COVID-19: Predicting Inhibition of Repurposed Drugs for SARS-Cov-2 Viral Activity and Cellular Entry Clifford Fong

COVID-19: Predicting inhibition of repurposed drugs for SARS-CoV-2 viral activity and cellular entry Clifford Fong

To cite this version:

Clifford Fong. COVID-19: Predicting inhibition of repurposed drugs for SARS-CoV-2 viral activity and cellular entry. [Research Report] Eigenenergy, Adelaide, Australia. 2020. hal-02963306

HAL Id: hal-02963306 https://hal.archives-ouvertes.fr/hal-02963306 Submitted on 10 Oct 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. COVID-19: Predicting inhibition of repurposed drugs for SARS-CoV-2 viral activity and cellular entry

Clifford W. Fong Eigenenergy, Adelaide, South Australia, Australia. Email: [email protected]

Keywords: COVID-2019 or SARS-CoV-2; SARS-CoV; MERS; ACE2 receptor binding, pro spike serine proteases, S-RBD, TMPRSS2, M , IC50, host cell membrane fusion or endocytosis, linear free energy relationships, HOMO-LUMO; quantum mechanics;

Abbreviations: Structure activity relationships SAR, ΔGdesolv,CDS free energy of water desolvation, ΔGlipo,CDS lipophilicity free energy, cavity dispersion solvent structure of the first solvation shell CDS, Dipole moment DM, Molecular Volume Vol, HOMO highest occupied molecular orbital, LUMO lowest unoccupied molecular orbital, HOMO-LUMO energy gap, linear free energy relationships LFER, Molecular docking binding energy BE, Receptor binding domain of S protein of SARS-CoV-2 S-RBD, transmembrane serine 2 protease TMPRSS2, angiotensin-converting enzyme 2 ACE2, main protease of SARS-CoV-2, Mpro.

Abstract A previously described linear free energy methodology has been effective in predicting the efficacy of a wide range of repurposed drugs against four known mechanisms that are critical to the infectiousness and pathogenesis of the SARS-CoV-2 virus: inhibition of the Mpro of SARS-CoV-2, inhibition of SARS-CoV-2 spike protein S-RBD, inhibition of ACE2 receptor, and inhibition of TMPRSS2 receptor.

These results cover experimental IC50 studies and computational docking binding energy/affinity studies. The overwhelming conclusion is that the HOMO, or LUMO or the HOMO-LUMO energy gap of the various inhibitors is the principal determinant of inhibition. The one exception found was the binding of inhibitors to the interface of the SARS-CoV-2 S- protein and the ACE2 receptor.

It appears that while docking studies can be used for rapid wide scale screening of potential inhibitors, these results do not necessarily agree with IC50 or other inhibitory experimental results. It is clear from these studies that the HOMO, or LUMO or HOMO-LUMO energy gap can play a useful role, alongside docking studies, in identifying inhibitors before further evaluation in preclinical trials.

Introduction

The SARS-CoV-2 virus enters cells via the ACE2 receptor. [1-3] ACE2 is present in many cell types and tissues including the lungs, heart, blood vessels, kidneys, liver and gastrointestinal tract. It is present in epithelial cells. The crystallization of SARS-CoV-2 spike protein complexed to ACE2 has been structurally determined. [PDB 6M18] The enzyme ACE2 is an important protein that promotes tissue regeneration. It is known that ACE inhibitors and angiotensin receptor blockers (ARB), which are standard treatments for hypertension and chronic heart failure, also lead to increased formation of ACE2. Unfortunately, the coronavirus SARS-CoV-2 binds to this protein and uses it to enter cells, where it multiplies, similarly to SARS-Cov. The triple combination of Cepharanthine/selamectin/mefloquine hydrochloride has been identified as a treatment for SARS-CoV-2. [3]

SARS-CoV-2 attaches to the host cell ACE2 receptor by way of the spike protein. Next, a part of the spike protein, called the fusion peptide, interacts directly with the host cell membrane and facilitates merging to form a fusion pore, or opening. The virus then transfers its genome into the host cell through this pore. These genomic instructions essentially commandeer the host’s machinery to produce more viruses. Calcium ions interacting with the fusion peptide can change the peptide’s structure, and how it interacts with membranes in ways that promote infection in MERS and SARS. SARS-CoV, MERS-CoV, and SARS-CoV- 2 entry (receptor binding and membrane fusion) is governed by the viral spike (S) protein.

The transmembrane serine 2 protease (TMPRSS2) on the cell surface is involved in S protein priming in lung cells. In other cells, such as kidney cells, the enzyme cathepsin L is required for the virus to infect them. TMPRSS2 and lysosomal cathepsins both have cumulative effects with furin on activating SARS-CoV-2 entry. The fusion peptide is well conserved across the CoV family, making it a good target for coronavirus antivirals. The biological sequences of the fusion peptides of SARS-CoV and SARS-CoV-2, are a 93% match. The virus can also enter the host cell by endocytosis after attaching itself to the ACE2 receptor and then engulfed by the cell membrane into the cellular cytoplasm. [1-7] Figure 1 illustrates membrane fusion and endocytosis mechanisms of viral cell entry.

