Two Receptor Binding Strategy of SARS-CoV-2 Is Mediated by Both the N-Terminal and Receptor-Binding Spike Domain

Abstract

It is not well understood why severe acute respiratory syndrome (SARS)-CoV-2 spreads much faster than other β-coronaviruses such as SARS-CoV and Middle East respiratory syndrome (MERS)-CoV. In a previous publication, we predicted the binding of the N-terminal domain (NTD) of SARS-CoV-2 spike to sialic acids (SAs). Here, we experimentally validate this interaction and present simulations that reveal a second possible interaction between SAs and the spike protein via a binding site located in the receptor-binding domain (RBD). The predictions from molecular-dynamics simulations and the previously-published 2D-Zernike binding-site recognition approach were validated through flow-induced dispersion analysis (FIDA)─which reveals the capability of the SARS-CoV-2 spike to bind to SA-containing (glyco)lipid vesicles, and flow-cytometry measurements─which show that spike binding is strongly decreased upon inhibition of SA expression on the membranes of angiotensin converting enzyme-2 (ACE2)-expressing HEK cells. Our analyses reveal that the SA binding of the NTD and RBD strongly enhances the infection-inducing ACE2 binding. Altogether, our work provides in silico, in vitro, and cellular evidence that the SARS-CoV-2 virus utilizes a two-receptor (SA and ACE2) strategy. This allows the SARS-CoV-2 spike to use SA moieties on the cell membrane as a binding anchor, which increases the residence time of the virus on the cell surface and aids in the binding of the main receptor, ACE2, via 2D diffusion.

Figure 1

Characterization of a SARS-CoV-2 spike region very similar to the SA binding site on the MERS-CoV spike. (a) Molecular surface representation of the N-terminal region of the MERS-CoV spike bound to SA, along with the corresponding representative 2D-patch disk. (b) Molecular surface representation of the N-terminal region of SARS-CoV-2, colored from white to red according to their shape similarity with the MERS-CoV binding region, along with the corresponding representative 2D-patch disk. (c) Same as in panel b, but for SARS-CoV. (d) Cartoon representation of the SARS-CoV-2 NTD with the predicted binding region highlighted in red. (e) As in d, but rotated.

Figure 2

Contact probability as determined with MD simulations of the S1, RBD, and NTD domains of the SARS-CoV-2 spike. (a) Contact probability between each residue and the SA molecule is shown for the three molecular systems: the S1 segment, RBD, and NTD of the spike protein. (b) Molecular structure of the spike in its trimeric form: one chain is depicted as a blue-colored cartoon, and the molecular surfaces are highlighted for the remaining two chains. The residues with the highest probability of interaction, as calculated from the MD trajectory of the largest system, S1, are represented in cyan and are located both in the NTD and in the RBD. (c) Box-and-whisker plot of the minimum distance of the SA molecule to the nearest protein residue during the SARS-CoV-2 NTD MD simulation as a function of the binding energies. The inset displays the distribution of binding energies of the SA molecule to the MERS-CoV NTD during the simulation, with the thick red line indicating the mean value and the thin red line the standard deviation.

Figure 3

FIDA of SARS-CoV-2 spike S1 segment binding to SA containing glycolipids. (a) Schematic of how complex formation by fluorescently labeled S1 will affect the Taylorgrams of FIDA measurements. A single species that includes a fluorescent label will give rise to a single Gaussian line shape, with a Gaussian width σ that increases with an increasing hydrodynamic radius. (b) Taylorgrams recorded for pure 100 nM spike S1 protein, with 200 μM DPPC vesicles, and with 1:9 GM1:DPPC vesicles, partly composed of the SA containing GM1 glycolipid. In the case of the glycolipid-containing vesicles, a double Gaussian line shape is observed, indicating the presence of both free S1 species and S1-vesicle complexes, as opposed to the single Gaussian lineshapes observed for the pure S1 for in the presence of the DPPC vesicles that do not contain glycolipids, indicating the absence of S1-binding in that case. (c) The relative number of bound species for the S1 segments of SARS-CoV, SARS-CoV-2, and MERS-CoV bound to (1:9)-GM1:DPPC and -GM3:DPPC lipid vesicles (lipid concentration in monomer units). Lines show best fit to eq 1. (d) Estimated dissociation constants (K_Ds).

Figure 4

Decreased binding of the SARS-CoV-2 full-length spike, S1, and RBD proteins to HEK-ACE2 cells upon inhibition of SA expression. (a–c) Example of an applied FCM gating strategy to assess full-length spike protein binding to single HEK-ACE2 cells, and the Alexa-488 fluorescence resulting from spike binding to the gated ensemble. (d) The normalized geometric mean of the Alexa-488 fluorescence intensity (gMFI) that indicates the relative binding strength of various segments of the SARS-CoV-2 spike protein to HEK-ACE2 cells, the Lectenz binding (which reports on the presence of SA groups on the cellular membrane), and ACE2 levels, as determined by FCM, with and without the addition of sialostatin (n = 6, 2-sided paired t-test). Statistics per experiment: SARS-CoV-2 spike S1 domain: (n = 5, 2-sided paired t-test), with one outlier (marked pink) was excluded based on a GRUBS test (alpha 0.05); SARS-CoV-2 spike RBD domain: n = 6, 2-sided paired t-test; SiaFind Pan-Specific Lectenz with and without the addition of sialostatin: n = 5, 2-sided paired t-test; ACE2 expression level of HEK-ACE2 cells: (n = 6, 2-sided paired t-test). ns: not significant.

Figure 5

Two receptor binding strategy of SARS-CoV-2 is mediated by both the N-terminal and receptor-binding spike domain. Computations and experiments reveal that both the NTD and RBD can bind to SA groups at the cell membrane. Both binding events have a positive effect on subsequent ACE2 binding. After a 3D diffusion process in which the spike proteins can interact with the omnipresent SA moieties on cellular membranes, SA binding will reduce the dimensionality of the search for the ACE2 to a 2D-diffusion process. As the latter binding triggers virion internalization, the two-receptor and two binding-domain strategy of SARS-CoV-2 strongly enhances its infection rate. Protein and virion renderings are reproduced from ref (57) under the CC-BY-4.0 license and the lipid rendering is created in UCSF Chimera.