Coronavirus (2019-nCoV) Deactivation via Spike Glycoprotein Shielding by Old Drugs, Bioinformatic Study

The disease of COVID-19 comprises the most serious against human health worldwide with a high rate of virulence and mortality. The disease is caused by the 2019-nCoV virus from the beta coronavirus family. The virus makes use of its surface glycoprotein named S protein or spike to enter the human cells. The virus attached to its receptor named angiotensin-converting enzyme 2 on host cells surface via its receptor-binding domain and its fusion is mediated by cleavage at S2' site that is carried out by surface protease. Vaccines or drugs interfering with S protein binding or cleavage sites could be considered as drugs to get rid of the infection. In the current work and though docking and molecular dynamic experiments we have checked more than 100 drugs with high enough molecular weights for their shielding potency toward S protein binding sites and processing S2' sites. Our results indicate the shielding potency of: fidaxomicin>ivermectin>heparin>azithromycin>clarithromycin>eryhthromycin>niclosamide>ritonavir. Considering affluent reports regarding the complex disturbance in the immune system and multi-organ Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 2 May 2020 doi:10.20944/preprints202005.0020.v1 © 2020 by the author(s). Distributed under a Creative Commons CC BY license. 2 involvement in the disease there is no single or binary drug regime for cure expectedly and instead, we claim the multi-drug regime should be the choice in this context. Accordingly, we suggest our extracted drugs as an adjuvant for clinical trials.

Considering the overall mushroom-like shape of the spike, this subunit places in the mushroom head with the RBD domain faced in such a way to interact with the cellular receptor of ACE2 [16][17][18]. The RBD domains of the spike trimer which are responsible for ACE2 binding adjust one of two conformations of up and down conformations. Up conformation corresponds to the receptor accessible state, while down conformation is inaccessible conformation [19][20][21][22].
The next subunit, S2 contains a signal sequence, a next cleavage site called S2′ for protease. This cleavage site becomes accessible for protease action upon receptor binding to the receptor and its consequent dissociation in the prefusion state. Fusion peptide (FP, residues 788-806) domain helps the virus to fuse host cells membrane and to form the post-fusion complex. Heptad repeat 1 (HR1, residues 912-984) central helix (CH), connector domain (CD), heptad repeat 2 (HR2, residues 1163-1213), transmembrane domain (TM, residues 124-1237) and cytoplasmic tail (CT, residues 1238-1273) are the rest domains of subunit S2 [16][17].
Upon S protein binding to ACE2 receptor and cleavage at S1/S2, subunit S1 undergoes vast structural rearrangements that eventually lead to its release from prefusion complex and ultimate fusion of the virus 4 with host cell [10][11][12][13][14][15][16][17][18]. The detailed scenario for the 2019-nCoV attack is as follows: RBD domain with up conformation preferentially binds to the ACE2 receptor. Simultaneous cleavage of S1/S2 site by protease triggers structural alterations in the S1 subunit destabilized the prefusion trimeric structure of S glycoprotein that leads to dissociation of S1 subunit and refolds S2 subunit to postfusion conformation [23][24][25]. Successful infection of host cells accomplished by S2' cleavage by furin protease and release of fusion peptide that is essential for postfusion state and virus entrance. The S2' site in the prefusion state is buried and inaccessible for furin but upon shedding of S1 in postfusion conformation become accessible for hydrolysis [26]. During this phenomenon heptad repeat, 1 (HR1) and heptad repeat 2 (HR2) interact with each other to form fusion core of a six-helical bundle which bringing viral and cellular membranes in close proximity for fusion. Currently, this hydrophobic core is considered as an ideal target for vaccine design or ligand interaction as effective tools to combat 2019-nCoV and COVID-19 treatment [10][11][12][13][14][15][16][17][18].
In the current work and through molecular dynamic/docking experiments we tried to enrollee different approved drugs to see if the can bind to RBD domain of spike protein in competition with ACE2 receptor or if they can bind to S2' region to mask it against hydrolysis by host cell protease and prevent human infections by this virus.

