Agaricus bisporus lectin binds mainly O-glycans but also N-glycans of human IgA subclasses

Differential contributions of recognition factors of two plant lectins -Amaranthus caudatus lectin and Arachis hypogea agglutinin, reacting with Thomsen-Friedenreich disaccharide (Galbeta1-3GalNAcalpha1-Ser/Thr)

Previous reports on the carbohydrate specificities of Amaranthus caudatus lectin (ACL) and peanut agglutinin (PNA, Arachis hypogea) indicated that they share the same specificity for the Thomsen-Friedenreich (T(alpha), Galbeta1-3GalNAcalpha1-Ser/Thr) glycotope, but differ in monosaccharide binding--GalNAc>>Gal (inactive) for ACL; Gal>>GalNAc (weak) with respect to PNA. However, knowledge of the recognition factors of these lectins was based on studies with a small number monosaccharides and T-related oligosaccharides. In this study, a wider range of interacting factors of ACL and PNA toward known mammalian structural units, natural polyvalent glycotopes and glycans were examined by enzyme-linked lectinosorbent and inhibition assays. The results indicate that the main recognition factors of ACL, GalNAc was the only monosaccharide recognized by ACL as such, its polyvalent forms (poly GalNAcalpha1-Ser/Thr, Tn in asialo OSM) were not recognized much better. Human blood group precursor disaccharides Galbeta1-3/4GlcNAcbeta (I(beta)/II(beta)) were weak ligands, while their clusters (multiantennary II(beta)) and polyvalent forms were active. The major recognition factors of PNA were a combination of alpha or beta anomers of T disaccharide and their polyvalent complexes. Although I(beta)/II(beta) were weak haptens, their polyvalent forms played a significant role in binding. From the 50% molar inhibition profile, the shape of the ACL combining site appears to be a cavity type and most complementary to a disaccharide of Galbeta1-3GalNAc (T), while the PNA binding domain is proposed to be Galbeta1-3GalNAcalpha or beta1--as the major combining site with an adjoining subsite (partial cavity type) for a disaccharide, and most complementary to the linear tetrasaccharide, Galbeta1-3GalNAcbeta1-4Galbeta1-4Glc (T(beta)1-4L, asialo GM(1) sequence). These results should help us understand the differential contributions of polyvalent ligands, glycotopes and subtopes for the interaction with these lectins to binding, and make them useful tools to study glycosciences, glycomarkers and their biological functions.

Evidence that Agaricus bisporus agglutinin (ABA) has dual sugar-binding specificity

Agaricus bisporus agglutinin (ABA) is known as a useful lectin to detect T-antigen (Core1) disaccharide (Galbeta1-3GalNAcalpha) and related O-linked glycans. However, a recent X-ray crystallographic study revealed the presence of another intrinsic sugar-binding site, i.e., for GlcNAc. To confirm this possibility, detailed analysis was performed using two advanced methods: lectin microarray and frontal affinity chromatography (FAC). In the lectin microarray, intense signals were observed on ABA spots for both N-glycanase-treated and O-glycanase/beta1-4galactosidase-treated Cy3-labeled asialofetuin. This indicates substantial affinity for both O-linked and agalactosylated (GlcNAc-exposed) N-linked glycans. A further approach by FAC using 20 pNP and 130 PA-oligosaccharides demonstrated that ABA bound to Core1 (K(d) = 3.4 x 10(-6) M) and Core2 (1.9 x 10(-5) M) but not to Core3 and Core6 O-linked glycans. It also showed substantial affinity to mono-, bi-, and tri-antennary agalactosylated complex-type N-linked glycans (K(d) > 1.8 x 10(-5) M). These results establish ABA as a lectin having dual sugar-binding sites with distinct specificity, i.e., for Gal-exposed O-linked glycans and GlcNAc-exposed N-linked glycans.

