Erythrocyte Receptor Research Articles

Partial characterization of the human erythrocyte receptor for rabbit haemorrhagic disease virus

An important, well known property of the rabbit haemorrhagic disease virus is its ability to agglutinate human red blood cells. Accordingly, red cells from human adult donors were agglutinated despite their blood group ABO status, and treatments with proteases or glycosidases did not prevent agglutination. However, we discovered that the cells from human umbilical cords or foetuses were not agglutinated. In order to identify the viral receptor on human erythrocytes, glycolipids and glycoproteins from adult red cells were separated and tested for their potency in inhibiting agglutination. The bulk of the biological activity was associated with the highly glycosylated glycolipids (polyglycosylceramides), whereas a lower but significant activity was also associated with neutral glycolipids. No activity was found in the lipid-free sialoglycoprotein fractions. All these data strongly suggest that the RHDV receptor on human red cells corresponds to a development antigen which is not expressed on foetal cells and is mainly carried by glycolipids. Faint activity was also found in membranes from sheep red cells, suggesting that a similar glycolipid component is carried by these animal cells.

Changes in insulin-like growth factor-I (IGF-I), IGF-binding protein-3, growth hormone (GH)-binding protein, erythrocyte IGF-I receptors, and growth rate during GH treatment.

To assess the relative determinants of growth rate, we measured serum levels of insulin-like growth factor-I (IGF-I), IGF-binding protein-3 (IGFBP-3), and GH-binding protein (GHBP) as well as IGF-I erythrocyte receptor specific binding (SB) in 14 prepubertal GH-deficient children before and during the first year of treatment with 0.043 mg/kg.day GH. Serum IGF-I and IGFBP-3 levels, measured by RIA, were significantly increased by 2 weeks and showed progressive increases throughout the year of GH therapy. Growth rate (height velocity SD score adjusted for bone age) correlated best with the 12 month changes in IGFBP-3 (r = 0.81; P < 0.001) and IGF-I (r = 0.72; P = 0.005), and to a lesser extent with the 12 month absolute IGFBP-3 (r = 0.58; P = 0.04) and the 6 month change in IGFBP-3 (r = 0.55; P = 0.05). The baseline IGF-I correlated inversely with the growth rate during GH therapy (r = -0.55; P = 0.05) and was the best pretreatment predictor of growth response. GHBP, as measured by ligand-mediated immunofunctional assay, showed no significant change during GH therapy and did not correlate with growth response. The baseline GHBP, however, did correlate with both the 12 month IGFBP-3 (r = 0.72; P = 0.006) as well as the 2 week change in IGFBP-3 (r = 0.63; P = 0.05). Erythrocyte IGF-I SB showed a significant decrease by 6 months secondary to a decrease in IGF-I receptor number, with no change in affinity. The 6 month IGF-I receptor binding correlated inversely with the increase in IGF-I (r = -0.88; P < 0.001). Erythrocyte IGF-I SB at baseline did not correlate with the growth response, although there was an inverse trend between the 6 month IGF-I receptor level and the growth rate. IGF-I and IGFBP-3 show progressive increases, whereas the erythrocyte IGF-I receptor-binding capacity decreases by 6 months, and GHBP shows little change during the first year of GH treatment. Data from this study suggest that changes in IGFBP-3 and, to a lesser extent, IGF-I are the major correlates of growth rate, and that down-regulation of the IGF-I receptor may have relatively little influence on growth rate compared with changes in IGFBP-3 and IGF-I.

Partial purification of the rat erythrocyte receptor for the channel‐forming toxin aerolysin and reconstitution into planar lipid bilayers

The cytolytic toxin aerolysin binds to a receptor on the surface of eukaryotic cells. Murine erythrocytes are among the most sensitive to the toxin. Here we describe the detergent solubilization and partial purification of the receptor from rat erythrocytes. We show that it can be successfully incorporated into planar lipid bilayers, greatly decreasing the concentration of aerolysin required to form channels. Exploiting the ability of the receptor to bind aerolysin after SDS electrophoresis and blotting, we obtain evidence that it is a 47 kDa glycoprotein that is sensitive to proteases and N-glycosidase. It may correspond to CHIP28, the water channel of the human erythrocyte.

Multiple ligands for cytoadherence can be present simultaneously on the surface of Plasmodium falciparum-infected erythrocytes.

