Abstract
Purpose: COVID-19, caused by SARS-CoV2 virus is a contagious disease affecting millions of lives throughout the globe. Currently, there are no clinically approved drugs for SARS-CoV2 although some drugs are undergoing clinical trials. The present study investigates the binding property of ivermectin on four important drug targets, spike protein, RNA-dependent RNA polymerase, 3-chymotrypsin- and papain-like proteases of SARS-CoV2. Methods: The 3D structure of ivermectin along with known antiviral drug lopinavir, simeprevir and four nucleotides ATP, GTP, CTP, and UTP were downloaded from PubChem database. Crystal structures of proteins were downloaded from PDB database. PDB files were converted into pdbqt file using AutoDock tools. After proper processing and grid formation, docking was carried out in AutoDock vina. Furthermore, the co-crystallized RNA and its binding interactions with RdRp were studied using various visualization tools including Discovery studio. Results: Docking study showed that ivermectin is the best binding drug compared to lopinavir and simeprevir. The best binding interaction was found to be -9.7kcal/mol with RdRp suggesting potential inhibitor of the protein. Twenty-one amino acid residues of RdRp were found to interact with ivermectin including the catalytic residue Asp760. Furthermore, RNA-RdRp complex revealed that the catalytic active residues Ser759 and Asp760 of RdRp formed strong interactions with RNA chain. Binding of ivermectin in the active site of RdRp make clash with the nucleotides of RNA chain suggesting the possible inhibition of replication. Conclusions: The present study suggests ivermectin as a potential inhibitor of RdRp which may be crucial to combat the SARS-CoV2.
Currently, there is no clinically approved drug for SARS-CoV2. However, recent publications reveal that several medications such as ivermectin, hydroxychloroquine, and remdesivir are being used to reduce the virus load and improve disease symptoms (Caly et al. 2020; McKee et al. 2020; Wang et al. 2020). Researchers around the world are working diligently to develop effective drugs based on the pathogenicity and molecular details of the SARS CoV-2. Four key targets proteins such as spike protein (S-protein), virus mail protease (3Clpro), papain-like protease (Plpro), and RNA-dependent RNA polymerases (RdRp) are being explored by researchers around the world to develop medicines against COVID-19. The SARS-CoV S-protein is an important drug target which is a quaternary protein composed of two subunits - S1 subunit that contains a receptor-binding domain (RBD) which engages with the host cell receptor angiotensin-converting enzyme-2 and the S2 subunit that mediates fusion between the viral and host cell membranes (Du et al. 2009; Zhou et al. 2019). 3Clpro and Plpro are two important virus proteases that catalyse the processing of polyproteins pp1a and pp1ab leading to the formation of RdRp and other important non-structural proteins (nsp) (Anand et al. 2003; Chen et al., 2020). The RNA-dependent RNA polymerase, also known as nsp12 is the central component of CoV replication and transcription machinery, and therefore, it appears to be a primary target for the antiviral drug (Xu et al. 2003). Given the pivotal roles of S-protein, 3Clpro, PL-pro, and RdRp in the infection, replication, and generation of viral particle, these proteins are widely regarded as an important and attractive target for the rational design of anti-SARS-CoV2 drugs.
Ligand Selection and Preparation
The 3D structure of ivermectin (Drugbank ID: DB00602) along with two antiviral drugs simeprevir (DB06290), and lopinavir (DB01601) from Drugbank database (https://www.drugbank.ca/). Four nucleotides ATP (PubChem CID: 5957), GTP (CID: 135398633), CTP (CID: 6176), and UTP (CID: 6133) were downoloaded from PubChem database. SDF files were converted into pdb file using PyMol software. All the PDB files of the ligands were processed and finally converted into .pdbqt file using AutoDock tool (Trott and Olson 2010).
Collection and Preparation of Proteins
Three-dimensional structures of S-protein (RBD) (PDB ID: 7BZ5), 3Clpro (PDB ID: 6M2N), Plpro (PDB ID: 7JN2), and RdRp (PDB ID: 6XQB) were downloaded from PDB database. The protein structures were cleaned by removing the water and other hetatms. Polar hydrogen atoms and Kollman charges were added to the structure and finally converted into .pdbqt file format for docking using AutoDock Tools.
