high affinity binding to NeuNAc-[alpha]2,3-Gal-[beta]1,4-GlcNAc termini and low affinity binding to unsialylated Gal-[beta]1,4-GlcNAc termini (Figure 5, Table V). A linear Scatchard plot was obtained for asialofetuin. The arginine in position 139 induces a preferential binding for sialic acid residues which is lost in the mutants SS and GS as reflected by a linear Scatchard plot for binding to fetuin. Evidently, serine at position 232 can be replaced by glycine or arginine (mutants RG and RR) without affecting the preference for NeuNAc-[alpha]2,3-Gal-[beta]1,4-GlcNAc structures. The behavior of double mutants is more difficult to interpret. However, it seems that a glycine at position 232 can rescue to some extent the preferential binding of sialylated structures in position 139 mutants, as in GG and SG mutant proteins. Arginine at position 232 is also able to accomplish this rescue for mutant GS, as in mutant GR (Table V). Interestingly, however, a positional switch of arginine and serine at positions 139 and 232, mutant SR, abolishes preferential binding to sialylated structures. The data from inhibition of binding assays obtained using NeuNAc-[alpha]2,3-lactose and lactose are broadly in agreement with direct binding assays for the single mutants. Single mutations at serine 232 have no or little effect on the relative binding of sialyl lactose and lactose whereas an arginine at position 139 is crucial for the discrimination between the two oligosaccharides. Similarly, a glycine at position 232 increases the selectivity of position 139 mutants for sialyl lactose 5- to 10-fold while arginine at position 232 converts to lesser extent position 139 mutants to selective lectins. . Representative Scatchard plots of the direct binding of fetuin (a, b) and asialofetuin (c) to galectin-3 and its single site amino acid mutants. See Table II for the mutations introduced at amino acids residues 139 and 232. Each binding isotherm was repeated three times using duplicate samples. For each data point SD < 10%

Comparison of theoretical and experimental binding data

The mutagenesis and binding data support generally the predictions from modeling studies concerning the importance of Arg139 residues in conferring a galectin-3 like carbohydrate binding specificity. The anomalous behavior of rat galectin-3 in this regard (Table I) is likely due to replacement of Arg139 and Ser 232 by Ser and Ile residues (Figure 2). However, the relative total protein-carbohydrate interaction energies computed for the model complexes (Table III) do not completely correlate with either the observed inhibition data or with the affinity coefficients (Table V). These differences between the models and the true protein complexes may possibly be attributed to solvent contributions (Lemieux et al., 1991; Lemieux, 1994) and to other approximations inherent in the docking and computational procedures. Within the set of mutant sugar complexes described here, the role of water was assumed as a first approximation to be constant.

Table V . Binding properties of wild type and mutant galectin 3
  Oligosaccharide inhibitor Glycoprotein substrate [Delta]kcal/mold
Lectina Lactose
1/KIb mM-1		 3[prime]SL ASF
Kac µM-1		 Fetuin  
        I II  
RS 1.0 3.3 0.6 0.6 6.3 -19.9
RR 1.3 5.8 0.7 0.8 4.3 -18.4 (R1), -15.4 (R2)
RG 1.4 7.7 0.7 0.8 3.1 -17.7
SS 1.1 1.1 1.0 1.2 - -6.6
GS 0.7 0.2 1.4 1.0 - -6.9
SR 1.1 1.7 0.5 0.4 - -11.3
GR 0.6 1.3 1.0 1.1 4.5 -8.0
GG 1.1 5.0 0.4 0.5 3.3 -19.5
SG 1.5 10.0 0.4 0.9 9.2 -19.5
aSee Table II for nomenclature.
bKI, Concentration required to give 50% inhibition of binding of lectin to asialofetuin-coated plates. 3[prime]SL = 3[prime]-sialylactose.
cKa, Affinity coefficients calculated from the direct binding of fetuin and asialofetuin to immobilized lectin. I and II, low and high affinity constants, respectively.
d[Delta], Differences in modeled interaction energies for 3[prime]sialylactose and lactose from Table III.

