HBD-2 binds SARS-CoV-2 RBD and blocks viral entry: Strategy to combat COVID-19

Abstract

New approaches to complement vaccination are needed to combat the spread of SARS-CoV-2 and stop COVID-19-related deaths and medical complications. Human beta defensin 2 (hBD-2) is a naturally occurring epithelial cell-derived host defense peptide that has anti-viral properties. Our comprehensive in-silico studies demonstrate that hBD-2 binds the site on the CoV-2-RBD that docks with the ACE2 receptor. Biophysical measurements confirm that hBD-2 indeed binds to the CoV-2-receptor-binding domain (RBD) (K_D ∼ 2μM by surface plasmon resonance), preventing it from binding to ACE2-expressing cells. Importantly, hBD-2 shows specificity by blocking CoV-2/spike pseudoviral infection, but not VSVG-mediated infection, of ACE2-expressing human cells with an IC₅₀ of 2.8 ± 0.4 μM. These promising findings offer opportunities to develop hBD-2 and/or its derivatives and mimetics to safely and effectively use as agents to prevent SARS-CoV-2 infection.

Keywords: Therapeutics; Virology.

Graphical abstract

Figure 1

Molecular dynamics simulations of RBD:ACE2 and RBD:hBD-2 (monomer) show stable protein complex (A) Comparison of the initial structures (shown in sand-color) and last structure (shown in raspberry for ACE2 and magenta for RBD) after 50 ns all-atom MD simulation for the RBD from SARS-COV-2 spike protein in complex with ACE2. (B) RMSD for RBD, ACE2, and RBD:ACE2 complex as a function of simulation time. The overall rms deviation (RMSD Ca) is ∼1.2 Å, for ACE2, ∼2.1 Å for the RBD and ∼2.4 Å for the complex. (C) Comparison of the initial (shown in sand-color) and last structures (shown in green for hBD-2 and magenta for RBD) after 500 ns all-atom MD simulations for the RBD:hBD2 complex. (D) RMSD of RBD, hBD-2, and RBD:hBD-2 complex as function of simulation time. (E and F) Distance map of inter-protein contacts in (E) the RBD:ACE2 complex and (F) in the RBD:hBD-2 complex with distances color coded by average proximity over the length of the simulations (see color scale in Å, right). Five regions of the RBD shaded in green are common regions for binding the two proteins; hBD-2 residues in contact with RBD are shaded yellow on y axis.

Figure 2

The RBD and hBD-2 proteins retain considerable dynamics as a complex (A) RMSF of RBD (left) and hBD-2 (right) in the complex over 500 ns in comparison with values for the unbound (free) proteins; the secondary structure of ACE2 and RBD are indicated (color shading as Figures 1E and 1F). (B) Cartoon representation of the complex showing difference in RMSF between bound and free proteins. The data are mapped to the cartoon representation of the complex with color bar (bottom) indicating the range of −0.5 Å (in blue) to 0.5 Å (in red). (C) Number of hydrogen bonds for the RBD bound to hBD-2 over the 500ns simulation. (D) Table of most prominent h-bonds and their occupancy.

Figure 3

Biophysical and biological assays demonstrate that hBD-2 binds to RBD (A) Concentration-dependent binding of recombinant hBD-2 (rhBD-2) to biotinylated recombinant RBD using surface plasmon resonance. The hBD-2 concentration ranged from 90 to 23,000 nM (see STAR Methods). Fitted data are from two experiments (the first experiment for SPR is shown giving a Kd of 2.8 μM). (B) Functional ELISA assay showing that rhBD-2 specifically binds to immobilized rRBD. LL37 was used as a negative control. (C) Recombinant His-RBD (0.19 μM) and hBD-2 (1.75 μM) were incubated as described in STAR Methods and precipitated with Ni-NTA beads to pull down His-tagged-RBD. Co-precipitation of hBD-2 was assessed by western blotting. Lane one shows 20% input of rhBD-2 and lane two shows Ni-NTA precipitation for background binding of rhBD-2 to the beads. Data are representative of three independent experiments. Control western blots showed modest background with hBD-2 alone. (D) ACE2-expressing HEK 293T cells were incubated with HIS-RBD containing culture supernatant, with and without hBD-2 at 0.7–2.1 μM concentrations. Ni-NTA immunoprecipitation was performed to precipitate ACE2 bound to HIS-RBD and to assess the effect of hBD-2 addition on RBD:ACE2 binding. Data are representative of three independent experiments. ns indicate non-specific band.

Figure 4

HBD-2 inhibits CoV-2 spike-pseudotyped viral entry into ACE2 293T cells (A and B) Cells were incubated with CoV-2 Spike/VSVG-pseudotyped virus media at different dilutions and luciferase activity was measured at 48 h post infection (n = 2). (C) Dose-response relationship of rhBD-2 against CoV-2 spike/VSVG-pseudotyped virus. Cells were co-infected with VSVG-pseudotyped (n = 4) or CoV-2 spike-pseudotyped (n = 4) virus along with varying amounts of rhBD-2 (0–25.6 μM) and luciferase activity was assessed. (D) Percent inhibition of spike-viral entry by rhBD-2 was calculated from RLU values in (C); IC₅₀ was calculated by plotting rhBD2 concentration (μM log) against % inhibition observed (n = 4). Values are Mean ± SEM ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05, and ns (non-significant) against CoV-2 spike-pseudotyped virus alone treated group.