TMPRSS2 inhibitors previously approved for clinical use may block entry to the receptor and can constitute a treatment option against infection. Camostat mesylate, Nafamostat mesylate and Arbidol are known inhibitors of the TMPRSS2 membrane fusion related protease. The lysosomal cathepsin inhibitor E64d, as well as Camostat and Nafamostat also inhibit SARS- CoV-2 entry. [2-6] The drugs A1AT, BHH, and AEBS have also been found to inhibit TMPRSS2. [7] Roomi [8] screened 4217 phytochemicals and flavanoids docked to the TMPRSS2 receptor (PDB O15393) and identified a promising range of drugs (see Table 5) with binding affinities from -13.4 to -7.4 kcal/mol. Notably comparing docking results with experimental inhibitory constants showed that Camostat mesylate with a binding affinity of -7.4 kcal/mol had an IC 3713 nM, which compared to the promising candidate Lutonarin -8.8 kcal/mol and 348 nM. Idris [9] docked 3000 compounds into the active sites of a homology modeled TMPRSS2 and found two promising candidates (-9.2 to -9.3 kcal/mol) compared to Nafamostat (-8.2 kcal/mol) and Camostat (-7.2 kcal/mol). (See Table 6)

The SARS-CoV-2 spike (S) protein plays a major role in viral attachment, fusion and entry, and consequently has been identified as a target for development of antibodies, entry inhibitors and vaccines. The receptor-binding domain (S-RBD) of the SARS-CoV-2 S protein binds strongly to the human and bat angiotensin-converting enzyme 2 (ACE2) receptors. SARS-CoV-2 S-RBD exhibits significantly higher binding affinity to the ACE2 receptor than the SARS-CoV RBD. [Tai 10]

Choudhary et al [11] have used a high throughput virtual screening approach to investigate FDA approved drugs against both the receptor binding domain of the virus spike protein (S- RBD) and the ACE2 host cell receptor. A homology model of the S-RBD was built on the crystal structure (PDB 6VSB) of the SARS-CoV-2 protein. The ACE2 receptor protein was PDB 2AJF. They found that GR 127935 hydrochloride hydrate, GNF-5, RS504393, TNP, and eptifibatide acetate bound to the virus binding motifs of ACE2 receptor, and KT203, BMS195614, KT185, RS504393, and GSK1838705A bound to the receptor binding site on the viral S-protein. Terali et al [12] screened an approved drug library of 7,173 ligands against the ACE2 receptor using molecular docking and energy minimization rescoring of docked ligands. ACE2 structures used were PDB 1R42 (free) and PDB 1R4L (inhibitor bound) proteins. Lividomycin, burixafor, quisinostat, fluprofylline, pemetrexed, spirofylline, edotecarin, and diniprofylline were identified as the most promising drug candidates.

Smith [13] developed an elegant unique ensemble interface docking homology model of the spike protein (S-protein) of SARS-CoV-2 interacting with the human ACE2 receptor. 9127 repurposed drugs were screened and identified that were predicted to bind to either the isolated Viral S- protein at its host receptor region or to the S protein-human ACE2 interface. Docking calculations were performed in two parts: (a) targeting the S-protein:ACE2 receptor interface (to identify small-molecules for interface disruption) and (b) focusing on preventing S-protein recognition by binding to the ACE2 recognition region of the isolated S-protein. The study also performed docking calculations targeting the S-RBD and attempted to differentiate between the interface being a “buried” pocket and a “shallow” protein surface of the S-RBD.

The aim of this paper is to examine the effect of repurposed inhibitors that have been identified as being effective against four known mechanisms that are critical to the infectiousness and pathogenesis of the SARS-CoV-2 virus (see Figure 1): (a) inhibition of the Mpro of SARS-CoV-2 (b) inhibition of SARS-CoV-2 spike protein S-RBD (c) inhibition of ACE2 receptor (d) inhibition of TMPRSS2 receptor The study uses a previously described linear free energy methodology that has been effective in predicting the inhibitory efficacy of a wide range of repurposed drugs against the Mpro of SARS-CoV-2, SARS, and HKU4-CoV. [14-18]

Results

(a) Inhibition of Mpro of SARS-CoV-2 and comparisons with SARS-Cov and MERS The evaluation method used in this study and others [14-18] was to calculate the molecular specifiers for the various inhibitors used on the 3C-like protease: (1) the free energy of water desolvation (ΔGdesolv,CDS), (2) the lipophilicity free energy (ΔGlipo,CDS) in n-octane, (3) the dipole moment in water, (4) the molecular volume in water, and (5) HOMO, (6) LUMO or (7) HOMO-LUMO energy gap in water. These independent variables values can be scaled to similar magnitudes so that the coefficients in the multiple linear regression equations can be directly compared to gauge the relative magnitudes of inhibitory sensitivity of these molecular variables. Stepwise multiple regression is then applied to seek out which of the seven drug molecular properties had the largest and most significant effect on the inhibition. Equations 1-7 show the most statistically significant relationship found after testing against all independent variables in a stepwise fashion. We have previously shown that equation 1 accurately describes the inhibition of the HKU4- CoV Mpro. HKU4 (HKU4-CoV) belongs to the same 2c lineage as MERS-CoV and shows high sequence similarity with MERS-CoV. HKU4-CoV Mpro shares high sequence identity (81%) with the MERS-CoV enzyme. Eq 1 Inhibition of Mpro protease of HKU4-CoV for 40 compounds was: pIC50 = 0.05ΔGdesolv,CDS - 0.11ΔGlipo,CDS - 0.08Dipole Moment – 0.23(HOMO-LUMO) + 6.63 2 Where R = 0.382, SEE = 0.39, SE(ΔGdesolvCDS) = 0.04, SE(ΔGlipoCDS) = 0.04, SE(Dipole Moment) = 0.025, SE(HOMO- LUMO) = 0.08, F=5.42, Significance= 0.0017

where ΔGdesolv,CDS is the free energy of water desolvation, ΔGlipo,CDS is the lipophilicity free energy, the dipole moment in water, and HOMO-LUMO is the energy gap in water.

The important finding was that pIC50 is dominantly related to the HOMO-LUMO energy gap of the inhibitors. The HOMO-LUMO gap is an inherent descriptor of the innate reactivity of the inhibitor, and is related to how the inhibitor binds to the protease. In particular, how the

HOMO of the protease (HOMOprot) interacts with the LUMO of the inhibitor (LUMOinhib), and how the HOMO of the inhibitor (HOMOinhib) interacts with the LUMO of the protease

(LUMOprot). These molecular interactions fundamentally define the inhibitor-protease binding interaction.

