key: cord-0736311-miamo46o
authors: Adamczyk, Zbigniew; Batys, Piotr; Barbasz, Jakub
title: SARS-CoV-2 Virion Physicochemical Characteristics Pertinent to Abiotic Substrate Attachment
date: 2021-06-02
journal: Curr Opin Colloid Interface Sci
DOI: 10.1016/j.cocis.2021.101466
sha: e28eafdc7afc290ee1f260c9fa226453dbfe2eaf
doc_id: 736311
cord_uid: miamo46o

The structure and main physicochemical characteristics of the SARS-CoV-2 virion with the spike transmembrane protein corona were discussed. Using these data, diffusion coefficients of the virion in aqueous media and in air were calculated. The structure and dimensions of the spike protein derived from molecular dynamic modeling and thorough cryo-EM measurements were also analyzed. The charge distribution over the molecule was calculated and shown to be largely heterogeneous. Whereas the stalk part is negatively charged, the top part of the spike molecule, especially the receptor binding domain, remains positively charged for a broad range of pH. It is underlined that such a charge distribution pattern promotes the spike corona stability and enhances the virion attachment to receptors and abiotic surfaces, mostly negatively charged. The review is completed by the analysis of experimental data pertinent to the spike protein adsorption at biotic surfaces comprising nanoparticle carrier particles. It is argued that these theoretical and experimental data can be used for developing quantitative models of virus attachment to receptors and abiotic surfaces facilitating adequate analysis of future experimental results.

Within a short time after the outbreak of the Covid-19 pandemic, an avalanche of publications have appeared, which are devoted to various aspects ranging from the SARS-CoV-2 virus transmission mechanisms [1] [2] [3] [4] its attachment to receptors and abiotic surfaces (fomites) [5] [6] [7] [8] , inactivation and removal [8, 9] to vaccination. [10] For sake of conciseness, in this fragmentary review attention is focused on the physicochemical characteristics of the virus particle (virion), especially the spike protein corona, which determines its attachment efficiency to various substrates. It is hypothesized that viruses can be treated as composite nanoparticles characterized by well-defined physicochemical properties. Therefore, it can be assumed that their transfer to various surfaces comprising airborne particulate matter (PM) surfaces is a phenomenon analogous to colloid particle deposition processes. This concept was quantitatively elaborated in a recent review [4] where it is shown that virus transmission via PM or aerosols droplets acting as a shuttle can play a significant role. Therefore, in analogy to the colloid deposition, one can distinguish three main steps of virus particle (referred to as virion) attachment [4] : (i) transfer over macroscopic distances to the vicinity of boundary surfaces either abiotic, e.g., masks, or cell membranes, which is governed by forced convection (flow) and diffusion, for example through mucus layer (ii) transfer through the thin surface layer adjacent to interfaces controlled by the electrostatic and van der Waals forces (the latter are often referred to as hydrophobic interactions) (iii) formation of a physical contact of the virion with the interface or the corresponding cellular receptor where except for the mentioned interactions the hydrogen bonding may play a significant role. One should mention that in the latter case there appears another important step consisting in the penetration of the virion through the membrane into the host cell. The overall virion attachment rate, quantified in terms of the solute flux, is controlled by the slowest step, often the transfer in the bulk phase, i.e., by the virus concentration in the ambient phases. However, the actual (maximum) number of attached virions is governed by the surface binding energy, i.e., affinity to receptors. The latter is controlled by the external layer of the virion, i.e., the corona of spike proteins bearing appropriate binding sites.

Considering this, one can argue that a quantitative analysis of the attachment kinetics requires thorough information about physicochemical parameters characterizing the virion and the spike protein. Therefore, in the first part of the review the SARS-CoV-2 virion architecture with emphasis on the spike protein structure and its physicochemical characteristics including the heterogeneity of charge distribution is analyzed. Recent results derived from molecular dynamic modeling and experimental investigations pertinent to the spike protein attachment to receptors and adsorption at biotic surfaces are discussed. It is argued that these theoretical and experimental data discussed in this work can be used for developing quantitative models of virus attachment kinetics to receptors and abiotic surfaces facilitating adequate analysis of future experimental data.

