Effectivity of repurposed drugs against SARS-CoV-2 infections, A hope for COVID 19: inhibitor modelling studies by docking and molecular dynamics

Abstract

In the present study, we have done a comparative study on the efficacy of some currently used repurposed drugs: Oseltamivir (O), Favipiravir (F) and Hydroxychloroquine (H) in individual and in their combinational mode against CoV-2 infections. The ADME analysis has helped us to identify the inhibitory possibility of the tested drugs towards receptor 3CL^pro protein of SARS-CoV-2. Various thermodynamical parameters obtained from Molecular Docking, Molecular dynamics (MD) and MMPBSA simulations like binding affinity, potential energy (E_pot), RMSD, RMSF, SASA energy, interaction energies, Gibbs free energy (ΔG_bind) etc. also helped us to verify the effectivity of mentioned drugs against CoV-2 protease.

SARS-CoV-2; COVID-19; 3CL^pro; Favipiravir; Hydroxychloroquine; Oseltamivir; RMSD; RMSF.

Graphical abstract

1. Introduction

In the 21^st century over the last three years, world is in the pandemic situation created by novel coronavirus. Sometime in December 2019, the virus first detected in China [1]. SARS-CoV-2, a subcategory of the novel coronavirus has played a key role in this catastrophic public health emergency throughout the world [2]. As per WHO, on 5^th December 2022, cumulative cases of COVID-19 disease are 640, 395, 351 and cumulative deaths are 6,618,579 [3]. CoV-2 infection can be easily spread by respiratory droplets from symptomatic, asymptomatic, and pre-symptomatic patients through surface, contact and aerosol transmission [4, 5, 6]. Community transmission and cluster formations are observed to be important factors for the rapid expansion of this disease. A lack of immunity in the majority of the population has already created a high incidence of life-threatening conditions in each and every corner of the world. Owing to these conditions there is an urgent need for a therapeutic solution and vaccine development with maximum efficacy against the disease as well as extensive research on its biology.

Only COVID-19 vaccines can provide strong resistance against death, hospitalization and serious illness. As no one is safe till everyone is safe, a mass vaccination programme has started early from December 2020. As on December 2022, around 10 vaccines are available in market for human use. There are more 31 vaccines are under different stages of human trial phases [7, 8, 9]. However even with massive efforts, till date none of the tested vaccine could significantly impact the course of COVID-19 pandemic. Also, it is a fact that some people are still getting ill from COVID-19 after vaccination. With the scientific idea obtained from recent outbreaks of Ebola virus disease, SARS-CoV and MERS-CoV, the researchers and medical practitioners have created a synchronized and fast-tracked answer to COVID-19, through the research into potential therapeutic pharmaceutical treatments. De novo development of therapeutics to bring a new drug into clinical practice has to pass through a laborious five step process. The steps include discovery & development, preclinical research, clinical research, FDA review and FDA post market drug safety monitoring. Usually, the whole process used to take a decade to act as effective option against any disease as drug invention and distribution may be refined endlessly even after the drug's final approval. So the whole procedure becomes time-consuming and expensive with a high possibility of failure. So as an alternative, another novel approach to identify new therapeutic option is drug repurposing [10, 11]. Drug repositioning is the main activity of drug repurposing which involves the investigation of new therapeutic indications of existing drugs. To combat against CoV-2 infections, many antiviral FDA approved drugs already used for known viral diseases like influenza, Ebola, HIV, hepatitis, Zika, Alzheimer's etc are being repositioning for the treatment of COVID-19 disease [12, 13]. Remdesivir, Sofosbuvir, Ribavirin etc are approved antiviral drugs for hepatitis C virus [14, 15, 16], are being repositioning for the treatment of COVID positive patients. Favipiravir and Oseltamivir, as originally approved drugs Ebola, Lassa and influenza viruses (A and B), now are being tested for the treatment of COVID-19. Antiviral drugs generally target viral proteins or cellular proteins where they do not destroy their target pathogens rather inhibit their development by any of the below mentioned actions. Drug can resist the viral protease from entering the host cell, secondly it can oppose the replication of viral protein or it can act as minimizing agent against the damages that can be created by infecting virus [14]. Not only antiviral drugs, many antibiotic drugs like Ivermectin and Doxycycline, antimalarial drugs Chloroquine and Hydroxychloroquine, also are being repositioning for COVID-19 treatment [17, 18]. Therapeutic switching is now considered as new effective approach to enhance the stereoisomers and metabolites of existing drugs. All of the above-mentioned individual antiviral and antibiotic drugs have been tested for the prevention and effective management of COVID-19 infection. Till now the available clinical data validated that most of the repurposed drugs tested for COVID-19 treatment, have shown lower or moderate effectiveness with many adverse side effects for the patient conditions like diabetes, hypertension, cardiac problems [19, 20]. Many repurposed combination drugs have also repositioned for the treatment of HIV, other coronaviruses and MERS-CoV infection [21]. Lopinavir + Ritonavir and Mycophenolic acid (MPA) + IFN-® combination are popularly used combination medicine for MERS-CoV infection [22, 23]. However, it is a big question that which repositioning showing better clinical benefits for hospitalized COVID-19 patients: individual or combination therapy [24]? It is a topic of argument that whether a repurposed drugs have shown any benefit in hospitalized COVID-19 patients or not, still they have been proved to be useful in slowing the development of illness, preventing severe disease and less hospitalization. Dose adjustments, acute/chronic toxicity, choice of appropriate delivery system and route of administration to deliver the drugs are some major challenges for the application of repurposed drugs which need a rigorous study. Also, there is an urgent need for assessment of repurposed drugs in terms of their binding efficiency with target virus, repurposing dynamics, limitations and delivery approaches to identify some effective antiviral drugs which will positively impact the COVID 19 therapy. Approved repurposed drugs have saved millions of human lives over the last decades, still it is a challenge to identify effective, low toxic, well tolerated drugs that can enhance patient recovery limit. This challenge has created the motivation for performing the present study. For the current study, the effectivity of some popularly used repurposed direct targeted and host targeted antiviral drugs in their individual and combinational modes have been checked with the help of some in silico methods like DFT, ADME, Molecular Docking, MD simulation and MMPBSA simulations against CoV-2. Our target individual drugs are Oseltamivir, Favipiravir, Hydroxychloroquine.

