Proposing a new anti-Covid-19 agent by using molecular docking and dynamics simulationsAbstract
The Covid-19 pandemic, caused by SARS-CoV-2, was responsible for millions of deaths worldwide. The main protease (Mpro) of SARS-CoV-2 is considered one of the important drug targets for the treatment of Covid-19. Recent studies have shown that anisotine should be a potent Mpro inhibitor. In the present work, four oxoquinoline derivatives are proposed as candidates for Mpro inhibitors. The main functional group of these derivatives shows similarity to anisotine, and they are active against the HSV-1, as well as the latter. Molecular docking studies evaluated whether these compounds could be active against Mpro of SARS-CoV-2. Structural modifications were proposed on the oxoquinoline derivative which formed a more stable complex with Mpro and this proposal formed an even more stable complex besides exhibiting improvements in the toxicological profile. Molecular dynamics simulations indicated that derivatives proposed promote greater stabilization by complexing with Mpro than anisotine.
Keywords:
SARS-CoV-2; Covid-19; Mpro; Oxoquinoline derivatives; Molecular docking; Molecular dynamics
INTRODUCTION
The pandemic due to Covid-19 was initially declared on March 11, 2020 by the World Health Organization. The first cases associated with the disease came from Wuhan, in China, in December 2019. The infected showed symptoms of pneumonia, caused by a new strain of coronavirus not yet identified in humans, which was later named SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) (Mslati et al., 2021).
Before the Covid-19 outbreak, other Coronaviruses had already caused epidemics, causing from common flu symptoms to serious SARS infections (which cause severe acute respiratory syndrome) and MERS-COV (which cause Middle East respiratory syndrome) (Ibrahim et al., 2020). Along with them, the SARS-CoV-2 virus belongs to the betacoronavirus genus, from the Coronaviridae family, having particular potential to promote more serious infections in humans. It was found that there is great structural and genomic similarity between SARS-CoV and SARS-CoV-2, mainly in the receptor-binding domain of the Spike protein, indicating that Coronaviruses have a high ability for genetic recombination, which in addition to some other factors, contribute to high transmissibility by increasing the efficiency of viral entry (Li et al., 2021).
At the end of 2023, more than 770 million cases of Covid-19 were confirmed worldwide and almost 7 million deaths were reported according to World Health Organization (2024). Measures were taken to contain the spread of the virus, such as social distancing and isolation, in addition to the use of masks. Mass vaccination has been the main tool to contain the spread of the disease, given that the drugs approved for treatment have side effects or are ineffective (Padasas et al., 2023).
While vaccines generally act specifically against the infectious agent, medications have a broader spectrum of effectiveness, making them quickly accessible for use against new infections. Therefore, the repositioning of pre-existing drugs that were developed to combat a series of other diseases can be considered a good strategy to alleviate the severity of pandemic conditions, offering advantages such as lower production time and cost (Vincenzi, Mercurio, Leone, 2023; Zarros, 2022).
Throughout the pandemic, several research groups estimated that existing drugs could be used to treat Covid-19 through computational studies using a series of different virtual chemical libraries that contain compounds approved by the Food and Drug Administration (FDA) or in the clinical testing phase, and other active substances (Bakowski et al., 2021; Franco, Maia, Barreiro, 2021).
To date, the FDA (2024) have already approved some repositioned drugs, such as Remdesivir, Baricitinib, Molnupiravir and Paxlovid. The efficacy of most of them against the SARS-CoV-2 virus was initially analyzed by in silico studies, which offered important tools to predict and investigate the mechanism of these compounds against Covid-19.
Although there are already approved antivirals for the treatment of Covid-19, due to the virus’s volutionary capacity, it is important to evaluate new drug candidates. In this context, the inhibitory capacity of several phytochemicals against the virus was studied (Jose et al., 2022; Saha et al., 2021; Swain Panda, Luyten, 2021). Anisotine (1, Figure 1) is one of the quinazolinone alkaloids derived from a medicinal plant native from Asia-named Justicia adhatoda-which is known for its anti-tuberculosis activity, in addition to exhibiting antiviral activity against the influenza virus and the Herpes Simplex type 1 (HSV-1) (Ghosh et al., 2021). This molecule was evaluated in several Molecular Modeling studies in relation to some target proteins of the SARS-CoV-2 virus, such as RNA-dependent RNA polymerase - RdRp, main protease - Mpro and serine protease TMPRS22, presenting promising results (Ghosh et al., 2021; Gowrishankar et al., 2021; Kar et al., 2020; Kar et al., 2022). Ghosh and collaborators (2021) performed a docking and molecular dynamics analysis between Mpro and the components of Justicia Adhatoda and two drugs that had already been proposed for the treatment of Covid-19, lopinavir and darunavir. They concluded that anisotine, which interacted with the two residues of the catalytic dyad (Cys145-His41), forms the most stable Mpro-ligand complex and should be tested in vitro (Ghosh et al., 2021).