Edible mushroom (Agaricus bisporus) lectin inhibits human retinal pigment epithelial cell proliferation in vitro

The retinal pigment epithelium (RPE) plays a major role in the development of the anomalous retinal scarring response termed proliferative vitreoretinopathy. The present study was undertaken to investigate whether agaricus bisporus lectin inhibited human RPE proliferation in vitro. Fluorescein isothiocyanate-labeled agaricus bisporus lectin was used to study binding of lectin to cultured human RPE. The effect of a 24-hour exposure of agaricus bisporus lectin on RPE proliferation was measured using (methyl-3H)-thymidine incorporation into DNA. Toxicity studies were assessed using morphologic evaluation, trypan blue exclusion, and a cell viability assay. Agaricus bisporus lectin bound to RPE cells and was inhibited by preincubation of lectin with asialomucin. Agaricus bisporus lectin caused a dose-dependent inhibition of RPE proliferation (one-way ANOVA, F = 94.470, p < 0.001) that was partially reversible on removal of the lectin. Compared with controls, cells remained viable and no morphological changes or trypan blue staining was noted in RPE exposed to agaricus bisporus lectin. Human RPE binds agaricus bisporus lectin and inhibits proliferation without apparent cytotoxicity. It therefore merits consideration as a potential antiproliferative agent in the prevention and treatment of proliferative vitreoretinopathy and other nonocular anomalous wound healing processes.

Effect of polyvalencies of glycotopes on the binding of a lectin from the edible mushroom, Agaricus bisporus

Agaricus bisporus agglutinin (ABA) isolated from edible mushroom has a potent anti-proliferative effect on malignant colon cells with considerable therapeutic potential as an anti-neoplastic agent. Since previous studies on the structural requirement for binding were limited to molecular or submolecular levels of Galbeta1-3GalNAc (T; Thomsen-Friedenreich disaccharide glycotope; where Gal represents D-galactopyranose and GalNAc represents 2-acetamido-2-deoxy-D-galactopyranose) and its derivatives, the binding properties of ABA were further investigated using our collection of glycans by enzyme-linked lectinosorbent assay and lectin-glycan inhibition assay. The results indicate that polyvalent Galbeta1-related glycotopes, GalNAcalpha1-Ser/Thr (Tn), and their cryptoforms, are the most potent factor for ABA binding. They were up to 5.5x10(5) and 4.7x10(6) times more active than monomeric T and GalNAc respectively. The affinity of ABA for ligands can be ranked as: multivalent T (alpha) (Galbeta1-3GalNAcalpha1-), Tn and I / II (Galbeta1-3GlcNac/Galbeta1-4GlcNAc, where GlcNAc represents 2-acetamido-2-deoxy-D-glucopyranose)>>>>monomeric T (alpha) and Tn > I >>GalNAc>>> II, L (Galbeta1-4Glc, where Glc represents D-glucopyranose) and Gal (inactive). These specific binding features of ABA establish the importance of affinity enhancement by high-density polyvalent (versus multiantennary I / II) glycotopes and facilitate our understanding of the lectin receptor recognition events relevant to its biological activities.

Edible mushroom (Agaricus bisporus) lectin modulates human retinal pigment epithelial cell behaviour in vitro

The retinal pigment epithelium (RPE) plays a major role in the development of proliferative vitreoretinopathy (PVR). In particular, RPE cells are implicated in generating the contraction forces seen. The present study was undertaken to investigate whether human RPE binds a lectin from the common edible mushroom, Agaricus bisporus, and to evaluate the effect of any binding on RPE-mediated matrix contraction in an in vitro model of PVR. Fluorescein isothiocyanate (FITC)-labelled Agaricus bisporus lectin (ABL) was used to study binding of lectin to normal retina, PVR scar tissue specimens and cultured human RPE. The effect of a 3-day exposure of ABL on human RPE-mediated contraction was evaluated using 2- and 3D RPE-populated collagen matrices. Effect of ABL on cell adhesion was measured using a collagen type I adhesion assay and determining the relative cellular attachment using absorbance readings. The normal RPE monolayer did not stain with FITC-ABL while PVR scar tissue stained intensely. Staining of in vitro RPE was characteristic but time-dependent. ABL caused a dose-dependent inhibition of RPE-mediated contraction of both 2D (one-way ANOVA, F = 7.94, p < 0.008) and 3D collagen matrices (one-way ANOVA, F = 164.955, p < 0.001). Pre-incubation of ABL with RPE in the 2D model caused a dramatic arrest of contraction (one-way ANOVA, F = 20.1, p < 0.001) that was due to a dose-dependent inhibition of adhesion (one-way ANOVA, F = 15.603, p < 0.001). Recovery of contraction was partially reversible on removal of ABL and was dependent on initial concentration of the lectin. ABL inhibits contraction and adhesion of human RPE cells in vitro without apparent cytotoxicity. It therefore deserves consideration as a potential therapeutic agent in the prevention and treatment of PVR and other non-ocular anomalous wound-healing processes.

Influence of terminal residue on adjacent disaccharide immunogenicity

Aberrant O-glycosylation of cell surface mucin antigens is characteristic of epithelial cancer cells. For example, Thomsen-Friedenreich disaccharide (TFD) is a chemically well-defined carbohydrate antigen with a documented link to malignancy. There have been many attempts to improve immune response to carbohydrate antigens, for use in immunotherapy. As part of an alternative strategy to improve carbohydrate immunogenicity, we studied the influence of terminal benzyl (Bzl) or p-nitrophenyl (pNP) residue on immunogenicity of adjacent TFD. Mice immunized with keyhole limpets hemocyanin-TFD (KLH-TFD), KLH-TFD(alpha)Bzl, or KLH-TFD(alpha)pNP produced anti-KLH antibodies, which were analyzed by enzyme-linked immunosorbent assay (ELISA). KLH-TFD did not give significant anti-TFD antibody titer, confirming the poor immunogenicity of TFD. Immunization with KLH-TFD(alpha)Bzl and KLH-TFD(alpha)pNP raised antibody titers against TFD(alpha)Bzl and TFD(alpha)pNP, respectively. KLH-TFD(alpha)Bzl also gave higher anti-TFD antibody response, whereas KLH-TFD(alpha)pNP did not, indicating that terminal Bzl residue improves immune response to adjacent carbohydrate. Analysis of anti-TFD(alpha)Bzl or anti-TFD(alpha)pNP IgG antibodies by competitive ELISA, using carbohydrate-related antigens as inhibitors, demonstrated their high specificity to their respective antigens. Anti-TFD(alpha)pNP antibody was not inhibited by TFD, but was significantly inhibited by GalNAc(alpha)pNP. The fact that p-nitrophenol (pNPol) has more competitive ability that GalNAc indicates that terminal polar residue is the main target antigen. In contrast, anti-TFD(alpha)Bzl antibody was inhibited to a similar degree by GalNAc(alpha)Bzl and TFD, confirming the carbohydrate recognition by antibodies yielded by terminal non-polar modification of the immunogen.

Structural requirements of carbohydrates to bind Agaricus bisporus lectin

Galbeta1-3GalNAc (T-disaccharide) and related molecules were assayed to describe the structural requirements of carbohydrates to bind Agaricus bisporus lectin (ABL). Results provide insight into the most relevant regions of T-disaccharide involved in the binding of ABL. It was found that monosaccharides bind ABL weakly indicating a more extended carbohydrate-binding site as compared to those involvedin the T-disaccharide specific lectins such as jacalin and peanut agglutinin. Lacto-N-biose (Galbeta1-3GlcNAc) unlike T-disaccharide, is unable to inhibit the ABL interaction, thus showing the great importance of the position of the axial C-4 hydroxyl group of GalNAc in T-disaccharide. This finding could explain the inhibitory ability of Galbeta1-6GlcNAc and lactose because C-4 and C-3 hydroxyl groups of reducing Glc, respectively, occupy a similar position as reported by conformational analysis. From the comparison of different glycolipids bearing terminal T-disaccharide bound to different linkages, it can be seen than ABL binding is even more impaired by an adjacent C-6 residual position than by the anomeric influence of T-disaccharide. Furthermore, the addition of beta-GlcNAc to the terminal T-disaccharide in C-3 position of Gal does not affect the ABL binding whereas if an anionic group such as glucuronic acid is added to C-3, the binding is partially affected. These findings demonstrate that ABL holds a particular binding nature different from that of other T-disaccharide specific lectins.