The model interaction energies between protein and NeuNAc-[alpha]2,3-lactose do indicate the RS and RR proteins to have the highest and the proteins GS and SS the lowest affinity, in agreement with the affinities found experimentally for fetuin. For these two extremes the computed model energies show about 19 kcal/mol difference in stability between the RR/RS and GS/SS complexes. The relative order of model structures to the measured affinities are however not in agreement. The observed Ka values towards fetuin are in the order: SG = RS > GR = RR > GG >RG >> SS = GS = SR while the modeled interaction energies towards 3-sialyl-lactose are RS = RR > RG = GG = SR = SG > GR >SS = GS. However, there is a stronger correlation between the inhibition and binding studies and the modeled interaction energies when one examines the relative preferences for each protein towards sialyl-lactose over lactose (Table V). The group of proteins containing Arg at position 139, i.e., RS, RR, and RG show a relatively large increase of binding energy when sialyl-lactose is modeled in the extended binding pocket rather than lactose which is in agreement with the experimental binding studies. The modeled structures for the SS and GS proteins show only a modest increase in binding interactions for sialyl lactose over lactose, while the proteins SR and GR show an intermediate increased preference for sialyl substitution. For the model proteins of GG and SG neither of their lactose nor sialyl-lactose complexes show the highest interaction energy within each series of sugar to protein complex. However, they both show the highest increase in computed binding energy for a sialyl group, indicating a strong preference for 3-sialyl-lactose over lactose that is reflected in the relative KI and Kavalues observed (Table V).

Concluding remarks

Taken together, these data show that modeling and oligosaccharide docking studies can provide valuable insights into the carbohydrate-binding specificities of the galectins and point to promising targets for more extensive mutagenesis of amino acid residues likely to be contact sites for binding of extended ligands by the galectin-3 CRD.

Previous studies (Ahmed and Vasta, 1994) have identified two broad types of CRDs in the galectins, based on differences in amino acid residues at key positions and the relative binding efficiencies (Table ) of key oligosaccharides including lactose (N-acetyllactosamine), NeuNAc-[alpha]2,3-lactose (N-acetyl-lactosamine), and the A-tetrasaccharide GalNAc-[beta]1,3-[Fuc-[alpha]1,2]-lactose (N-acetyl-lactosamine). These authors suggested that substitution or deletion of residues at positions 161 and 162 of Type II CRDs such as galectin-3 is responsible for the ability of these CRDs to accommodate bulky nonreducing sugar residues. By contrast, we propose a more positive role for this region, especially Asn 162, in positioning key residues in the sugar-binding pocket. In galectin-2 this region is extensively deleted and the equivalent galectin-3 region, i.e., Asn161-Asn162 must have different orientations of these side-chains compared to the galectin-1 CRD which has the His.Gly.Asp sequence at this equivalent position. Other workers (Solis et al., 1996) in a study of two chicken galectins, galectins-1 and -1m (see Figure 2 for explanation of nomenclature) have proposed residue 165, Leu57 or Thr57 in chicken galectins-1m and -1, respectively (Figure 2), as a possible residue that can fine-tune binding specificity. However, we find in the modeled structures that position 165 is not significant; the observed differences between chicken galectin-1m (CL14) (Ohyama and Kasai, 1988) and chicken galectin-1 (CL16) (Sakakura et al., 1990) could arise from a combination of their differences at residues 139, 141, 160, 230 and 232 and in the pre-Trp176 turn. Thus, they differ at every point we propose as additional sugar binding residues.

Materials and methods

Carbohydrate model binding

Starting coordinates for the sugars were taken from the small molecule x-ray crystal structure of lactose (Noordik et al., 1984) with torsion angles [Phi], O(5)-C(1)-O(4[prime])-C(4[prime]) -93.5 and [Psi], C(1)-O(4[prime])-C(4[prime])-C(5[prime]) -143.5°. A complex was generated by first fixing the galactose ring for each sugar in the position defined by the x-ray structure of the lactose-galectin-1 complex, i.e., stacking the apolar face of the galactose to the conserved aromatic Trp176. Sugar units were built on to the lactose using the program spartan (Hehre and Huang, 1995). Initial docked positions were found manually using interactive graphics. The space available for the additional nonreducing end sugars is restricted and starting positions were obvious. Automatic docking using the program Dock (Meng et al., 1992, 1993) gave sugar positions that were almost identical to the manually found positions. The potential energy of the entire system was evaluated using the function and energy parameters given in Encad for protein, and energy parameters for carbohydrate atom types appropriate for protein/carbohydrate complexes (Imberty et al., 1991). The process was then repeated for each sugar/mutant complex.