For SARS-CoV-2 drug development, the intriguing drug target is broad -spectrum class of viral RdRP inhibitors. The most promising RdRp inhibitors are nucleoside analogues (NAs). Favipiravir is one of popularly used RdRP inhibitor or NA [25]. NAs are incorporated into the nascent viral RNA by the RdRp. RdRP inhibitor is mistakenly considered as purine nucleotide by receptor viral protease. As a result of that target protein makes the RdRP: receptor complex structure in the form of exonuclease (ExoN) [26]. The main problem of NA incorporation by RdRp into nascent RNA is that they many causes disturbance in RNA synthesis. NAs can non-obligate the RNA synthesis process by perturbing the product RNA structure enough to stop further synthesis by polymerase which may definitely reduce their antiviral effects. NAs also can obligate the RNA chain extension as they are unable to be recognized as regular Watson-Crick nucleobases. As a result of that production of non-viable genomes occurs by a process ‘lethal mutagenesis. Favipiravir belongs to this class of antivirals which can act actively against variety of viruses (Ebola, Lassa and influenza viruses). Though for CoVs, certain NAs insertion into nascent RNA have shown their reduced antiviral effects, Favipiravir is widely and successfully used to cure COVID-19 patients with a good efficacy [27, 28, 29]. In India and Chaina, Favipiravir is being used as an approved drug for COVID 19 treatment [30]. Another popular antiparasitic drug showing anti SARS-CoV 2 activity is Hydroxychloroquine. It is an antimalarial drug which also used for the treatment many other inflammatory diseases like rheumatoid arthritis [31, 32], systemic lupus erythematosus [33]. The utility of Hydroxychloroquine for COVID-19 treatment has been described in several research papers [32, 34], but majority of studies reported no benefit with Hydroxychloroquine treatment for hospitalized patients. Moreover, the concern has been raised about the cardiac toxicity effects of Hydroxychloroquine. As a result, use of Hydroxychloroquine as proposed drug against COVID treatment has been terminated in many countries but later on emergency use authorization have been again revived by FDA [35]. Another commonly used antiviruses in the regimen is Oseltamivir. Oseltamivir is a commonly used drug for prophylaxis of infection due to influenza viruses A and B [36]. Oseltamivir is a most accepted drug for MERS-CoV and SARSCoV-2 therapy in Korea and China [37]. It is an analog to sialic acid which acts to inhibit the neuraminidase enzyme for viral replication. A large number of reported data suggested that for the hospitalized patients with several respiratory tract infections administration of Oseltamivir has reduced their of length of stay in hospitals [38].

