The HCoV-HKU1 N-Terminal Domain Binds a Wide Range of 9-O-Acetylated Sialic Acids Presented on Different Glycan Cores

Abstract

Coronaviruses (CoVs) recognize a wide array of protein and glycan receptors by using the S1 subunit of the spike (S) glycoprotein. The S1 subunit contains two functional domains: the N-terminal domain (S1-NTD) and the C-terminal domain (S1-CTD). The S1-NTD of SARS-CoV-2, MERS-CoV, and HCoV-HKU1 possesses an evolutionarily conserved glycan binding cleft that facilitates weak interactions with sialic acids on cell surfaces. HCoV-HKU1 employs 9-O-acetylated α2-8-linked disialylated structures for initial binding, followed by TMPRSS2 receptor binding and virus-cell fusion. Here, we demonstrate that the HCoV-HKU1 NTD has a broader receptor binding repertoire than previously recognized. We presented HCoV-HKU1 NTD Fc chimeras on a nanoparticle system to mimic the densely decorated surface of HCoV-HKU1. These proteins were expressed by HEK293S GnTI- cells, generating species carrying Man-5 structures, often observed near the receptor binding site of CoVs. This multivalent presentation of high mannose-containing NTD proteins revealed a much broader receptor binding profile compared to that of its fully glycosylated counterpart. Using glycan microarrays, we observed that 9-O-acetylated α2-3-linked sialylated LacNAc structures are also bound, comparable to OC43 NTD, suggesting an evolutionarily conserved glycan-binding modality. Further characterization of receptor specificity indicated promiscuous binding toward 9-O-acetylated sialoglycans, independent of the glycan core (glycolipids, N- or O-glycans). We demonstrate that HCoV-HKU1 may employ additional sialoglycan receptors to trigger conformational changes in the spike glycoprotein to expose the S1-CTD for proteinaceous receptor binding.