We previously showed that eq 2 describes the inhibition for 35 aromatic disulphides drugs of the Mpro protease of SARS-CoV: Eq 2

IC50 = -0.74ΔGdesolv,CDS - 0.30ΔGlipo,CDS - 0.21Dipole Moment + 0.25(HOMO-LUMO) - 2.31 2 Where R = 0.725, SEE = 0.69, SE(ΔGdesolvCDS) = 0.11, SE(ΔGlipoCDS) = 0.075, SE(Dipole Moment) = 0.07, SE(HOMO- LUMO) = 0.38, F=19.75, Significance=0.000000 Similarly we also showed that a different series of 25 inhibitors of the SARS-CoV Mpro

protease yielded eq 3(a) and 3(b) (although the accuracy of the experimental IC50 values were of lower accuracy that those analyzed in eqs 1 and 2):

Eq 3(a)

IC50 = -25.2ΔGdesolv,CDS + 19.7ΔGlipo,CDS + 5.3Dipole Moment + 24.1HOMO + 192.1 2 Where R = 0.347, SEE = 102.8, SE(ΔGdesolvCDS) = 10.6, SE(ΔGlipoCDS) = 15.4, SE(Dipole Moment) = 7.8, SE(HOMO) = 41.8, F=2.79, Significance=0.053 Eq 3(b)

IC50 = -29.2ΔGdesolv,CDS + 15.2ΔGlipo,CDS + 7.2Dipole Moment + 27.7LUMO + 4.3 2 Where R = 0.350, SEE = 102.5, SE(ΔGdesolvCDS) = 12.3, SE(ΔGlipoCDS) = 15.5, SE(Dipole Moment) = 7.2, SE(HOMO) = 42.3, F=2.82, Significance=0.051 Using the same methodology, stepwise analysis of 23 inhibitors [19] (see Table 1 and Figure 1) of the Mpro of SARS-CoV-2 yields eq 4(a), 4(b) and 4(c): Eq 4(a)

IC50 = 11.65ΔGdesolv,CDS - 5.75ΔGlipo,CDS – 4.66Dipole Moment + 56.68(HOMO-LUMO) +118.83 2 Where R = 0.28, SEE = 193.56, SE(ΔGdesolvCDS) = 15.29, SE(ΔGlipoCDS) = 19.50, SE(Dipole Moment) = 8.25, SE(HOMO- LUMO) = 38.24, F=1.33, Significance=0.300

Eq 4(b) Eliminating the ΔGlipo,CDS and Dipole Moment variables from eq 4(a) as they show the weakest correlations

IC50 = 13.00ΔGdesolv,CDS + 66.55(HOMO-LUMO) +60.40 2 Where R = 0.21, SEE = 186.35, SE(ΔGdesolvCDS) = 10.30, SE(HOMO-LUMO) = 34.39, F=2.57, Significance=0.101 Eq 4(c) Finally it is clear that the dominant correlation is between inhibitory activity and the HOMO-LUMO energy gap

IC50 = 64.71(HOMO-LUMO) +60.40 Where R2 = 0.14, SEE = 189.96, SE(HOMO-LUMO) = 34.84, F=3.49, Significance=0.077 It is noted that bepridyl, oxiconazole, nelfinavir, trihexyphenidyl, clemstine and metixicene were obtained as ionic salts but treated as neutral species and the ionic species in equal proportions as the inhibitors were initially dissolved in DMSO (pH ca 10.7) then added to the protease at pH 7.8 buffer with 20% DMSO at 37C. [19] Duloxetine was a clear outlier in all analyses.

Molecular docking is currently the mainstay of predictive computational methods to evaluate new and repurposed ant-virals for coronaviruses, and there have been some reports [19,20] that inhibitory structure activity relationships (SARs) do not always agree with docking results for the Mpro. We have tested this observation using the data of Vatansever [19], and find that eqs 5(b) and (c) shows that the docking binding energy is a function of the LUMO and the molecular volume. Such LFERs may be more cost effective means of screening new drugs than the more intensive docking method.

Since Vatansever et al [19] noted that their docking results did not necessarily correlate well

with their IC50 results, an analysis of the docking binding energy against the HOMO-LUMO gap or LUMO results in eq 5(a) and 5(b), where the molecular volumes in water has been multiplied by 100 to allow a direct comparison of the magnitude of the coefficients. The results indicate that the LUMO is the dominant molecular specifier that determines docking binding energy, with the molecular volume being about a third in magnitude. The direct correlation between the docking binding energy and IC50 is poor (significance F 0.22).

Eq 5(a) for 22 inhibitors from Vatansever [19] Table 1 : Docking Binding Energy = 0.45(HOMO-LUMO) - 0.12 Molecular Volume -11.07 Where R2 = 0.21, SEE = 0.81, SE(HOMO-LUMO) = 0.24, SE(Molec Vol) = 0.19, F=2.53, Significance=0.105 Eq 5(b) Docking Binding Energy = 0.55LUMO - 0.18 Molecular Volume - 8.16 Where R2 = 0.34, SEE = 0.74, SE(LUMO) = 0.19, SE(Molec Vol) = 0.16, F=4.88, Significance=0.019 Eq 5(b) is clearly a better correlation than eq 5(a), but the best correlation is found in eq 5(c)

Eq 5(c) Docking Binding Energy = 0.58LUMO - 8.72 Where R2 = 0.31, SEE = 0.75, SE(LUMO) = 0.19, F=8.65, Significance=0.008 It is noted that the corresponding correlation with HOMO is much poorer than that for LUMO in eq 5(c).

Fischer et al [21] computationally screened a library of over 606 million compounds for binding to the main protease (Mpro) of SARS-CoV-2 (PDB 6LU7) and identified 15 available drugs (see Table 2) that could be repurposed as inhibitors for the Mpro. The docking conditions were at pH 7.4 so 15 protonated and neutral drugs were tested in eq 6(a), (b) and (c).