2. Materials and methods

2.1. Identification of target protein structure of SARS-CoV-2 and its preparation

The structures of newly invented CoV-2 virus and already known CoV virus is very similar [39, 40]. Till now there are more than 160 reported compounds (co-crystallized with SARS-CoV-2 Pro/CL^pro) proteins have been identified. Inhibitors of the SARS-CoV (2003) M^pro can act through a irreversible inactivation mechanism where the inhibitor associates with the M^pro to form enzyme-inhibitor complex through stable covalent bonding [41]. Among many available SARS-CoV-2 M^pro/CL^pro reported [42, 43] structures, 6LU7 is used as main protease in CoV enzyme. 6LU7 plays a crucial role in mediating viral replication and transcription, making it an better drug target for this virus. We have used 6LU7, a 3CL^pro proteases as main target protein. The 3D structure of the 6LU7 was retrieved from the Protein Data Bank website (https://www.rcsb.org) [44] shown in Table 1 and used as a receptor. To prepare the protein structure for simulation ready, the existing water molecules were removed and then polar hydrogens were added in protein structure by using AutoDock and MGL tools. Also, the inbuilt ligand was removed from the protein structure by Discovery studio 2020 [45, 46, 47]. The output protein structure was saved in PDB format.

Table 1.

Structures of receptor protein (6LU7), and targeted drugs: Oseltamivir (O), Favipiravir (F) and Hydroxychloroquine (H).

Compound Name	Structure
Protease:6LU7
Oseltamivir (O) (C16H28N2O4)
Favipiravir (F) (C5H4FN3O2)
Hydroxychloroquine(H) (C18H26ClN3O)
Red color-oxygen atom, blue-nitrogen atom, green – chlorine atom, light-blue-fluorine atom

2.2. Procedure of ligand drug molecules preparations

For the screening of O, F and H, the SWISS ADME (https://www.swissadme.ch) and ADMET (https://vnnadmet.bhsai.org/) software's were used to verify Drug-likeness rules like Lipinski's rule of five (Ro5) or ‘‘a rule of thumb, Veber's rule, MDDR-like rule, Egan rule, Ghose filter, Muegge rule etc., [48, 49].

For Drug preparation, the ligands in ‘SDF’ format were obtained directly from the PubChem (National Library of Medicine) (https://pubchem.ncbi.nlm.nih.gov/) and converted to ‘PDB’ format with the help of Autodock tools [46]. All ligands have been used at their optimized structures for further simulation studies. All molecular structures of the ligands and appo protein were optimized by using density functional theory (DFT) with the basis set 6.311G (d,p) [50] using the Gaussian 09 program [51]. The optimized structures were visualized with the help of Gauss View 5 molecular visualization program [52].

2.3. Molecular docking

By using Autodock Vina and Discovery studio visualizer [45, 46], the docking-based studies and analysis of the recommended inhibitors against protease of SARS-CoV-2 have been performed. The parameters for configuration file were as follows: total no. of binding modes was taken as 9, exhaustiveness as 8 and maximum energy difference was considered as 3 kcal/mol. Among all of the observed 9 poses, the best pose was chosen on the basis of maximum non bonded interaction, minimum binding affinity (kcal/mol), dreiding energy, dipole moment and also minimum inhibition constant. For inhibition constant (K_i), following equation was used

$${{\mathit{K}}_{\mathit{i}}=\mathit{e}}^{\frac{\Delta \mathit{G}}{\mathit{R}\mathit{T}}}$$ (1)

where, binding affinity is G, universal constant is R and T temperature (300K). Lowest value of K_i validates the strong interaction between ligand and receptor protein 6LU7. For combinational mode of ligands sequential docking were performed.

2.4. Molecular dynamic (MD) and MMPBSA

For calculating thermodynamics parameters of the ligand: protein complex structure through MD simulation, the LINUX based platform ‘‘GROMACS 5.1 Packageˮ with GROMOS43A2 force field was used [53, 54]. For topology creation of protein: ligand complex structure or appo protein structures, TIP3P water model was used with adding 4Na⁺ ions to maintain the neutrality of the structure. Simulation has been performed starting with steepest descent algorithm for achieving the optimized structure with minimum potential energy (E_pot) under equilibrium condition. The algorithm has a cut off up to 239.006 kcal/mol for minimizing the steric clashes which includes two phases each having 500,000 steps. The first phase includes constant particles no., volume and temperature (NVT) and the second phase includes, constant pressure and temperature (NPT). On the optimized lower energy structure MD simulation has been run ranges from 1 ns-100 ns and various thermodynamical parameter have been derived within the mentioned time scale. Some important thermodynamic parameters are root mean square deviation (RMSD), root mean square fluctuation (RMSF), E_pot, radius of gyration (R_g), inter-molecular H-bonds, solvent accessible surface area (SASA) and various non-bonded interaction energies for protein:ligand complex structures. Short range Coulomb interaction (Coul-SR) and short range Lennard–Jones interaction (SR-LJ) were used to compute nonbonded interaction between protein and ligand. Different terms of energies can be expressed in additive forms as:

$${\mathit{E}}_{\mathit{t}\mathit{o}\mathit{t}\mathit{a}\mathit{l}}={\mathit{E}}_{\mathit{b}\mathit{o}\mathit{n}\mathit{d}\mathit{e}\mathit{d}}+{\mathit{E}}_{\mathit{n}\mathit{o}\mathit{n}\mathit{b}\mathit{o}\mathit{n}\mathit{d}\mathit{e}\mathit{d}}$$ (2)

$${\mathit{E}}_{\mathit{b}\mathit{o}\mathit{n}\mathit{d}\mathit{e}\mathit{d}}={\mathit{E}}_{\mathit{b}\mathit{o}\mathit{n}\mathit{d}}+{\mathit{E}}_{\mathit{a}\mathit{n}\mathit{g}\mathit{l}\mathit{e}}+{\mathit{E}}_{\mathit{d}\mathit{i}\mathit{h}\mathit{e}\mathit{d}\mathit{r}\mathit{a}\mathit{l}}$$ (3)

$$E_{nonbonded}=E_{hydrogenbond}+E_{electrostatic}+E_{vanderwaals}$$ (4)

$${\mathit{E}}_{\mathit{e}\mathit{l}\mathit{e}\mathit{c}\mathit{t}\mathit{r}\mathit{o}\mathit{s}\mathit{t}\mathit{a}\mathit{t}\mathit{i}\mathit{c}}={\mathit{E}}_{\mathit{c}\mathit{o}\mathit{u}\mathit{l}\mathit{o}\mathit{m}\mathit{b}\mathit{i}\mathit{c}}+{\mathit{E}}_{\mathit{l}\mathit{e}\mathit{n}\mathit{a}\mathit{r}\mathit{d}\mathit{J}\mathit{o}\mathit{n}\mathit{e}\mathit{s}}$$ (5)

To find the binding affinities of inhibitors towards receptor protein we have applied the MM/PBSA method [55]. To calculate the interaction free energies (ΔG_bind) for the protein:ligand complex structure, the MMPBSA (Molecular Mechanics Poisson-Boltzmann Surface Area) method [56] sourced from the APBS and GROMACS packages have been used within 0 ns–100 ns. In MM/PBSA, method the van der Waals energy components (E_vdw) and electrostatic energy components (E_{electrostatic}) between inhibitor and receptor are used to determine the stability/binding affinity for all complex structures. The same process is applied for single and multiple ligand system. The ΔE_MM was calculated by using molecular mechanics (MM) force field with polar part of the solvation free energy calculation by linearized Poisson-Boltzmann model and apolar part by using surface area approach. In aqueous environment, the binding free energy of the ligand: receptor complex can be computed by following equations.

$${\Delta \mathit{G}}_{\mathit{b}\mathit{i}\mathit{n}\mathit{d},\mathit{a}\mathit{q}\mathit{u}}=\Delta \mathit{H}-\mathit{T}\Delta \mathit{S}\approx {\Delta \mathit{E}}_{\mathit{M}\mathit{M}}+{\Delta \mathit{G}}_{\mathit{b}\mathit{i}\mathit{n}\mathit{d},\mathit{s}\mathit{o}\mathit{l}\mathit{v}}-\mathit{T}\Delta \mathit{S}$$ (6)

$${\Delta E}_{MM}={\Delta E}_{covalent}+{\Delta E}_{electrostatic}+{\Delta E}_{vanderwaals}$$ (7)

$${\Delta E}_{covalent}={\Delta E}_{bond}+{\Delta E}_{angle}+{\Delta E}_{torsion}$$ (8)

$${\Delta G}_{bind,solv}={\Delta G}_{polar}+{\Delta G}_{nonpolar}$$ (9)

where, −TΔS, Δ and Δ, are the molecular conformational energy due to binding and solvation free energy and mechanical energy changes in gas phase, respectively.

