Heparin binding lectin of human placenta as a tool for histochemical ligand localization and isolation.

Biotinylated heparin has been used to detect the presence of specific binding sites in sections of human placenta, which has prompted demonstration of expression of lectin activity for this proteoglycan. Purification of this lectin from full-term placenta facilitates the synthesis of its biotinylated derivative, using biotin-amidocaproyl hydrazide, without affecting its activity. It also enables immunization to obtain antibodies. The labeled lectin is shown to bind specifically to nuclear and cytoplasmic locations in various cell types of human placenta, nuclear expression of lectin binding sites being more pronounced at the full-term stage than after 8 weeks of development. The structurally related histone H2B exhibits obvious differences in its binding pattern. The presence of ligands accessible to the lectin whose binding activity can be inhibited by addition of an excess of heparin correlates in most instances with the level of lectin expression detected immunohistochemically. Biochemical information on the nature of the glycohistochemically inferred lectin-specific ligand(s) is obtained by affinity chromatography on resin-immobilized lectin. It leads to isolation of a proteoglycan with similar electrophoretic mobility in agarose-polyacrylamide gel electrophoresis relative to the independently purified heparan sulfate-containing fibronectin binding proteoglycan from human placenta. Both fractions inhibit binding of heparin to the lectin and contain immunologically detected co-purified lectin, emphasizing their ligand properties. Application of labeled tissue lectins in conjunction with lectin-specific antibodies is proposed to obtain valuable insights into the expression of the receptor as well as the ligand part of protein-carbohydrate recognition.

Heparin-binding lectin from human placenta: further characterization of ligand binding and structural properties and its relationship to histones and heparin-binding growth factors.

We have previously demonstrated that the heparin-binding lectin of human placenta dissociates into up to four distinct polypeptides with molecular weights of 14,400, 15,000, 16,200, and 16,700 (Kohnke-Godt, B., & Gabius, H.-J. (1989) Biochemistry 28, 6531-6538). Stable complexes to ligands can shift the molecular weight appearance of the lectin to higher values. They can be dissociated in the additional presence of 9 M urea or by enzymatic degradation of heparin in model studies. The binding of heparin is rather stable over a range of salt concentrations from 1 to 3 M NaCl. Chemical modification with group-specific reagents to arginine, lysine, histidine, tyrosine, and tryptophan results in substantial inactivation of binding activity. Further amino-terminal sequence analyses point to a high-scoring relationship in this region to histone sequences, namely, histone H2B, but to no published sequences for any heparin-binding growth factor. Calculation of relatedness on the basis of differences in amino acid composition corroborates the conclusion of molecular distinction between the lectin, histones H2A and H2B, and the fibroblast growth factor as well as angiogenin. Histones only weakly agglutinate type II erythrocytes in contrast to the lectin. The immobilized lectin exhibits two classes of binding sites with KD values of 3 and 110 nM in contrast to one estimated KD value of 250 nM with a commercially available histone fraction. Both fractions retain binding activity to biotinylated heparin in transblots and are immunologically cross-reactive to antibodies, raised against the lectin as antigen. Subcellular fractionation clearly demonstrates that heparin-inhibitable hemagglutination activity and immunologically cross-reactive protein bands, characteristic for the lectin, but not unequivocally distinguishable from certain histone fractions in blots, are not confined to the nuclear fraction in the human placenta.

Tissue distribution of radioiodinated neoglycoproteins and mammalian lectins.

Quantitation of tissue distribution of radioiodinated neoglycoproteins 1 h after intravenous injection into mice allowed to evaluate their suitability to uncover potential selectivity in tracer retention. Variations within the panel of neoglycoproteins were introduced to the carbohydrate determinant, its density and linkage to the carrier. Five arrays of neoglycoproteins, encompassing up to twelve different carbohydrate moieties were used. The individual response on the level of organ content showed differences, accounted for by carbohydrate structure and density. However, increase in sugar density eventually caused general decrease in tissue retention, emphasizing the importance of synthetic parameters. Attachment of sugar residues to the spacer via primarily the C-6 group of monosaccharides led to rather prolonged survival in circulation of the resulting neoglycoprotein compared to the application of neoglycoproteins with p-aminophenyl glycosides as derivatives for coupling. Besides applying neoglycoproteins tissue uptake was also measured for several organs, when four mammalian lectins were employed as radiotracers. These lectins bind to cellular carbohydrate ligands, namely beta-galactosides, alpha-fucosides or heparin. Differences were measured for retention in liver, kidneys, spleen, stomach, thymus and bone marrow. The distinct properties of different tissues with respect to binding of neoglycoproteins as well as to endogenous lectins, exhibiting a certain degree of selectivity, are a step within the framework to attempt to therapeutically exploit the carrier potential of probes by recognitive protein-carbohydrate interactions.

Heparin-binding lectin from human placenta: purification and partial molecular characterization and its relationship to basic fibroblast growth factors.

The heparin-binding lectin from human placenta is isolated on the basis of its tendency to form large aggregates by gel filtration and on the basis of its affinity for heparin by affinity chromatography. The purified lectin dissociates into up to four distinct polypeptides with molecular weight values of 14,400, 15,000, 16,200, and 16,700 and a single isoelectric point of 9.0. Molecular heterogeneity is not due to different degrees of glycosylation, as evidenced by gel electrophoretic analysis after extensive treatment with various endoglycosidases. Despite its similarities of affinity to heparin, molecular size, and isoelectric point to the basic fibroblast growth factor (bFGF), the comparatively high yield of the lectin (approximately 1.5 mg/100 g of placenta), the occurrence of proteolytic fragmentation in the presence of heparin, and the lack of homology to the amino-terminal sequence of the lectin argue against any notable relationship to bFGF. Most importantly, the lack of mitogenic activity in a commonly used bioassay with quiescent 3T3 fibroblasts rules out any FGF-like activity on cell proliferation. The heparin-binding lectin is thus clearly distinguishable from heparin-binding growth factors. By employing biotinylated heparin as labeled ligand to visualize and quantify heparin binding, hapten inhibition in a solid-phase assay reveals that except for heparin no other vertebrate glycosaminoglycan but the sulfated fucan fucoidan can effectively reduce the Ca2+-independent ligand binding. Proteolytic fragmentation by chymotrypsin in two independent assays demonstrates that a fragment of Mr 7800 still retains ability to bind heparin. The interaction of this lectin with naturally occurring heparin-like molecules may physiologically be involved in modulatory regulation of heparin-mediated processes.