Quinolone derivatives are associated with antibacterial, antifungal, antitumor, antiviral and antiparasitic activity (Dube, Legoabe, Beteck, 2022). Yoneda and co-workers (2014) synthesized three compounds containing the 1,4-dihydro-4-oxoquinoline nucleus (quinolone derivatives 2-4, Figure 2) active against Human Immunodeficiency Virus type 1 (HIV-1). They were evaluated via molecular docking as candidates for HSV-1 DNA polymerase inhibitors, and compound 2, must present good activity against HSV-1. Based on the docking results, structural modifications were proposed with the aim of increasing interactions with active site and such modifications resulted in the structure of compound 5 (Figure 2), which was then proposed as a new candidate for a dual HSV/HIV inhibitor (Yoneda et al., 2014).
It is possible to observe a structural similarity between the quinolone fragment of compounds 2-5 and the quinazolinone fragment of anisotine (1). Furthermore, the quinolone derivatives in Figure 2 were shown to be potential candidates for anti-HSV-1 agents, while anisotine is active against the same virus. As data from the literature indicate that anisotine should be a potent inhibitor of the Mpro of SARS-CoV-2, in this work, molecular docking was carried out between compounds 2-5 and the main protease of SARS-CoV-2, in order to verify whether quinolone derivatives would present interactions similar to those presented by anisotine, and if more stable complexes would be formed with the enzyme, being an option for inhibiting the new Coronavirus. In this way, it was possible to evaluate the proposal of these compounds as potential candidates for SARS-CoV-2 inhibitors and propose structural modifications that led to a compound with greater affinity for the enzyme, seeking the formation of more interactions with key residues of the active site and the improvement of pharmacokinetic properties.
MATERIAL AND METHODS
As every computational approach, the one used here have some limitations but are the starting point to structure-based drug design.
The effectiveness of a docking program hinges on two primary factors: the scoring function and the search method, both of which navigate the conformational space. This space encompasses all potential conformations and orientations of proteins and ligands. Exhaustive exploration of this space is impractical because there are current computational limitations (Chaudhary, Tyagi, 2024). However, molecular docking simulations are highly valuable because they can make fundamental contributions to structure-based drug design, especially when combined with molecular dynamics simulations.
Some limitations of molecular dynamics simulations include the high computational cost of simulating large systems and the fact that some receptor conformational changes in molecular recognition occur on a timescale that exceeds current computational capacities (Ferreira et al., 2015). Despite these limitations molecular dynamics simulations bring the possibility of obtaining models of molecules very close to biological reality, as it directly includes characteristics such as molecular flexibility (through the temporal variation of properties) and temperature (through the dynamics of atoms) (Verli, Barreiro, 2005).
Molecular Docking
The crystallographic structure of Mpro available in the Protein Data Bank (Berman, 2000) with the code 6LU7 was used in the studies, in accordance with the methodology used by Ghosh and co-workers (2021) in the study of anisotine, used as positive control in this work. This structure contains the peptide-type covalent inhibitor known as N3 complexed with the enzyme, and also solvation water molecules.
The standard protonation states of amino acid residues were confirmed by the PDB2PQR server (Jurrus et al., 2018) and calculated at pH 7.4, which is the pH at which maximum Mpro activity is observed (Tan et al., 2005). Water molecules were removed and polar hydrogens were added based on the predicted protonation states. In Autodock Tools (Morris et al., 2009) Gasteiger charges were computed and the file was converted to PDBQT format.
The structures of the studied ligands were constructed and optimized with DFT B3LYP/6-31G** method, using the Spartan’10 software (Wavefunction Inc., 2011), after conformational analysis with MMFF force field.
Molecular docking was run with the genetic algorithm and local search optimization of Autodock Vina program (Trott, Olson, 2010). Initially, the co-crystallized ligand (N3) was redocked with different exhaustivity values (8, 16, 32 and 64) to obtain the RMSD and choose the protocol to be used.