2.5. Computational details

MD simulations and corresponding energy calculations have been computed in a single system using HP Intel Core i5 - 1035G1 CPU and 8 GB of RAM with Intel UHD Graphics and a 512 GB SSD.

3. Results and discussion

3.1. Screening and analysis of drug likeness properties of targeted drugs

All the targeted repurposed drugs: F, H, O we have considered for current study follow the Lipinski's rule of five (Ro5) and veber rule (Table 2). All of them also satisfy all bioavailability conditions [49]. All the targeted drugs satisfy all required conditions for Physiochemical Properties, Fraction CSp³, Water Solubility and Medicinal Chemical properties. Results from ADMET analysis (https://vnnadmet.bhsai.org/) shows that none of the tested drug has any cyto-toxicity effect and maximum suggested dose for O, F and H were obtained as 165mg/day, 170mg/day and 478mg/day.

Table 2.

Physiochemical, Drug-likeness, water solubility properties, Medicinal chemistry and toxicity of Oseltamivir, Favipiravir and Hydroxychloroquine.

Name of Ligand	Oseltamivir	Favipiravir	Hydroxychloroquine
Physiochemical Properties
Molecular Formula	C16H28N2O4	C5H4FN3O2	C18H26ClN3O
Molecular Weight	312.40 g/mol	157.10 g/mol	335.87 g/mol
Hydrogen Bond Donor Count	2	2	2
Hydrogen Bond Acceptor Count	5	4	3
Topological Polar Surface Area	90.65 Å2	88.84 Å2	48.39 Å2
Fraction CSP3	0.70	0	0.50
Water Solubility
Log S (SILICOS-IT)	-2.47	-1.42	-6.35
Class	soluble	Soluble	Poorly soluble
Solubility	1.05e+00 mg/ml; 3.37e-03 mol/l	6.04e+00 mg/ml; 3.85e-02 mol/l	1.50e-04 mg/ml; 4.46e-07 mol/l
Drug Likeness
Lipinski Rule	Yes; 0 violation	Yes; 0 violation	Yes; 0 violation
Ghose Filter	Yes	No; 4 violations:MW < 160, WLOGP < -0.4, MR < 40, #atoms<20	Yes
Veber (GSK) Rule	Yes	Yes	Yes
Egan (phatmacial) Filter	Yes	Yes	Yes
Muegge (Bayer) Filter	Yes	No; 1 violation:MW < 200	Yes
Bioavailability (Abbott) Score	0.55	0.55	0.55
Medicinal Chemistry
PAINS (Pan Assey Interference Structures)	0 alert	0 alert	0 alert
Brenk	1 alert: phosphor	0 alert	0 alert
Leadlikeness	No; 1 violation:Rotors>7	No; 1 violation:MW < 250	No; 2 violations:Rotors>7, XLOGP3>3.5
Synthetic accessibility	4.449	2.08	2.82
Toxicity
Cyto-toxicity	No	No	No
MRTD(mg/day)	165	170	478

3.2. Analysis of molecular docking results

As per molecular docking results for O, among 9 pose structures pose 1 can be considered as the stable complex (O:6LU7) structure having a lowest value of binding energy (−6.2 kcal/mol), inhibition constant (2.8 × 10⁻⁵ M) as obtained from Eq. (1) and maximum number of intermolecular hydrogen bonded interactions, dreiding energy (91.62 kcal/mol), dipole moment (4.039 Debye) at room temperature (300 K) (Figure 1a, Table 3) [57]. Similarly in case of F and H also for pose 1 we obtained the stable complex structures (F: 6LU7, H:6LU7) at 300 K (Figures 1b, 1c and Table 3). Based on their crystal structure it was identified that CoV enzyme 6LU7 have some binding pockets within its protease which act as active sites to bind with various ligands [58, 59]. The binding sites are known as S1, S2, S3 and S4 which comprise of some side chain residues and backbone of the base protein. From molecular docking study it is observed that during individual modes of interaction O gets attached with GLY143, GLN189, GLU166 chain residues of active site S4, F gets attached with GLY143, SER144, CYS145 chain residues of S2 binding sites whereas H gets attached with GLU166, MET165, ARG188, THR190 chain residues of active sites of S4 (Table 3).

Figure 1.

Binding energies and active binding sites for drugs a) O, b) F and c) H towards protein 6LU7 by individual docking.

Table 3.

Various parameters like binding affinity, Hydrogen bonded interaction, dipole moment, dreiding energy, Inhibition constant for the docked structure of O, F and H with receptor protein 6LU7.