Eq 6(a)

Docking BE = -2.21ΔGdesolv,CDS + 3.85ΔGlipo,CDS - 3.33LUMO - 57.4

2 Where R = 0.540, SEE = 7.68, SE(ΔGdesolvCDS) = 1.04, SE(ΔGlipoCDS) = 1.12, SE(LUMO) = 4.34, F=4.25, Significance=0.031 Eq 6(b) using the HOMO instead of the LUMO specifier gives a similar correlation:

Docking BE = -2.55ΔGdesolv,CDS + 3.93ΔGlipo,CDS + 2.13HOMO – 42.0

2 Where R = 0.515, SEE = 7.85, SE(ΔGdesolvCDS) = 1.32, SE(ΔGlipoCDS) = 1.15, SE(HOMO) = 8.67, F=3.89, Significance=0.040 Eq 6(c) using the HOMO-LUMO gives an improved correlation over eq 6(a) and (b):

Docking BE = -2.63ΔGdesolv,CDS + 3.93ΔGlipo,CDS – 5.68HOMO-LUMO – 30.76

2 Where R = 0.561, SEE = 7.47, SE(ΔGdesolvCDS) = 1.02, SE(ΔGlipoCDS) = 1.09, SE(HOMO-LUMO) = 5.11, F=4.69, Significance=0.024

(b) Inhibition of SARS-CoV-2 spike protein S-RBD

Examination of Choudhary’s docking study of S-RBD inhibitors [11] (see Table 3) gives eq 7, which is only indicative since the number of inhibitors is limited to 6:

Eq 7 Docking BE = 2.59HOMO + 6.84 Where R2 = 0.958, SEE = 0.22, SE(HOMO) = 0.27, F=90.98, Significance=0.0007 Correlations with other molecular specifiers were poor.

Examination of Choudhary’s docking study of 10 ACE2 inhibitors [11] (see Table 3) gives eq 8:

Eq 8

Docking BE = -0.67ΔGlipo,CDS – 4.69HOMO - 42.8

2 Where R = 0.565, SEE = 1.66, SE(ΔGlipoCDS) = 0.26, SE(HOMO) = 2.79, F=4.53, Significance=0.055 Terali has investigated the binding of 9 repurposed drugs [12] (see Table 4) to the ACE2 receptor eq 9:

Eq 9 Docking BE = -0.81HOMO - 0.18 Molecular Volume - 8.16

Where R2 = 0.52, SEE = 0.45, SE(HOMO) = 0.56, SE(Molec Vol) = 0.09, F=3.27, Significance=0.109

(d) Inhibition of TMPRSS2 receptor Analysis of Roomi’s docking of 24 repurposed drugs [8] (see Table 5) to the TMPRSS2 receptor gives eqs 10(a) or (b):

Eq 10(a) Docking BE = -2.26LUMO - 0.43 Molecular Volume - 10.64

Where R2 = 0.658, SEE = 1.07, SE(LUMO) = 0.57, SE(Molec Vol) = 0.16, F=20.20, Significance=0.00001 The correlation with HOMO was very poor.

Eq 10(b) eliminating the weaker molecular volume specifier gives: Docking BE = -2.97LUMO - 13.87 Where R2 = 0.547, SEE = 1.20, SE(LUMO) = 0.58, F=26.55, Significance=0.00003 Analysis of the binding energy of 10 inhibitors of TMPRSS2 from Idris [9] (see Table 6) gives eqs 11(a) or (b):

Eq 11(a) Docking BE = 1.09HOMO-LUMO - 0.49 Molecular Volume - 10.64 Where R2 = 0.715, SEE = 0.85, SE(HOMO-LUMO) = 0.62, SE(Molec Vol) = 0.38, F=8.75, Significance=0.012 Eq 11(b) Docking BE = 1.67HOMO-LUMO – 15.00 Where R2 = 0.649, SEE = 0.88, SE(HOMO-LUMO) = 0.62, F=14.77, Significance=0.005

Analysis of Smith’s data [13] (see Table7) for 14 repurposed inhibitors of the S-protein of SARS-CoV-2 interacting with the human ACE2 receptor:

Eq 12

Docking BE = -0.40ΔGlipo,CDS + 0.017Dipole Moment - 7.50

2 Where R = 0.462, SEE = 0.107, SE(ΔGlipoCDS) = 0.016, SE(DM) = 0.007, F=4.53, Significance=0.055 No correlations were found with other molecular specifiers, eg HOMO, LUMO or HOMO- LUMO, molecular volume or water desolvation.

Discussion

It is clear that the application of the LFER model to the in vitro inhibition of (a) Mpro of SARS-CoV-2, (b) SARS-CoV-2 spike protein S-RBD, (c) ACE2 receptor, and the (d) TMPRSS2 receptor is well described over a wide range of repurposed drugs, as shown in eqs

1-11. These results cover experimental IC50 studies and computational docking binding energy/affinity studies. The overwhelming conclusion is that the HOMO, or LUMO or the HOMO-LUMO energy gap of the various inhibitors is the principal determinant of inhibition.

As previously described, the IC50 of a wide range of inhibitors applies to SARS-CoV-2, SARS, and HKU4-CoV [14-18] with more enhanced contributions from the molecular

specifiers, ΔGdesolv,CDS, ΔGlipo,CDS, the dipole moment, and the molecular volume compared to much weaker contributions from these specifiers found in the correlations with binding energies/affinities. With the exception of eq 6 and 8, the binding energies for all receptors are dominated by either the HOMO, or LUMO or HOMO-LUMO with a minor contribution from the molecular volume. Eq 6 and 8 show dependencies on HOMO and HOMO-LUMO but also on water desolvation and lipophilicity as well.

Eq 12 derived from a docking analysis of the S-protein of SARS-CoV-2 interacting with the human ACE2 receptor is an exception to the rule that HOMO, or LUMO or HOMO-LUMO is the principal determinant(s). It is unclear whether this result is due to the interface between the S- protein and the ACE2 being a “shallow” binding pocket as opposed to the usual “buried” binding pocket, or possibly due to counterposing contributions from inhibitorHOMO à receptorLUMO or receptorHOMO à inhibitorLUMO dominated interactions for both S-protein and ACE2.

