Suramin, Penciclovir and Anidulafungin Bind Nsp12

Page No: 1 Title Page

Full title: Suramin, Penciclovir and Anidulafungin bind nsp12, which governs the RNA-dependent-RNA polymerase activity of SARS-CoV-2, with similar interaction energy as Remdesivir-triphosphate, indicating potential in the treatment of COVID-19 infection

Running Head: Repurposed drugs for RdPp inhibition targeting COVID-19

Author 1: Sanjay Kumar DEY; BSc, MSc, PhD; [email protected]; Post-doctoral

Associate; 1Center for Advanced Biotechnology and Medicine, Rutgers University, New Jersey-

08854, USA; 2Department of Biochemistry, University of Delhi South Campus, New Delhi-

110021, India;

Author 2: Manisha SAINI; BTech, MTech; [email protected]; PhD Student; 2Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi- 110021, India Author 3: Chetna DHEMBLA; BSc, MSc; [email protected]; PhD Student; 2Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi- 110021, India Author 4: Shruti BHATT; BSc, MSc; [email protected]; PhD Student; 2Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi-110021, India Author 5: A. Sai RAJESH; BSc, MSc; [email protected]; PhD Student; 2Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi-110021, India and 3Department of Biosciences and Biotechnology, Fakir Mohan University, Odisha-756020, India Author 6: Varnita ANAND; BSc, MSc; [email protected]; PhD Student; 2Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi-110021, India Author 7: Hirendra Kumar DAS; BSc, MSc, PhD; [email protected]; Distinguished Professor; 4School of Biotechnology, Jawaharlal Nehru University, New Delhi-110067, India Author 8 (corresponding author): Suman KUNDU*; BSc, MSc., PhD; [email protected]; Professor; Department of Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi-110021, India

Page No: 2 Author for correspondence and requests for reprints: *Prof. Suman Kundu, Department of

Biochemistry, University of Delhi South Campus, Benito Juarez Road, New Delhi-110021, India;

Tel: +91-1124112081; E-mail: [email protected]

Word count: 7000

Number of tables: 3

Number of figures: 9

Number of supplementary digital content files: None

Conflicts of Interest and Source of Funding: The authors declare no conflict of interest.

SKD is thankful to Professor Eddy Arnold, Rutgers University, USA, for a post-doctoral associateship. MS and SB acknowledge the Department of Biotechnology (DBT), Government of India for their project and research fellowships, respectively. CD appreciates the financial support from Lady Tata Memorial Trust in the form of a research fellowship. ASR is thankful to the Government of Odisha for the Biju Patnaik research fellowship. HKD acknowledges Indian

National Science Academy for Emeritus Fellowship. SK appreciates financial support over the years from DBT, University Grants Commission (UGC), Defense Research and Development

Organization (DRDO) and Department of Science and Technology (DST), Government of India.

SK is thankful to the University of Delhi for support in all forms. Page 1

Full title: Suramin, Penciclovir and Anidulafungin bind nsp12, which governs the RNA- dependent-RNA polymerase activity of SARS-CoV-2, with similar interaction energy as Remdesivir-triphosphate, indicating potential in the treatment of COVID-19 infection

Running Head: Repurposed drugs for RdPp inhibition targeting COVID-19

Word count: 7000 Number of tables: 3 Number of figures: 9 Number of supplementary digital content files: None

Structured abstract: Introduction: COVID-19, for which no vaccine or confirmed therapeutic agents are available, has claimed over 7,30,000 lives globally. A feasible and quicker method to resolve this problem may be ‘drug repositioning’. Areas covered: We investigated selected FDA and WHO-EML approved drugs based on their previously promising potential as antivirals, antibacterials or antifungals. These drugs were docked onto the three-dimensional structure of nsp12 protein, which reigns the RNA-dependent RNA polymerase activity of SARS-CoV-2 and is one of the major therapeutic targets for corona viruses. Inhibitor-protein complexes were also subjected to molecular dynamics simulation. The binding energies and the mode of interaction of the active site of the protein with the drugs were evaluated. Results: Suramin, Penciclovir and Anidulafungin were found to bind to nsp12 with similar binding energies as that of Remdesivir, which is currently being used in the treatment of COVID-19. In addition, recent experimental evidences indicate that these drugs exhibit antiviral efficacy against SARS-CoV-2. Thus, they might have a prospective therapeutic potential against the key viral enzyme. Expert opinion: Repurposed drugs will provide viable options for the treatment of COVID-19 and insight into the molecular mechanisms by which these potential drug candidates exhibit anti-SARS- CoV-2 activity.

Keywords: SARS-CoV-2, RdRp, Non-Structural Protein 12, FDA approved Drugs, WHO-EML, COVID-19, Suramin, Penciclovir, Anidulafungin, and Remdesivir-TP

Article highlights: • The nsp12 protein of SARS-CoV-2 is a potential drug target as it harbors the RdRp activity which helps in the viral replication. The present study identified repurposed drugs of various origins capable of inhibiting nsp12 of SARS-CoV-2 to combat the viral infection. • Penciclovir, Anidulafungin and Suramin, the drugs identified in this study by computational methods, binds to nsp12 with a similar affinity like Remdesivir, which is currently being used to treat COVID-19. Page 2

• Existing experimental evidences support our findings; Suramin is a known non-nucleoside inhibitor of norovirus RdRp51 and exhibits antiviral activity against the SARS-CoV-2 cell lines; penciclovir is shown to reduce the COVID-19 infection in cell lines; Anidulafungin inhibits viral infection with lower half maximal inhibitory concentration compared to that of Remdesivir against SARS-CoV-2. • The three drugs with the potential to treat COVID-19 are FDA or WHO approved and thus can be taken to clinical trials directly without the need for preclinical toxicity evaluation. • The study also offers avenues for lead optimization of the three compounds or their analogue design to produce highly potent inhibitors of nsp12.

