Polyunsaturated ω-3 fatty acids inhibit ACE2-controlled SARS-CoV-2 binding and cellular entry

Abstract

The strain SARS-CoV-2, newly emerged in late 2019, has been identified as the cause of COVID-19 and the pandemic declared by WHO in early 2020. Although lipids have been shown to possess antiviral efficacy, little is currently known about lipid compounds with anti-SARS-CoV-2 binding and entry properties. To address this issue, we screened, overall, 17 polyunsaturated fatty acids, monounsaturated fatty acids and saturated fatty acids, as wells as lipid-soluble vitamins. In performing target-based ligand screening utilizing the RBD-SARS-CoV-2 sequence, we observed that polyunsaturated fatty acids most effectively interfere with binding to hACE2, the receptor for SARS-CoV-2. Using a spike protein pseudo-virus, we also found that linolenic acid and eicosapentaenoic acid significantly block the entry of SARS-CoV-2. In addition, eicosapentaenoic acid showed higher efficacy than linolenic acid in reducing activity of TMPRSS2 and cathepsin L proteases, but neither of the fatty acids affected their expression at the protein level. Also, neither reduction of hACE2 activity nor binding to the hACE2 receptor upon treatment with these two fatty acids was observed. Although further in vivo experiments are warranted to validate the current findings, our study provides a new insight into the role of lipids as antiviral compounds against the SARS-CoV-2 strain.

Figure 1

Effects of PUFAs on binding of RBD sequence of SARS-CoV-2 to human ACE2 receptor. (A) Binding of RBD sequence of SARS-CoV-2 spike protein to immobilized hACE2 receptor. HRP-conjugated RBD sequence was treated with indicated FAs at different concentrations for 30 min. followed by incubation for 15 min. with hACE2 receptors immobilized on plate. HRP signal was measured at 450 nm. (B) Binding of A549 cells expressing SARS-CoV-2 eGFP-spike protein in the present of selected FAs at different concentrations to soluble hACE2 receptor. Binding was performed on plates with immobilized human monoclonal antibody against hACE2 receptor at 10 µg/ml concentration and detected as green fluorescence signal. Data are presented as % of control ± SD; #p ≤ 0.05, ∆p ≤ 0.01, *p ≤ 0.001. Control—0.025% DMSO, positive and negative controls were provided by the manufacturer.

Figure 2

Effects of linolenic acid and EPA on viability of ACE2 expressing A549 cells and on human ACE2 receptor. (A) Viability of A549/hACE2 cells after 1 h, 3 h, and 48 h incubation with linolenic acid and EPA at different concentrations using MTT assay. Cell viability is expressed as change in absorbance at 570 nm in % compared to lipid-free control ± SD; positive control—100% dead cells, negative control—addition-free sample. (B) Binding of linolenic acid and EPA at indicated concentrations to hACE2 receptor (1.0 μg/ml) immobilized on the plate using human primary anti-ACE2 antibody at 1:500 dilution and HRP-conjugated secondary antibody at 1:1000 dilution, and measuring chemiluminescence signal. Data are presented as % of lipid-free control ± SD; control—0.025% DMSO, positive control—50% DMSO. (C) Activity of hACE2 upon treatment with selected FAs and indicated concentrations. Purified hACE2 enzyme at 0.1 ng/µl was incubated with linoleic acid and EPA at different concentrations for 1 h at RT followed by addition of 25 µl fluorogenic substrate for 30 min. Fluorescence signal was measured at Ex/Em = 535/595 nm using spectrofluorimeter. Data are presented as % of lipid-free control ± SD; *p ≤ 0.001. Control—0.025% DMSO, positive control—10% DMSO.

Figure 3

Effects of linolenic acid and EPA on SARS-CoV-2 pseudo-virion binding to human ACE2 receptor. Binding SARS-CoV-2 spike protein encapsulated pseudo-virions to A549 cells stably overexpressing human ACE2 receptor was evaluated with monoclonal antibody. (A) Spike-pseudo-virions were treated with indicated FAs at different concentrations for 1 h before inoculation, added simultaneously or 1 h after inoculation of the virions into hACE2/A549 cells. Next, wells were incubated for 1 h at 37^OC, washed and binding was evaluated by adding primary anti-spike protein monoclonal antibody at 1:1000 dilution followed by secondary HRP-conjugated antibody at 1:2500 dilution and signal measurement at 450 nm. (B) Spike-pseudo-virions were treated with indicated FAs at different concentrations for 1 h before inoculation, added simultaneously or 1 h after inoculation of the virions into hACE2/A549 cells. Next, wells were incubated for 3 h at 37^OC, washed and binding was evaluated by adding primary anti-spike protein monoclonal antibody at 1:1000 dilution followed by secondary HRP-conjugated antibody at 1:2500 dilution and signal measurement at 450 nm. Data are presented as % of control ± SD; #p ≤ 0.05, ∆p ≤ 0.01, *p ≤ 0.001. Controls—0.025% DMSO, positive and negative controls were provided by the manufacturer.