SARS-CoV-2 is an enveloped virus belonging to the coronaviridae family that cause respiratory and enteric system infections in humans and many animals. It shows similarity to SARS-CoV, a coronavirus known since 2003 [11] in respect to the structure of the membrane and the spike protein shell. [5, 6, 12, 13] . Also, it uses the same ACE2 receptor for the attachment to cells. This similarity allows drawing some conclusions about its structure and mechanism of action J o u r n a l P r e -p r o o f Figure 1. (a) A schematic representation of a SARS-CoV-2 particle obtained using CellPAINT software [14] . (b) Slices through tomographic reconstructions of SARS-CoV-2 virions, reprinted with permission from Ref. [15] Scale bar, 30 nm.

The SARS-CoV-2 virion schematically shown in Fig. 1 contains a single-stranded RNA genome composed of ca. 30,000 base pairs. The genome is encapsidated by the nucelocapsid N protein. On the other hand, the envelope of the virion comprises the M (membrane) protein that plays an important role in all coronavirus assemblies, [11] and the E protein. The spike (S) protein incorporated into the membrane (see Fig 1) is the virus receptor protein also responsible for the virus and the cell membrane fusion. Molar masses of proteins are as follows E -8.4, M -25.1, N -45.6, and S -141.2 kg mol -1 . [11, 16] The main structural and conformational information about the virion stems from thorough cryo-TM imaging. [15, 17, 18] On the other hand, the virion micromechanical properties comprising its compliance and temperature resistance to temperature perturbation were determined by atomic force microscopy (AFM) imaging and force spectroscopy measurements. [19] These J o u r n a l P r e -p r o o f 6 experiments enabled to establish that the virion are spherical in shape (see Fig. 1 ) characterized by the core external dimeter equal to 91 ± 11 nm (an average value taken from 179 single particles [18] ) with the membrane thickness of 5 ± 1 nm. [15, 20] The AFM measurements are less accurate due to tip convolution effects yielding the adsorbed virion height equal to 60 nm. [19] Assuming a spherical shape of the virion, the core part surface area is equal to 2.6 ×10 4 nm 2 (see Table 1 ). One can also calculate the virion diffusion coefficients in various media using the Stokes-Einstein relationship,

where k is the Boltzmann constant, T is the absolute temperature, η is the medium viscosity, and H d is the hydrodynamic diameter.

Assuming that the hydrodynamic diameter of the virion corresponds to its core part diameter one obtains from Eq.(1) D = 2.7 ×10 -10 and 5.4 ×10 -12 m 2 s -1 for the air and an aqueous phases, respectively (at the temperature of 298 K). Obviously, in other media characterized by a larger visocity, such as for example sliva or the mucus [21] the virion diffusion coefficient may significantly decrease creating additional resistance to the overall virion transfer rate.

As mentioned, the virion membrane comprises a number of transmembrane spike proteins, which exhibit a large conformational flexibility as demonstrated experimentally [15] and by extensive molecular dynamics modeling. [22] The distribution of the spike protein molecules over the core and their number were quantitatively evaluated by cryo-TM [15, 18] and by AFM. [19] It is confirmed that the number of molecules per virion denoted by N is characterized by a large spread with average values ranging between 26 [18] to 61 [19] . This effect may be caused by the disruption of the protein/membrane contact upon freezing. It is J o u r n a l P r e -p r o o f 7 interesting to mention that the spike protein molecules orientation is also rather variable with the average angle of the longer axis toward the membrane equal to approximately 50 deg. [18] Using the N numbers one can calculate that the spike protein surface density varies between 1×10 -3 and 2.3×10 -3 nm 2 . However, one should mention that at present our knowledge about the membrane surface charge density as a function of pH and its ion adsorption properties is fairly limited. 

aqueous media (298 K) calculated from Stokes-Einstein relationship for equivalent sphere

Ke et al [18] , cryo-EM Amaro et al [22] , "Open" ("Closed") state S-protein number, per virion, N

Ke et al [18] , cryo-EM Turoňová et al [15] , cryo-EM Kiss et al [19] , AFM S-protein surface concentration, Summarizing the above data one can approximate SARS-CoV-2 virion as spherical core/shell nanoparticle with heterogeneous charge distribution as demonstrated in Refs. [24, 25] The structure and physicochemical parameters pertinent to the spike protein are analyzed in more detail in the next section.