1. Introduction: The novel coronavirus disease 2019 (COVID-19) was declared a global pandemic on the 11th of March 2020 by WHO, barely within 3 months of the emergence of its first case [1, 2]. Since that initial instance, it has kept shifting its epicenter through various continents and has now spread to 213 countries and territories. It has already infected 1,90,00,000 people across the globe, claimed more than 7,00,000 lives, gravely impacted the countries socio-economically and even the most advanced healthcare systems have crumbled under its weight. It is a highly contagious pneumonia caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the symptoms include complications in the gastrointestinal tract, respiratory problems, like cough, cold, shortness of breath, fever, and may even lead to fatality in some cases. Family of this virus have inflicted the world with similar outbreaks, not the least of which was the Spanish Flu, which during the year 1918-1920 had infected one-fourth of the world’s population and killed approximately 50 million (Centers for Disease Control and Prevention, USA). Recent examples of such epidemics caused by the closely related strains include Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS), each one spanning to more than 20 countries and killing approximately 1600 people in their wake, with a fatality rate of 10% and 35%, respectively [3]. COVID-19 has a global fatality rate of 4.6%, and if the outbreak reaches anywhere near the previous scales of pandemics, this rate would translate to millions of deaths [3]. These closely related strains are members of the coronavirus (CoV) family and belong to the class of positive-stranded RNA viruses with a huge 30 kb polycistronic genome. The viral replicase polyproteins namely pp1a and pp1ab are translated by the proximal two-thirds of the 5’ end of CoV genome (ORFs 1a and 1b) [4]. These two polyproteins are cleaved into 16 nonstructural proteins (nsps), which are essential for viral replication and transcription, thus being regarded as a potential virulence factor and drug targets for CoV [5]. ORFs near the 3’ end encodes the structural proteins namely S, M, E and N which are the spike, membrane, envelope, and nucleocapsid proteins, respectively [6]. Out of the 16 nsps encoded by the 5’ end, nsp12 houses the RdRp (RNA-dependent RNA polymerase) activity. The central catalytic subunit of the RNA-synthesizing machinery for viral replication and transcription is the RdRp, which has seven catalytic motifs (A-G). Each motif has a particular function from selection and correct positioning to the addition of the newly incorporated NTPs (Nucleoside triphosphates) to the growing chain which is carried out with the help of a set of Page 3 conserved amino acid residues. RdRp is an essential viral enzyme in the life cycle of RNA viruses and has been targeted in various viral infections, including the hepatitis C virus (HCV), the Zika virus (ZIKV), coronaviruses (CoVs) etc. [7-12] It has continuously served as a predominant drug target in SARS and MERS research with minimal cytotoxic effects on host cells [13]. Thus, nsp12 holds potential as an attractive target to develop therapeutic agents against SARS-CoV-2. In addition, human homologs of nsp12 are not known to exist and thus drugs that target nsp12 of COVID- 19 are not expected to cause major disruption in physiological processes [13]. Of late, attempts had been made to identify small molecules that could bind to nsp12 with the potential of therapeutic intervention for COVID-19 affected patients. Investigations have revealed vitamin B12, valproic acid co-A, Remdesivir, Sofosbuvir, Tenofovir and Ribavirin to be effective inhibitors of nsp12 [14-19]. While in vitro validation and clinical trial of these molecules, except Remdesivir, are awaited for COVID-19, it is necessary to identify additional molecules with similar objectives to populate the pipeline of potential drug candidates against COVID-19. The previous investigations utilized a homology model of the protein for small molecule screening, and with the experimental (cryo-electron microscopy) three-dimensional structures of the protein having just been reported [20], the exercise of small molecule screening was necessary to be re-visited. Novel small molecules would have to undergo validation and toxicity tests and the initial rounds of clinical trials, resulting in a long-drawn process of drug discovery. To address the current pandemic swiftly, we propose drug repositioning as a feasible method to tackle the problem of unavailability of vaccines or therapeutic agents against novel coronavirus. FDA and WHO-essential medicine list 2019 approved drugs, along with some widely prescribed and investigational drugs, provide a safer alternative as there is no requirement to test the toxicity of these drugs. Hence, in this report, we have used molecular docking and simulations to identify potent drugs capable of inhibiting nsp12 of SARS-CoV-2. The exercise revealed three FDA- approved drugs with the potential to treat COVID-19.

2. Materials and methods: 2.1. Sequence alignment: The fasta sequence of nsp12 protein of SARS-CoV (6NUR_SARSCoV) and the newly emerged cryo-EM (6M71) structure of nsp12 of SARS-CoV-2 (6M71_SARSCoV2) were retrieved from the Research Collaboratory for Structural Bioinformatics (RCSB) database PDB (Protein Data Bank) and aligned using MultAlin software [21]. Further graphical enhancement and secondary structure elements were added using ESPript 3.0 [22]. For the homology model of SARS-CoV-2 nsp12, sequence from National Center for Biotechnology Information (NCBI) nucleotide database with accession number NC045512.2 was used and aligned with the other two using the same software [21]. 2.2. Modelling and preparation of target protein: The amino acid sequence of nsp12 of SARS-CoV-2 was retrieved from NCBI and a homology model was built based on the cryo-EM structure of RdRp of SARS-CoV (PDB ID 6NUR) using SWISS-MODEL [23], since the earlier investigations on drug screening were based on such a Page 4 homology model. The template shared 96.53% sequence identity with the input query. The model developed was validated by using PROCHECK [24], ERRAT [25] and Verify 3D [26]. However, unwanted water and cofactors were removed from the structure before commencing molecular docking. In the course of completion of our manuscript, the cryo-EM structure of nsp12 of SARS-CoV- 2 was reported [20] at 2.9 Å resolution in complex with nsp7 and nsp8 proteins (PDB ID 6M71). The coordinates for nsp7 and nsp8 were removed from the PDB file and the resulting PDB file was used for all further analysis of nsp12 protein structure. The structure was further prepared as above for the subsequent docking studies. The homology model was superimposed on the cryo-EM structure of nsp12 and RMSD was calculated for structural comparison using PyMOL [27]. 2.3. Ligand preparation: A total of 43 diverse arrays of drugs (Table 1) exhibiting medicinal values were downloaded from PubChem databank in .sdf format and converted to .mol2 extension prior to docking using UCSF Chimera [28]. 2.4. Molecular docking: To evaluate the best binding energy and pose of the ligands potentially capable of inhibiting nsp12 of SARS-CoV-2 (both the model and the cryo-EM structure), molecular docking was performed with the help of SwissDock [29], an automated web-based tool capable of predicting the molecular interactions that may occur between a target protein and a small molecule. Before initiation of docking both the proteins and the ligands were minimized using CHARMM forcefield [23]. Docking experiments were executed keeping the accurate default parameters of SwissDock with no prefecture of defined regions (blind docking). After the completion of docking, the protein-ligand complex was examined using the UCSF Chimera [28]. Many of the docking results were also verified using the professional and licensed software Schrodinger [30]. 2.5. Molecular dynamics simulations: Docked protein-ligand complexes with the best pose showing lowest binding energy (against the SARS-CoV-2 cryo-EM structure (PDB 6M71)) [20], namely, nsp12-anidulafungin (-11.44 kcal/mol), nsp12-penciclovir (-11.7 kcal/mol), and nsp12-suramin (-12.09 kcal/mol) were subjected to molecular dynamics simulations using the Molecular Modelling Toolkit or MMTK software [31] integrated by V. Munoz-Robles and J. D. Marechal of the Computational Biotechnology Chemistry team into UCSF Chimera [28]. Cryo-EM protein structure of nsp12 (PDB ID: 6M71), was processed by the ProteinPrep wizard of Schrodinger software [30] to complete missing amino acid side chains, proper configuration of N- and C-terminal amino acids and initial energy minimization using OPLS2005 forcefield for the protein [32-34]. Each protein-ligand complex was individually subjected to molecular dynamics simulations (MD simulations) using the MMTK module of UCSF Chimera. Complexes were pre- processed using the Dock-Prep tool of UCSF Chimera to remove all hydrogens and replaced with AMBER compliant atomic nomenclatures, followed by calculation of charges to all atoms including the ligands using the AMBER ff14sb and AM1-BCC forcefields based on the ANTECHAMBER algorithm [35]. After adding proper hydrogens and charge calculations, protein-ligand complexes were solvated Page 5 using three-point TIP3PBOX type water molecules surrounding 10 Å in each of the three directions of the complex forming a cubic box system for simulation. A cubic box system of MD simulation was generated using a 10 Å distance from each side of the protein surface. Noncovalent interactions were measured with a cut off distance of 14 Å. The particle mesh Ewald (PME) [36] method was used to calculate electrostatic interactions while Lennard-Jones interaction method [37] was used to constrain all hydrogen bonds. The system first performed 300 steps of steepest descent with energy minimization. Then, the minimized system was used to perform simulations using an NVT ensemble. The Nosé–Hoover method [38] was used to maintain a fixed temperature while Anderson Barostat [39] was used to maintain a constant atmospheric pressure (i.e. 1 ATM) using NPT ensemble. The system was heated from 0 K to 298 K using an increment of 10 K/ps time frame. The time step of each simulation was set to 2 fs and finally, production MD was run for 1000 ps. Visualizations, RMSD as well as energy calculation and data analysis were performed with UCSF Chimera software [28].