Expression plasmids

The cDNA (clone L1) encoding the full-length sequence of hamster galectin-3 (Mehul et al., 1994) was mutated at positions 139 and 232 by site-specific PCR-mediated mutagenesis using the plasmid templates and primers indicated in Table II. The starting L1 plasmid was constructed (Mehul et al., 1994) for expression of wild type galectin-3 and its secretion into the periplasmic space of transformed bacteria. PCR used to introduce mutations at position 232 was performed using appropriate templates (Table II) and Amplitaq DNA polymerase as follows: one cycle at 93°C for 3 min, 30 cycles at 93°C for 30 s, 45°C for 1 min, and 72°C for 2 min, and 1 cycle for 5 min at 72°C. PCR products were digested by NCoI and EcoRI then ligated into the PTM-N vector as described previously (Mehul et al., 1994). PCR to mutate the Arg139 to Ser (clones SS and SR) was performed using appropriate templates (Table II) and a mixture of Amplitaq DNA polymerase and Taq-extender (Stratagene) as follows: one cycle at 93°C for 3 min, 30 cycles at 93°C for 30 s, 44°C for 1 min and 72°C for 2 min, and 1 cycle for 10 min at 72°C. PCR to mutate the Arg139 to Gly (clone GS) was performed using rTth DNA polymerase (GeneAmp. XL PCR Kit, Perkin Elmer) and the following conditions; one cycle at 93°C for 2 min, 16 cycles at 93°C for 30 s and 58°C for 10 min, 12 cycles at 93°C for 30 s and 58°C for 10 min (increased by 15 s per cycle) and 1 cycle for 10 min. at 72°C. These PCR products were incubated with DpnI and pfu DNA polymerase to digest template DNA plasmid and blunt end the PCR products. After agarose-gel purification, ligation and transformation of E.coli BL21(DE3)(pLysS) was performed as described previously (Mehul et al., 1994). Nucleotide sequences were verified by the dideoxy-chain termination method using a Sequenase Kit, version 2.0 (U.S. Biochemical Corp.) to ensure proper mutagenesis and the absence of second-site mutations. Purification of the galectin-3 mutants was carried out by affinity chromatography on lactose-Agarose as described for the wild type galectin-3 (Mehul et al., 1994). Purity was verified by SDS-PAGE and Western blotting with specific antibodies as described previously (Mehul et al., 1994).

Binding assay

Fetuin and asialofetuin (10 mg) were tritiated with 1 mCi [H3]-acetic anhydride (Amersham, 500 mCi/mmol) at pH 8.5, in carbonate buffer (0.5 ml), for 3 h at 4°C, and then gel-filtered on Biogel P60. The resulting specific activity was 2-4 × 103 d.p.m./pmol. Protein A Sepharose CL-4B (Pharmacia Biotech.) was cross-linked to antibodies raised against N-terminal domain epitopes of hamster galectin-3 (Mehul et al., 1994) as follows: beads (1 ml) were incubated with antiserum (3 ml) for 60 min at room temp in PBS, washed three times in PBS, resuspended in borate buffer pH 9 (3 ml), and cross-linked by addition of dimethylpimelidate according to the manufacturer's instructions (Pierce, Rockland, IL). Beads (1 ml packed volume) were then mixed with galectin-3 solutions (1 mg) in PBS (1 ml) and kept at room temp. for 1-2 h. For direct binding assays, aliquots (50 µl) of the washed bead suspension (20% in PBS, EDTA 5 mM) were incubated with known amounts of labeled fetuin or asialofetuin (0.01 - 20 µM) in a total volume of 100 µl, at 23°C for 2-3 h. For the determination of nonspecific binding, the mixture contained an excess (200 µM) of unlabeled ligand. After centrifugation, the supernatant was tested for unbound radioactivity. The beads were washed four times with ice-cold phosphate buffer and the bound ligand was released with 0.2 M lactose (2 h at 23°C) and counted. The affinity constants were calculated from the Scatchard plots of the binding isotherms.