A grid box of 0.375 Å spacing was positioned to encompass the largest binding site of the enzyme, where the inhibitor (N3) was complexed with 6LU7 (Ghosh et al., 2021). The residues of the catalytic dyad, Cys145 and His41, are located in this site. All rotatable torsions of the ligands were freed during docking, and conformational changes were evaluated using Vina’s scoring function in several runs starting from random conformations. The lowest binding energy ligand conformation was chosen to evaluate the enzyme-ligand complex, where Pymol (Schrödinger, 2020) was used to visualize them and Discovery Studio Visualizer (Dassault Systèmes, 2019) was used to analyze interactions with binding site residues.
The OSIRIS® Property Explorer server (Sander, 2001) was used to evaluate toxicity risks, pharmacokinetic properties and fragment-based druglikeness score of the studied compounds.
Molecular Dynamics Simulations
Since the conformation with the minimum energy score in a docking result may be assigned as the preferable binding mode (Neves et al., 2021), the lowest energy complexes of the best candidate among the quinolone compounds, the compound originating from structural modifications and anisotine were subjected to molecular dynamics simulations using the GROMACS software (version 2021.5) (Abraham et al., 2023) with GROMOS96 54a7 force field (Schmid et al., 2011). The ligand topologies were obtained using the Automated Topology Builder (ATB) server (Canzar et al., 2013; Malde et al., 2011) and the Mpro topology was generated using the ‘pdb2gmx’ script. Ligands were assembled to the processed Mpro structure and then cubic unit cells were prepared with the complex centered at a distance of 0.9 nm from the edges. The complex was then solvated using the SPC model of water and Na+ ions were also added to make the system electrically neutral (Kar et al., 2020). Energy minimization of the Mpro/ligand complexes was achieved in several steps using the steepest descent algorithm, to relax the system in a protected volume. The Particle Mesh Ewald (PME) method was used to treat the long-range electrostatic potential with a cutoff of 1.2 nm, and the same value was applied for the short-range van der Waals cutoff. The equilibration run of each system was carried out in two main steps, by positional restriction of the complexes. First, with the NVT ensemble using the v-rescale algorithm, the systems were progressively heated to a temperature of 300 K on a time scale of 10 ns and, to set the pressure at 1 bar, a 10 ns long NPT run was then performed. The equilibrated systems were then subjected to unrestrained production MD simulations of 100 ns each, following the procedure described by Ghosh and co-workers (2021), with a time step of 1 fs, maintaining the same pressure (1 bar) and temperature (300 K). The RMSD and root mean square fluctuation (RMSF) for each system were calculated with the GROMACS ‘gmx rms’ and ‘gmx rmsf’ modules from the MD trajectories (Ghosh et al., 2021).
RESULTS AND DISCUSSION
Molecular Docking
Initially, N3 was redocked with 6LU7 enzymes by using different exhaustivities. The results on Table I indicates that the lowest energy position found with the implementation of exhaustivity equal to 32 leads to an overlap closest to that of the crystallized ligand, presenting a deviation of only 2.19 Å. This result is in agreement with data obtained in a similar study carried out with other docking program, whose RMSD value of complexes where the non-covalent ligand is located within the Mpro active site was 3.43 Å for the pose with best score (Bassani et al., 2022). Furthermore, the score found was very close to that presented in the work of Ahamad and collaborators (2021), which was-8.3 kcal.mol-1, where they used the same crystallographic structure and also carried out docking with Autodock Vina. Based on this analysis, molecular docking tests were carried out using a protocol where exhaustivity was set to 32, since this configuration proved to be more capable of reproducing experimental data, validating the docking strategy.
Affinity energy and RMSD values referring to docking simulations of the Mpro/N3 complex at different exhaustivities
The docking results (Table II) indicates that the ligands (2-5) interact with at least one of the components of the Cys145-His41 catalytic dyad, which is an interesting point for inhibition of Mpro activity.
Interactions and affinity energy values obtained with the Autodock Vina program for Mpro-ligand complexes
Anisotine, which was chosen as a positive control, performed 5 interactions involving residues His41, Cys145, Met165, Glu166 and Arg188. Previous work that evaluated the inhibition potential of anisotine against Mpro found a similar pattern of interactions, identifying all of the previously mentioned residues (Ghosh et al., 2021) or at least two of them, His41 and Met165 (Kar et al., 2020). Besides the three hydrophobic interactions with His41, Cys145 and Met165, and an electrostatic one with Glu166, a strong hydrogen bond was found at a distance of 2.43 Å from the carbonyl oxygen atom of Arg188 (the strong hydrogen bonds are at a distance of around 2.2 to 2.5 Å and moderate ones are between 2.5 to 3.2 Å) (Dannenberg, 1998).