Detailed information on structure and function of the S protein are given in Refs. [15, 22, 26] In Protein and N-glycans are highlighted orange and blue, respectively.

The molecular structure obtained in Ref. [22] for open and closed conformation is used to visualize and determine the dimensions of the SARS-CoV-2 spike protein, shown in Fig. 2 

This enables estimation of the most relevant physiochemical parameters, such as the protein number surface density, cross-section area or its surface coverage, see Table 1 . These primary physicochemical data can also be used for calculating several derivative parameters characterizing the protein, which facilitate a proper interpretation of experimental results. 

where a,b are the longer and the shorter spheroid semi-axes connected with the spheroid volume (equal to the molecule volume) through the dependence Beside the charge distribution at physiological conditions, i.e., pH 7.4, it is interesting to know how the change in pH can influence the spike protein charge. It was shown that stability and infectivity of SARS-CoV-2 is significantly suppressed at pH < 5 and pH > 9. [34] Interestingly, similar observations were reported for SARS-CoV virus. [35] Recently, Zhou et al [36] reported that at endosomal pH 4.0-5.5, the spike evades potentially neutralizing antibody through a pH-dependent mechanism of conformational masking. The dependence of the nominal (titrable) charge of protein molecules can be conveniently calculated using, e.g., the PROPKA algorithm [27] . In Fig. 4 we have presented the nominal charge of the receptor binding domain (RBD) of the spike protein (PDB: 7BZ5) [37] (highlighted red in the inset) plotted as a function of pH [38] . As can be seen, the nominal charge of RBD domain remains positive for the entire pH range up to 9 that agrees with the charge distribution at pH 7.4 presented in Fig. 3(c) . Interestingly, the myoglobin molecule exhibits analogous dependence of the nominal charge for acidic and neutral pH range.

Considering this analogy, one can calculate the electric potential distribution applying the method previously developed for myoglobin [38] exploiting the nonlinear Poisson-Boltzmann model where the formation of two electric double-layers within and outside the molecule were considered. The electric potential at the shear plane, experimentally accessible via electrophoretic mobility measurements [38] , affects the RBD interactions (and in consequence the entire virion) with receptors and abiotic surfaces. This suggests, that results of experimental measurements performed for myoglobin [38] , especially the dependence of the zeta potential on pH, ionic strength and temperature, can serve as a useful reference data for predicting the attachment of the virion to receptors and abiotic surfaces.

Molecular dynamics and free energy modeling can also be used to derive quantitative information about the spike protein binding to the ACE2 receptor. In order to increase the efficiency of calculations only small fragments of the protein are considered, usually the receptor binding domain (RBD) or the receptor binding motif (RBM). In ref. [39] , the dynamic trajectory method was applied to calculate the binding free energy. It is confirmed in this way that the binding strength of the SARS-CoV-2 RBD to the receptor is 10 to 15 times larger than that of the SARS-CoV RBD, which may contribute to increased infection rate.

Interestingly, it is argued that the difference is due to enhanced electrostatic complementarity of the SARS-CoV-2 RBD to the ACE2 caused by the mutation in the hydrophobic residue and the removal of four proline residues. However, on should mention that in these calculations the influence of ionic strength, pH and electric potential appearing due to doublelayer formation and specific ion adsorption were not considered.

The influence of ionic strength varied by NaCl between 0.15 and 1 M on the free energy of SARS-CoV-2 RBD binding to the ACE2 receptor was studied in Ref. [40] by molecular dynamics modeling and experimentally by the Surface Plasmon Resonance (SPR) measurements. Using the free energy, it is predicted that the dissociation constant is equal to 7×10 -10 , 2.9×10 -8 and 1×10 -9 M for NaCl concentration equal to 0.15, 0.5 and 1 M, respectively. On the other hand, the experimental measurements showed that the RBD dissociation constant monotonically increases with NaCl concentration from 1.1×10 -9 , to 8×10 -9 for 0.15 and 1 M, respectively. Given that the dissociation constant assumes very low values, it is concluded that that the increase in the electrolyte concentration does not break the SARS-CoV-2 RBD/hACE2 complex. However, the influence of pH and electrostatic potential were not considered in these calculations.