Inhibition assay

Microtiter wells of Nunc-[Igr]mmunoplates were coated with asialofetuin ( 0.1 µg/well) at 37°C for 60 min in 0.01 M carbonate buffer pH 9.6. After blocking with 2% of bovine serum albumin, fixed amounts of lectin (40 µl, 10-50 µg/ml) were added in the presence of varying concentrations(from 0.01 to 20 mM) of lactose or 3[prime]-sialyllactose (Dexra Laboratories) and incubated for 2.5 h at 23°C. The plates were washed four or five times with ice-cold 5 mM Tris-HCl pH 7.2, 15 mM NaCl, 0.1% Nonidet P40 buffer. Bound lectin was determined by a double-antibody method, using rabbit polyclonal antiserum directed against N-terminal domain epitopes of hamster galectin-3 (Mehul et al., 1994) as the primary antibody, and goat anti-rabbit IgG conjugated to alkaline phosphatase as the secondary antibody. The plates were read at 405 nm on a Titertek Multiscan enzyme linked immunosorbent assay reader.

References

Abbott,W.M. and Feizi,T. (1991) Soluble 14 kDa [beta]-galactoside binding specific bovine lectin: evidence from mutagenesis and proteolysis that almost the complete polypeptide chain is necessary for integrity of the carbohydrate recognition domain. J. Biol. Chem., 266, 5552-5557. MEDLINE Abstract

Abbott,W.M., Hounsell,E.F. and Feizi,T. (1988) Further studies of oligosaccharide recognition by the soluble 13 kda lectin of bovine heart-muscle-ability to accommodate the blood group-H and blood group-B related sequences. Biochem. J., 252, 283-287. MEDLINE Abstract

Ahmed,H. and Vasta,G.R. (1994) Galectins: conservation of functionally and structurally relevant amino acid residues defines two types of carbohydrate recognition domains. Glycobiology, 4, 545-549. MEDLINE Abstract

Ahmed,H., Allen,H.J., Sharma,A. and Matta,K.L. (1990) Human splenic galaptin carbohydrate-binding specificity and characterization of the combining site. Biochemistry, 29, 5315-5319. MEDLINE Abstract

Ahmed,H., Fink,N.E., Pohl,Y. and Vasat,G.R. (1996) Galectin 1 from bovine spleen; biochemical characterisation, carbohydrate specificity and tissue-specific isoform profiles. J. Biochem., 120, 1007-1019. MEDLINE Abstract

Barondes,S.H., Castronovo,V., Cooper,D.N.W., Cummings,R.D., Drickamer,K., Feizi,T., Gitt,M.A., Hirabayashi,J., Hughes,R.C., Kasai,K., Leffler,H., Liu,F.-T., Lotan,R., Mercurio,A.M., Monsigny,M., Pilli,S., Poirer,F., Raz,A., Rigby,P.W.J., Rini,J.M. and Wang,J.L. (1994a) Galectins:A family of animal [beta]-galactoside-binding lectins. Cell, 76, 597-598. MEDLINE Abstract

Barondes,S.H., Cooper,D.N.W., Gitt,M.A. and Leffler,H. (1994b) Galectins. Structure and function of a large family of animal lectins. J. Biol. Chem., 269, 20807-20810. MEDLINE Abstract

Barton,G. J. (1993) ALSCRIPT, a tool to format multiple sequence alignments. Protein Eng., 6, 37-40. MEDLINE Abstract

Bourne,Y., Rouge,P. and Cambillau,C. (1990) X-Ray structure of a ([alpha]-Man(1,3)-[beta]-Man(1,4)GlcNAc)-lectin complex at 2.1 resolution. The role of water in sugar lectin interaction. J. Biol. Chem., 265, 18161-18165. MEDLINE Abstract

Bourne,Y., Bolgiano,B., Liao,D.-I., Strecker,G., Cantau,P., Herzberg,O., Feizi,T. and Cambillau,C. (1994) Crosslinking of mammalian lectin (galectin-1) by complex biantennary saccharides. Nature Struct. Biol., 1, 863-870.