A major virulence factor of Plasmodium falciparum is the adherence of parasitized erythrocytes to the wall of postcapillary venules via a specific interaction between parasite-derived erythrocyte surface ligands and receptors on endothelial cells. To study this phenomenon in vitro, we selected a parasite population that expressed at least two different ligands and demonstrated that parasitized cells may coexpress ligands with specificity for multiple receptors. This selected parasite line had several antigenic and cytoadherence characteristics that were different from those of the parent line. Single parasitized erythrocytes were able to adhere to three distinct receptors via at least two separate ligands; a trypsin-sensitive molecule mediated cytoadherence to CD36 and intercellular adhesion molecule 1 and a trypsin-insensitive molecule(s) was responsible for adherence to a third receptor on the surface of melanoma cells. We present evidence that this newly discovered receptor for cytoadherence is an N-linked glycosaminoglycan, as treatment of melanoma cells with endoglycosidase H abolished cytoadherence. These observations emphasize the adaptability of P. falciparum and the complexity of the cytoadherence phenomenon.

Receptor and Ligand Domains for Invasion of Erythrocytes by Plasmodium falciparum

A 175-kilodalton erythrocyte binding protein, EBA-175, of the parasite Plasmodium falciparum mediates the invasion of erythrocytes. The erythrocyte receptor for EBA-175 is dependent on sialic acid. The domain of EBA-175 that binds erythrocytes was identified as region II with the use of truncated portions of EBA-175 expressed on COS cells. Region II, which contains a cysteine-rich motif, and native EBA-175 bind specifically to glycophorin A, but not to glycophorin B, on the erythrocyte membrane. Erythrocyte recognition of EBA-175 requires both sialic acid and the peptide backbone of glycophorin A. The identification of both the receptor and ligand domains may suggest rational designs for receptor blockade and vaccines.

Simultaneous expression of CD2 on a B‐cell, small lymphocytic neoplasm

An unusual neoplasm, best classified as a B-cell chronic lymphocytic leukemia on the basis of cytofluorographic, immunohistologic, and gene rearrangement analysis also co-expressed the T-cell associated marker CD2 (sheep erythrocyte receptor), but without other cell markers restricted to T cells. This case was associated with a more aggressive course than typically seen with B-CLL or malignant lymphoma, small lymphocytic type, implying along with previous reports that CD2 expression may serve as a marker of small lymphocytic tumor cell behavior and the ultimate clinical course. Thus, surface phenotypic analysis of these particular hematopoietic neoplasms may serve not only for identification of the malignancy, but could also render prognostic information.

Glycophorin B as an EBA-175 independent Plasmodium falciparum receptor of human erythrocytes

Invasion of erythrocytes by malaria parasites involves multiple receptor-ligand interactions. To elucidate these pathways, we made use of four parasite clones with differing specificities for invasion, erythrocytes that are mutant for either glycophorin A or B, and enzyme modification of the erythrocyte surface with neuraminidase and trypsin. Neuraminidase alone abolishes invasion of two parasite clones (Dd2, FCR3/A2); these invade after trypsin treatment alone. A third clone (7G8) is unable to invade trypsin-treated erythrocytes. The fourth clone (HB3) can invade after either neuraminidase or trypsin treatment. The receptor for invasion of trypsin-treated erythrocytes was explored in two ways: treatment of trypsin-treated normal cells with neuraminidase and trypsin treatment of glycophorin B-deficient cells. Both treatments eliminated invasion by all clones, indicating that the trypsin-independent pathway uses sialic acid and glycophorin B. To identify parasite proteins involved in the different pathways, erythrocyte binding assays were performed with soluble parasite proteins from each clone. Based on binding assays using erythrocytes that lack glycophorin A, the parasite protein known as EBA-175 appears to bind predominantly to glycophorin A. In contrast, the glycophorin B pathway does not appear to involve EBA-175, as binding of EBA-175 was similarly reduced to trypsin-treated normal and trypsin-treated glycophorin B-deficient erythrocytes. Thus, the glycophorin B-dependent, sialic acid-dependent invasion of trypsin-treated normal erythrocytes uses a different parasite ligand, indicating two or more sialic-dependent pathways for invasion. Clone 7G8, which cannot invade trypsin-treated erythrocytes, may be missing the ligand for invasion via glycophorin B. Redundancy in the invasion process may give a selective advantage to parasites that must survive in polymorphic human populations.