Figure 4

SARS-CoV-2 eGFP-luciferase-pseudo-virion binding and cellular entry. Binding and entry of SARS-CoV-2 pseudo-virions with encapsulated eGFP-luciferase spike protein was evaluated without spinfaction and with spinfaction. (A) A549 cells stably overexpressing hACE2 receptor were inoculated (no spinfection) with SARS-CoV-2 pseudo-virions treated with indicated FAs at different concentrations and time. After 48 h post-inoculation time, transduction efficacy was measured according to luciferase activity. (B) Binding and entry of SARS-CoV-2 pseudo-virions with encapsulated eGFP-luciferase Spike protein. A549 cells stably overexpressing hACE2 receptor were inoculated (spinfection) with SARS-CoV-2 pseudo-virions treated with indicated FAs at different concentrations and time. After 48 h post-inoculation time, transduction efficacy was measured according to luciferase activity. Data are presented as % of control ± SD; #p ≤ 0.05, ∆p ≤ 0.01, *p ≤ 0.001. Control—0.025% DMSO, positive control—bald SARS-CoV-2 eGFP-luciferase-pseudo-virions, negative control—ΔG-luciferase rVSV pseudo-typed particles; red fame—concentrations that showed 85–100% cytotoxicity.

Figure 5

Effect of selected FAs on fusion to human ACE2 receptor overexpressing A549 cells. (A). Cell–cell fusion of A549 cells expressing eGFP spike protein with A549 cells stably expressing human ACE2 receptor. A549 cells expressing eGFP spike protein were pre-treated with indicated FAs at different concentrations for 1 h at 37 °C and co-cultured for additional 4 h at 37 °C with A549 cells stable expressing human ACE2 receptor. The scale bar indicates 250 µm. (B) Quantitative analysis of formed syncytia. Experiments were done in triplicate and repeated three times. Data are presented as % of control ± SD; *p ≤ 0.001. Control—0.025% DMSO, positive control—20 μg/ml anti-ACE2 antibody.

Figure 6

Effect of selected FAs on TMPRSS2 and Cathepsin L proteases. (A) Purified TMPRSS2 enzyme at 1 µM was incubated with selected FAs at different concentrations for 1 h at RT followed by addition of fluorogenic substrate at 10 µM concentration for 30 min. Fluorescence signal was measured at Ex/Em = 360/440 nm using spectrofluorimeter (upper panel). A549 cells were treated with indicated FAs at different concentrations for 3 h and 48 h at 37 °C and enzymatic activity was measured by addition of 0.2 mM fluorogenic substrate and incubation for 30 min. at 37 °C. Fluorescence signal was measured at Ex/Em = 360/440 nm using spectrofluorimeter (lower panel). Data are presented as % of control ± SD; #p ≤ 0.05, ∆p ≤ 0.01, *p ≤ 0.001. Control—0.025% DMSO, positive control—50–100 μM camostat mesylate. (B) Purified cathepsin L enzyme at 0.02 ng/µl was incubated with selected FAs at different concentrations for 1 h at RT followed by addition of fluorogenic substrate at 10 µM concentration for 30 min. Fluorescence signal was measured at Ex/Em = 360/440 nm using spectrofluorimeter (upper panel). A549 cells were treated with indicated lipids at different concentrations for 24 h at 37 °C and enzymatic activity was measured by addition of 0.2 mM fluorogenic substrate and incubation for 30 min. at 37 °C. Fluorescence signal was measured at Ex/Em = 360/535 nm using spectrofluorimeter (lower panel). Data are presented as % of control ± SD; #p ≤ 0.05, *p ≤ 0.001. Control—0.025% DMSO, positive control—0.1 μM E-64. (C) Western blot analysis of TMPRSS2 and cathepsin L expression in A549 cells treated with indicated FAs with different concentration for 48 h. Detection was done using rabbit anti-TMPRSS2 monoclonal antibody at 1:1000 and mouse anti-cathepsin L antibody at 1:200. Experiments were done in triplicate and repeated three times. Data are presented as % of lipid-free control ± SD.