Feizi,T., Solomon,J.C., Yuen,C.T., Jeng,K.C.G., Frigeri,L.G., Hsu,D.K. and Liu,F.T. (1994) The adhesive specificity of the soluble human lectin, IgE-binding protein, toward lipid-linked oligosaccharides-presence of the blood group-A, group-B, group-B-like, and group-H monosaccharides confers a binding-activity to tetrasaccharide (lacto-N-tetraose and lacto-N-neotetraose) backbones. Biochemistry, 33, 6342-6349. MEDLINE Abstract

Frishman,D. and Argos,P. (1995) Knowledge-based secondary structure assignment. Proteins, 23, 566-579. MEDLINE Abstract

Frishman,D. and Argos,P. (1996) Incorporation of nonlocal interactions in protein secondary structure prediction from the amino acid sequence. Protein Eng., 9, 133-142. MEDLINE Abstract

Geourjon,C. and Deleage,G. (1994) SOPMA: a self optimised prediction method for protein secondary structure prediction. Protein Eng., 7, 157-164. MEDLINE Abstract

Gitt,M.A., Wiser,M.F., Leffler,H., Herrmann,J., Xia,Y.-R., Massa,S.M., Cooper,D.N.W., Lusis,A.J. and Barondes,S.H. (1995) Sequence and mapping of galectin-5, a [beta]-galactoside binding lectin found in rat erythrocytes. J. Biol. Chem., 270, 5032-5038. MEDLINE Abstract

Green,E.D., Adelt,G., Wilson,S., Baenziger,J.U. and van Halbeck,H. (1988) The asparagine-linked oligosaccharides on bovine fetuin. Structural analysis of N-glycanase-released oligosaccharides by 500-megahertz NMR spectroscopy. J. Biol. Chem., 34, 18253-18268.

Hehre,W.J. and Huang,W.W. (1995) Chemistry with Computation: An Introduction to SPARTAN. Wavefunction, Inc.

Hirabayashi,J. and Kasai,K.-I. (1992) Effect of amino acid substitution by site directed mutagenesis on the carbohydrate recognition and stability of human 14-kDa [beta]-galactoside binding lectin. J. Biol. Chem., 266, 23648-23653.

Hirabayashi,J. and Kasai,K.-I. (1993) The family of metazoan metal-independent [beta]-galactoside-binding lectins: structure, function and molecular evolution. Glycobiology, 3, 297-304. MEDLINE Abstract

Hirabayashi,J. and Kasai,K.-I. (1994) Further evidence by site-directed mutagenesis that conserved hydrophilic residues form a carbohydrate binding site of human galectin-1. Glycoconjugate J., 11, 437-442.

Hughes,R.C. (1994) Mac2: a versatile galactose-binding protein of mammalian tissues. Glycobiology, 4, 5-12. MEDLINE Abstract

Hutchinson,E.G. and Thornton,J.M. (1996) PROMOTIF: a program to identify and analyze structural motifs in proteins Protein Sci., 5, 212-220. MEDLINE Abstract

Imberty,A., Delage,M.-M., Bourne,Y., Cambillau,C. and Perez,S. (1991\) Data bank of three dimensional structures of disaccharides. 2. N-Acetyllactosaminic type N-glycans. Comparison with the crystal structure of a biantennary octa-saccharide. Glycoconjugate J., 8, 456-483.

Imberty,A., Hardman,K.D., Carver,J.P.C. and Perez,S. (1991bb) Molecular modeling of protein-carbohydrate interactions. Docking of monosaccharides in the binding site of concanavalin A. Glycobiology, 1, 631-642. MEDLINE Abstract

Imberty,A., Casset,F., Gegg,C.V., Etzler,M.E. and Perez,S. (1994) Molecular modeling of the Dolichos biflorus seed lectin and its specific interactions with carbohydrates: alpha-d-N-acetyl-galactosamine, Forssman disaccharide and blood group A trisaccharide. Glycoconjugate J., 11, 400-413.

Imberty,A., Mikros,E., Koca,J. Mollicone,R., Oriol,R. and Perez,S. (1995) Computer-simulation of histo-blood group oligosaccharides-energy maps of all constituting disaccharides and potential-energy surfaces of 14 ABH and Lewis carbohydrate antigens. Glycoconjugate J., 12, 331-349.