Eqs 4 and 5 which are derived for the same set of inhibitors of Mpro, most clearly show that

IC50 values are strongly correlated with HOMO-LUMO, while the binding energies are strongly correlated with the LUMO.

Binding energies would be expected to differ somewhat from IC50 studies, as the binding energies/affinities represent the calculated interaction between the inhibitor in the binding pocket of the protein receptor, while IC50 covers the much wider processes that are involved as the inhibitor moves from the bulk solvent (ε 80), to ε 20-30 at the protein surface into the binding pocket, then involves dynamic processes involved in interacting with the receptor in the pocket (ε 6-7).

In many studies potential inhibitors identified by rapid virtual screening docking studies were further characterized by molecular dynamics studies to more accurately determine binding energies to the receptor. [8,9,11,12,13,21,22] In an analysis of the molecular dynamics energy terms that dominate the protein-ligand interaction, it was found that none of the van der Waals (ΔEVDW), electrostatics (ΔEVDW + ΔGPB), nonpolar solvation term (ΔGSA), or entropy (TΔS) terms was dominant. In addition it was noted where the charged form of an inhibitor was dominant under physiological pH conditions, both the neutral and charged forms need to be evaluated. For example the neutral form of streptomycin has a MM-PBSA- WSAS binding free energy to Mpro of -7.92 kcal/mol, much better than the charged form (- 3.82 kcal/mol) even though the charged form (3+) is dominant at physiological pH. The difference is caused by the distinct electrostatic properties between the neutral and charged molecules. [22] The LFER analyses used in this study has included both neutral and charged form of inhibitors where both are present at physiological pH levels. The use of the HOMO, or LUMO or HOMO-LUMO of potential therapeutics incorporates all contributions of inherent reactivity, be they inhibitorHOMO à receptorLUMO or receptorHOMO à inhibitorLUMO dominated interactions.

Conclusion A previously described linear free energy methodology has been effective in predicting the efficacy of a wide range of repurposed drugs against four known mechanisms that are critical to the infectiousness and pathogenesis of the SARS-CoV-2 virus: inhibition of the Mpro of SARS-CoV-2, inhibition of SARS-CoV-2 spike protein S-RBD, inhibition of ACE2 receptor, and inhibition of TMPRSS2 receptor.

Experimental Methods

All calculations were carried out using the Gaussian 09 package. Energy optimizations were at the DFT/B3LYP/6-31G(d) (6d, 7f) level of theory for all atoms in water. Selected optimizations at the DFT/B3LYP/6-311+G(d,p) (6d, 7f) level of theory gave very similar results to those at the lower level. Optimized structures were checked to ensure energy minima were located, with no imaginary frequencies. Energy calculations were conducted at the DFT/B3LYP/6-31G(d,p) (6d, 7f) for neutral and cationic compounds with optimized geometries in water, using the IEFPCM/SMD solvent model. With the 6-31G* basis set, the SMD model achieves mean unsigned errors of 0.6 - 1.0 kcal/mol in the solvation free energies of tested neutrals and mean unsigned errors of 4 kcal/mol on average for ions. [23] The 6-31G** basis set has been used to calculate absolute free energies of solvation and compare these data with experimental results for more than 500 neutral and charged compounds. The calculated values were in good agreement with experimental results across a wide range of compounds. [24,25] Adding diffuse functions to the 6-31G* basis set (ie 6- 31+G**) had no significant effect on the solvation energies with a difference of less than 1% observed in solvents, which is within the literature error range for the IEFPCM/SMD solvent model. HOMO and LUMO calculations included both delocalized and localized orbitals (NBO). It is noted that high computational accuracy for each species in different environments is not the focus of this study, but comparative differences between various species is the aim of the study. Experimental errors in inhibitory and docking binding studies are substantially higher than those in calculated molecular specifiers. Figure 1. Membrane fusion and endocytosis mechanisms of viral cell entry showing potential receptor inhibitory sites: S-RBD, TMPRSS2, ACE2 and MPro