Ligand	Binding affinity (kcal/mol)	Hydrogen bonded interaction (donor - acceptor, distance in Å) (A-Amino acid, drug:ligand)	Dreiding energy (ligand) (kcal/mol)	Dipole moment of ligand (Debye)	Inhibition Constant (M) Ki = ${\mathit{e}}^{\Delta \mathit{G}/\mathit{K}\mathit{T}}$ from equ (1)
O	-6.2	(A:GLY143:HN -::O:O, 1.92) (A:GLU166:HN -::O:O, 1.99) (:O:H -: A:GLN189:OE1, 2.2)	91.62	4.039	2.8 × 10−5
F	-4.9	(A:GLY143:HN -:F:O, 2.57) (A:SER144:HN -:F:O, 2.15) (A:CYS145:HN-:F:O, 2.20)	55.0	2.460	2.5 × 10−4
H	-4.8	(:H:N -:A:GLU166:O, 3.23) (:H:O -:A:MET165:SD, 3.61) (:H:O -:A:ARG188:O, 3.29) (:H:C -:A:THR190:O, 3.59)	126.57	2.461	3.0 × 10−4

To check the effectivity of combination drugs against CoV 2 target protein, sequential molecular docking mechanism has been used for inspection of the binding affinity of combination drugs (O + F, O + H, F + H, F + H + O) as inhibitors towards receptor 6LU7. The computed binding affinity for O + F, O + H, F + H, F + H + O combinational inhibitors and individual inhibitors towards receptor protein of 6LU7 have shown that O in individual mode and O + F + H in combination mode exhibited better affinity (−6.2, -5.3 kcal/mol) towards 6LU7 (Figure 2a,b, Table 3, SD1). For every best pose, the donor–acceptor surface are shown in Figure 3(a-f) in 3D and 2D view with their possible hydrogen bonding (Table 3 and supporting document (SD)1). Further through MD simulation results, we have tried to check the effectivity of ligand: receptor complex formation for our proposed individual and combinations drugs.

Figure 2.

Binding energies and binding sites for drug a) O, b) O + F + H towards protein 6LU7 by sequential docking.

Figure 3.

Donor: acceptor surface for best pose in terms of H-bond interaction a), b), c) O: 6LU7, F: 6LU7, H: 6LU7 d), e), f) Possible types of interaction in pose obtained from O:6LU7, F:6LU7, H:6LU7.

3.3. Analysis of MD simulation results

MD simulation of appo 6LU7 and for ligand:6LU7 complexes have been carried out for the time scale of 1 ns–100 ns (Eqs. (1), (2), (3), (4), and (5)). Before MD simulation to work every structure have to be optimized by potential energy (E_pot) minimization process. For energetically stable structure the E_pot should be minimum with a maximum force value. For appo 6LU7 and ligand:6LU7 complexes we have obtained steady convergence of potential energy (Figure 4). Average energy of appo 6LU7 was obtained as -0.3 × 10⁶ kcal/mol. The optimized potential energies were computed at the order of ∼ -0.1 × 10⁶ kcal/mol for all complex structures (in individual and combination forms) (Figure 4).

Figure 4.

Potential energy of receptor 6LU7 and complex structures of O, F, H and O + F + H with 6LU7.

To run the MD simulation in aqueous solution a well-known water model: TIP3P has been used. In aqueous environment before MD simulation, the stability of each complex structures has been verified for a short interval of time (100 ps) under the equilibrium condition (NVT and NPT). It was observed that at 300 K during stabilization process, each structure individually has achieved its stable equilibrated phase after first few ps (∼10 ps) of computation and maintained that stable configuration by the obtained simulated data like density (D), pressure (P) and temperature (T) throughout the whole process (SD2 a-c). After achieving the fully equilibrated phase for both appo protein and ligand: protein complex forms, MD simulations were operated for complete 100 ns time trajectory. Output for all thermodynamic parameters obtained from MD simulation are shown in Table 4, SD3.

Table 4.

Data obtained by MD simulations for protein 6LU7 in its bare state and for the O:6LU7, F:6LU7 and H:6LU7 complex structure.