Jones,T.A., Zou,J.Y., Cowan,S.W. and Kjeldgaard,M. (1991) Improved methods for the building of protein models in electron density maps and the location of errors in these models, Acta Crystallogr., A47, 110-119.

Kneller,D.G., Cohen,F.E. and Langridge,R. (1990) Improvements in protein secondary structure prediction by an enhanced neural network. J. Mol. Biol., 214, 171-182. MEDLINE Abstract

Knibbs,R.N., Agrwal,N., Wang,J.L. and Goldstein,J. (1993) Carbohydrate binding proteins 35. Interaction of the recombinant polypeptide with saccharides. J. Biol. Chem., 268, 14940-14947. MEDLINE Abstract

Kraulis,P.J. (1991) MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures, J. Appl. Crystallogr., 24, 946-950.

Laskowski,R.A., MacArthur,M.W., Moss,D.S. and Thornton,J.M. (1993) PROCHECK a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr., 26, 283-291.

Lee,C. (1994) Predicting protein mutant energetics by self-consistent ensemble optimization. J. Mol. Biol., 236, 918-939. MEDLINE Abstract

Lee,C. and Subbiah,S. (1991) Prediction of protein side-chain conformation by packing optimization. J. Mol. Biol., 217, 373-388. MEDLINE Abstract

Lee,R.T., Ichikawa,Y., Allen,H.J. and Lee,Y.C. (1990) Binding characteristics of galactoside-binding lectin (Galaptin) from human spleen. J. Biol. Chem., 265, 7864-7871. MEDLINE Abstract

Leffler,H. and Barondes,S.H. (1986) Specificity of binding of three soluble rat lung lectins to substituted and unsubstituted mammalian beta-galactosides. J. Biol. Chem., 261, 10119-10126. MEDLINE Abstract

Lemiux,R.U. (1994) The hydrophilic character of the polyamphiphilic surfaces of oligosaccharides recognised by lectins and antibodies. Biophys. J., 66, A26.

Lemieux,R.U., Delbaere,L.T., Beierbeck,H. and Spohr,U. (1991) Involvement of water in host-guest interactions. Ciba Found. Symp., 158, 231-248. MEDLINE Abstract

Levitt,M. (1983) Protein folding by restrained energy minimisation and molecular dynamics. J. Mol. Biol., 170, 723-764. MEDLINE Abstract

Levitt,M. (1992) Accurate modeling of protein conformation by automaticsegment matching. J. Mol. Biol., 226, 507-533. MEDLINE Abstract

Liao,D.-I., Kapadia,G., Ahmed,H., Vasta,G.R. and Herzberg,O. (1994) Structure of S-lectin, a developmentally regulated vertebrate [beta]-galactoside binding protein. Proc. Natl. Acad. Sci. USA, 91, 1428-1432.

Lobsanov,Y., Gitt,M.A., Leffler,H., Barondes,S.A. and Rini, J.M. (1993) X-Ray crystal structure of the human dimeric S-lac lectin, L-14-II, in complex with lactose at 2.9 resolution. J. Biol. Chem., 268, 27034-27038. MEDLINE Abstract

Martin,A.C.R. (1995) PROFIT, protein least squares fitting program. SciTech Software.

Massa,S.M., Cooper,D.N.W., Leffler,H. and Barondes,S.H. (1993) L-29 an endogenous lectin, binds to glycoconjugate ligands with positive cooperativity. Biochemistry, 32, 260-267. MEDLINE Abstract

Mehul,B., Bawumia,S., Martin,S.R. and Hughes,R.C. (1994) Structure of baby hamster kidney carbohydrate binding protein CBP30, an S-type animal lectin. J. Biol. Chem., 269, 18250-18258. MEDLINE Abstract

Meng,E.C., Shoichet,B.K. and Kuntz,I.D. (1992) Automated docking with grid-based energy evaluation. J. Comp. Chem., 13, 505-524.