Dimorphism and intergenic recombination within the microneme protein (MP-1) gene family of Plasmodium knowlesi

The microneme protein-1 (MP-1) of Plasmodium knowlesi and Plasmodium vivax facilitates merozoite invasion of the erythrocyte by binding to Duffy blood group antigens. Since this protein is important in the invasion process and is a potential vaccine candidate, it is important to understand the nature of diversity within the MP-1 gene. Nine MP-1 gene sequences were compared from 2 isolates of P. knowlesi and a laboratory strain of P. vivax. The MP-1 genes of P. knowlesi were dimorphic based upon the central hydrophilic regions (III and IV) that were well conserved as α and β types. Other regions were conserved among all P. knowlesi genes except for the amino cysteine-rich region (region II), a region predicted to be the initial contact site of the erythrocyte binding domain. Two distinct sequence motifs and part of a third were identified in region II that had a common identity of 68%. In some MP-1 genes recombination had occurred to create hybrids of the two sequence types. All cysteines and aromatic amino acids of region II were conserved in all genes or within a sequence type. There were 2 apparent recombination points within region II where switching occurred between sequence types. Another possible recombination site, identified as a common sequence motif, was identified in the middle of the hydrophilic region, at the beginning of regions III or IV. Nonsynonymous mutations within region II were biased towards radical amino acid changes, especially towards the carboxyl third, where there were 3 distinct types of sequence. Most synonymous and nonsynonymous nucleotide mutations within regions I, V, and VI were infrequent, individual events and not associated with any particular sequence type. Cysteinerich regions of the P. vivax MP-1 gene compared to the P. knowlesi genes were characterized by an increased number of synonymous and nonsynonymous changes. This data identifies 2 mechanisms for generation of diversity in the MP-1 gene family, intergenic recombination and nucleotide mutations. Both may be mechanisms the parasite uses to evade the host immune response or to alter erythrocyte receptor specificity.

Analysis of CD36 binding domains: ligand specificity controlled by dephosphorylation of an ectodomain.

The protein CD36 is a membrane receptor for thrombospondin (TSP), malaria-infected erythrocytes, and collagen. Three functional sequences were identified within a single disulfide loop of CD36: one that mediates TSP binding (amino acids 87 to 99) and two that support malarial cytoadhesion (amino acids 8 to 21 and 97 to 110). One of these peptides (p87-99) is a consensus protein kinase C (PKC) phosphorylation site. Dephosphorylation of constitutively phosphorylated CD36 in resting platelets and a megakaryocytic cell line led to the loss of collagen adhesion and platelet reactivity to collagen, with a reciprocal increase in TSP binding. PKC-mediated phosphorylation of this ectodomain resulted in a loss of TSP binding and the reciprocal acquisition of collagen binding. In site-directed mutagenesis studies, when the threonine phosphorylation site was changed to alanine, CD36 was expressed in a dephosphorylated state and bound to TSP constitutively.

Adherence to human cells of a cryptic Haemophilus genospecies responsible for genital and neonatal infections.

Haemophilus strains usually identified as Haemophilus influenzae biotype IV belonging to a cryptic genospecies are responsible for genital and neonatal infections. As a first approach to identifying the bacterial factors involved in the pathogenesis of these unusual diseases, we studied the piliation, adherence, and invasion properties of 17 strains assigned to this cryptic genospecies. Twelve strains spontaneously displayed abundant peritrichous piliation, and two strains expressed peritrichous pili after enrichment procedures. For virtually all strains, piliation correlated with adhesion to cultured HeLa cells of genital origin and to a lesser extent with adhesion to HEp-2 cells of laryngeal origin. A variation in the adherence properties of the various strains was observed: all piliated strains except one adhered to 50 to 100% of HeLa cells, with a mean number of bacteria per cell varying from 4 to 50. Adherence was not dependent on the state of growth for most strains, was more pronounced with HeLa cells than with HEp-2 cells for 10 of the 12 highly adherent strains, was time and inoculum dependent, and was not followed by significant invasion of cells. Most of the strains belonging to this unusual Haemophilus clone possess adhesins that do not recognize erythrocyte receptors, since agglutination of human erythrocytes was observed with only 3 of the 14 piliated strains.

A receptor for the malarial parasite Plasmodium vivax: the erythrocyte chemokine receptor.

