SARS-CoV-2 Binding to Terminal Sialic Acid of Gangliosides Embedded in Lipid Membranes

Abstract

Multiple recent reports indicate that the S protein of SARS-CoV-2 specifically interacts with membrane receptors and attachment factors other than ACE2. They likely have an active role in cellular attachment and entry of the virus. In this article, we examined the binding of SARS-CoV-2 particles to gangliosides embedded in supported lipid bilayers (SLBs), mimicking the cell membrane-like environment. We show that the virus specifically binds to sialylated (sialic acid (SIA)) gangliosides, i.e., GD1a, GM3, and GM1, as determined from the acquired single-particle fluorescence images using a time-lapse total internal reflection fluorescence (TIRF) microscope. The data of virus binding events, the apparent binding rate constant, and the maximum virus coverage on the ganglioside-rich SLBs show that the virus particles have a higher binding affinity toward the GD1a and GM3 compared to the GM1 ganglioside. Enzymatic hydrolysis of the SIA−Gal bond of the gangliosides confirms that the SIA sugar unit of GD1a and GM3 is essential for virus attachment to the SLBs and even the cell surface sialic acid is critical for the cellular attachment of the virus. The structural difference between GM3/GD1a and GM1 is the presence of SIA at the main or branched chain. We conclude that the number of SIA per ganglioside can weakly influence the initial binding rate of SARS-CoV-2 particles, whereas the terminal or more exposed SIA is critical for the virus binding to the gangliosides in SLBs.

Giant Plasma Membrane Vesicles as Cellular‐Mimics for Probing SARS‐CoV‐2 Binding at Single Particle Level

“…This labelling approach for enveloped viruses has been applied by multiple groups including us. [11,12,15,39,41] We characterized the labelled viruses upon fluorescence imaging with wide-field illumination and compared the signal intensity using a control system (liposomes labelled with DiI). The particle intensity analysis reveals that < 20 % of the labelled viruses are likely to be in aggregated states (Figure S2 for details).…”

“…The purified virus sample was then genome‐inactivated with UV treatment for performing imaging experiments with a relatively lower biosafety requirement. Recently, we have implemented this UV‐inactivation approach (also by other groups [40] ) to examine the binding of SARS‐CoV‐2 to supported lipid bilayers and have shown that controlled UV exposure affects the viral genome without perturbing the intact viral structure [39] (data in Figures S1A–C). The S protein remains functional for antibody binding as detected upon immunoblotting (Figure 1A).…”

“…Binding assay of labelled virus : Virus stock solution (protein content 0.5 mg/mL) was labelled using membrane‐anchoring fluorophore DiI as reported earlier [39] and detailed in SI section 1.4. The labelled virus sample was used for examining virus binding to the adherent cells, cell‐attached GPMVs, and soluble GPMVs.…”

“…[35] For binding studies, we have genome-inactivated the virus upon controlled UV-treatment. [36][37][38] This approach has been discussed in detail in our recently published article [39] and described in the SI section 1.4. Both the live and genome-inactivated viruses were characterized with total protein content, Western blot, and immunostaining (fluorescence imaging).…”

Kinetics of Ganglioside-Rich Supported Lipid Bilayer Formation with Tracer Vesicle Fluorescence Imaging

“…Even pathogens utilize gangliosides as membrane receptors or attachment factors for their cellular attachment and entry. For example, GM1 is a cellular receptor of cholera toxin and E. coli , tetanus, and botulinum neurotoxins bind to GT1b ganglioside, viruses like influenza A virus, Sendai virus, and polyomaviruses bind to membrane gangliosides, ,, and recently, we showed that gangliosides are attachment factors of SARS-CoV-2 …”