Meng,E.C., Gschwend,D.A., Blaney,J.M. and Kuntz,I.D. (1993) Orientation sampling and rigid-body minimization in molecular docking. Proteins, 17, 266-278 MEDLINE Abstract

Merritt,E.A. and Murphy,M.E.P. (1994) Raster3D version 2.0, a program for photorealistic molecular graphics. Acta Crystallogr., D50, 869-873.

Morris,A.L., MacArthur,M.W., Hutchinson,E.G., Moss,D.S and Thornton,J.M. (1992) Stereochemical quality of protein structure coordinates. Proteins, 12, 345-364. MEDLINE Abstract

Noordik,J.H., Beursken,P.T., Benneman,P., Visser,R.A. and Gould,R.O. (1984) Crystal structure, polarity and morphology of 4-O-[beta]-d-galactopyranosyl-[alpha]-d-glucopyranose monohydrate, [alpha]-lactose monohydrate. A redetermination. Z. Kristallogr., 168, 49-65.

Ohyama,Y. and Kasai,K.-I. (1988) Isolation and characterisation of the chick 14k beta-galactoside binding lectin gene. J. Biochem., 104, 173-177. MEDLINE Abstract

Powell,M.J.D. (1977) Restart procedures for the conjugate gradient method. Math. Program., 12, 241-252.

Ramkumar,R., Surolia,A. and Podder,S.K. (1995) Energetics of carbohydrate binding by a 14 kDa S-type mammalian lectin. Biochem. J., 308, 237-241. MEDLINE Abstract

Rost,B. and Sander,C. (1994) Combining evolutionary information and neural networks to predict protein secondary structure. Proteins, 19, 55-77. MEDLINE Abstract

Sakakura,Y., Hirabayashi,J., Oda,Y., Ohyama,Y. and Kasai,K. (1990) Structure of chicken 16-kDa [beta]-galactoside binding lectin. J. Biol. Chem., 265, 21573-21579. MEDLINE Abstract

Sato,S. and Hughes,R.C. (1992) Binding specificity of a baby hamster kidney lectin for H Type I and II chains, polylactosamine glycans and appropriately glycosylated forms of laminin and fibronectin. J. Biol. Chem., 267, 6983-6990. MEDLINE Abstract

Solis,D., Romero,A., Kaltner,H., Gabius,H.-J. and Diaz-Maurino,T. (1996) Different architecture of the combining site of the two chicken galectins revealed by chemical mapping studies with synthetic ligand derivatives. J. Biol. Chem., 271, 12744-12748. MEDLINE Abstract

Solomon,J.C., Stoll,M.S., Penfold,P., Abbott,W.M., Childs,R.A., Hanfland,P. and Feizi,T. (1991) Studies of the binding specificity of the soluble 14,000-dalton bovine heart muscle lectin using immobilised glycolipids and neoglycolipids. Carbohydr. Res., 213, 293-307. MEDLINE Abstract

Sparrow,C.P., Leffler,H. and Barondes,S.H. (1987) Multiple soluble [beta]-galactoside binding lectins from human lung. J. Biol. Chem., 262, 7383-7390. MEDLINE Abstract

Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673-4680. MEDLINE Abstract

Wilmot,C.M. and Thornton,J.M. (1990) [beta]-Turns and their distortions: a proposed new nomenclature. Protein Eng., 3, 479-493. MEDLINE Abstract

1Present address: European Bioinformatics Institute, EMBL Outstation, Hinxton, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, United Kingdom
2Permanent address: Department of Human Biopathology, University of Rome 'La Sapienza," Rome 06100, Italy
3Present address: Oncogene Laboratory, Institute for Cancer Research, 94801 Villejuif, France
4To whom correspondence should be addressed
Dedicated to Roger W. Jeanloz on the occasion of his 80th birthday