3. Results and discussion: 3.1. Sequence alignment of nsp12 proteins of SARS-CoV and SARS-CoV-2 do not reveal major differences in their key amino acid residues that govern RdRp activity: The novel coronavirus (CoV-2) achieved its ability to infect humans primarily due to specific mutations in its spike protein that endowed the virus with high affinity binding to the corresponding human receptor. It is thus worthwhile to examine whether any of the non-structural proteins of CoV-2 also evolved mutations, making them distinct from other homologues in the corona virus family, which in turn may necessitate design of drugs unique to CoV-2. On the alignment of amino acid sequence of nsp12 of SARS-CoV-2 [homology model and cryo-EM structure [20] (PDB ID: 6M71)] with its closest homologue nsp12 of SARS-CoV (PDB ID: 6NUR), it was evident that the RdRp domain is conserved throughout comprising of three subdomains, the palm, thumb, finger and the seven motifs (A-G) as shown in Figure 1. The motif A of nsp12 SARS-CoV-2 spans residues 611-626 with an asparagine instead of threonine at position 614 (Figure 1) and forms a β-strand. Motif B spans the region from 680-710 (Figure 1) and forms an α-helix. Motif C includes the residues from 753-767 which are present in the turn between the two β-strands. Motif D is assigned from 775-796 and forms an α helix with alanine instead of serine at residue position 786. Motif E spanning residue 810-820 forms a tight loop. Motif F from residue 538-560 forms an ordered loop (β-loop) connecting two anti-parallel β- strands. Lastly, motif G spans the residues from 500-518 forming an α-helix. It is thus evident that the key amino acid chains in the RdRp domains of nsp12 in the SARS family are fairly conserved. The asparagine and alanine mutations in motif A and motif D, respectively, do not seem to directly influence either the structure or the function of these motifs in a perceptible manner. Nonetheless, in vitro validation of the role of these mutations in RdRp activity may be worth investigating [40]. 3.2. The homology model of nsp12 of SARS-CoV-2 is similar to the experimental structure of the protein including the conformational architecture of the RdRp motifs: The previous investigations on screening of drugs or small molecules with nsp12 CoV-2 as the target were performed with homology models in absence of experimental structures [14-16]. For parity with such studies, we had also initiated investigation initially with a homology model of the Page 6 protein. The cryo-EM experimental structure released subsequent to our initiation of investigation was in complex with nsp7 and nsp8, which may induce conformational changes not present in the structure of nsp12 alone. Hence, it seemed imperative to compare the homology and experimental models and screen drugs using both the models to examine the docking results for any significant differences in binding energies between the models. The homology model for nsp12 of CoV-2 was built (Figure 2A) by taking the structure of SARS-CoV nsp12 (PDB ID: 6NUR) as a template. The robustness of the model was validated by its Ramachandran plot displaying ~93% of the residues in the most favored region (data not shown; average from three validation software). The experimental structure of nsp12 CoV-2 (extracted from the PDB ID: 6M71, which was in complex with two helper proteins), also displayed predominantly alpha-helical globular fold (Figure 2B). The homology model when superimposed on the experimental structure revealed an RMSD of 0.48 Å, indicating that the two structures were significantly similar (Figure 2C). Both the models were analyzed and the seven RdRp motifs (A-G), indicated in the sequence of the protein (Figure 1), were mapped onto the structures (Figure 3). It was evident that the motifs exist in a similar and analogous manner in both the models. 3.3. Selection of FDA and WHO-EML 2019 approved drugs for repurposing for COVID-19 treatment: A total of 43 drugs of different physiognomies belonging to various classes like antivirals, antimalarials, antibacterials, anti-helminths, antifungals, anticancer, antiarrhythmic, receptor agonist, calcium channel blockers, immunosuppressant, antidiarrheal, anti-asthma and anti-inflammation were selected from the available literature on their previously demonstrated ability to treat diseases that are closely related to COVID-19 like SARS and MERS and other viral diseases (Table 1). The drugs were either FDA or WHO-EML 2019 approved for human use or widely prescribed in different parts of the world or are investigational drugs in clinical trials based on FDA recommendation (Table 1). With the status of the drugs as suitable for human consumption, the need to test toxicity of the compounds do not arise, and the drugs can be suitably repurposed for treatment of COVID-19, provided they bind the target protein with high affinity. The selected drugs were screened in silico in search of a potential candidate capable of inhibiting nsp12 of SARS-CoV-2 and, hence restricting the contagion. Remdesivir, a recent investigational antiviral, which entered Phase-3 clinical trial (NCT04292730) to treat COVID-19 on the basis of its ability to inhibit SARS-CoV-2 RdRp [17] was used as a positive control in the docking study. Tetrandrine was chosen as a negative control because while it is known to reduce the expression of Human coronavirus OC43 (HCoV-OC43) spike and nucleoside capsid protein, it is not related with RdRp as of yet. 3.4. Molecular docking identified three drugs with the potential to bind to nsp12 with binding energy similar to Remdesivir-TP: Docking, an established strategy for structure-based drug discovery, was utilized to screen the 43 drugs (Table 1) for their ability to dock or bind to nsp12 SARS-CoV-2. A drug that binds to the active site of the target protein or interacts with it allosterically by binding at a site away from the active site will be capable of arresting the RdRp activity and thus CoV-2 spread. Binding energy (in kcal/mol) was used as an indicator of binary complex formation, with high negative values indicating Page 7 high affinity binding (Figure 4). The positive control Remdesivir inhibits the RdRp by mimicking as a nucleoside analogue. De-phosphorylated form of Remdesivir is an inactive prodrug. Remdesivir triphosphate (Remdesivir-TP), the active form of the drug, is incorporated into nascent RNA strand as Remdesivir monophosphate (Rem-MP) while its cleaved pyrophosphate remains close to it and both of them together inhibits the replication [41]. Remdesivir-TP is expected to yield a high binding energy, while the negative control Tetrandrine, which is not known to bind to nsp12, is expected to yield low binding energy in the docking experiments. RemdesivirTP indeed exhibited a high binding energy of -15.79 kcal/mol with the homology model as the target protein and -14.82 kcal/mol with the cryo-EM structure as the target protein. Tetrandrine, on the contrary, exhibited binding energies of - 5.58 and -5.86 kcal/mol against the homology model and experimental structure, respectively (Figure 4, Table 1). Thus, drugs with binding energy equivalent to that of Remdesivir-TP were considered as potential drugs, with -10 kcal/mol as the stringent cut off binding energy for consideration as a candidate drug for COVID-19. Molecular docking of all the 43 drugs to nsp12 SARS-Cov-2 generated the best-fit pose of the ligand and protein and the resulting binding energies are shown in Figure 4 (and Table 1). Penciclovir (Penciclovir-TP is the active form), an antiviral, Anidulafungin, an antifungal and Suramin, an antimicrobial showed binding energies of -11.71, -11.44 and -12.09 kcal/mol respectively (Table 1; Figure 4), as compared to the positive control (-14.82 kcal/mol) considering the cryo-EM structure as the model. The three drugs above are thus expected to inhibit nsp12 of SARS-Cov-2 in a similar manner to Remdesivir-TP. The binding energies of the drugs with the homology model and cryo-EM structures showed differences in their values but the trend overall was similar in both the cases. In addition to the three drugs indicated above with binding energies equivalent to Remdesivir- TP, the drugs Eltrombopag, Ombitasvir, Digitoxin and Ribavirin also displayed binding energies close to -10 kcal/mol, which may be considered to be significant as well (Figure 4; Table 1). These drugs thus may also be considered for treatment of COVID-19 along with Penciclovir, Anidulafungin and Sumarin. The fact that Eltrombopag has recently been reported [42] as a potent inhibitor of nsp12 validates our study. 3.5. The lead drugs interact with nsp12 in the RdRp domain involving multiple motifs, with motifs A and F being common to all the three drugs: Molecular docking provided the best-fit pose of the ligand in the protein structure and this allowed analysis of the interactions of the drugs with the motifs and the amino acid side chains lining the motifs. Such analysis assures that the drugs will inhibit RdRp activity, if they bind in this domain consisting of seven motifs (Figure 3). The three best drugs indeed bound to the RdRp domain using more than one motif in each case (Figure 5). The side chains that interact with the drug molecules were also identified (Figure 6) indicating the ability of these molecules to fit inside the protein complex comprising of nsp7, 8 and 12 (Figure 7). Such interaction analysis would also provide scope for lead optimization in the future, if it is necessary to augment the binding of the drugs to nsp12; however, the optimized molecules would need to be subjected to toxicity tests and clinical trials yet again. Penciclovir-TP binds to motifs A and F where it forms hydrogen bonds with THR556, ARG553 and CYS622 (Figure 5A; Figure 6A). Similarly, Anidulafungin forms hydrogen bonds with ASP760, Page 8

