Influence of Cholesterol on the Insertion and Interaction of SARS-CoV-2 Proteins with Lipid Membranes

Abstract

Cholesterol is an essential sterol in cell membranes that regulates organization and fluidity. This biomolecule has been identified as one of the critical factors in the internalization process of various viruses in human cells. Therefore, understanding these mechanisms is crucial for a deeper comprehension of viral pathogenicity in the search for practical therapeutic approaches against viral diseases. The biochemical and biophysical processes related to these diseases are highly complex. For this reason, studying model systems capable of mimicking the interaction of lipid membranes with cholesterol and proteins is fundamental. In this work, we propose to study the structural and elastic changes in mono-, bi-, and tridimensional lipid systems composed of dipalmitoylphosphatidylcholine (PC) with varying amounts of cholesterol in the presence and absence of the S protein (Spike) and its receptor-binding domain (RBD) from SARS-CoV-2. To characterize these systems, we used both experimental and theoretical approaches such as Langmuir trough, atomic force microscopy (AFM), small-angle X-ray scattering (SAXS), electrochemical methods, and molecular dynamics (MD) simulations. With the interpretation of all results obtained in this work, it was possible to propose a structural model of the membrane in the presence of cholesterol and the interaction with the Spike protein and RBD. The behavior of the adsorption isotherm and SAXS data, together with the results provided by MD simulations, led us to conclude that cholesterol in PC monolayers promotes local alterations, inducing the formation of more rigid membrane regions. More importantly, cholesterol plays a crucial role in facilitating the allocation of SARS-CoV-2 proteins in lipid systems. This is especially true for the Spike protein, which displayed a non-ACE2 mediated stable binding to the lipid membrane with high internalization.

Keywords: Langmuir monolayer; SARS-CoV-2; atomic force microscopy; cholesterol dynamics; dipalmitoylphosphatidylcholine; lipid membranes; molecular dynamics simulations; small angle X-ray scattering.

1

Results from small angle X-ray scattering (SAXS) experiments on model membranes made of PC, both with and without cholesterol. (A) Scattering for lipid vesicles comprising PC and Spike protein. (B) PC with RBD. (C) Combination of [Chol/PC] with Spike. (D) Combination of [Chol/PC] with RBD. The solid lines overlaying the data points represent fits using the Gaussian model.

2

Structural and elastic parameters of vesicles with a [Chol/PC] ratio of 0.25 were examined across varying protein concentrations ranging from ng mL^–1 to mg mL^–1. (A) Lamellar periodicity (D); (B) Membrane thickness (δ_M); (C) Number of unilamellar vesicles N _single; (D) Caillé parameter (η) and (E) Electron density profile (Δρ). Blue and red lines are provided to guide the eye in tracking the behavior of the parameters.

3

Surface pressure isotherms plotted against the molecular area. A solid black line represents PC alone; a dotted black line indicates varying [Chol/PC] ratios. In (A) and (B), the blue lines show the effect of adding Spike protein, and the red lines indicate the influence of RBD, respectively. In (C) and (D), isotherms of surface pressure per average molecular area are shown for different [Chol/PC] ratios, namely 0.10, 0.25, and 0.42. (C) Blue dashed lines represent Spike protein at a 0.1 mg/mL concentration peak, while (D) red dashed lines represent RBD at the same concentration.

4

AFM images illustrate the influence of cholesterol on the lipid monolayer. (A) Pure PC monolayer, (B) [Chol/PC] = 0.25 monolayer, (C) Pure PC monolayer with Spike protein, (D) [Chol/PC] = 0.25 monolayer with Spike protein, (E) Pure PC monolayer with RBD, (F) [Chol/PC] = 0.25 monolayer with RBD.

5

Electrochemical study of ITO modified with [Chol/PC] + Spike monolayers in the presence and absence of anti-RBD antibodies. (A) Cyclic voltammogram and (B) Nyquist diagram of the [Chol/PC] + Spike-0.1 modification. (C) Cyclic voltammogram and (D) Nyquist diagram of the [Chol/PC] + Spike-1 modification. Electrolyte: K4Fe­(CN)­6/K3Fe­(CN)­6 5 mmol L^–1 solution in KCl 0.1 mol L^–1. ITO modified with [Chol/PC] + RBD monolayers in the presence and absence of anti-RBD antibodies. (E) Cyclic voltammogram and (F) Nyquist diagram of the [Chol/PC] + RBD-0.1 modification. (G) Cyclic voltammogram and (H) Nyquist diagram of the [Chol/PC] + RBD-1 modification. Electrolyte: K₄Fe­(CN)₆/K₃Fe­(CN)₆ 5 mmol L^–1 solution in KCl 0.1 mol L^–1.

6

Time evolution of SARS-CoV-2 RBD (A) and SARS-CoV-2 Spike (B) membrane penetration (in Å) over the MD trajectories for the lipid monolayers at [Chol:PC] ratios of 0 (red graph), 0.10 (green graph) and 0.30 (blue graph).

7

Models of SARS-CoV-2 RBD (A) and SARS-CoV-2 Spike protein (B) in complex with the [Chol/PC] = 0.30 lipid monolayer at the end of the MD simulations. Three rotated side views are portrayed for each system. PC carbon atoms are colored grey, PC oxygen atoms are colored red, and cholesterol molecules are colored green. Each RBD domain in Spike is drawn in a different color. Yellow: chain A domain, orange: chain B domain, violet: chain C domain.