Plasmodium vivax and P. falciparum are the major causes of human malaria, except in sub-Saharan Africa where people lack the Duffy blood group antigen, the erythrocyte receptor for P. vivax. Duffy negative human erythrocytes are resistant to invasion by P. vivax and the related monkey malaria, P. knowlesi. Several lines of evidence in the present study indicate that the Duffy blood group antigen is the erythrocyte receptor for the chemokines interleukin-8 (IL-8) and melanoma growth stimulatory activity (MGSA). First, IL-8 binds minimally to Duffy negative erythrocytes. Second, a monoclonal antibody to the Duffy blood group antigen blocked binding of IL-8 and other chemokines to Duffy positive erythrocytes. Third, both MGSA and IL-8 blocked the binding of the parasite ligand and the invasion of human erythrocytes by P. knowlesi, suggesting the possibility of receptor blockade for anti-malarial therapy.

Effects of exogenous porcine somatotropin on the carcass composition, hormonal and metabolic profiles, lipogenic capacity, and binding of insulin to erythrocyte receptors of fast- versus slow-growing swine

Twenty barrows were designated as fast-growing (FG) and their littermates designated as slow-growing (SG) based on birth, weaning, and 56-d weight. Half of each group received 70 micrograms of porcine somatotropin (pST)/kg BW daily beginning at 40 kg BW. At 60 and 105 kg BW, blood was collected every .5 h for 12 h beginning 1 h before pST injection, fat biopsies were taken for in vitro lipogenic activity, and insulin erythrocytes were isolated for receptor binding. Swine treated with pST had elevated ADG (.95 vs .88 kg/d; P < .1) and reduced days to slaughter (61 vs 67; P < .1). The pST-treated pigs had less average backfat (2.73 vs 3.96 cm; P < .01), larger longissimus muscle areas (32.3 vs 28.2 cm2; P < .05), and a higher percentage of muscle (56.3 vs 50.3%; P < .01) than control pigs. Exogenous pST increased protein (17.4 vs 13.2%; P < .05) and decreased fat (22.9 vs 37.1%; P < .05). The FG pigs had higher ADG (.98 vs .86 kg/d; P < .01) and required fewer days to slaughter (57 vs 71; P < .01) than SG pigs. Administration of pST increased (P < .01) average pST levels (1.7 vs 14.0 ng/mL) in FG and SG pigs at 60 kg BW. At 105 kg BW, pST was higher (P < .01) in pST-FG than in pST-SG swine (46.0 vs 19.3 ng/mL) but was not different between FG and SG control swine (1.9 vs 1.8 ng/mL). Administration of pST increased concentrations of IGF-I (510.0 vs 160.0 ng/mL) and nonesterified fatty acids (182 vs 109 muEq/L, P < .01) in FG and SG swine. Over sample periods and growth rates, pST reduced (P < .05) CO2 production and lipid synthesis (.345 and 1.85 vs .575 and 2.71 mumol of glucose incorporated.g-1.2 h-1). At 60 kg BW, FG swine had less (P < .01) CO2 production and lipid synthesis (.299 and 1.83 vs .921 and 3.61 mumol.g-1.2 h-1) than did SG swine. Exogenous pST increased (P < .05) binding to insulin erythrocyte receptors (7.25 vs 6.34%).

Isolation, expression, and nucleotide sequencing of the pilin structural gene of the Brazilian purpuric fever clone of Haemophilus influenzae biogroup aegyptius

In this study we isolated the pilin gene from the Brazilian purpuric fever (BPF) clone of Haemophilus influenzae biogroup aegyptius, expressed the gene in Escherichia coli, and determined its nucleotide sequence. Comparison of the nucleotide sequence of the BPF pilin gene with the sequences of pilin genes from strains of H. influenzae sensu stricto demonstrated a high degree of identity. Consistent with this observation, hemagglutination inhibition studies performed with a series of glycoconjugates indicated that BPF pili and H. influenzae type b pili possess the same erythrocyte receptor specificity.