SER759 and THR556 and van der Waals interactions with ALA580, LEU758, SER 659 and ASP623, all of which lie in motifs A, C and F (Figure 5B; Figure 6B). Suramin forms van der Waals interactions with LYS577, ASN497, ALA688, ASP623, PRO620, THR556, ASP164 and ALA554 and hydrogen bonds with ASN496 in motifs A, B and F (Figure 5C; Figure 6C). However, Remdesivir-TP interacts with THR556, SER682, VAL557 and ASP499. It is interesting to note that the motifs A and F are common in interactions of all the three repurposed drugs with nsp12, while B and C vary between the three. None of these drugs, however, interact with motif G as in the case of Remdesivir-TP. Motifs A and F are at the centre of the nsp12 protein and on the top segment of the RdRp domain, while motif G is more toward the surface (Figure 3). The central interactions of the three drugs make them unique compared to Remdesivir-TP and the potential exists for these drugs to be used in combination with Remdesivir-TP for the complete inhibition of RdRp activity. Motifs A, C and F help in correct positioning of incoming NTPs for RNA synthesis, while motif B helps in proper positioning of the ribose moiety of the NTP by forming hydrogen bond with the 2’ OH group of the ribose [6, 41]. Motif G primarily helps in forming polar interaction with the template RNA [40]. The three drugs are thus expected to inhibit RdRp activity thus blocking replication and preventing transcription and thus viral propagation. 3.6. MD simulation of the lead drugs bound to SARS-CoV-2 nsp12 revealed local conformational changes that might stabilize drug binding: Figure 8 depicts the RMSD changes against the time of MD simulations. Penciclovir attained the structural stability earlier than Anidulafungin and Suramin at ~390 ps, ~400 ps and ~450 ps, respectively. The overall structural changes were minimal indicating that the docked structures were similar to the highest energy minimized conformations on the ps time scale. The longer time needed to stabilize Suramin bound nsp12 structure was probably due to the complex chiral structure of the drug undergoing a number of alternative bond breaking and formation to finally obtain its lowest energy structure. Energy parameters after the MD simulation indicated that Suramin-nsp12 interaction was the most stable (ΔG = -848.06 kcal/mol) followed by the interaction of the drugs Penciclovir (ΔG = -846.39 kcal/mol) and Anidulafungin (ΔG = -845.87 kcal/mol) (Table 2). While Penciclovir displayed highest van der Waals interactions with nsp12, Anidulafungin exhibited a -10 kcal/mol extra electrostatic interaction which confirmed and corroborated well with the interaction maps of each of these drugs interacting with the various amino acid side chains of SARS-CoV-2 nsp12 (Figure 9). These interaction maps revealed that Suramin preserved its H-bonds with THR556 but formed a new one with ARG553. Penciclovir formed additional H-bonds with ASP623 and ARG553 as compared to the docking results. Anidulafungin stabilized its binding mainly by electrostatic interactions with ASP202, ASN204, ASN206, ASP209, GLN108 AND ARG109 probably due to the presence of multiple oppositely charged atoms present in the structure of the drug near its surrounding protein surface, while it lost its H-bonds with the target (Figure 9). Overall, the MD simulation results confirmed the stable binding of these top three drugs candidates with nsp12, which may be validated in vitro in terms of evaluating their inhibitory efficacies or solving the drug-bound nsp12 complex cryo- EM or co-crystal structures thereby facilitating lead optimization to improve their efficacies and specificities against the target, if any, and understanding of the molecular mechanism of how these Page 9 drugs exhibit anti-SARS-CoV-2 activity.

3.7. FDA or WHO-approved drugs Penciclovir, Anidulafungin and Suramin are potential inhibitors of nsp12 and suitable for clinical trials for COVID-19 treatment: Penciclovir, Anidulafungin and Suramin, identified in the present study, have immense potential for the inhibition of SARS-CoV-2 RdRp activity as evident from the above findings. The former two are FDA approved and the latter (Suramin) is listed as a drug in WHO Essential medicine list-2019. These drugs can thus be tested directly in humans inflicted with COVID-19. Various characteristics of the drugs along with their dosage regimen are summarized in Table 3 for due consideration by the scientific and medical community. Penciclovir (trade name-Denavir) [43] is a synthetic nucleoside analogue, an acyclic guanine derivative. It is an antiviral drug whose active form Penciclovir-TP is used for the treatment of various herpesvirus infections. In its active form, Penciclovir binds with a much higher affinity to viral polymerase than human polymerase, thus impairing viral replication. This accounts for the low cytotoxicity of healthy host cells [44]. Anidulafungin (trade name-Eraxis and Ecalta) is a synthetic antifungal drug, generally used for the treatment of Candida infections. It disrupts fungal cell wall synthesis by inhibiting the enzyme complex 1,3-β-D-glucan synthase which is involved in the synthesis of 1,3-β-D-glucan, a component of the fungal cell wall. This leads to cell death and inhibition of fungal growth [45]. Suramin (trade names -309 F, Antrypol, Bayer 205, Belganyl, Fourneau 309, Germanin, Moranyl, Naganin, Naganol, Naphuride) is a synthetic sulphated naphthylamine with multiple biological effects. In the 1920s it was found as a treatment for African trypanosomiasis (African sleeping sickness), wherein it acts by inhibiting the enzymes of energy metabolism of the parasite. It can inhibit many other enzymes and proteins like RNA polymerase, reverse transcriptase, thymidine kinase, dihydrofolate reductase, urease, hexokinase etc. Hence, the precise mechanism of action remains unclear [46]. Suramin has been identified with various other biological functions, for example, it is used in the treatment of cancer [47] as well as an antiviral agent against HIV [48, 49].