References

[1] T Tang, M Bidon, JA Jaimes, GRWhittaker, S Daniel, Coronavirus membrane fusion mechanism offers a potential target for antiviral development, Antiviral Research, 2020, 178, 104792 [2] J Shang, Y Wan, C Luo, G Ye, Q Geng, A Auerbach, F Li, Cell entry mechanisms of SARS-CoV-2, PNAS, 2020, 117, 11727–11734 [3] DL McKee, A Sternberg, U Stange, S Laufer, C Naujokat, Candidate drugs against SARS-CoV-2 and COVID-19, Pharmacol Res, 2020, 157, 104859 [4] M Wang, R Cao, L Zhang et al, Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro, Cell Research 2020, 30, 269–271, doi:10.1038/s41422-020-0282- 0 [5] M Hoffmann, H Kleine-Weber, S Schroeder, et al. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor, Cell, 2020, 181, 271–280.e8. [6] X Wang , S Xia, Q Wang, W Xu, W Li, L Lu, S Jiang, Broad-Spectrum Coronavirus Fusion Inhibitors to Combat COVID-19 and Other Emerging Coronavirus Diseases, Int. J. Mol. Sci. 2020, 21, 3843; doi:10.3390/ijms21113843 [7] NP Azouz, AM Klinger, ME Rothenberg, Alpha 1 Antitrypsin is an Inhibitor of the SARS-CoV2–Priming Protease TMPRSS2, bioRxiv preprint, 2020, doi: https://doi.org/10.1101/2020.05.04.077826. [8] MS Roomi, YD Khan, Potential Compounds for the Inhibition of TMPRSS2, 2020, https://doi.org/10.26434/chemrxiv.12727787.v1 [9] MO Idris , AA Yekeen , OS Alakanse, OA Durojaye, Computer-aided screening for potential TMPRSS2 inhibitors: a combination of pharmacophore modeling, molecular docking and molecular dynamics simulation approaches, J Biomol Struct Dyn, DOI: 10.1080/07391102.2020.1792346 [10] W Tai, L He, X Zhang, et al. Characterization of the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development of RBD protein as a viral attachment inhibitor and vaccine, Cell Mol Immunol 2020, 17, 613–620 https://doi.org/10.1038/s41423- 020-0400-4 [11] S Choudhary, YS Malik, S Tomar, Identification of SARS-CoV-2 Cell Entry Inhibitors by Drug Repurposing Using in silico Structure-Based Virtual Screening Approach, Front. Immunol., 2020, https://doi.org/10.3389/fimmu.2020.01664 [12] K Teralı, B Baddal, HO Gülcan, Prioritizing potential ACE2 inhibitors in the COVID-19 pandemic: Insights from a molecular mechanics-assisted structure-based virtual screening experiment, J Molec Graph Model, 2020, 100, 107697-107707 [13] M Smith, JC Smith, Repurposing Therapeutics for COVID-19: Supercomputer-Based Docking to the SARS-CoV-2 Viral Spike Protein and Viral Spike Protein-Human ACE2 Interface. ChemRxiv. 2020, Preprint. https://doi.org/10.26434/chemrxiv.11871402.v4 [14] CW Fong, Inhibition of COVID-2019 3C-like protease: structure activity relationship using quantum mechanics, hal archives 2020, hal-02529030v1 [15] CW Fong, Screening potential repurposed COVID-2019 3C-like protease inhibitors, hal archives 2020, hal-02663287v1 [16] CW Fong, Screening potential anti-virals for the main protease of the Coronaviridae family including SARS-CoV-2, SARS-CoV, MERS, hal archives 2020, hal-02897882v1 [17] CW Fong, COVID-19: predicting inhibition of the main protease and therapeutic intracellular accumulation and plasma and lung concentrations of repurposed inhibitors, hal archives 2020, hal-02917312v1 [18] CW Fong, Covid-19: Predicting inhibition of SARS-CoV-2 in Caco-2 cells: viral cellular entry, hal archives 2020, hal-02943341v1 [19] EC Vatansever, K Yang, KC Kratch, A Drelich, et al, Targeting the SARS-CoV-2 Main Protease to Repurpose Drugs for COVID-19, bioRxiv Preprint 2020 Jul 27, doi: 10.1101/2020.05.23.112235 [20] T Bobrowski, V Alves, CC Melo-Filho, et al, Computational Models Identify Several FDA Approved or Experimental Drugs as Putative Agents Against SARS-CoV-2, ChemRxiv 2020 Preprint https://doi.org/10.26434/chemrxiv.12153594.v1 [21] A Fischer, M Sellner, S Neranjan, M Smieško, MA Lill, Potential Inhibitors for Novel Coronavirus Protease Identified by Virtual Screening of 606 Million Compounds, Int J Mol Sci 2020, 21, 3626; doi:10.3390/ijms21103626 [22] J Wang, Fast Identification of Possible Drug Treatment of Coronavirus Disease -19 (COVID-19) Through Computational Drug Repurposing Study, J Chem Inf Model 2020, 60, 3277–3286 [23] AV Marenich, CJ Cramer, DJ Truhlar, Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Deﬁned by the Bulk Dielectric Constant and Atomic Surface Tensions, J Phys Chem B, 2009, 113, 6378 -96 [24] S Rayne, K Forest, Accuracy of computational solvation free energies for neutral and ionic compounds: Dependence on level of theory and solvent model, Nature Proceedings, 2010, http://dx.doi.org/10.1038/npre.2010.4864.1. [25] RC Rizzo, T Aynechi, DA Case, ID Kuntz, Estimation of Absolute Free Energies of Hydration Using Continuum Methods: Accuracy of Partial Charge Models and Optimization of Nonpolar Contributions, J Chem Theory Comput. 2006, 2, 128-139 Table 1. Inhibition and docking binding energies of SARS-CoV-2 Mpro inhibition [19] and calculated molecular specifiers of repurposed drugs

Inhibitor IC50 ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- µM kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO D cm3/mol eV Pimozide 42.2 -10.01 -7.72 -12.06 4.26 357 -5.64 -0.31 5.33 Ebastine 57 -10.62 -8.27 -13.9 5.17 370 -5.77 -1.54 4.22 Bepridyl 72 -8.31 -6.69 -10.44 3.62 326 -5.26 -0.12 5.14 Bepridyl Ion 72 -8.31 -9.5 -10.54 15.53 283 -5.44 -0.18 5.26 Sertaconazole 76 -8.77 -6.02 -11.3 2.76 315 -6.06 -0.94 5.12 Rimonabant 85 -11.23 -6.54 -11.17 6.57 343 -6.23 -1.33 4.90 Oxiconazole Ion 99 -9.18 -6.57 -10.85 13.31 267 -6.42 -2.92 3.50 Oxiconazole 99 -9.18 -5.82 -10.67 9.38 297 -6.21 -1.37 4.84 Itraconazole 111 -8.44 -11.02 -20.68 8.75 563 -4.77 -0.96 3.80 Tipranavir 180 -10.74 -12.86 -11.47 7.91 391 -6.24 -1.95 4.30 Nelfinavir 234 -9.67 -11.77 -13.78 11.8 390 -5.70 -0.93 4.77 Nelfinavir Ion 234 -9.67 -12.25 -13.77 12.71 463 -5.77 -1.00 4.77 Zopiclone 349 -10.1 -0.35 -7.79 8.27 260 -5.98 -2.32 3.66 Trihexyphenidyl 370 -8.72 -3.14 -8.99 2.77 292 -5.74 -0.03 5.71 Trihexyphenidyl 370 -8.72 -6.59 -9.12 15.22 237 -6.58 -0.12 6.46 Ion Saquinavir 411 -10.37 -11.96 -14.84 10.08 501 -5.92 -2.02 3.89 Isavuconazole 438 -8.77 -8.7 -9.05 8.08 296 -6.15 -1.71 4.44 Lopinavir 486 -8.91 -13.93 -16.77 5.82 491 -5.84 -0.08 5.76 Clemastine 497 -8.36 -4.7 -9.37 5.60 237 -5.71 -0.37 5.34 Clemstine Ion 497 -8.36 -8.39 -9.56 27.20 293 -6.41 -0.40 6.00 Metixene 635 -9.01 -2.7 -8.8 3.10 244 -5.51 -0.34 5.16 Metixene Ion 635 -9.01 -6.71 -9 14.64 178 -5.60 -0.41 5.19 Rupintrivir 68 -16.39 -11.88 10.94 482 -6.37 -1.62 4.75 Duloxetine 3047 -8.79 -4.55 -7.68 1.65 185 -5.54 -0.92 4.63 Duloxetine Ion 3047 -8.79 -7.8 -7.97 20.40 203 -5.61 -0.96 4.65