Ligand 2, which received the lowest energy score, performed a conventional hydrogen bond with His164 and a non-conventional one (where the π orbitals of one of the quinolone rings participate in the bond) with Glu166. The remaining interactions were predominantly hydrophobic.
Ligand 3 interacted with residues Phe140, Met165 and Gln189 via hydrogen bonding at distances of approximately 2 Å between atoms, participating with the amine of the quinolone group and the oxygen and carbon atoms of the acylhydrazone group, respectively. Two electrostatic interactions were formed between the oxygens of the Glu166 side chain and the π electrons of the quinolone rings, and a π-sulfur type bond.
Ligand 4 received the best energy score (-8.0 kcal. mol-1), creating an even more stable complex than Mpro-anisotine. Furthermore, this compound made the largest number of conventional hydrogen bonds (between O and H atoms), with residues His41, Tyr54 and Gly143 at distances of 2-3 Å. Gly143 is an important residue that is located in the cleft between domains I and II of Mpro together with the His41-Cys145 catalytic dyad, helping the enzymatic activity (Kar et al., 2020). There was also a hydrogen bond involving the α-carbon atom of Arg188. Some of the oxygen atoms present in the structure of compound 4 acted as hydrogen acceptors, mainly those from the nitro group, which coordinated three of these bonds.
Compound 5 was the only one that did not form hydrogen bonds, showing only hydrophobic interactions with residues His41, Cys145, Met165, Thr190 and Ala191. However, its predicted docking energy was very close to that of Mpro-anisotine complex.
All quinolone derivatives 2-5 exhibited more intermolecular interactions with the active site of Mpro than anisotine, many of them occurring with the same residues. The π orbitals of the aromatic rings of these compounds contributed significantly to the hydrophobic and electrostatic interactions, while O and N atoms of the structures acted as acceptors and hydrogen donors. However, due to the greater affinity between compound 4 and the enzyme, which was evidenced by the higher affinity energy and the greater number of strong interactions with key residues of the active site, this was considered the most promising inhibitor candidate among the studied molecules.
In the search for potential anti-Covid-19 candidates, a structural modification on compound 4 was also evaluated considering: (i) the compound resulting from the modifications must contain the main groups of compound 4 responsible for interactions with Mpro; (ii) it must have functional groups similar to those of anisotine; (iii) to justify the proposal, the new molecule must present better binding affinity than compound 4 and carry out a greater or equal number of short-distance interactions within the active site; (iv) it must have pharmacokinetic properties better than or comparable to those of anisotine and compound 4. Taking these considerations into account, compound 6 (Figure 3) was proposed. It has a spatial arrangement close to that of anisotine due to the 5-membered ring that was built in the same position as that found in its structure, with the aim of fitting in a similar way to the Mpro active site. Furthermore, the nitro group was kept as a substituent not only because of the stabilizing interactions performed by this group in compound 4, but also because it seems to contribute to activity enhancement against SARS-CoV-2 main protease (Niranjana et al., 2023; Jia et al., 2023).
Structural modification and planned chemical structure of compound 6, arising from the structural modification in compound 4.
Lipinski’s rule-of-five (Lipinski, 2004) proposes that poor absorption or cell permeability of a drug occurs when its chemical structure fulfils more than one of the following criteria: the molecular weight (MW) is greater than 500 Daltons; the calculated LogP is greater than 5; the number of hydrogen bond donors (HBD) are more than 5; and the number of hydrogen bond acceptors (HBA) are more than 10. According to the Veber’s rule (Veber et al., 2002), for good oral availability, the polar surface area (PSA) value must be less than or equal to 140 A2. The physicochemical properties calculated for the studied compounds fit these parameters (Table III).
Pharmacokinetic properties of quinolone derivative 4, proposed compound 6 and anisotine calculated with the OSIRIS server
Regarding toxicity risks, according to the OSIRIS Property Explorer server it was found that compound 4 presented high mutagenic and tumorigenic risks because of the nitro-furan fragment connected to the hydrazone. It was also possible to verify that anisotine presented a medium tumorigenic risk associated with the fragment that contains a secondary amine linked to a benzene ring. Furthermore, druglikeness value for compound 6 was 0.81, which is higher than that of anisotine. The toxicity risks of this molecule only pointed to a medium mutagenic risk, representing an improvement in relation to compound 4, which had presented high mutagenic and tumorigenic risks.