Mutations adjacent to the dimple of the canine parvovirus capsid structure affect sialic acid binding

The erythrocyte receptor on rhesus macaque erythrocytes used by canine parvovirus (CPV) for binding in hemagglutination (HA) was examined. Erythrocyte membrane proteins were electrophoresed and blotted to nitrocellulose and probed with [ 125I]-labeled CPV capsids, showing seven virus-binding proteins. Treatment of erythrocytes or isolated membranes with Clostridium perfringens neuraminidase virtually abolished virus binding. Binding was also affected by treatment with potassium periodate and inhibited by wheat germ agglutinin, but was not significantly affected by concanavalin A, peanut agglutinin, or soluble N-acetyl-neuram in lactose. A non-HA mutant of CPV failed to bind to erythrocytes or to blotted erythrocyte membrane proteins. The mutation was a single Arg-Lys difference of VP2 amino acid residue 377. The pH dependence of binding of the closely related feline panleukopenia virus was shown to result from a decreased binding in buffers with pH values of 6.8 or greater. The VP2 residues responsible for that difference have been shown to be 323 and 375. The sequences affecting binding were all adjacent to the dimple in the capsid, implicating that region of the capsid as the sialic acid binding site. The role of sialic acid in virus-host cell interactions was not defined, but the plaque sizes of the non-HA mutant and wild type CPV were indistinguishable.

Involvement of CD36 on erythrocytes as a rosetting receptor for Plasmodium falciparum-infected erythrocytes

Plasmodium falciparum-infected erythrocytes (parasitized red blood cells [PRBCs]) can adhere to uninfected erythrocytes (RBCs) to form rosettes, and adhere to the endothelial cell (EC) surface antigen CD36. These adherence phenomena have previously been considered quite different. We show that anti-CD36 monoclonal antibodies (MoAbs) reverse rosetting of PRBCs from both a culture-adapted line (Malayan Camp [MC] strain) and a natural isolate, GAM425. Three MoAbs that block adherence of PRBCs to ECs or C32 melanoma cells also reversed rosetting by greater than 50% at levels of less than 1 microgram/mL (OKM5, OKM8, and 8A6). Two other MoAbs that react with purified CD36 (1D3 and 1B1), but do not react with the surface of C32 cells, failed to reverse rosetting. When rosettes were disrupted and the RBCs and PRBCs were pretreated separately with antibodies before mixing to allow rosette reformation, only pretreatment of RBCs had an effect. MoAb 8A6 pretreatment of RBCs blocked rosette reformation, while MoAb 1B1 pretreatment did not. Rosetting was also reversed by purified human platelet CD36. In conjunction with evidence that CD36 is expressed on normal human erythrocytes (van Schravendijk et al, Blood 80:2105, 1992), we conclude that this CD36 is able to act as a host receptor for rosetting in the MC strain and some natural isolates of P falciparum.

Involvement of CD36 on Erythrocytes as a Rosetting Receptor for Plasmodium falciparum-Infected Erythrocytes

AbstractPlasmodium falciparum-infected erythrocytes (parasitized red blood cells [PRBCs]) can adhere to uninfected erythrocytes (RBCs) to form rosettes, and adhere to the endothelial cell (EC) surface antigen CD36. These adherence phenomena have previously been considered quite different. We show that anti-CD36 monoclonal antibodies (MoAbs) reverse rosetting of PRBCs from both a culture-adapted line (Malayan Camp [MC] strain) and a natural isolate, GAM425. Three MoAbs that block adherence of PRBCs to ECs or C32 melanoma cells also reversed rosetting by greater than 50% at levels of less than 1 µg/mL (OKM5, OKM8, and 8A6). Two other MoAbs that react with purified CD36 (1D3 and 1B1), but do not react with the surface of C32 cells, failed to reverse rosetting. When rosettes were disrupted and the RBCs and PRBCs were pretreated separately with antibodies before mixing to allow rosette reformation, only pretreatment of RBCs had an effect. MoAb 8A6 pretreatment of RBCs blocked rosette reformation, while MoAb 1B1 pretreatment did not. Rosetting was also reversed by purified human platelet CD36. In conjunction with evidence that CD36 is expressed on normal human erythrocytes (van Schravendijk et al. Blood 80:2105, 1992), we conclude that this CD36 is able to act as a host receptor for rosetting in the MC strain and some natural isolates of P falciparum.

Group C rotavirus requires sialic acid for erythrocyte and cell receptor binding.