Until recently, few theoretical investigations were devoted to the important issue of the spike protein interaction with abiotic surfaces. One of the exemptions represents Ref. [41] where the all atom MD modeling was used to determine the interactions of the SARS-CoV-2 trimeric spike protein with cellulose (assumed to be capable of forming strong hydrogen bonding) and graphite, a hydrophobic material. The protein with total charge of -23 e (corresponding to pH 7) was inserted in a water droplet, while the substrate was neutral and not hydrated. The system was neutralized by the 23 Na + ions, which in result gives a very low ionic strength.

The progress of the protein attachment (with the RBD directed toward the substrate surface) J o u r n a l P r e -p r o o f 17 was monitored as a function of time by the number of residues in contact with the surface and the rms factor. It is observed that in the case of graphite significant deformation of the protein was observed that decreases its binding strength.

One should underline that such molecular modeling is a powerful tool capable resolving structure-function relationship, which is often beyond the scope of experiments and there is a need for particular effort in this direction. for pure water was equal to ca. 530 s, which agrees with the above estimate. However, the adsorption kinetics was much slower for the 0.165 NaCl concentration, especially for the alumina substrate where the monolayer formation time increased to 4000 s. As argued in Ref. [42] where an analogous effect was observed for the human serum albumin adsorption on a silica substrate, the increased adsorption time can be attributed to the protein aggregation at such a large electrolyte concentration. Nevertheless, the results reported in Ref. [29] unequivocally confirmed a significant adsorption of the spike protein on the oxide surfaces, which are negatively charged at pH of 7.5. This suggests that the protein exhibited a positive electrokinetic charge, at least on the RBD domain that qualitatively agrees with the theoretically predicted charge distribution (see Fig. 4 ). However, because of the lack of the information about the zeta potential of the S protein and the substrates, a more quantitative analysis of these results obtained in Ref. [29] is not feasible.

J o u r n a l P r e -p r o o f 

where mx  is the maximum volume fraction of nanoparticles in the suspension, m  is the particle density, c m is the particle concentration in the suspension, d m is the particle diameter Ref. [44] is not feasible.

Considering these experimental data one can argue that in order to unequivocally elucidate the spike protein adsorption mechanisms on fomite surfaces thorough experimental measurements are needed. They should be performed for well-characterized systems using a combination of complementary methods, for example the DLS, LDV and concentration depletion methods as previously done for globular proteins. [15, 28] 

Basic physicochemical data collected for the SARS-CoV-2 virion in this work can be used or predicting its transfer and attachment to abiotic surfaces. It is argued that in analogy to colloid particles the overall virion attachment rate is controlled by its transfer through the bulk phase, usually through air or aqueous media in experimental investigations. However, the maximum number of attached virions is governed by its affinity to a substrate controlled by the corona of spike proteins. Therefore, quantitative information about the spike protein, especially its interactions with various substrates is of primary importance.

The analysis of published data derived from advanced molecular dynamic modeling and experimentally from cryo-EM confirmed that the charge distribution over the spike molecule was heterogeneous with the receptor binding domain positively charged for a broad range of pH. It is suggested that such a charge distribution pattern enhances the virion binding strength to receptors and abiotic surfaces, usually negatively charged. As far as theoretical investigation are concerned, one may suggest that future studies should be focused on elucidating the role of electric double layers, specific ion adsorption and local pH changes in spike protein adsorption at various substrates.

There are no conflicts of interest to declare.