Molecular docking of compound 6 was carried out with Mpro, obtaining an energy score of-8.2 kcal.mol-1. Table IV lists the interactions carried out by this compound and also by 4 and anisotine, comparing which were gained and which were lost in relation to them. Figure 4 shows the types of interactions displayed and the lowest energy conformation assumed by compound 6 and by anisotine within the Mpro active site.
Docking result for anisotine (red) and proposed compound 6 (green) in the active site of MPro.
It is possible to verify that the complex formed with the proposed compound was even more stable than that of compound 4 (Table IV), which suggests that compound 6 is, like anisotine, a potential candidate for an Mpro inhibitor.
The presence of hydrogen bonds with four different residues (Gly143, Ser144, Cys145 and Glu166) is observed, at distances in the range of 1.9 - 3.3 Ǻ, which can be characterized as strong to moderate. In addition to the hydrogen bond, there was a hydrophobic interaction with the Cys145 residue, and also a hydrophobic interaction with His41. These interactions with the two residues of the catalytic dyad were also observed for anisotine, and are important for the inhibition of Mpro activity. Another interaction in common with anisotine was that carried out with the residue Met165, which participated in two hydrophobic interactions involving the π orbitals of the aromatic systems of compound 6. Based on this analysis, compound 6 originated more interactions similar to those of anisotine than the quinolone derivatives, and obtained the same interaction profile as anisotine, except for the Arg188 residue, with the addition of hydrogen bond-type interactions with residues Gly143 and Ser144. Given this, it is possible to suggest that it will be more potent in inhibiting Mpro than compounds 2-5 and anisotine, and with less toxicity effects than the latter and compound 4.
Molecular Dynamics Simulations
The analysis of the trajectories of molecular dynamics simulations for the Mpro structure and its complexes with compounds 4, 6 and anisotine allowed obtaining information about the stability of these structures from the calculation of the RMSD at each 1 ns step of the simulation. The backbone group of the original (input) structures of the dynamics was selected, which comprises the basic unit of all the amino acids that make up the protein (NH - CαH - C’=O) as a reference for the least squares adjustment in relation to the protein-ligand complex after each step of the dynamics. Figure 5 graphically shows the results obtained.
RMSD versus time curves obtained by molecular dynamics with Gromacs software for unbound Mpro (blue) and for its complexes with anisotine (orange), ligands 4 (green) and 6 (magenta).
The unbonded Mpro structure experienced some fluctuations during dynamics. From 2 to 20 ns, the RMSD value grew slightly from 0.28 - 0.37 nm. In periods between 25 and 34 ns, and between 51 and 59 ns, peaks were observed where the maximum RMSD values were located in the curve, above 0.50 nm. The other peaks caused smaller variations, of up to 0.08 nm in relation to the equilibrium values. After 65 ns of dynamics, the deviation value remained in equilibrium at around 0.40 nm until the end of the simulation.
The RMSD curve for the Mpro/anisotine complex showed a certain stability around 0.30 nm until reaching a maximum of 0.50 nm between 76 and 79 ns. After this maximum, the RMSD values dropped slightly until they balanced around 0.36 - 0.37 nm, maintaining in this range after this time.
The Mpro/4 complex reached equilibrium after approximately 20 ns of simulation, remaining around 0.30 nm until 70 ns, at which point its deviation becomes a little higher, but returns to the previous value until the end of the dynamic. As can be seen from the Mpro/4 RMSD graph in Figure 5, this ligand caused the least disturbance in the dynamic conformation of the complex, as the RMSD spectrum showed fewer fluctuations compared to the other ligands.
The deviation of the Mpro/6 complex showed similar behavior to that of ligand 4 up to 70 ns, remaining stable around 0.30 nm after reaching equilibrium. However, after this period of time, an increase in RMSD values was noted with many fluctuations, which lasted until the end of the simulation.