S. No.	Parameter	Bare 3CLpro protease (6LU7)		O:6LU7		F:6LU7		H:6LU7
		Mean	Range	Mean	Range	Mean	Range	Mean	Range
MD Simulation Result
1.	RMSD (nm)	0.22	0.13–0.32	0.23	0.09–0.48	0.22	0.10–0.40	0.24	0.23–0.24
2.	RMSF (nm)	0.23	0.05–0.4	0.33	0.07–0.58	0.29	0.06–0.52	0.16	0.05–0.28
3.	Inter H-Bonds	NA	NA	2	0–3	5	0–9	3	0–5
4.	Radius of gyration	2.18	2.13–2.24	2.18	2.16–2.4	2.21	2.20–2.21	2.21	2.20–2.22
5.	SASA (nm2)	33	30–35	8	6.5–9	8	7–9	8.2	7.3–9
MM/PBSA Results
6.	Potential Energy (kcal/mol)	-0.30×106$\pm$ 8.8	-7.0×105 - -1.3×106	-0.1×106$\pm$ 4.94	-3.4×104 - -2.8×105	-0.1×106$\pm$ 4.94	-3.4×104 - -2.8×105	-0.1×106$\pm$ 4.94	-3.4×104 - -2.8×105
7.	Binding energy(ΔG) (kcal/mol)	NA	NA	-6.2	NA	-4.1	NA	-49.4	NA

Stability of any complex structure is generally checked by its RMSD value with respect to the reference carbon backbone structure. For the present case we have used the reference backbone structure of receptor protein 6LU7. Figure 5a has shown the 3D view of RMSD values for carbon backbone (6LU7) and of O:6LU7. Figure 5b has shown the 2D view of RMSD values for all ligand: protein complex structures (O:6LU7, F:6LU7, H:6LU7, O + F + H:6LU7) with respect to 6LU7. RMSD variations were observed between 0.09-0.48 nm, 0.10–0.40 nm 0.23–0.24 nm for O:6LU7, F:6LU7 and H:6LU7 complex structures compared to appo 6LU7 variation of 0.13–0.32 nm. Similarly, for O + F + H:6LU7 combination mode, the RMSD variation was observed between 0.10-0.40 nm. The RMSD data indicates the existence of less fluctuation and better stability of the ligand: 6LU7 complex formation (Table 4, SD3). The flatness in 2D contour for H:6LU7 complex structure reveal that in presence of H, during complexation host 6LU7 does not show any significant change (Figure 5b).

Figure: 5.

a) Root mean square deviation (RMSD) graphs for bare state of receptor protein 6LU7 and in complex (O:6LU7) with 6LU7 3D view up to 10000ps and b) 2D for complex (O:6LU7, F:6LU7 and H:6LU7and O + F + H: 6LU7) view up to 10000ps.

RMSF has shown average values of 0.33, 0.29, 0.16, 0.18 nm for O:6LU7, F:6LU7, H; 6LU7 and O + F + H:6LU7 complex structures whereas appo 6LU7 has shown the average value as 0.23 nm, which indicates the protein backbone (Figure 6a) was never get effected due to complexation with any of the selected inhibitor drugs. For further verification of the stability and compressed nature of each complex structures, we have tested the R_g data also for all proposed inhibitors [60]. R_g data has shown the fluctuations between 2.16-2.24 nm, 2.20–2.21 nm, 2.20–2.22 nm and 2.12–2.25 nm for O:6LU7, F:6LU7, H:6LU7 and O + F + H:6LU7 complex structures whereas 2.13–2.24 nm for bare 6LU7 structure. The average R_g were observed in between 2.18 to 2.21nm (Figure 6b, Table 4). Small variation of R_g values showed that all our tested drugs in complex structure with target protein 6LU7 are quite stable and compressed in nature (Figure 6b, Table 4). Among all of them O:6LU7 shows better stability after complex formation due to its compactness with respect to bare protein 6LU7 (2.18 nm).

Figure: 6.

a): RMSF of 6LU7 in its bare state and of complex (O:6LU7, F:6LU7, H:6LU7 and O + F + H: 6LU7) structure, b) R_g of 6LU7 in its bare state and of complex (O:6LU7, F:6LU7, H:6LU7 and O + F + H: 6LU7) structure.