Current immunological and biochemical information regarding the hemagglutinin and virus-cell interactions of rotavirus is obtained exclusively from studies with group A rotaviruses. In this study, I report that the immunologically and genetically distinct group C rotavirus also possesses a hemagglutinin. The viral hemagglutinin was identified on a cultivable porcine group C rotavirus strain (strain AmC-1) by using agglutinated human and guinea pig erythrocytes. Neuraminidase treatment of fresh human erythrocytes or blocking with glycophorin A or fetuin prevented hemagglutination. Infection of swine testicular cells with group C AmC-1 virus was also prevented by glycophorin A, fetuin, and neuraminidase treatment, suggesting that sialic acid constitutes an essential part of the cell receptor.

Plasmodium falciparum-infected erythrocyte receptor(s) for CD36 and thrombospondin are restricted to knobs on the erythrocyte surface.

Adherence of Plasmodium falciparum-infected RBCs (PRBC) to endothelial cells causes PRBC sequestration in cerebral microvessels and is considered to be a major contributor to the pathogenesis of cerebral malaria. Both CD36 and thrombospondin (TSP) are glycoproteins that mediate PRBC adherence to endothelial cells in vitro. Because they are both expressed on the surface of endothelial cells, they probably contribute to PRBC sequestration and vascular occlusion in vivo. By applying affinity labeling of receptor binding sites with purified ligands, we showed for the first time that both CD36 and TSP can bind independently to the PRBC surface and that the PRBC receptor(s) for CD36 and TSP are localized specifically to the electron-dense knob protrusions of the PRBC surface. These findings may help in efforts to develop a malaria vaccine to prevent cerebral malaria.

Sialoadhesin on macrophages: its identification as a lymphocyte adhesion molecule.

In this study we present evidence that the mouse and rat sialoadhesin (originally named sheep erythrocyte receptor) on macrophages can function as a lymphocyte adhesion molecule. Lymphocytes were shown to bind to the splenic marginal zone, and lymph node subcapsular sinus and medulla in a frozen section assay. Selective depletion experiments showed that binding was mediated by macrophages. Adhesion was blocked by preincubation of the sections with monoclonal antibodies against mouse or rat sialoadhesin. Binding was temperature dependent, divalent cation independent, and involved sialic acid residues on the lymphocyte, as it could be inhibited by prior neuraminidase treatment or addition of the ganglioside GD1a. Binding to sialoadhesin was confirmed using the purified receptor and was observed among T cells, T blasts, B cells, and B blasts. Isolated macrophages or dendritic cells showed little binding. Sialoadhesin provides the first example of a macrophage-restricted lymphocyte adhesion molecule.

Crystal structure of a wheat germ agglutinin/glycophorin-sialoglycopeptide receptor complex. Structural basis for cooperative lectin-cell binding.

The crystal structure of wheat germ agglutinin isolectin 1 (WGA1) complexed with a tryptic sialoglycopeptide fragment (T-5) from its erythrocyte receptor glycophorin A, which contains the O-linked tetrasaccharide NeuNAc-alpha 2,3-Gal-beta 1,3-(alpha 2,6-NeuNAc) Gal-NAc-alpha 1-O-Thr, has been determined by molecular replacement techniques and refined at 2.0-A resolution (R = 18.1%). The structure reveals that association between WGA1 dimers, composed of two identical four-domain (A-D) monomers, and T-5 is asymmetric and involves sialic acid binding at three nonequivalent aromatic residue-rich sites. Two independent binding modes are observed. In the dominant ("major") binding mode, the two highest affinity sites are utilized to cross-link neighboring crystallographically related WGA1 dimers. The branched tetrasaccharide has an extended rigid conformation, and its terminal alpha 2,6-NeuNAc and alpha 2,3-NeuNAc residues occupy specificity sites in domains B1 (monomer 1) and C2 (monomer 2) on opposing dimers, respectively. This asymmetric selection of binding sites leads to infinite open-ended arrays of interlinked lectin molecules. In the subsidiary "minor" binding mode, only the terminal alpha 2,6-NeuNAc, anchored to the aromatic residue-rich binding site in domain A2, is clearly visible. The remaining portion of T-5 is disordered. This structure presents the first evidence for NeuNAc binding in the aromatic residue-rich sites of domains A and C and suggests a preference of WGA for alpha 2,6-linked NeuNAc. Moreover, the unusual asymmetric WGA1-tetrasaccharide association, involving domain binding sites that differ in their binding affinities for NeuNAc, offers explanations for the widely observed cooperative cell binding behavior of WGA.