4. Conclusion: The nsp12 protein of SARS-CoV-2 is a potential drug target as it houses the RdRp activity which helps in the viral replication utilizing the coordinated functions of motifs (A-G). These motifs are conserved across the various classes of RNA viruses and drugs against this target may be useful for treatment of viral infection inflicted by any of these virus family members. The present study was designed to identify drugs of various origins capable of inhibiting nsp12 of SARS-CoV-2 and hence limiting the spread of the viral infection. The top three drugs - Penciclovir, Anidulafungin and Suramin – identified in this study binds to the nsp12 motifs harbouring the RdRp activity interfering with the positioning of NTPs and can be of use against this novel corona virus. These drugs have similar binding energies for nsp12 as Remdesivir-TP, which is in clinical trials for treating COVID-19, and thus have the potential in the treatment of COVID-19 infection. Since these drugs are FDA or WHO approved, they could be taken to clinical trials directly without the need for preclinical toxicity evaluation. At the same time, it also opens up avenues for lead optimization in Page 10 the search of highly potent novel entities to bind specifically to nsp12. The lead optimized drugs, however, will need to be subjected to toxicity evaluation and clinical trials. Existent experimental and clinical studies fortify our findings. Amongst the selected drugs, suramin, the only reported non-nucleoside inhibitor of norovirus RdRp [50], have also recently shown to exhibit antiviral activity against the SARS-CoV 2 cell lines (VERO E6) by targeting initial stages of replication in the viral life cycle [51]. The clinical benefits of suramin against COVID-19 infection in humans are currently being explored in clinical trials as well (ChiCTR2000030029). Furthermore, we demonstrated that Penciclovir potentially binds to the catalytic center of the SARS-CoV 2 RdRp and it has been previously established that it reduces the COVID-19 infection but with a higher half-maximal effective concentration (EC50=95.96 μM) compared to that of Remdesivir (0.77 μM). This indicates that it might require high concentrations of the nucleoside analog to reduce the viral infection; nonetheless the drug provides an option for treating the deadly disease [52]. Finally, experimental evidences have also revealed that Anidulafungin inhibits viral infection with lower half maximal inhibitory concentration (IC50=4.64 µM) compared to that of Remdesivir (11.41 µM) against SARS- CoV-2 and is a potential candidate [53]. Though claims above are promising but the results predicted and discussed in the present investigation are solely based on in silico methods and further experimental evaluation into their molecular mode of action would be required.

5. Expert opinion: COVID-19 has affected humanity like never before. Immediate solutions are needed to treat this devastating disease. Worldwide efforts are ongoing in the area of prevention, defining modes of precautions, diagnostics, as well as drug and vaccine development. While many vaccines are in Phase II or Phase III clinical trials, they will certainly need time to be clinically relevant Hence, the feasible, immediate solution to curb the menace of the pandemic is to repurpose or reposition existing drugs to treat the infection. Approved drugs already in clinical practice, which possess the ability to prevent SARS-CoV-2 entry, replication or repackaging in human host, will allow them to be used directly for the treatment of COVID-19, without the necessity for expensive and time consuming clinical trials. Effort to repurpose drugs to treat COVID-19 began in parallel with the outbreak of the disease. As such, Remdesivir, which is a NUC inhibitor identified to treat Ebola virus, is now successful in treating COVID-19 [54, 55]. We opine that many such repurposed drugs will provide solutions at the speed of the current need. The present investigation supports this claim with Penciclovir, Anidulafungin and Suramin having displayed the ability to bind nsp12 of SARS-CoV-2 in a comparable affinity as Remdesivir, thus allowing these repurposed drugs to be used for COVID-19 treatment. Other studies demonstrating the utility of repurposed drugs in treating COVID-19 has emerged of late [14-19]. However, many of these findings are outcome of computational study and there is an urgent need to subject the putative drugs to in vitro, ex vivo and in vivo evaluations to validate their anti-COVID-19 efficacies and estimate their dose requirements. Separate set of experts should perform such experiments simultaneously and in tandem to expedite the process of drug validation. The assembly line should be in action with immediate effect for an emerging solution. It is Page 11 expected that in the next few months to a year such efforts will borne fruits. In addition, efforts should immediately be focused on synthesis protocols to optimize rapid and cost effective production of the drugs to be useful for the humanity at large, especially when economies are in shambles to support the escalating health cost. It is imperative that efforts in various laboratories should also be geared towards screening many other existing approved drugs against not only nsp12 but also against all plausible drug targets (proteins) of SARS-COV-2 to outwit the virus spread. All existing resources should be geared towards this end.

6. Acknowledgements: SKD is thankful to Professor Eddy Arnold, Rutgers University, USA, for a post-doctoral associateship. MS and SB acknowledge the Department of Biotechnology (DBT), Government of India for their project and research fellowships, respectively. CD appreciates the financial support from Lady Tata Memorial Trust in the form of a research fellowship. ASR is thankful to the Government of Odisha for the Biju Patnaik research fellowship. HKD acknowledges Indian National Science Academy for Emeritus Fellowship. SK appreciates financial support over the years from DBT, University Grants Commission (UGC), Defence Research and Development Organization (DRDO) and Department of Science and Technology (DST), Government of India. SK is thankful to the University of Delhi for support in all forms.

7. References: 1. Bogoch, I.I., A. Watts, A. Thomas-Bachli, et al., Pneumonia of unknown aetiology in Wuhan, China: potential for international spread via commercial air travel. J Travel Med, 2020. 27(2). 2. Hui, D.S., E.I. Azhar, T.A. Madani, et al., The continuing 2019-nCoV epidemic threat of novel coronaviruses to global health - The latest 2019 novel coronavirus outbreak in Wuhan, China. Int J Infect Dis, 2020. 91: p. 264-266. 3. Morens, D.M., P. Daszak, and J.K. Taubenberger, Escaping Pandora's Box - Another Novel Coronavirus. N Engl J Med, 2020. 382: p. 1293–1295 4. Kirchdoerfer, R.N. and A.B. Ward, Structure of the SARS-CoV nsp12 polymerase bound to nsp7 and nsp8 co-factors. Nat Commun, 2019. 10(1): p. 2342. 5. te Velthuis, A.J., J.J. Arnold, C.E. Cameron, et al., The RNA polymerase activity of SARS-coronavirus nsp12 is primer dependent. Nucleic Acids Res, 2009. 38(1): p. 203- 14. 6. Mirza, M.U. and M. Froeyen, Structural Elucidation of SARS-CoV-2 Vital Proteins: Computational Methods Reveal Potential Drug Candidates Against Main Protease, Nsp12 RNA-dependent RNA Polymerase and Nsp13 Helicase. Preprints, 2020. 7. Elfiky, A.A., W.M. Elshemey, and W.A. Gawad, 2ʹ-Methylguanosine Prodrug (IDX- 184), Phosphoramidate Prodrug (Sofosbuvir), Diisobutyryl Prodrug (R7128) Are Better Than Their Parent Nucleotides and Ribavirin in Hepatitis C Virus Inhibition: A Molecular Modeling Study. Journal of Computational and Theoretical Nanoscience, 2014. 12: p. 376–386. 8. Elfiky, A.A. and W.M. Elshemey, Molecular dynamics simulation revealed binding of nucleotide inhibitors to ZIKV polymerase over 444 nanoseconds. J Med Virol, 2018. 90(1): p. 13-18. Page 12