In general, all the ligands analyzed provided extra stability to the Mpro structure, which can be seen by the position of the RMSD curves of the complexes, which were below the unbound Mpro curve. The average RMSD values for unbound Mpro, and for the Mpro/anisotine, Mpro/4 and Mpro/6 complexes were 0.36 nm, 0.32 nm, 0.30 nm and 0.31 nm, respectively. These RMSD values show that once the ligands reach a stable conformation within the active site, their positions do not change much (Dagur et al., 2023). Furthermore, they suggest that compounds 4 and 6 reached stable and reliable dynamic equilibria, as they were very close to that of the Mpro/ anisotine complex, which reinforced the docking results (Ahamad et al., 2021). All average values were very close to those found by Ghosh and collaborators for the Mpro/anisotine, Mpro/darunavir and Mpro/lopinavir complexes, which varied between 0.31 and 0.37 nm (Ghosh et al., 2021), validating the molecular dynamics simulation strategy.
It was also found that ligands 4 and 6, proposed in this work, were able to reduce the RMSD of the protein to a lower level when bound to it, than anisotine.
Root mean square fluctuation (RMSF) analysis of residues of a protein-ligand complex is effective for addressing local changes along a protein chain and deducing its conformational stability (Kar et al., 2020). The RMSF was calculated over the simulation time for each studied system. The results were plotted in Figure 6.
RMSF (nm) profiles obtained by molecular dynamics with the Gromacs software for the unbound Mpro (blue) and for the complexes with the anisotine (orange), ligands 4 (green) and 6 (magenta).
The RMSF profiles in Figure 6 showed that all systems showed high fluctuations in a certain portion of domain I (formed by residues 8 - 100) of the protein. In a very similar way, the Mpro, Mpro/anisotine and Mpro/4 systems had this highest conformational fluctuation of domain I at residues 48 and 49, unlike the Mpro/6 complex, for which it was observed at residues 62 - 64. In domain III (formed by residues 200 - 303) a similarity was also observed in coincident regions, seen by the fluctuations of residues 279 (Mpro) and 277 (anisotine, 4 and 6). It can be observed that the Mpro/4 and Mpro/6 complexes showed smaller fluctuations (mainly in domains I and II, where binding occurs) compared to the Mpro (unbound) and Mpro/anisotine complexes. The average RMSF of the Mpro/4 and Mpro/6 complexes was 0.17 and 0.19 nm, while that of the Mpro (unbound) and Mpro/anisotine complexes was 0.24 and 0.20 nm. respectively. These values were also very close to those found by Ghosh and collaborators for the Mpro/anisotine, Mpro/darunavir and Mpro/lopinavir complexes, which were 0.179 nm for the first and 0.195 nm for the last two (Ghosh et al., 2021).
Given the results obtained with the RMSD and RMSF averages, it is possible estimate that compounds 4 and 6, suggested in this work as potential candidates for Mpro inhibitors, should form more stable complexes with this enzyme than anisotine.
CONCLUSIONS
Molecular docking studies for quinolone derivatives 2 - 5 and anisotine were carried out with SARS-CoV-2 main protease. Evaluating the obtained results, structural modifications were proposed based on the structures of compound 4 and anisotine, searching for a new molecule that presented a better toxicological and pharmacokinetic profile, and also that had greater affinity for Mpro. Thus, compound 6 was designed and molecular dynamics simulations were run. According to the obtained trajectories, it was possible to hypothesize that compounds 4 and 6 can form more stable complexes with Mpro than anisotine and, therefore, like the latter, could be tested in vitro to better explore their inhibition potential, and confirm whether they could become good candidates for anti-Covid-19 agents.
ACKNOWLEDGEMENTS
To FAPERJ (Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro) and to PROPPI-UFF (Pró Reitoria de Pesquisa, Pós-graduação e Inovação da Universidade Federal Fluminense).
REFERENCES
- Abraham M, Alekseenko A, Berg C, Blau C, Briand E, Doijade M, et al. GROMACS 2023.1.2023.
- Ahamad S, Kanipakam H, Birla S, Ali MS, Gupta D. Screening Malaria-box compounds to identify potential inhibitors against SARS-CoV-2 Mpro, using molecular docking and dynamics simulation studies. Eur J Pharmacol. 2021;890:73664-173676.
- Bakowski MA, Beutler N, Wolff KC, Kirkpatrick MG, Chen E, Nguyen TH, et al. Drug repurposing screens identify chemical entities for the development of COVID-19 interventions. Nat Commun. 2021;12:3309-3323.
- Bassani D, Pavan M, Bolcato G, Sturlese M, Moro S. Re-Exploring the Ability of Common Docking Programs to Correctly Reproduce the Binding Modes of Non-Covalent Inhibitors of SARS-CoV-2 Protease Mpro. Pharmaceuticals. 2022;15:180-198.
- Berman HM. The protein data bank. Nucleic Acids Res. 2000;28:235-242.
