Azole-acridine hybrids as potential enzymatic inhibitors of coronavirus-2 main protease and RNA polymerase by molecular modeling strategy

SARS-CoV-2 has been identied as the cause of the current outbreak of coronavirus disease (COVID-19). As part of the efforts to develop potential drugs with promise for clinical use, a molecular docking study on azole (triazole and pyrazole) based molecules on the main protease M pro and RNA polymerase was conducted, as possible inhibitors that could be elected for further experimental bioassays. Autodock has been employed to identify azole derivatives 1-6 preferred conformations in the active site of the enzyme and to estimate their binding anities to the protease and RNA polymerase targets. From the molecular docking strategy, these new azole compounds though nonpeptides in nature display possible inhibition of M pro activity with comparable anities (-4.7 kcal/mol to -6.5 kcal/mol) to the recently reported peptide-like inhibitors such as α-ketoamide inhibitor 13b (-5.0 k/cal/mol). They also exhibit improved binding anities to RNA polymerase (-6.3 to -7.1 kcal/mol) comparable to remdesivir (-6.6 kcal/mol). Based on the observed binding energies, these compounds may possess anti-coronavirus bioactivity through inhibition of the virus main protease as well as RNA polymerase activities in living cells.


Introduction
Coronavirus disease  outbreak is currently a pandemic as declared by WHO in March 2020. 1 It is caused by the new severe acute respiratory syndrome coronavirus-2, SARS-CoV-2. Owing to the novelty of the coronavirus, neither vaccines nor de ned treatments for coronavirus infections are available at the moment which has contributed to increased efforts to control its global spread. There were over 4.4 million globally con rmed cases of COVID-19 as of 16 th May 2020 since the rst case was reported in December 2019 in Wuhan City, China. 2 The successful isolation and genomic sequence of SARS-CoV-2 has been critical in understanding the novel corona virus, identi cation of diagnostic kits and establishment of the key pathways involved in the development and spread of the virus. The recent breakthrough in identi cation of the virus pathway of action including receptors and the viral proteins determinants in transmission has provided a signi cant advancement in the current search for treatment. [3][4][5][6] Main protease (M pro ) has been established as one of the key enzymes involved in the viral replication and maturation of the coronavirus proteome. 4, 7-9 It plays a fundamental role in the processing of two large polyproteins (PP1a and PP1b) which are required for the virus to mature and cause infections. M pro has at least 11 cleavage sites on the polyprotein and a unique cleavage site sequence Leu-Gln↓(Ser,Ala,Gly)) recognition, and is thus a vital molecular target in the development of anti-coronavirus therapy. 4,10 On the other hand, RNA polymerase mediates the replication of the SARS-CoV-2 virus through a complex of non-structural proteins (nsp) for the virus. 6,[11][12] Inhibition of the RNA-dependent RNA polymerase (RdRp) activity will rationally hinder binding of the template and reduce/limit its replication process, and is therefore a fundamental molecular target for antiviral drugs. As part of our contribution to the scienti c efforts to develop novel therapeutic agents against coronavirus (CoV-2), we sought to evaluate new azole based molecules recently developed in our laboratory 13 for their possible anti-coronavirus activity through molecular docking strategy. Herein, we provide a discussion on the bioinformatics of the triazole and pyrazole derivatives containing acridine frameworks as possible inhibitors of SARS-CoV-2 M pro and RNA-dependent RNA-polymerase (RdRp). An insight into the kind of conformations adapted by the new azole hybrids in the active site of M pro and RdRp, binding a nities and key interactions has been provided through computational methods, mainly molecular docking.

Methodology
Ligand and protein preparation For docking studies, the ligands were converted into autodock input le using Open Babel. 14 Ligand preparation involved geometry optimization, energy minimization and assignment of partial charges; a prerequisite le format for autodock input. The crystal structure of the protein of (un)liganded main protease, M pro (ID: 6lu7), main protease M pro co-crystallized with α-ketoamide inhibitor 13b (6Y2F/6Y2G) for binding site identi cation and SARS-CoV-2 RNA-dependent RNA-polymerase, RdRp, (ID: 7BV1; for apoRdRp, and ID: 7BV2; for the Remdesivir liganded-RdRp in complex with a template-primer) were obtained from the protein data bank (https://www.rcsb.org/).. Each of these molecular targets was transformed into docking input les using autodock program.
Docking and analysis: Autodock les for the ligands and the proteins were transferred to the autodock module in PyRx software. [15][16] Docking grid parameters (Table S1) were de ned based on the residues of the binding site/pockets of the reported inhibitors co-crystallized with protease M pro 4, 17 and RdRp (remdesivir) in the crystal structures. 6 Docking studies and analysis of the binding poses/a nities were performed using autodock vina program in PyRx software. 15 The binding conformations of the ligands in the active site of the SARS-CoV-2 M pro and RdRp, and their corresponding protein-ligand interactions were visualized and processed using Discovery Studio 4.118 or PyMol program. 19 Results And Discussions