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 25 Jan 1998
Copyright© Oxford University Press, 1998.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
H. Ideo, A. Seko, and K. Yamashita
Recognition Mechanism of Galectin-4 for Cholesterol 3-Sulfate
J. Biol. Chem., July 20, 2007; 282(29): 21081 - 21089.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
M. Demers, K. Biron-Pain, J. Hebert, A. Lamarre, T. Magnaldo, and Y. St-Pierre
Galectin-7 in Lymphoma: Elevated Expression in Human Lymphoid Malignancies and Decreased Lymphoma Dissemination by Antisense Strategies in Experimental Model
Cancer Res., March 15, 2007; 67(6): 2824 - 2829.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
M. G. Chiu, T. M. Johnson, A. S. Woolf, E. M. Dahm-Vicker, D. A. Long, L. Guay-Woodford, K. A. Hillman, S. Bawumia, K. Venner, R. C. Hughes, et al.
Galectin-3 Associates with the Primary Cilium and Modulates Cyst Growth in Congenital Polycystic Kidney Disease
Am. J. Pathol., December 1, 2006; 169(6): 1925 - 1938.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
H. Stalz, U. Roth, D. Schleuder, M. Macht, S. Haebel, K. Strupat, J. Peter-Katalinic, and F.-G. Hanisch
The Geodia cydonium galectin exhibits prototype and chimera-type characteristics and a unique sequence polymorphism within its carbohydrate recognition domain
Glycobiology, May 1, 2006; 16(5): 402 - 414.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
A. Lagana, J. G. Goetz, P. Cheung, A. Raz, J. W. Dennis, and I. R. Nabi
Galectin Binding to Mgat5-Modified N-Glycans Regulates Fibronectin Matrix Remodeling in Tumor Cells
Mol. Cell. Biol., April 15, 2006; 26(8): 3181 - 3193.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
T. K. van den Berg, H. Honing, N. Franke, A. van Remoortere, W. E. C. M. Schiphorst, F.-T. Liu, A. M. Deelder, R. D. Cummings, C. H. Hokke, and I. van Die
LacdiNAc-Glycans Constitute a Parasite Pattern for Galectin-3-Mediated Immune Recognition
J. Immunol., August 1, 2004; 173(3): 1902 - 1907.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
I. Pelletier, T. Hashidate, T. Urashima, N. Nishi, T. Nakamura, M. Futai, Y. Arata, K.-i. Kasai, M. Hirashima, J. Hirabayashi, et al.
Specific Recognition of Leishmania major Poly-{beta}-galactosyl Epitopes by Galectin-9: POSSIBLE IMPLICATION OF GALECTIN-9 IN INTERACTION BETWEEN L. MAJOR AND HOST CELLS
J. Biol. Chem., June 13, 2003; 278(25): 22223 - 22230.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
S. Sassi, M. Sweetinburgh, J. Erogul, P. Zhang, P. Teng-umnuay, and C. M. West
Analysis of Skp1 glycosylation and nuclear enrichment in Dictyostelium
Glycobiology, April 1, 2001; 11(4): 283 - 295.
[Abstract] [Full Text] [PDF]

Home page
 J. Am. Soc. Nephrol.Home page
S. L. BULLOCK, T. M. JOHNSON, Q. BAO, R. C. HUGHES, P. J. D. WINYARD, and A. S. WOOLF
Galectin-3 Modulates Ureteric Bud Branching in Organ Culture of the Developing Mouse Kidney
J. Am. Soc. Nephrol., March 1, 2001; 12(3): 515 - 523.
[Abstract] [Full Text]

Home page
 GlycobiologyHome page
E. A. M. Barboni, S. Bawumia, K. Henrick, and R.C. Hughes
Molecular modeling and mutagenesis studies of the N-terminal domains of galectin-3: evidence for participation with the C-terminal carbohydrate recognition domain in oligosaccharide binding
Glycobiology, November 1, 2000; 10(11): 1201 - 1208.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
C. Fradin, D. Poulain, and T. Jouault
beta -1,2-Linked Oligomannosides from Candida albicans Bind to a 32-Kilodalton Macrophage Membrane Protein Homologous to the Mammalian Lectin Galectin-3
Infect. Immun., August 1, 2000; 68(8): 4391 - 4398.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Kleene, H. Yang, M. Kutsche, and M. Schachner
The Neural Recognition Molecule L1 Is a Sialic Acid-binding Lectin for CD24, Which Induces Promotion and Inhibition of Neurite Outgrowth
J. Biol. Chem., June 8, 2001; 276(24): 21656 - 21663.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 (34)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Henrick, K.
Right arrow Articles by Hughes, R. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henrick, K.
Right arrow Articles by Hughes, R. C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?