9. Elfiky, A.A. and A.M. Ismail, Molecular modeling and docking revealed superiority of IDX-184 as HCV polymerase inhibitor. Future Virology, 2017. 12(7): p. 339–347 10. Elfiky, A.A. and A.M. Ismail, Molecular docking revealed the binding of nucleotide/side inhibitors to Zika viral polymerase solved structures. SAR QSAR Environ Res, 2018. 29(5): p. 409-418. 11. Elfiky, A.A., S.M. Mahdy, and W.M. Elshemey, Quantitative structure-activity relationship and molecular docking revealed a potency of anti-hepatitis C virus drugs against human corona viruses. J Med Virol, 2017. 89(6): p. 1040-1047. 12. Ganesan, A. and K. Barakat, Solubility: A speed‒breaker on the drug discovery highway. MOJ Bioequivalence & Bioavailability 2017. 3(3): p. 56-58. 13. Canrong, W., L. Yang, Y. Yueying, et al., Analysis of therapeutic targets for SARS-CoV- 2 and discovery of potential drugs by computational methods. Acta Pharmaceutica Sinica B, 2020. 14. Narayanan, N. and D.T. Nair, Vitamin B12 May Inhibit RNA-Dependent-RNA Polymerase Activity of nsp12 from the SARS-CoV-2 Virus. Preprints, 2020. 15. Ahmad, M., A. Dwivedy, R. Mariadasse, et al., Prediction of small molecule inhibitors targeting the novel 1coronavirus (SARS-CoV-2) RNA-dependent RNA polymerase2. OSF Preprints, 2020. 16. Patra, A. and N.S. Bhavesh, Virtual screening and molecular dynamics simulation suggest Valproic acid Co-A could bind to SARS-CoV2 RNA depended RNA polymerase. Preprints, 2020. 17. Deng, C.X., The global battle against SARS-CoV-2 and COVID-19. Int J Biol Sci. , 2020. 16(10): p. 1676–1677. 18. Ju, J., X. Li, S. Kumar, et al., Nucleotide Analogues as Inhibitors of SARS-CoV Polymerase bioRxiv 2020. 19. Elfiky, A.A., Ribavirin, Remdesivir, Sofosbuvir, Galidesivir, and Tenofovir against SARS- CoV-2 RNA dependent RNA polymerase (RdRp): A molecular docking study. Life Sci, 2020. 20. Gao, Y., L. Yan, Y. Huang, et al., Structure of RNA-dependent RNA polymerase from 2019-nCoV, a major antiviral drug target. bioRxiv, 2020. 21. Corpet, F., Multiple sequence alignment with hierarchical clustering. Nucl. Acids Res., 1988. 16(22): p. 10881-10890. 22. Robert, X. and P. Gouet, Deciphering key features in protein structures with the new ENDscript server. Nucl. Acids Res., 2014. 42(W1): p. W320-W324. 23. Waterhouse, A., M. Bertoni, S. Bienert, et al., SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res, 2018. 46(W1): p. W296-W303. 24. Laskowski, R.A., M.W. MacArthur, D.S. Moss, et al., PROCHECK - a program to check the stereochemical quality of protein structures. . J. App. Cryst., 1993. 26: p. 283-291. 25. Colovos, C. and T.O. Yeates, Verification of protein structures: patterns of nonbonded atomic interactions. . Protein Science, 1993. 2(9): p. 1511-1519. 26. Bowie, J.U., R. Lüthy, and D. Eisenberg, A Method to Identify Protein Sequences That Fold Into a Known Three-Dimensional Structure. Science, 1991. 253(5016): p. 164- 170. 27. Schrodinger, L., The Pymol Molecular Graphics System, Version 2.0. 2015. 28. Pettersen, E.F., T.D. Goddard, C.C. Huang, et al., UCSF Chimera—a visualization system for exploratory research and analysis. Journal of computational chemistry, 2004. 25(13): p. 1605-12. Page 13

29. Grosdidier, A., V. Zoete, and O. Michielin, SwissDock, a protein-small molecule docking web service based on EADock DSS. Nucleic Acids Res., 2011. 39(Web Server issue): p. W270–W277. 30. Schrodinger, L., Schrodinger software suite. New York: Schrödinger, LLC. 2011. 670. 31. Hinsen, K., The molecular modeling toolkit: a new approach to molecular simulations. . Journal of Computational Chemistry, 2000. 21(2): p. 79-85. 32. Robertson, M.J., J. Tirado-Rives, and W.L. Jorgensen, Improved peptide and protein torsional energetics with the OPLS-AA force field. . Journal of chemical theory and computation, 2015. 11(7): p. 3499-509. 33. Shivakumar, D., E. Harder, W. Damm, et al., Improving the prediction of absolute solvation free energies using the next generation OPLS force field. . Journal of chemical theory and computation, 2012. 8(8): p. 2553-8. 34. Guvench, O. and A.D. MacKerell, Comparison of protein force fields for molecular dynamics simulations. . Molecular modeling of proteins, 2008. 443: p. 63-88. 35. Wang, J., W. Wang, P.A. Kollman, et al., Automatic atom type and bond type perception in molecular mechanical calculations. . Journal of molecular graphics and modelling, 2006. 25(2): p. 247-60. 36. Darden, T., D. York, and L. Pedersen, Particle mesh Ewald:An N⋅ log (N) method for Ewald sums in large systems. . The Journal of chemical physics, 1993. 98(12): p. 10089-92. 37. Boulanger, E., L. Huang, A.D. MacKerell, et al., Improved Lennard-Jones Parameters for Accurate Molecular Dynamics Simulations. . Biophysical Journal, 2016. 110(3): p. 646a. 38. Mackowiak, S., S. Pieprzyk, A. Branka, et al., A Nosé-Hoover Thermostat Adapted to a Slab Geometry. . CMST, 2017. 23(3): p. 211-218. 39. Bereau, T., Multi-timestep integrator for the modified Andersen barostat. . Physics Procedia, 2015. 68: p. 7-15. 40. Venkataraman, S., B. Prasad, and R. Selvarajan, RNA Dependent RNA Polymerases: Insights from Structure, Function and Evolution. Viruses, 2018. 10(2): p. 76. 41. Ruan, Z., C. Liu, Y. Guo, et al., Potential inhibitors targeting RNA-dependent RNA polymerase activity (NSP12) of SARS-CoV-2. Preprints, 2020. 42. Warren, T.K., R. Jordan, M.K. Lo, et al., Therapeutic efficacy of the small molecule gs- 5734 against Ebola virus in rhesus monkeys. Nature, 2016. 531: p. 381-385. 43. Selisko, B., N. Papageorgiou, F. Ferron, et al., Structural and Functional Basis of the Fidelity of Nucleotide Selection by Flavivirus RNA-Dependent RNA Polymerases. Viruses, 2018. 10(2): p. 59. 44. National Center for Biotechnology Information. PubChem Database. Penciclovir, CID=135398748. 45. Bacon, T.H., M.J. Levin, J.J. Leary, et al., Herpes simplex virus resistance to acyclovir and penciclovir after two decades of antiviral therapy. Clin Microbiol Rev., 2003. 16(1): p. 114–128. 46. Kathryn, S. and T. Gumbo, Anidulafungin in the treatment of invasive fungal infections. Therapeutics and clinical risk management, 2008. 4(1): p. 71-8. 47. Stein, C.A., Suramin: A Novel Antineoplastic Agent with Multiple Potential Mechanisms of Action Cancer Res 1993. 53(10): p. 2239-2248. Page 14