Molecular Docking
After the ligand drugs and the target enzymes were prepared docking was carried out in AutoDock Vina (Trott and Olson 2010). The grid box parameters for docking of all the four proteins are presented in table 1. The docking algorithm was carried out by keeping the default exhaustiveness at 8. After docking, the pose scoring the lowest binding energy (kcal/mol) was selected and visualize in Discovery Studio.
Results
The present study investigates the binding affinity of ivermectin, a broad-spectrum antiparasitic FDA approved drug against four key enzymes of SARS-CoV2, the spike protein, 3Clpro, Plpro, and RdRp enzymes. Figure 1 showed the binding energies of ivermectin with all the four protein. Ivermectin showed the best and strongest binding affinity with RdRp (-9.7kcal/mol) followed by S-protein, Plpro, and 3Clpro. The reference antiviral drug simeprevir also showed strong affinity to all the proteins. Highest binding affinity was observed in Plpro. Lopinavir showed the weakest affinity among all the three drugs. To compare the affinity of nucleotides to the active site of RdRp enzymes, all the four nucleotides ATP, GTP, CTP, and UTP were also docked with the protein. All the four nucleotides showed weaker affinity to the enzyme compared to ivermectin. ATP showed the highest binding energy -8.1 kcal/mol followed by UTP (-8.0 kcal/mol), CTP (-7.8 kcal/mol), and lowest in GTP (-7.6 kcal/mol). Fig. 2 showed the binding interactions of the ivermectin with the RdRp of SARS-CoV2 and Ramachandran plot of the amino acid residues. A total of 21 amino acid residues (Asn497, Arg555, Thr556, Val557, Leu576, Lys577, Ala580, Ile589, Gly590, Cys622, Asp623, Ser682, Gly683, Asp684, Ala685, Thr687, Ala688, Tyr689, Asn691, Leu758, and Ser759) were found to be interacting with the ivermectin. The ligand-binding site in the RdRp protein and the schematic view of the complex is shown in Fig. 2a. Five conventional H-bonds were seen between the ligand-protein complexes while majority of the binding interactions were found to be van der Waal’s interaction (Fig. 2b,d). The H-bond donor and acceptor and the hydrophobicity property of the ligand surrounding amino acid residues were shown in Fig. 2e and 2f. Most of the surrounding amino acid residues showed hydrophilicity while eight residues showed hydrophobic property. Ramachandran plot study revealed that 11 amino acid residues (all glycine) of RdRp apoprotein were found to be distributed outside the allowed region of the plot. The various amino acid residues interacting with the replicating RNA in the active site of RdRp enzyme is presented in Fig. 3. The co-crystallized structure of RNA-RdRp complex (Pdb ID: 6xqb) has been dissected to visualize the amino acids attaching to the dsRNA chain. A total of 41 amino acid residues were found to form the surrounding surface of the dsRNA (Fig. 3c). Out of 41 amino acid residues, 23 amino acids form interactions with the template RNA strand (3’ – 5’ strand, Fig. 3d) while 20 residues interact with the newly replicating strand (5’- 3’ strand, Fig. e) of RNA. Two amino acids, Thr687 and Glu857 formed interactions with both the stands of dsRNA. Fig. 3f showed the H-bonding amino acid residues with the dsRNA strand. Out of 41 interacting residues, only 15 residues formed hydrogen bonding with the RNA chain. Ten amino acid residues formed eleven H-bonds with the template strand (3’-5’ strand) while five amino acids form seven H-bonds with the complementary new strand of RNA. Asn497 of template strand form 2-H-bonds with the uracil (U2) nucleotide while Arg836 and Ser759 formed 2H-bonds each with cytosine (C7) and adenine (A9) nucleotides of new RNA strand (Fig. 3f). It is also observed that out of 21 amino acid residues of RdRp that are interacting with the ivermectin, 16 residues also made interactions with the dsRNA. The interaction of ivermectin with active site amino acid and RNA strands is shown in Fig. 4. The RNA-ligand interactions and surrounding amino acids were shown in Fig. 4a and 4b. The co-crystallized dsRNA structure of RdRp and the best docking pose of ivermectin when fitted into the binding site five nucleotides – G1, U2, G3, and G4 from template strand (3’ – 5’ strand) and adenine-9 from newly synthesized RNA strand made unfavourable interactions with the ivermectin (Fig. 4c). Fig. 4d showed that 6 active site amino acid residues (out of 15) that were making H-bonds with dsRNA were also found to make interactions with the ligand.
Credited to Bio Science