Methods and Materials:
Spike Coordinate: Coordinate structures of 2019-nCoV and SARS S protein with PDB ID 6VYB and 6CRZ as well as coordinate structure of ACE2 receptor with PDB ID 1O8A were retrieved from protein data bank (https://www.rcsb.org/). The structures were obtained by the X-ray diffraction and refined at the resolutions of 3.46Å, 3.30Å and 2.0Å respectively. The structures were energy-minimized in 12.85×13.13×17.12 nm, 14.68×14.28×17.72nm and 7.21×8.30×7.75nm separate rectangular boxes. The simulated boxes were filled with SPCE water with shells of 1.0-nm thickness. Energy minimization algorithm of Steepest descent was used to minimize the system energy to lower than 100 kJ/mol. Neutral 5 pH ( given Asp, Glu, Arg and Lys ionized), temperature of 37°C and one atmospheric pressure were used as energy minimization conditions [27][28].
In order to study the dynamic behavior of spike proteins especially at RBD and fusion core we performed molecular dynamic simulations using double-precision MPI version of GROMACS 4.5.5 installed on UBUNTU version 16.04 with GROMOS force field for 20 ns at 37degrees centigrade and 1 atmosphere [29].
Sequence Alignment: given the binding property of RBD domains for SARS and 2019-nCoV is determined by their amino acid sequences we compared the RBD sequence with the same sequence of SARS-CoV through sequence alignment on EMBOSS Stretcher (www.ebi.ac.uk), scheme 2, to pick up the underling principles for their different pathogenesity comparatively [30][31]. Blind Docking experiments: to survey the potential binding potency of available drugs with high enough molecular weight and binding energy we carried out blind docking experiments in Hex 8.0.0 (http://www.loria.fr/~ritchied/hex/) using 2019-nCoV spike protein as receptor against enrolled drugs as lignds [33]. The mode of Sahpe+Electrostatic with macro sampling was used as docking parameters and the best 100 poses were analyzed accordingly.
Data Handling and Analysis: all the numerical data were exploited in Excel and SPSS software. P value under .05 was considered as the significance level.

Results and Discussion:
Studies on S proteins from SARS-CoV and 2019-nCoV origins indicated that despite large differences seen in their whole sequences and also at their domains including RBD which is determinant for receptor recognition and consequent virus infectivity, there seem to be structural similarities between these two proteins especially at their domains of NTD, RBD SD1 and SD2 from S1 subunit as well as domains of FP, HR1, HR2, and S2' cleavages site from S2 subunits with RMSD differences less than 4Å [11][12][13][14][15].
Henceforth it expected that these two proteins should behave similarly in their functions i.e. receptor recognition and host cell infection.
Given the much higher affinity of 2019-nCoV for ACE2 than SARS-CoV indicate that there should be detailed differences between these two S proteins that play a vital role in more severity of COVID-19 outbreak with high rates of virulence and mortality. Structural optimization and molecular dynamic simulations for S proteins from these two origins reveal detailed structural differences 2019-nCoV and SARS-CoV spike proteins. Figure 1  affinity. Our data also indicates that S protein of SARS-CoV mainly binds to ACE2 receptor by up conformation of RBD domain with a much less binding energy of -379.66kJ/mol (p-value<0.01). This finding may be partially helpful in understanding the higher affinity of 2019-nCov for the ACE2 receptor and its more severe virulence [19][20][21][22].  and ACE2 receptor upon binding [35][36][37]. This is the next factor that may interpret the higher affinity of 2019-nCoV S protein to host cell receptors we postulate. Isoelectric pH (pI) is the pH in which the protein has no net charge or the total charge of the protein is zero. Using protein sequence we have  addition to its anti-protease activity. It is very important to mention that even though the binding energies of our shielding candidates except fidaxomicin are significantly lower than the binding energy for ACE2 receptor in 1:1 competition ratio but we should remember that in pharmacological dosage the ratio of drug/ACE2 receptor is far from unity and so we can expect that the total binding energies of shielding candidates are much higher than ACE2 receptor and hence they will comprise logic shield toward viral infections.

Conclusion:
To this end, the disease of COVID-19 with a high rate of virulence and fast spread in the human body with multi-organ involvement and high rate of mortality comprises the greatest problem since the Second World War with more than 3 millions infected cases and more than 218,000 deaths by April 2020 [50][51][52]. Decreased lymphocytes, increased C reactive protein (CRP), and pro-inflammatory cytokines as well as hypercoagulability with increased d-dimer lead to lung lesions with infiltrated immune [53][54][55][56]. It seems credulous to think that in a battle against COVID-19 that invades multi organs and disturbs the immune defensive system in a short period to be achieved by one or two drugs, especially at an advanced state. Based on plentiful reports in this context and considering our current and previous work [57] we imagine that a successful treatment regime should contain multi drugs of protease inhibitors, spike shielding drugs, and immunomodulatory drugs in early steps of the disease. Ivermectin>heparin (as intravenous or nebulized)>macrolides seem to be good adjuvant candidates in all anti 2019-nCov regimes to shield S protein even for prophylactic purposes.