Chemoenzymatic Synthesis of Heparan Sulfate Oligosaccharides having Domain Structures

Abstract

Heparan sulfate (HS) can interact with various proteins that mediate various biological processes. The biosynthesis of HS is incomplete leading to molecular diversity to obtain well-defined compounds from natural sources difficultly. There is data to support that the domain structures of HS can regulate protein binding, however, such a binding mode has been difficult to probe because of a lack of methodologies to prepare well-defined HS oligosaccharides having domain structures. A chemoenzymatic methodology is described that can provide well-defined HS mimetics that have multiple NS domains separated by NA domains of different length. It is based on the chemical synthesis of a sulfated HS oligosaccharide that enzymatically could be extended by GlcA-GlcNAc moieties and a terminal GlcNAc-6N3 moiety. The reducing end of HS oligosaccharide could be functionalized by an alkyne moiety and the resulting compound could be coupled with the afore GlcNAc-6N3 HS oligosaccharide by CuAAC reaction to give compounds having two NS domains separated by an NA domain. The process of enzymatic NA introduction and click reaction could be repeated to give mimetics having three sulfated domains. SPR competition studies indicate that the length of NA domain influences the binding of chemokines CXCL8 and CCL5 in complex manners. We prepared heparin mimetics by head-to-tail CuAAC-mediated coupling of a sulfate oligosaccharide having an alkyne at the reducing end and azide moiety at the nonreducing end. By repeating the process of enzymatically introducing an azido containing GlcNAc moiety and chemically attachment of an anomeric alkyne containing moiety followed by CuAAC mediated coupling, well-defined multimeric compounds could be prepared composed of as many as 27 monosaccharides. We examined mono-, di-, tri-, and tetravalent heparin mimetics for their ability to inhibit the binding of the spike of SARS-CoV-2 to immobilized heparin. It was found that increasing the number of repeating units resulted in large increases in inhibitory potential and a tetravalent compound had similar potency compared to heparin. Similar inhibition was observed for the binding of trimeric RBD to Vero cells. The data support that well-defined heparin mimetics can be developed to inhibit the attachment of SARS-CoV-2 to the target cell. We described a modular synthetic methodology that can provide libraries of HS oligosaccharide bearing glucosamine residues modified by N-acetyl and N-sulfate moieties. It was found that the amino protecting groups trifluoromethylphenyl-methanimine and azide, as well as the hydroxyl protection by Lev, TDS, Alloc and Fmoc allow for selective manipulation of functionalities in the synthesis of high complex HS oligosaccharides. The microarray studies highlight key requirements of chemokines to engage with HS. The methodology will make it possible to prepare panels of compounds for structure-activity relationship investigations for a better understanding the biology of HS and provide the possibility to assemble HS oligosaccharides having NS and NA domains chemically. In summary, we have developed efficient methodology for the synthesis of HS derivatives of unprecedented complexity. Although further developments are needed to access the immense complexity found in natural HS, we were able to capture the domain architecture of HS in synthetic oligosaccharides.