Table 2. Docking binding energies of SARS-CoV-2 Mpro inhibitors [21] and calculated molecular specifiers of repurposed drugs

Inhibitor ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO D cm3/mol eV Apixaban -84 -10.92 -10.55 7.16 299 -5.91 -1.36 4.55 Nelfinavir -80.6 -11.77 -13.78 11.8 390 -5.70 -0.93 4.77 Nelfinavir -80.6 -12.25 -13.77 12.8 463 -5.77 -1.00 4.77 Ion Glecaprevir -80.3 -13.8 -13.31 13.85 556 -6.53 -2.17 4.37 Lorecivivint -79.7 -5.75 -10.44 8.49 406 -5.87 -1.66 4.20 Rivaroxban -77.2 -9.09 -9.93 6.88 289 -5.97 -1.56 4.41 Betrixaban -73.3 -8.95 -10.29 15.64 363 -5.89 -1.61 4.28 Betrixaban -73.3 -9.98 -10.56 37.64 311 -6.16 -2.06 4.10 Ion Saquinavir -71.5 -11.96 -14.84 10.08 501 -5.92 -2.02 3.89 Saquinavir -71.5 -14.04 -14.9 17.37 396 -6.35 -2.05 4.30 Ion Voxilaprevir -66.5 -15.09 -12.35 12.07 594 -6.68 -2.19 4.49 Amprenavir -66.5 -10.22 -9.62 5.79 365 -5.90 -0.56 5.35 Taxifolin -53.3 -7.49 -6.84 9.14 171 -5.96 -1.67 4.29 Rhamnetin -52.4 -8.79 -6.64 7.2 186 -5.48 -1.82 3.66 N3 -59.3 -15.57 -12.5 7.63 443 -6.48 -1.64 4.84

Table 3. Docking binding energies of ACE2 and S-RBD inhibitors [11] and calculated molecular specifiers of repurposed drugs

ACE2 Inhibitor ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO D cm3/mol eV GR127935 Rot -11.23 -6.68 -11.5 4.42 360 -5.36 -1.68 3.68 Conformer GR127935RotIon -11.23 -10.67 -11.69 38.94 334 -5.54 -1.70 3.85 GR127935 -11.23 -6.17 -12.22 2.75 322 -5.21 -1.61 3.60 GR127935 Ion -11.23 -10.46 -11.36 24.61 344 -5.41 -1.64 3.77 GNF-5 -7.57 -8.55 -10.38 11.94 243 -5.65 -1.54 4.12 RS504393 -8.32 -5.94 -9.98 10.36 315 -5.72 -1.16 4.56 TNP -7.42 -6.99 -13.51 11.81 266 -5.18 -3.05 2.13 TNP Rot Conformer -7.42 -7.96 -14.2 6.36 228 -5.30 -3.02 2.28 Eptifibatide Ion -6.05 -14 -16.12 46.92 531 -5.65 -1.06 4.59 Eptifibatide -6.05 -12.8 -15.72 6.76 637 -5.64 -1.04 4.61

S-RBD Inhibitor ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO D cm3/mol eV KT203 -8.73 -9.73 -13.02 3.88 322 -5.91 -1.53 4.38 BMS195614 -8.25 -9.03 -12.01 15.5 322 -5.80 -1.77 4.03 KT185 -8.16 -8.53 -14.47 12.6 454 -5.87 -1.25 4.62 RS504393 -7.67 -5.94 -9.98 10.36 315 -5.72 -1.16 4.56 GSK1838705A -6.46 -8 -14.6 12 402 -5.12 -1.07 4.04 GSK1838705A -6.46 -12.18 -14.85 39.12 329 -5.15 -1.09 4.06 Ion

Table 4. Docking binding energies of ACE2 inhibitors [12] and calculated molecular specifiers of repurposed drugs

ACE2 ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- Inhibitor kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO D cm3/mol eV Lividomycin -2.145 -11.6 -8.35 15.8 462 -5.71 1.29 7.00 Burixafor -2.108 -2.87 -16.07 8.73 404 -5.45 0.02 5.47 Quisinostat -1.999 -7.05 -12.83 9.69 307 -5.32 -1.04 4.28 Fluprofylline -1.785 -6.84 -13.34 8.7 291 -5.83 -1.64 4.19 Pemetrexed -1.603 -9.86 -12.12 26.1 298 -5.44 -1.16 4.28 Spirofylline -1.542 -7.83 -14.38 8.95 295 -5.92 -1.04 4.89 Edotecarin -1.312 -10.84 -13.3 8.61 305 -5.40 -2.30 3.10 Diniprofylline -1.292 -8.76 -12.07 5.07 341 -6.13 -1.74 4.39 MLN-4760 -0.309 -8.19 -9.66 11.82 277 -5.92 -0.85 5.07 Table 5. Docking binding energies of TMPRSS [8] and calculated molecular specifiers of repurposed drugs