48. Li, H., H. Li, H. Qu, et al., Suramin inhibits cell proliferation in ovarian and cervical cancer by downregulating heparanase expression. Cancer Cell International, 2015. 15(1): p. 52. 49. Cheson, B.D., A.M. Levine, and D. Mildvan, Suramin Therapy in AIDS and Related Disorders: Report of the US Suramin Working Group JAMA, 1987. 258(10): p. 1347- 1351. 50. Croci, R., M. Pezzullo, D. Tarantino, et al., Structural bases of norovirus RNA dependent RNA polymerase inhibition by novel suramin-related compounds. PLoS One, 2014. 9(3): p. e91765. 51. da Silva, C.S.B., M. Thaler, A. Tas, et al., Suramin inhibits SARS-CoV-2 infection in cell culture by interfering with early steps of the replication cycle. Antimicrobial Agents and Chemotherapy, 2020. 52. Wang, M., R. Cao, L. Zhang, et al., Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro. Cell research, 2020. 30(3): p. 269-271. 53. Jeon, S., M. Ko, J. Lee, et al., Identification of antiviral drug candidates against SARS- CoV-2 from FDA-approved drugs. Antimicrobial Agents and Chemotherapy, 2020. 54. Beigel, J.H., K.M. Tomashek, L.E. Dodd, et al., Remdesivir for the treatment of Covid- 19—preliminary report. New England Journal of Medicine, 2020. 55. Grein, J., N. Ohmagari, D. Shin, et al., Compassionate use of remdesivir for patients with severe Covid-19. New England Journal of Medicine, 2020. 382(24): p. 2327- 2336.

Table 1: List of various classes of FDA-approved, WHO essential medicine list 2019, investigational and other widely used prescription drugs screened for repositioning against nsp12 of SARS-CoV-2 along with their binding energies.

No. Drug Structure Estimated ΔG (kcal/mol) Cryo-EM Model Structure CONTROLS 1.a Remdesivir -9.23 -10.45 (Positive Control)

FDA approved for COVID- 19 under emergency use authorization (EAU)

1.b Remdesivir -14.82 -15.79 Triphosphate (Rem-TP)

2. Tetrandrine -5.86 -5.58 (Negative Control)

Non-FDA Under COVID-19 trial Phase 4

ANTIVIRALS 3. Ribavirin -8.76 -10.04

NDA#020903

4. Dasabuvir -8.64 -8.64

NDA#206619

5. Sofosbuvir -9.01 -8.98

NDA#208341

6. Ombitasvir -10.10 -9.17

NDA#206619

7. Bictegravir -7.85 -8.52

NDA#210251

8. Raltegravir -9.15 -8.09

NDA#203045

9. Elivitegravir -7.94 -8.13

NDA#207561

10. Dolutegravir -7.57 -7.61

NDA#211994

11. Penciclovir -11.71 -13.63 NDA#020629

Penciclovir Triphosphate (Pen-Tp)

12. Tilorone -7.68 -8.00

(Widely used for broad antiviral activity)

13. Galidesivir -7.98 -9.16

Broad spectrum antiviral Under COVID-19 Trials

14. Grazoprevir -8.30 -9.29

NDA#208261

15. Paritaprevir -9.30 -9.73

NDA#206619

16. Simeprevir -9.67 -9.56

NDA#205123

17. Saquinavir -8.89 -9.70

NDA#021785

18. Delaviridine -8.49 -8.58

NDA#020705

19. Bevirimat -6.78 -8.49

Investigational antiviral in Phase 2b

20. Darunavir -9.16 -9.03

NDA#202118

21. Calanolide A -7.10 -7.64

Investigational anti-HIV drug

ANTIMALARIALS 22. Mefloquine -7.57 -7.58

ANDA#076392

23. Amodiaquine -7.62 -7.88

NDA#006441

24. Quinine -7.25 -7.53

NDA#021799

25. Artemisinin -6.74 -7.19

(WHO essential medicine list 2019)

26. Hydroxy -7.73 -7.94 chloroquine

NDA#009768

ANTIMICROBIALS 27. Salinomycin -8.98 -9.04

Approved*

28. Hexachlorophene -6.96 -7.70

NDA#017433

29. Suramin -12.09 -13.18

WHO Essential Medicine list 2019

ANTIHELMINTH 30. Niclosamide -7.52 -7.07

NDA#018669

31. Oxyclozanide -7.28 -7.60

Approved*

ANTIFUNGAL 32. Anidulafungin -11.44 -10.57

NDA#021632

33. Itraconazole -9.17 -9.09

ANDA#076104

ANTICANCER 34. Bazedoxifene -8.04 -8.78

NDA#022247

35. Gilteritinib -8.29 -9.21

NDA#211349

ANTIARRYTHYMIC AGENTS 36. Proscillaridin -8.10 -8.55

Approved*

37. Digoxin -9.13 -9.66

NDA#021648

38. Digitoxin -9.47 -10.27

NDA#009436 (Discontinued)

39. Ouabain -8.30 -8.27

Approved*

RECEPTOR AGONIST 40. Eltrombopag -10.10 -9.73

NDA#022291

41. OHPC -7.43 -7.65 (Hydroxyprogesterone caproate)

NDA#021945

ANTIASTHMATIC 42. Ciclesonide -8.18 -8.44

NDA#021658

ENZYME INHIBITOR 43. Abemaciclib -8.11 -9.26

NDA#208716

RECEPTOR POTENTIATOR 44. Ivacaftor -9.17 -8.30

NDA#203188

ANTIDIARRHEAL 45. Loperamide -8.63 -8.92

NDA#019487

*Approved by FDA for: Veterinary use, over the counter and as prescription based drug. Table 2: Energy parameters after 1000 ps MD simulation of the best docked protein-inhibitor complexes of top three leads for SARS-CoV-2

Sl. SARS-CoV-2 Energy parameters after 1000 ps/1 ns production MD No. nsp12 complex names (cal/mol)

Electrostatic van der Waals Kinetic Potential Total energy interactions interactions energy energy 1 SARS-CoV-2 -314691.1 53560.3 -453243.3 -393147.9 -846391.2 nsp12- penciclovir 2 SARS-CoV-2 -324826.3 52662.2 -462110.7 -383563.5 -845874.2 nsp12- anidulafungin 3 SARS-CoV-2 -316611.1 51923.9 -451243.1 -396814.8 -848057.9 nsp12- suramin