A quantitative physicochemical concept of virus transmission via airborne particulate matter and aerosols droplets was elaborated and thoroughly discussed

The structural analysis of the SARS-CoV-2 virion attachment to the human ACE2 receptor

The SARS-CoV-2 virion structure, dimensions, the number of spike proteins and their distribution over the membrane, their size and orientations were determined by cryo-EM

The SARS-CoV-2 virion dimensions, the number of spike proteins, their orientation angle distribution, size and shape were determined by cryo-EM

The SARS-CoV-2 S protein molecular structure derived from all-atom molecular dynamics simulations

The SARS-CoV-2 virion recognition of ACE2 receptor is experimentally determined yielding binding constants, antibody neotralisation of the S protein by antibodies is theoretically analyzed

On the concentration of SARS-CoV-2 in outdoor air and the interaction with pre-existing atmospheric particles

A systematic review of possible airborne transmission of the COVID-19 virus (SARS-CoV-2) in the indoor air environment

Soft matter science and the COVID-19 pandemic

Chemodynamic features of nanoparticles: Application to understanding the dynamic life cycle of SARS-CoV-2 in aerosols and aqueous biointerfacial zones

Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2

Structural basis of receptor recognition by SARS-CoV-2

Contamination of inert surfaces by SARS-CoV-2: Persistence, stability and infectivity. A review

Make it clean, make it safe: A review on virus elimination via adsorption

Hard Nanomaterials in Time of Viral Pandemics

Covid-19 vaccine frontrunners and their nanotechnology design

Supramolecular Architecture of the Coronavirus Particle

The pivotal link between ACE2 deficiency and SARS-CoV-2 infection

Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV

In situ structural analysis of SARS-CoV-2 spike reveals flexibility mediated by three hinges

A SARS-CoV-2 protein interaction map reveals targets for drug repurposing

Molecular Architecture of the SARS-CoV-2 Virus

Structures and distributions of SARS-CoV-2 spike proteins on intact virions

Topography, Spike Dynamics, and Nanomechanics of Individual Native SARS-CoV-2 Virions

SARS-CoV-2 structure and replication characterized by in situ cryo-electron tomography

Biotransport: Principles and Applications

Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein

Average protein density is a molecular-weightdependent function

Core/shell nanoparticles: Classes, properties, synthesis mechanisms, characterization, and applications

Albumin adsorption at solid substrates: A quest for a unified approach

Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein

PROPKA3: Consistent treatment of internal and surface residues in empirical p K a predictions

Protein adsorption: A quest for a universal mechanism

Adsorption of SARS-CoV-2 Spike Protein S1 at Oxide Surfaces Studied by High-Speed Atomic Force Microscopy

Structure and catalytic behavior of myoglobin adsorbed onto nanosized hydrotalcites

Adsorption of Myoglobin and Corona Formation on Silica Nanoparticles

Cies la M. Human fibrinogen adsorption on positively charged latex particles

Mechanisms of Fibrinogen Adsorption on Silica Sensors at Various pHs: Experiments and Theoretical Modeling

Factors affecting stability and infectivity of SARS-CoV-2

Inactivation of the coronavirus that induces severe acute respiratory syndrome, SARS-CoV

A pH-dependent switch mediates conformational masking of SARS-CoV-2 spike

A noncompeting pair of human neutralizing antibodies block COVID-19 virus binding to its receptor ACE2

Myoglobin molecule charging in electrolyte solutions

Enhanced receptor binding of SARS-CoV-2 through networks of hydrogen-bonding and hydrophobic interactions

Molecular Dynamics Reveals Complex Compensatory Effects of Ionic Strength on the Severe Acute Respiratory Syndrome Coronavirus 2 Spike/Human Angiotensin-Converting Enzyme 2 Interaction

Computer simulations of the interaction between SARS-CoV-2 spike glycoprotein and different surfaces

Human Serum Albumin Adsorption Kinetics on Silica: Influence of Protein Solution Stability

Quantum Dot-Conjugated SARS-CoV-2 Spike Pseudo-Virions Enable Tracking of Angiotensin Converting Enzyme 2 Binding and Endocytosis

Nano-size dependence in the adsorption by the SARS-CoV-2 spike protein over gold colloid

This work was supported by the statutory activity of the Institute of Catalysis and Surface Chemistry Polish Academy of Sciences. This research was supported in part by PLGrid Infrastructure.