- Canzar S, El-Kebir M, Pool R, Elbassioni K, Mark AE, Geerke DP, et al. Charge group partitioning in biomolecular simulation. J Comput Biol. 2013;20:188-198.
- Chaudhary M, Tyagi K. A review on molecular docking and it’s application. Int J Adv Res. 2024;12:1141-1153.
- Dagur P, Rakshit G, Sheikh M, Biswas A, Jha P, Al-Khafaji K, et al. Target prediction, computational identification, and network-based pharmacology of most potential phytoconstituent in medicinal leaves of Justicia adhatoda against SARS-CoV-2. J Biomol Struct Dyn. 2023;41:3926-3942.
- Dannenberg JJ. An introduction to hydrogen bonding. J Am Chem Soc. 1998;120:5604-5604.
- Dassault Systèmes BIOVIA, Discovery Studio Visualizer, v20.1, San Diego: Dassault Systèmes, 2019.
- Dube PS, Legoabe LJ, Beteck RM. Quinolone: a versatile therapeutic compound class. Mol Divers. 2022;17:1-26.
- Ferreira LG, dos Santos RN, Oliva G, Andricopulo AD. Molecular docking and structure based drug design strategies. Molecules. 2015;20:13384-13421.
-
Food & Drug Administration. FDA. Coronavirus (COVID-19) | Drugs: Coronavirus (COVID-19) | Drugs: https://www.fda.gov/drugs/emergency-preparedness-drugs/coronavirus-covid-19-drugs Accessed in may, 2024.
» https://www.fda.gov/drugs/emergency-preparedness-drugs/coronavirus-covid-19-drugs - Franco LS, Maia RC, Barreiro EJ. Identification of LASSBio-1945 as an inhibitor of SARS-CoV-2 main protease (M PRO) through in silico screening supported by molecular docking and a fragment-based pharmacophore model. RSC Med Chem. 2021;12:110-119.
- Ghosh R, Chakraborty A, Biswas A, Chowdhuri S. Identification of alkaloids from Justicia adhatoda as potent SARS CoV-2 main protease inhibitors: An in silico perspective. J Mol Struct. 2021;1229:129489-129497.
- Gowrishankar S, Mutyhumanickam S, Kamaladevi A, Karthika C, Jothi R, Boomi P, et al. Promising phytochemicals of traditional Indian herbal steam inhalation therapy to combat COVID-19 - An in silico study. Food Chem Toxicol. 2021;148:111966-111980.
- Ibrahim MAA, Abdelrahman AHM, Hussien TA, Badr EAA, Mohamed TA, El-Seedi HR, et al. In silico drug discovery of major metabolites from spices as SARS-CoV-2 main protease inhibitors. Comput Biol Med. 2020;126:104046-104056.
- Jia Z, Lan X-W, Lu K, Meng X, Jing W-J, Jia S-R, et al. Synthesis, molecular docking, and binding Gibbs free energy calculation of β-nitrostyrene derivatives: Potential inhibitors of SARS-CoV-2 3CL protease. J Mol Struct . 2023;1284:135409.
- Jose S, Gupta M, Sharma U, Quintero-Saumeth J, Dwivedi M. Potential of phytocompounds from Brassica oleracea targeting S2-domain of SARS-CoV-2 spike glycoproteins: Structural and molecular insights. J Mol Struct . 2022;1254:132369-132383.
- Jurrus E, Engel D, Star K, Monson K, Brandi J, Felberg LE, et al. Improvements to the APBS biomolecular solvation software suite. Prot Sci. 2018;27:112-128.
- Kar P, Kumar V, Vellingiri B, Sen A, Jaishee N, Anandraj A, et al. Anisotine and amarogentin as promising inhibitory candidates against SARS-CoV-2 proteins: a computational investigation. J Biomol Struct Dyn . 2020;40:1-11.
- Kar P, Saleh-E-In MM, Jaishee N, Anandraj A, Kormuth E, Vellingiri B, et al. Computational profiling of natural compounds as promising inhibitors against the spike proteins of SARS-CoV-2 wild-type and the variants of concern, viral cell-entry process, and cytokine storm in COVID-19. J Cell Biochem. 2022;123:964-986.
- Li, X, Chang J, Chen S, Wang L, Yau TO, Zhao Q, et al. Genomic Feature Analysis of Betacoronavirus Provides Insights into SARS and COVID-19 Pandemics. Front Microbiol. 202;12:614494.