Table 3: Repositioned drugs as potential inhibitors of nsp12/RNA-dependent RNA polymerase (RdRp) of SARS-CoV-2 and related information Drug Mechanism Of Physical Properties Dosing Regimen Side effects Current Action application Penciclovir • (C10H15N5O3) Nucleoside A. Physical State: White to • Administra Negligible to Approved for analog inhibits pale yellow solid tion: Oral (tablet normal cells. Herpes Simplex viral B. Solubility: form) and cream virus (Herpes replication** i) Penciclovir • Regimen: Some trials liabilis) • 0.2 mg/mL in Herpes liabilis suggest methanol, 1.3 mg/mL in • Famciclovi headache, propylene glycol, and 1.7 r (Prodrug) oral/pharyng eal edema, mg/mL in water • Initial episode: 250 mg PO and ii) Famciclovir (Prodrug) parosmia.** Penciclovir-triphosphate q8hr for 7-10 days • Soluble in acetone * and methanol. • Suppressiv • Hygroscopic below e therapy: 250 mg 85% relative humidity PO q12hr for 12 • Soluble (>25% months w/v) in water (25°C)*** • Recurrent episodes: 1500 mg

PO once. Interacting Residues PDB-ID 6M71-THR556, • Penciclovir ARG553 and CYS622, ALA547 • 0.1%(10m g/g)-apply q2hr while awake for 4 days*** Anidulafungin (C58H73N7O17) Repress fungal A. Physical State: White to Administration: Not Approved for cell wall off-white powder Intravenous (3.33 determined candedemia formation by mg/mL) extensively. inhibiting 1→3- Some β-D glucan Regimen: reported synthase B. Solubility: Eraxis • Day 1: 200 hepatic and ** formulation of mg IV infusion, hypersensitiv Anidulafungin is soluble in THEN e adverse 0.9% saline *** events.*** • Day 2 and thereafter: 100

mg/day IV Interacting Residues Continued for 14 PDB-ID 6M71-ASP760, days after last SER759, THR556, ALA580, positive culture*** ASP623, LEU758

Suramin Mechanism A. Physical state: A white to Administration: Nausea, Approved for the (C51H40N6O23S6) unknown. off-white powder Intravenous vomiting and treatment of Trypanocidal discomfort African activity; inhibits B. Solubility: Soluble in *** trypanosomiasis

enzymes involved DMSO, water *** Regimen: and river with the Trypanosomiasis blindness oxidation of reduced NADH** • 100-200 mg (test dose) IV, then 1 g IV on days

1, 3, 7, 14, 21.***

Interacting Residues PDB-ID 6M71- ASN496, ASN497, LYS577, ALA688, ASP623, PRO620, THR556, ALA554, ASP164

*Hydrogen bonding side chains are in bold font

**Information taken from Drugbank (https://www.drugbank.ca/)

***Information taken from Medscape (https://www.medscape.com/)

Figure 1: Amino acid sequence alignment of nsp12 protein of SARS-CoV (6NUR_SARSCov) with the sequence used for homology model (COVID19_NSP12model) and that from the cryo- EM (6M71) structure of nsp12 of SARS-CoV-2 (6M71_SARSCov2) using ESPript 3.022. The conserved as well as the non-conserved residues are highlighted in red and white, respectively, and the RdRp motif sequences (Motif A-G) are highlighted with coloured boxes.40

B A ) )

C) RMSD 0.48 Å

Figure 2: Structure of the nsp12 protein of the novel coronavirus (nCoV-19) depicted as cartoon topology. A) Cryo-EM structure of nsp12-SARS-CoV-2 (PDB ID:6M71), B) Homology model of nsp12, C) Superimposed cryo-EM (pink) and modelled structure (cyan) of nsp12, with a root mean square deviation (RMSD) of 0.48 Å.

Figure 3: The structural correlation and disposition of the seven motifs (A-G) that dictate RdRp activity in A) Cryo-EM structure of nsp12 of SARS-CoV-2 (PDB ID:6M71) and B) Homology model of nsp12.

Model Structure

Figure 4: Bar graph representing the binding energies (kcal/mol) of the selected FDA-approved, WHO-EML 2019, investigational and widely prescribed drugs when docked to nsp12 of SARS-CoV-2 (model (green) and cryo-EM structure (yellow)) estimated by SwissDock. Remdesivir-TP and Tetrandrine, FDA investigational antivirals, were used as positive and negative controls, respectively, for the docking experiments in silico.

A) B)

Figure 5: Top three leads docked to the RdRp region of nsp12 (SARS-CoV-2, PDB ID: 6M71) A) Penciclovir, B) Anidulafungin, and C) Suramin. The motifs that the drugs interact with are indicated in the figures.

Figure 6: Interaction maps of A) Penciclovir, B) Anidulafungin and C) Suramin depicting the amino acid side chains (magenta) in nsp12 (PDB ID:6M71) that interact with the drugs (blue). Solid green lines depict hydrogen bonds.

A )

Figure 7: Top three leads docked to the nsp7-nsp8-nsp12 complex (SARS-CoV-2, PDB ID: 6M71), where gray chain represents nsp12, yellow and blue chains represent two copies of nsp8 and green chain represents one copy of nsp7. A) Penciclovir, B) Anidulafungin, and C) Suramin bound to nsp7-nsp8-nsp12 complex of SARS-CoV-2, PDB ID: 6M71.

A) 0.7 0.6 0.5 0.4 0.3 6M71 + Suramin 6M71 + Penciclovir RMSD RMSD (Å) 0.2 6M71 + Anidulafungin 0.1 0 0 200 400 600 800 1000 Time (ps)

B) 0.7 0.695 0.69 0.685

0.68 6M71 + Suramin 6M71 + Penciclovir RMSD RMSD (Å) 0.675 6M71 + Anidulafungin 0.67 0.665 0.66 0 200 400 600 800 1000 Time (ps)

Figure 8: The root-mean-square deviations (RMSD) of the backbone atoms of SARS-CoV-2 nsp12 structure (PDB ID: 6M71) relative to their energy minimized complex structures as a function of time for Suramin (green), Penciclovir (red) and Anidulafungin (blue). A) RMSD view from initial MD trajectory of 0 ps to final trajectory at 1000 ps showing almost similar but a low total change in RMSD of just ~0.7 Å for complexes of each of the top three leads against SARS-CoV-2 nsp12. The low RMSD changes also indicates that there were no overall structural changes in the protein but rather local changes around the bound compound/s took place in the current MD simulation time frame. B) Enlarged view of RMSD changes indicates that Penciclovir attained the structural stability earlier than Anidulafungin and Suramin at ~390 ps, ~400 ps and ~450 ps, respectively. The overall changes are minimal indicating that the docked structures mimic the highest energy minimized conformations.

A) B)

Figure 9: Interaction maps of the three leads with SARS-CoV-2 nsp12 (PDB ID:6M71) after 1000 ps MD simulation depicting the amino acid residues in nsp12 that interact with the drugs. Magenta lines depict hydrogen bonds. A) Penciclovir, B) Anidulafungin and C) Suramin interacting with SARS-CoV-2 nsp12 (PDB ID:6M71). As compared to the docking results, while Suramin has preserved its hydrogen bonds with THR556, it forms one with ARG553 as well; Penciclovir forms additional hydrogen bonds with ASP623 and ARG553 and Anidulafungin on the contrary stabilized its binding mainly by electrostatic interactions with ASP202, ASN204, ASN06, ASP209, GLN108 and ARG109 due to the presence of oppositely charged atoms in the structure of the drug.