Chemistry of the compounds
The compounds evaluated in this study include azole-acridine molecular hybrids of triazole and pyrazole containing acridines designed rationally for targeted therapy with nucleic acid as one of the molecular targets. Figure 1 shows the molecular triazole and pyrazole hybrids elected for inhibition of main protease SARS-CoV M pro by molecular docking. The synthesis of the acridine based triazole hybrids has been reported previously with 1,3-dipolar cycloaddition as a key reaction in the triazole heterocycle synthesis. 13 A convergent reaction between 3-phenylpyrazole and acridine in the presence of sodium hydride and 2bromoethylamine hydrobromide salt afforded the ethyl linked acridine based pyrazole hybrid 6 analogous to triazole 2.
Molecular docking studies: To determine the binding a nity, BE, of the azole derivatives To probe the potential of the synthesized compounds for anti-coronavirus activity, a docking strategy was employed to not only determine their enzymatic binding a nities and preferred conformations on the SARS-CoV-2 main protease (M pro ) and RdRp but also to identify the key interactions involved in their binding sites.
Notably, the selected binding site was derived from the liganded crystal structures of SARS-CoV-2 M pro (PDBID: 6lu7) which co-crystallized with peptide-like compounds as candidate inhibitors. 4, 17 Experimentally determined α-ketoamide inhibitor 20 13b was used as a reference inhibitor to corroborate the methodology used in this docking study as well as for comparative assessment of the potential of these new compounds as potential inhibitors of SARS-CoV-2 M pro through their binding a nities. For RdRp, the binding conformations and a nities of the ligands were determined relative to the crystallographic site of remdesivir. 6 In addition, other drugs such as chloroquine (CQ), 21 hydroxychloroquine (HCQ) 22 and remdesivir, 6,23 which are currently being used or under clinical trials were included in the docking study for comparative purposes. Table 1 shows the binding a nities of the most preferred binding poses obtained for each of the compounds evaluated in this study based on the docking scoring function.
It is notable that for acridine containing azole hybrids 1-4 and 6, the presence of acridine moiety orients the molecule in such a way that the 9-amino group forms at least two key H-bonds with thiol and imidazole groups of Cys145 and His 41 amino acid residues, respectively. In a similar fashion, the two amino acids In a similar fashion to the inhibition activity against protease M pro , the azole compounds 1 -6 have exhibited good binding a nities to SARS-CoV-2 RNA-dependent RNA polymerase (RdRp) in the range of -5.6 to -7.1 kcal/mol (Table 1). All the azole-acridine hybrids 1-4 and 6 displayed good a nities for RdRp active site comparable to that of remdesivir (-6.6 kcal/mol) with binding energies below -6.3 kcal/mol. Their binding a nities are interestingly better than that of CQ (-5.4 kcal/mol) and HCQ (-6.1 kcal/mol). Also, azoleacridine hybrids exhibit better a nities for RdRp active site than the non-acridine containing triazole 5 (--5.6 kcal/mol) implying that the acridine moiety plays an important role in the activity of the azole derivatives.
The presence of the phenyl group on the triazole and pyrazole moieties appears to improve the binding a nities of the azole-acridine hybrids (2, 4 and 6) as evidenced by binding energies below -6.6 kcal/mol (Table 1) in comparison azole-acridines 1 (--6.6 kcal/mol) and 3 (--6.3 kcal/mol) with hydroxymethyl and cyclopropyl groups respectively. Besides, the chloro and methoxy groups on the acridine framework contribute to improved binding a nities of the azole-acridine evidenced by favorable binding of azoleacridine 2 and 6 (--7.1 kcal/mol each) compared to 4 (--6.8 kcal/mol). While the acridine framework in each of the azole-acridine hybrids occupy the same pocket in the RdRp, it is interesting to note that the preferred conformation of azole-acridine 1 adapts an opposite orientation to that of the other azole-acridines 2, 3, 4 and 6 ( Figure S1) in the RdRp-ligand complexes. This could be attributed to the polarity of the binding pocket of the triazole moiety in 1 as evidenced by hydrogen bonding of the hydroxyl group with amino acid residues Asp618/760/761 as compared to the hydrophobic binding pocket of the phenyl group of azole-acridines 2, 4, and 6 ( Figures S1 and S3). The observed better binding a nities suggest that the azole-acridine hybrids may disrupt the activities of RNA-dependent RNA-polymerase by forming signi cant local interactions with the active site limiting the replication activity of the viral RNA polymerase. 6 The results from these docking strategies provide preliminary evidence to support the rationale for anticoronavirus assays of these reported azole compounds developed from our laboratory 13 and suggest further study on their in vitro and/or in vivo assays are necessary to validate these models.
The cheminformatics described herein imply that these compounds may hold promise for optimization as lead compounds and possibly in the development of drug candidates for development into anti-coronavirus agents.