Recombinant Rod Domain of Vimentin Reduces SARS-CoV-2 Viral Replication by Blocking Spike Protein-ACE2 Interactions

Abstract

Although the SARS-CoV-2 vaccination is the primary preventive intervention, there are still few antiviral therapies available, with current drugs decreasing viral replication once the virus is intracellular. Adding novel drugs to target additional points in the viral life cycle is paramount in preventing future pandemics. The purpose of this study was to create and test a novel protein to decrease SARS-CoV-2 replication. We created the recombinant rod domain of vimentin (rhRod) in E. coli and used biolayer interferometry to measure its affinity to the SARS-CoV-2 S1S2 spike protein and the ability to block the SARS-CoV-2-ACE2 interaction. We performed plaque assays to measure rhRod's effect on SARS-CoV-2 replication in Vero E6 cells. Finally, we measured lung inflammation in SARS-CoV-2-exposed K18-hACE transgenic mice given intranasal and intraperitoneal rhRod. We found that rhRod has a high affinity for the S1S2 protein with a strong ability to block S1S2-ACE2 interactions. The daily addition of rhRod decreased viral replication in Vero E6 cells starting at 48 h at concentrations >1 µM. Finally, SARS-CoV-2-infected mice receiving rhRod had decreased lung inflammation compared to mock-treated animals. Based on our data, rhRod decreases SARS-CoV-2 replication in vitro and lung inflammation in vivo. Future studies will need to evaluate the protective effects of rhRod against additional viral variants and identify the optimal dosing scheme that both prevents viral replication and host lung injury.

Keywords: COVID-19; SARS-CoV-2; acute lung injury; vimentin.

Figure 1

In silico prediction modeling of spike protein RBD, ACE2, and rhRod proteins. (a) Structure of spike protein RBD (cyan) in complex with its receptor ACE2 (salmon) (PDB: 6LZG). The spike protein RBM (blue) and N-terminal helix of ACE2 (red) are highlighted. (b) Overlay of rhRod (tan) between spike protein RBD (cyan) and ACE2 (salmon) proteins based on structural homology between ACE2 and rhRod (PDB:1GK7). The spike protein RBM (blue) and ACE2 N-terminal helix (red) are highlighted. (c) Representative model of the top-scoring predictions of rhRod monomer and spike protein RBD. (d) Representative model of the top scoring predictions of rhRod dimer and spike protein. Orientation of the predicted model in (d) closely matched the overlay of rhRod, as seen in (b).

Figure 2

SPOT peptide array of (a) SARS-CoV-2 spike protein RBD or (b) human vimentin protein, with each spot consisting of 20 aas and adjacent spots frame-shifted by 3 aas. Detection was either with (a) IR800-rhRod or (b) IR800-S1S2ECD-His. The letters and numbers correspond to the rows and columns, respectively, found in Supplementary Table S1.

Figure 3

Representative BLI fitting view and steady state analyses of (a,b) rhRod and (c,d) HI rhRod. Note the significantly smaller scale for the ordinates for HI rhRod (0–0.1 nm; (c,d)) compared to rhRod (0–0.6 nm (a,b)). The colors represent the different rhRod and HI rhRod concentrations (as listed in panel (a)). (e) Combined non-linear regression curve based on the lot-specific Req at each concentration. Error bars represent SEM and shaded areas indicate the 95% CI of the best-fit line.

Figure 4

Non-linear regression analyses of the normalized response using BLI to calculate the IC₅₀ of rhRod to block spike (S1S2ECD-His)–ACE2 protein interaction. (a) ACE2 was immobilized on AR2G sensors and then placed into wells containing 125 µM S1S2ECD-His with increasing concentrations of rhRod (0–500 nM). (b) S1S2ECD-His was immobilized on AR2G sensors and first placed into wells containing increasing concentrations of rhRod (0–1000 nM) before wash and placement into wells containing ACE2 (200 nM each). Experiments were performed in duplicate and data are normalized to the highest and lowest response values. Error bars represent SEM and shaded areas represent 95% CI of the best-fit line.

Figure 5

In vitro viral infection studies. Vero E6 cells were cultured in multi-well plates and infected with SARS-CoV-2 (MOI 0.1). (a,b) Cells treated with a single dose of rhRod at increasing concentrations did not have significant reductions in viral titers at 72 h; however, there was a decrease in plaques after 48 h. (c) Cells treated daily with rhRod started having decreased plaque counts after 48 h at rhRod doses of 1 µM and higher. (d) At 96 h, daily treatment with rhRod of 1 µM or greater lead to significant decreases in viral titers. Heat inactivated rhRod at the highest dose (4 µM) was similar to no rhRod. # indicates the slopes that were significantly different than the other doses. These data represent analysis in aggregate of two independent infection studies. * p < 0.05, *** p < 0.001, and **** p < 0.0001 using Dunnet’s multiple comparison test versus 0 µM.

Figure 6

(a) Representative lung histology from control- (left) and rhRod (right)-treated SARS-CoV-2-infected K18-hACE2 mice. Randomly selected sections (yellow boxes) of the lung were used to analyze the degree of acute lung injury at 4 days post-infection. Whole lung section scale bar = 500 µm; Fields of view scale bar = 50 µm. (b) Mice treated with rhRod had significantly less lung inflammation than control animals. Red = female and blue = male; * p < 0.05.