Number of nonbonded interactions are directly related to the stability of ligand: receptor protein complex structure. For present MD simulation work we have used a 3.5 Å cut-off condition for identification of proper nonbonded hydrogen bonded interaction. We have observed the presence of 0–5 number of hydrogen bonded interactions for different complex combinations (Table 4, Figure 7a, SD3). For the present work solvent Accessible Surface areas (SASA) also have been computed to get the information about the area of receptor contact to the solvents (Figure 7b, Table 4). Smaller value of SASA means ligand is mostly surrounded and covered by the receptor protein means better inhibitor: receptor complexation. For different complex structures O:6LU7, F:6LU7, H:6LU7 and O + F + H:6LU7, the SASA values were observed between 6.5-9, 7–9, 7.3–9 and 6–11 nm² with average value 8, 8, 8.2 and 8.3 nm². For appo protein structure it was computed between 30-35 nm² with a mean value of 33 nm². Oseltamivir and Favipiravir in complex forms have shown the minimum SASA values. Favipiravir already had proved its good clinical efficacy against COVID-19 [61]. Similarly clinical trial data shows Oseltamivir is one of the effective drugs used for treating pregnant COVID Patients [62].

Figure 7.

a): Intermolecular Hydrogen bond numbers for complex (O:6LU7) structure for the total time trajectory 0–10000 ps b) SASA area for protein 6LU7 in its bare state and for complex (O:6LU7, F:6LU7, H:6LU7 and O + F + H:6LU7) structure.

MM/PBSA method has been applied for the present study for checking the inhibitor's nonbonding binding affinity towards target protein (Table 4). By MM/PBSA method, change in free binding energy (ΔG_bind) indicate the nonbonding interaction energies of the binding region for the complex formation which is dependent on energies of the complex in vacuum, in nonpolar solvation state and in polar solvation states (Eqs. (6), (7), (8), and (9)). The average ΔG_bind values of -6.2, -4.1, -49.4 and -51.3 kcal/mol were computed for O:6LU7, F:6LU7, H:6LU7 and O + F + H:6LU7. Simulation result on binding energy verified that in individual and combination mode, presence of H has showed a stronger affinity towards 3CL^pro protease. Several reported clinical trial data also validated the positive effectivity of H for the treatment of hospitalized COVID-19 patients. Submitted report by Mokhtari et al. [63] suggested that those COVID positive patients who were treated with H have shown reduced period of hospital stay.

4. Conclusion

For the present study we have tried to recognize the mechanism of action and effectivity of some repurposed common antiviral and antibiotic individual and combination drugs Favipiravir, Hydroxychloroquine and Oseltamivir against COVID-19 disease. The properties like physiochemical, medical chemistry and pharmacokinetics from ADME analysis have found a strong inhibitory opportunity of O, F, H towards SARS-CoV-2 protein 3CL^pro. The results of molecular docking validated the good binding affinity of all tested individual drugs: O, F and H and combination drugs: F + O, F + H and O + F + H towards CoV-2 virus. Existence of strongest binding affinity (−6.2 kcal/mol) and lowest inhibition constant (2.8 × 10⁻⁵ M) established the possibility of better complexation of O with 6LU7 protease. Molecular dynamics simulation data for the whole-time scale (0–100 ns) validated the stability of all tested repurposed inhibitors in terms of their different thermodynamical parameters (E_pot, T, V, D, interaction energies, R_g, ΔG_bind, SASA energy). Trajectories analysis showed that all studied complexes have displayed structural stability during the molecular docking and MD runs. Among all, the perfect closeness of average RMSD of O:6LU7, F:6LU7 (0.23 nm, 0.22 nm) and host 6LU7 (0.22 nm) confirmed the total inheritance of Favipiravir and Oseltamivir inside host protein whereas RMSF variation data clearly showed that H:6LU7 and F:6LU7 complex structures does not affected the protein backbone. Lowest SASA energy for all tested inhibitors in their individual modes (∼8 nm²) validated the best stability of these individual drugs F, O, H towards receptor 6LU7 compared to their combination modes. From the simulated data of binding affinity by MM/PBSA method it is observed that H showed higher affinity than F and O against COVID 19 main protease. Supported by previous literatures and our in silico results we may conclude that all our tested repurposed drugs may act as an inhibitor for the main protease of SARS-CoV-2 since they all have considerable effect but none of them has shown to be fully effective against the CoV-2 infection. We hope our current comparative study of repurposed drugs: Favipiravir, Oseltamivir and Hydroxychloroquine in their individual and combinational modes against CoV-2 infections will help the current clinical research to develop an effective drug for the cure for COVID -19 treatment.

Declaration

Author contribution statement

Pooja Yadav: Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Papia Chowdhury: Conceived and designed the experiments; Analyzed and interpreted the data; Wrote the paper.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

Data included in article/supplementary material/referenced in article.

Declaration of interests statement

The authors declare no competing interests.

Additional information

No additional information is available for this paper.

Appendix A. Supplementary data

The following is the supplementary data related to this article: