Human galectin-1 recognition of poly-N-acetyllactosamine and chimeric polysaccharides

Human galectin-1 is a dimeric carbohydrate binding protein (Gal-1) (subunit 14.6 kDa) widely expressed by many cells but whose carbohydrate binding specificity is not well understood. Because of conflicting evidence regarding the ability of human Gal-1 to recognize N-acetyllactosamine (LN, Galbeta4GlcNAc) and poly-N-acetyllactosamine sequences (PL, [-3Galbeta4GlcNAcbeta1-]n), we synthesized a number of neoglycoproteins containing galactose, N-acetylgalactosamine, fucose, LN, PL, and chimeric polysaccharides conjugated to bovine serum albumin (BSA). All neoglycoproteins were characterized by MALDI-TOF. Binding was determined in ELISA-type assays with immobilized neoglycoproteins and apparent binding affinities were estimated. For comparison, we also tested the binding of these neoglycoconjugates to Ricinus communis agglutinin I, (RCA-I, a galactose-binding lectin) and Lycopersicon esculentum agglutinin (LEA, or tomato lectin), a PL-binding lectin. Gal-1 bound to immobilized Galbeta4GlcNAcbeta3Galbeta4Glc-BSA with an apparent K(d) of approximately 23 micro M but bound better to BSA conjugates with long PL and chimeric polysaccharide sequences (K(d)'s ranging from 11.9 +/- 2.9 microM to 20.9 +/- 5.1 micro M). By contrast, Gal-1 did not bind glycans lacking a terminal, nonreducing unmodified LN disaccharide and also bound very poorly to lactosyl-BSA (Galbeta4Glc-BSA). By contrast, RCA bound well to all glycans containing terminal, nonreducing Galbeta1-R, including lactosyl-BSA, and bound independently of the modification of the terminal, nonreducing LN or the presence of PL. LEA bound with increasing affinity to unmodified PL in proportion to chain length. Thus Gal-1 binds terminal beta4Gal residues, and its binding affinity is enhanced significantly by the presence of this determinant on long-chain PL or chimeric polysaccharides.     

Purification and cDNA cloning of UDP-GlcNAc:GlcNAcbeta1-3Galbeta1-4Glc(NAc)-R [GlcNAc to Gal]beta1,6N-acetylglucosaminyltransferase from rat small intestine: a major carrier of dIGnT activity in rat small intestine

A rat intestinal beta1,6N-acetylglucosaminyltransferase (beta1-6GnT) responsible for the formation of the beta1,6-branched poly-N-acetyllactosamine structure has been purified to apparent homogeneity by successive column chromatographic procedures using an assay wherein pyridylaminated lacto- N-triose II (GlcNAcbeta1-3Galbeta1-4Glc-PA) was used as an acceptor substrate and the reaction product was GlcNAcbeta1-3(GlcNAcbeta1-6)Galbeta1-4Glc-PA. The purified enzyme catalyzed the conversion of the polylactosamine acceptor GlcNAcbeta1-3'LacNAc into GlcNAcbeta1-3'(GlcNAcbeta1-6') LacNAc (dIGnT activity), but it could not transfer GlcNAc to LacNAcbeta1-3'LacNAc (cIGnT activity). This enzyme could also convert mucin core 1 and core 3 analogs, Galbeta1-3GalNAcalpha1-O-paranitrophenyl (pNP) and GlcNAcbeta1-3GalNAcalpha1-O-pNP, into Galbeta1-3(GlcNAcbeta1-6) GalNAcalpha1-O-pNP (C2GnT activity) and GlcNAcbeta1-3(GlcNAcbeta1-6)GalNAcalpha1-O-pNP (C4GnT activity), respectively. Based on the partial amino acid sequences of the purified protein, the cDNA encoding this enzyme was cloned. The COS-1 cells transiently transfected with this cDNA had high dI/C2/C4GnT activities in a ratio of 0.34:1.00:0.90, compared with non- or mock-transfected cells. The primary structure shows a significant homology with human and viral mucin-type core 2 beta1-6GnTs (C2GnT-Ms), indicating that this enzyme is the rat ortholog of human and viral C2GnT-Ms. This is the first identification and purification of this enzyme as a major carrier of dIGnT activity in the small intestine. This rat ortholog should mostly be responsible for making distal I-branch structures on poly-N-acetyllactosamine sequences in this tissue, as well as making mucin core 2 and core 4 structures, given that it also has high C2/C4GnT activities.     

Altered T cell surface glycosylation in HIV-1 infection results in increased susceptibility to galectin-1-induced cell death

The massive T cell death that occurs in HIV type 1 (HIV-1) infection contributes profoundly to the pathophysiology associated with AIDS. The mechanisms controlling cell death of both infected and uninfected T cells ("bystander" death) are not completely understood. We have shown that HIV-1 infection of T cells results in altered glycosylation of cell surface glycoproteins; specifically, it decreased sialylation and increased expression of core 2 O-glycans. Galectin-1 is an endogenous human lectin that recognizes these types of glycosylation changes and induces cell death of activated lymphocytes. Therefore we studied the possible contribution of galectin-1 in the pathophysiology of AIDS. O-glycan modifications were investigated on peripheral lymphocytes from AIDS patients. Oligosaccharides from CD43 and CD45 of CEM cells latently infected with HIV-1 were chemically analyzed. Consistent with our previous results, we show that HIV-1 infection results in accumulation of exposed lactosamine residues, oligosaccharides recognized by galectin-1 on cell surface glycoproteins. Both latently HIV-1-infected T cell lines and peripheral CD4 and CD8 T cells from AIDS patients exhibited exposed lactosamine residues and demonstrated marked susceptibility to galectin-1-induced cell death, in contrast to control cultures or cells from uninfected donors. The fraction of cells that died in response to galectin-1 exceeded the fraction of infected cells, indicating that death of uninfected cells occurred. Altered cell surface glycosylation of T cells during HIV-1 infection increases the susceptibility to galectin-1-induced cell death, and this death pathway can contribute to loss of both infected and uninfected T cells in AIDS.     

Differential Expression Patterns of N-Acetylglucosaminyl Transferases and Polylactosamines in Uterine Lesions

Polylactosamine (polyLacNAc) is a fundamental structure in glycoconjugates and it is expressed in specific cells/tissues associated with the development and carcinogenesis. β1,3-N-acetylglucosaminyl transferases ((β3GnTs) play an important role in polyLacNAc synthesis, however the roles of these glycosyltransferases and their products in cancer progression are still unclear. In this sense, this work aimed to evaluate differential expression pattern of the N-acetylglucosaminyl transferases and polylactosamines in invasive and premalignant lesions of the uterus cervix. The expression of β3GnT2 and β3GnT3 were evaluated in normal (n=10) and uterine cervix lesions (n=120), both malignant [squamous carcinoma (SC)] and premalignant [cervical intraepithelial neoplasia (CIN), grades 1, 2 and 3] using immunohistochemistry. Besides, lectin histochemistry with Phytolacca americana lectin (PWM) and Wheat germ agglutinin (WGA) was also carried out to observe the presence of polyLacNAc chains and N-acetylglucosamine (GlcNAc), respectively. The β3GnT3 was expressed in almost all samples (99%) and β3GnT2 was higher expressed in disease samples mainly in CIN 3, when compared with normal (P=0.002), CIN 1 (P=0.009) and CIN 2 (P=0.03). The expression of polyLacNAc was higher is SC samples, when compared with normal (P=0.03), CIN 1 (P=0.02) and CIN 3 (P=0.004), and was observed only nuclear expression in nearly 50% of the SC samples, showing a statistically significant when compared with normal (P=0.01), CIN 1 (P=0.002), CIN 2 (P=0.007) and CIN 3 (P=0.04). Deferring from transferases and polyLacNAc chains, GlcNAc (WGA ligand) reveals a gradual staining pattern decrease with the increase of the lesion degree, being more expressed in CIN 1 lesions when compared with normal (P<0.0001), CIN 2 (P<0.0001), SC (P<0.0001) and CIN 3 (P=0.0003). Our data reveal that β3GnT2 and polyLacNAc may be involved in the progression of the pre-malignant lesions of the human uterine cervix. In addition, polyLacNAc expression only in the nucleus can be associated a poor prognostic in uterine lesions.