- Lipinski CA. Lead-and drug-like compounds: The rule-of-five revolution. Drug Discov Today Technol. 2004;1:337-341.
- Malde AK, Zuo L, Breeze M, Stroet M, Poger D, Nair PC, et al. An Automated Force Field Topology Builder (ATB) and Repository: Version 1.0. J Chem Theory Comput. 2011;7:4026-4037.
- Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J Comput Chem. 2009;30:2785-2791.
- Mslati H, Gentile F, Perez C, Cherkasov A. Comprehensive consensus analysis of SARS-CoV-2 drug repurposing campaigns. J Chem Inf Model. 2021;6:3771-3788.
- Neves BJ, Mottin MM, Moreira-Filho JT, Sousa BKP, Mendonça SS, Andrade, CH. Chapter 4-Best Practices for Docking-Based Virtual Screening. Ed. Mohane S. Coumar, Molecular Docking for Computer-Aided Drug Design, Academic Press, 2021, 75-98.
- Niranjana SV, Chethan BS, Mahesha, Chethan Prathap KN, Hema MK, Lokanath NK. In-silico investigation of 4-nitro-N-1H-pyrazol-3-ylbenzamide towards its potential use against SARS-CoV-2: a DFT, molecular docking and molecular dynamics study, J Biomol Struct Dyn . 2023;28:1-21.
- Padasas BT, Espanõ E, Kim S-H, Song Y, Lee C-K, Kim J-K. COVID-19 Therapeutics: An Update on Effective Treatments Against Infection with SARS-CoV-2 Variants. Immune Netw. 2023;23:e13-37.
- Saha P, Bose S, Srivastava AK, Chaudhary AA, Lall R, Prasad S. Jeopardy of COVID-19: Rechecking the Perks of Phytotherapeutic Interventions. Molecules . 2021;26:6783-6807.
-
Sander T. OSIRIS Property Explorer. SwitzerlandIdorsia Pharmaceuticals Ltd, 2001. https://www.organic-chemistry.org/prog/peo/ Accessed in: march, 2024.
» https://www.organic-chemistry.org/prog/peo/ - Schmid N, Eichenberger AP, Choutko A, Riniker S, Winger M, Mark AE, et al. Definition and testing of the GROMOS force-field versions 54A7 and 54B7. Eur Biophys J. 2011;40:843-856.
- SPARTAN’10, Wavefunction Inc. Irvine, CA 92612, U.S.A., 2011.
- Swain SS, Panda SK., Luyten W. Phytochemicals against SARS-CoV as potential drug leads. Biomed J. 2021;44:74-85.
- Tan J, Verschuere KHG, Anand K, Shen, J, Yang M, Xu Y, et al. pH-dependent Conformational Flexibility of the SARS-CoV Main Proteinase (Mpro) Dimer: Molecular Dynamics Simulations and Multiple X-ray Structure Analyses. J Mol Biol. 2005;354:25-40.
- The PyMOL Molecular Graphics System, Version 2.4 Schrödinger, LLC, 2020, [s.d.].
- Trott O, Olson AJ. AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem . 2010;31:455-461.
- Veber DF, Johnson SR, Cheng H-Y, Smith BR, Ward KW, Kopple KD. Molecular Properties That Influence the Oral Bioavailability of Drug Candidates. J Med Chem. 2002;45:2615-2623.
- Verli H, Barreiro E J. Um paradigma da química medicinal: a flexibilidade dos ligantes e receptores. Química Nova. 2005;28: 95-102.
- Vincenzi M, Mercurio FA., Leone M. Looking for SARS-CoV-2 therapeutics through computational approaches. Curr Med Chem. 2023;30:3158-3214.
-
World Health Organization. WHO. Coronavirus disease (COVID-19) pandemic: Coronavirus disease (COVID-19) pandemic: https://www.who.int/emergencies/diseases/novel-coronavirus-2019 Accessed in march, 2024.
» https://www.who.int/emergencies/diseases/novel-coronavirus-2019 - Yoneda JD, Albuquerque MG, Leal KZ, Santos FC, Batalha PN, Brozeguini L, et al. Docking of anti-HIV-1 oxoquinoline-acylhydrazone derivatives as potential HSV-1 DNA polymerase inhibitors. J Mol Struct . 2014;1074:263-270.
- Zarros A. Computational drug screening in COVID-19 drug repurposing research: Encouraging findings and limitations. J Expl Res Pharmacol. 2022;7:3-4.