TMPRSS Inhibitor ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- kcal/mol kcal/mol kcal/mol Moment D Vol eV eV LUMO cm3/mol eV 6-Deacetylnimbinene -13.4 -10.92 -6.9 9.62 313 -5.96 -0.68 5.27 2',3'-Dehydrosalannol -12.9 -12.49 -8.06 7.95 412 -5.89 -1.25 4.65 Deacetylsalannin -11.9 -13.7 -8.23 7.59 433 -6.04 -1.05 4.99 Salannin -11.2 -15.54 -8.54 8.58 523 -5.99 -0.92 5.07 Salannol acetate -10.7 -13.65 -8.35 6.14 397 -5.88 -0.51 5.36 17-epi-17-Hydroxyazadiradione -10 -12.86 -7.39 7.87 311 -6.16 -1.56 4.59 17-Hydroxyazadiradione -9.7 -12.58 -7.83 3.9 384 -5.73 -1.66 4.07 Nobiletin -9.2 -13.65 -4.25 5.67 338 -6.09 -1.96 4.12 5,6,7,8,3',4',5'-Heptamethoxyflavone -9.5 -14.97 -3.94 4.61 302 -5.97 -1.89 4.08 Pinostrobin -9.1 -7.34 -6.75 6.16 226 -6.15 -1.43 4.72 Sakuranetin -8.9 -7.85 -6.85 6.43 184 -5.99 -1.41 4.58 Homoeriodictyol -8.8 -8.25 -6.56 7.81 194 -5.95 -1.41 4.55 Umuhengerin -8.75 -12.53 -5.09 8.56 291 -6.07 -1.84 4.23 Eucalyptin -8.5 -9.08 -7.07 10.64 242 -5.79 -1.97 3.82 Lutonarin -8.8 -12.51 -11.65 10.51 385 -5.83 -2.04 3.79 3-Hydroxy-3',4',5,7- -7.8 -14.98 -3.99 4.56 333 -5.97 -1.89 4.08 tetramethoxyflavone 3-Hydroxy-4'-methoxyflavone -8.4 -7.62 -6.9 3.56 148 -5.59 -1.96 3.63 5,6,5'-Trihydroxy-3,7,2',4'- -7.6 -11.59 -5.06 9.06 244 -5.74 -1.91 3.82 tetramethoxyflavone Glabone -7.6 -8.09 -7.11 7.74 188 -5.98 -1.93 4.05 Isothymonin -7.5 -10.84 -5.73 6.37 268 -5.69 -1.85 3.83 3-Acetyltricin -7.5 -11.12 -6.86 14.12 205 -5.90 -1.83 4.07 Camostat Mesylate -7.4 -10.69 -10.7 43.64 274 -6.69 -2.03 4.66 Camostat -7.4 -9.3 -10.48 11.41 256 -5.78 -1.41 4.36 7,3',4',5'-Tetramethoxyflavanone -7.3 -11.05 -5.43 5.02 236 -6.16 -1.63 4.53

Table 6. Docking binding energies of TMPRSS [9] and calculated molecular specifiers of repurposed drugs

TMPRSS Inhibitor ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- kcal/mol kcal/mol kcal/mol Moment D Vol eV eV LUMO cm3/mol eV ZINC64606047 -9.3 -7.49 -12.87 5.66 254 -5.65 -1.35 4.30 ZINC05296775 -9.2 -7.27 -10.13 5.61 314 -5.49 -1.40 4.09 Nafamostat -8.2 -6.33 -9.19 4.18 256 -5.87 -1.83 4.04 Nafamostat DiIon -8.2 -9.51 -0.38 19.85 227 -6.19 -2.13 4.06 Camostat -7.2 -9.3 -10.48 11.41 256 -5.78 -1.41 4.36 Camostat Mesylate -7.2 -10.69 -10.7 43.64 274 -6.69 -2.03 4.66 Baricitinib -6.9 -3.23 -6.96 7.23 259 -5.85 -1.26 4.59 Ruxolitinib -6.7 -4.86 -7.79 7.76 261 -5.77 -1.30 4.47 Pefabloc -5.6 -2.89 -3.22 7.93 145 -6.23 -1.34 4.89 Phenylmethyl -5 -3.41 -3.88 5.68 143 -6.94 -0.59 6.35 sulfonylfluoride Table 7. Docking binding energies of S-Protein of SARS-Cov-2 to ACE2 interface [13] and calculated molecular specifiers of repurposed drugs

Virus S-Protein - ΔGbinding ΔGdesolvCDS ΔGlipoCDS Dipole Molec HOMO LUMO HOMO- ACE2 Interface kcal/mol kcal/mol kcal/mol Moment Vol eV eV LUMO Inhibitor D cm3/mol eV pemirolast -7.4 -3 -3.5 7.19 167 -5.96 -1.86 4.10 benserazide -7.3 -6.89 -3.56 11.78 184 -5.80 0.17 5.97 pyruvic acid calcium -7.3 -7.26 -4.85 2.26 288 -6.61 -1.84 4.77 isoniazid quercitin -7.3 -7.43 -7.04 6.57 181 -5.54 -2.00 3.53 protirelin -7.3 -7.43 -8.2 4.73 254 -5.93 -0.17 5.76 carbazochrome -7.08 -5.89 -5.19 16.45 164 -5.78 -2.49 3.29 nitrofurantoin -7.2 -9.79 -5.04 8.18 135 -6.41 -3.37 3.04 sapropterin -7.1 -2.47 -6.07 17.75 151 -4.57 0.08 4.65 Vidarabine -7.1 -2.56 -8.23 6.57 177 -5.90 -0.54 5.36 eriodictyol -7.1 -7.2 -7.02 9.04 184 -5.83 -1.42 4.40 tazobactum ion -7.1 -5.22 -3.63 32.88 169 -5.74 -0.46 5.28 phenformin ion -7 -3.31 -8.79 22.5 138 -6.49 -0.33 6.16 vildagliptin -7 -4.71 -5.78 12.9 210 -5.84 0.24 6.08 demethyl-coclaurine -7 -5.29 -7.29 3.65 216 -5.52 0.01 5.53