The biological activity of basic fibroblast growth factor (bFGF) is influenced greatly by direct binding to heparin and heparan sulphate (HS). Heparin-derived oligosaccharides have been utilized to determine the structural requirements present in the polymer that account for binding to bFGF. We had previously demonstrated that fragments > 6 mer can inhibit the interaction between cell surface heparan sulphate proteoglycan (HSPG) and bFGF, and bFGF-induced proliferation of adrenocortical endothelial (ACE) cells. In contrast, oligosaccharides > 10 mer can enhance the binding of bFGF to its high-affinity receptor or support bFGF-induced mitogenesis in ACE cells (Ishihara et al., J. Biol. Chem., 268, 4675-4683, 1993). We have extended these studies to size- and structure-defined oligosaccharides from heparin, 2-O-desulphated (2-O-DS-) heparin, 6-O-desulphated (6-O-DS-) heparin, carboxy-reduced (CR-) heparin and carboxy-amidomethylsulphonated (AMS-) heparin. Oligosaccharides from these polymers were fractionated on a bFGF-affinity column and were assessed as inhibitors or enhancers of specific bFGF-derived biological activities. The results of these studies indicate that both 2-O-sulphate and the negative charge of the carboxy group [L-iduronic acid (IdoA) residues] are required for specific interactions of heparin-derived oligosaccharides with bFGF and for modulation of bFGF mitogenic activity. In addition, the charge of the carboxy groups in uronic acids can be replaced by other functional groups with a negative charge, such as the amidomethyl sulphonate moiety described here.

Heparin, carboxy-group-reduced heparin, several sulphated monosaccharides and disaccharides formed from heparin, and a tetrasaccharide prepared from chondroitin sulphate were treated at 100°C with hydrazine containing 1% hydrazine sulphate for periods sufficient to cause complete N-deacetylation of the N-acetylhexosamine residues. Under these hydrazinolysis conditions both the N-sulphate and the 0-sulphate substituents on these compounds were completely stable. However, the uronic acid residues were converted into their hydrazide derivatives at rates that depended on the uronic acid structures. Unsubstituted L-iduronic acid residues reacted much more slowly than did unsubstituted D-glucuronic acid or 2-O-sulphated L-iduronic acid residues. The chemical modification of the carboxy groups resulted in a low rate of C-5 epimerization of the uronic acid residues. The hydrazinolysis reaction also caused a partial depolymerization of heparin but not of carboxy-groupreduced heparin. Treatment of the hydrazinolysis products with HNO2 at either pH4 or pH 1.5 or with HI03 converted the uronic acid hydrazides back into uronic acid residues. The use of the hydrazinolysis reaction in studies of the structures of uronic acid-containing polymers and the implications of the uronic acid hydrazide formation are discussed.

Binding of HSV-1 to cells is mediated by interactions of virion glycoproteins gC and/or gB with heparin sulfate (HS) glycosaminoglycans on cell surface proteoglycans. HS and the related glycosaminoglycan, heparin, comprise a family of heterogeneous carbohydrates composed of long, unbranched polysaccharides modified, for example, by sulfations and acetylations. To define the specific features of HS important for viral binding, we took advantage of the structural similarities between heparin and cell surface HS and compared the ability of chemically modified heparin compounds to inhibit the binding of viral particles to the cell surface and subsequent plaque formation. Because binding presumably involves multiple, complex interactions between both known heparin-binding glycoproteins, gC and gB, and cell surface HS, we compared the effects of modified heparin compounds on the binding and subsequent plaque formation of wild-type and gC-negative strains of HSV-1 and, in select cases, the binding of gB-negative virus to cells. We identified specific structural features of heparin essential for the inhibition of viral binding. For example, both N-sulfation and 6-O-sulfation must be important determinants since desulfation of heparin at these sites abolished or decreased the antiviral activity of heparin. Moreover, we found that the antiviral activity of heparin was independent of its anticoagulant activity. Carboxyl-reduced and 2,3-O-desulfated heparin selectively inhibited binding of gC-positive viruses (wild-type or a gB-negative strain) to cells, but had little or no inhibitory effect on binding and subsequent plaque formation for a gC-deletion virus. These results suggest that gC and gB interact with different structural features of HS.

No abstract