key: cord-0858118-tvszfaw0 authors: Saraswat, Juhi; Riaz, Ufana; Patel, Rajan title: In-silico study for the screening and preparation of ionic liquid-AVDs conjugate to combat COVID-19 surge date: 2022-05-03 journal: J Mol Liq DOI: 10.1016/j.molliq.2022.119277 sha: 1c8be3e53821f224d5b202ec8137aa2f3bd75150 doc_id: 858118 cord_uid: tvszfaw0 The pandemic due to COVID-19 caused by SARS-CoV-2 has led to the recorded deaths worldwide and is still a matter of concern for scientists to find an effective counteragent. The combination therapy is always been a successful attempt in treating various threatful diseases. Recently, Ionic liquids (ILs) are known for their antiviral activity. Fascinating tunable properties of ILs make them a potential candidate for designing the therapeutic agent. The concern while using ILs in biomedical field remains is toxicity therefore, choline-based ILs were used in the study as they are considered to be greener as compared to other ILs. In the present study strategically, we performed the blind molecular docking of antiviral drug (Abacavir, Acyclovir, and Galidesivir)-choline based ILs conjugates with the target protein (Mpro protease). The molecules were screened on the basis of binding energy. The data suggested that the combination of AVDs-ILs have greater antiviral potential as compared to the drugs and ILs alone. Further, the ADME properties and toxicity analysis of the screened conjugates was done which revealed the non-toxicity of the conjugates. Additionally, the energetic profiling of the ILs drugs and their conjugates was done using DFT calculations which revealed the stability of the conjugates and have a better option to be developed as a therapeutic agent. Also, from molecular dynamic simulation was done and results showed the stability of the complex formed between target protein and the designed conjugates of AVDs and ILs. It's been almost and year and half, the outbreak of severe acute respiratory syndrome is still uncontrolled rather the situation in some part of the world has become even worst specially in the Asian countries [1] . As per the World Health Organisation, 1,49,432,808 coronavirus (COVID- 19) cases have been recorded (on 8 th June, 2021, while writing the manuscript) and is expected to increase more intensively and exponentially by the end of the May, 2021. The total death recorded till now is 34,49,189 worldwide however, 127,111,632 have been recovered [2] . The high mortality and morbidity are been seen in this pandemic due to coronavirus spread and is even more in 2021 as compared to 2019-20. The destruction is comparatively high as compared to the 2019-20. The reason being the continuous change of virus mutation which is more devastating than 2019-20 [3] . As previously reported, the spread of infection from host mediates via respiratory droplets, smog and fomites [4] . Structurally, the coronavirus is a single stranded RNA which directly affects the lungs and causes the lethal impact. The mechanism of action of the spike protein is already been discussed in our earlier publication [5] . Recently, in the March, 2021 (Phase 2) in India National Centre for Disease Control (NCDC) reported that corona test carried out in Maharashtra, Delhi and Punjab showed new variant of coronavirus [6] . When genome sequencing was done by Indian SARS-CoV-2 Consortium on Genomics (INSACOG) it was found two important mutations in single variant and is termed as "double mutant" [7] . The rapid growth of this variant caused large destruction in India and still the panic situation continues and this clearly indicates the role of this variant in India's surge and the destruction is even far more as compared to first wave of corona infection in the year 2020 [8] . Other reason of India's surge could be the people lowered the guards and did not follow the COVID-19 protocol after the first wave. Opening of every sector to pre-covid level and inappropriate behaviour which was no longer taking care of the adverse effect exposed the susceptible population in a very big way [9] . Although we are in a mid of era where, urgent need of effective treatment is urgently needed. The whole world's scientist and doctors are working to find the way and stop the pandemic situation. India has developed the coronavirus drug named, Covid shield, Covaxin and a Russian drug, Sputnik-V have emerged as a hope to combat the battle against coronavirus [10, 11] . Many other trials are in a pipeline to find the way out; however, still urgent drug development is needed. In our previous work we explored the antiviral potential of ILs against Covid-19 computationally [5] . The combination therapy is always been a successful attempt in treating various threatful diseases. Therefore, in the present work we aimed to create the combination of ILs and conventional antiviral drugs against a key regulator of the life cycle of coronavirus, main protease (M pro or 3CL pro ) enzyme [12] . M pro plays a crucial role in the replication of the coronavirus as it promotes the translation of viral RNA to viral protein and causes a growth of coronavirus infection [13] . Hence, study on M pro enzyme becomes an important drug target to study on to combat covid-19 surge [14] . ILs are the molten salts below 100 o C and composed of organic cation and organic/inorganic anions. ILs are known as designer solvents because of their tunable characteristic feature [15] [16] [17] . Though, the toxicity of ILs always has been a topic of concern for the researchers specially when designing the drug molecule. In this regard, we have chosen choline based-ILs in our study because of their greener property as well as low toxicity [18] . Recent studies showed the antiviral application of the ILs [19] . Palanisamy et. al. in their work showed that ILs enhance the stability of Mpro protein and have have significant contributions to the proteindrug binding, which may be useful in drug development for COVID-19. [20, 21] . After extensive literature survey, we have computationally screened some choline based-ILs and antiviral drugs (abacavir, acyclovir, galidesivir that are clinically tested (chemical structures given on Table 1 ). Also, the toxicity and drug likeliness assay were performed computationally to further filter out our findings for their application as a potential inhibitor of M pro . Moreover, the density functional theory (DFT) was employed to determine the stability of IL_AVDs conjugates. With employing these advanced computational approaches, designed compounds might help in significant improvement in the antiviral activity and can have a better chance to be developed as drug leads against covid-19 virus. Computation approach for designing a molecule plays an important role in developing the effective molecule against target molecules which causes severe illness. One of them nowadays everybody countering with is Covid-19 disease. Repeated experiments and trials in laboratories and failures in the study is always been a problematic situation for the scientist working hard to combat in the pandemic situation. To overcome the chances of failure, computational approach plays an important role in biomedical field. M pro protease is reported as a regulatory enzyme important role in coronavirus infection and its progression in the body [22] . The three-dimensional structure of M pro protease was obtained from protein data bank (PDB) (PDB ID: 6Y84) from https://www.rcsb.org/ [5] . To study the effect of ILs as well as its conjugates with clinically tested antiviral drugs the computational study was done on M pro protease. We have taken Abacavir (ABA), Acyclovir (ACY), and Galidesivir (GAL) as antiviral drugs (AVDs) which was found to be active against covid-19 [23] [24] [25] . The biological importance and action of selected AVDs are listed in Table 1 . In our previous studies ILs were found to be antiviral and toxicity remained an issue therefore Choline based ILs because of their green characteristic feature is found to be a potential candidate for the present study [26] . AVDs, choline-based-ILs namely N- (2-hydroxyethyl hydroxyethyl)-N,N-dimethyldodecan-1-aminium and its conjugates with AVDs was drawn using ChemDraw 12.0 software and were saved as pdb format. The energy minimisation of all the structure was done using spdv.exe 4.10 software. The binding affinity of ILs, AVDs and IL-AVD conjugates towards M pro protease was evaluated using AutoDock 1.5.6 software. The structure of the protein was optimised, water molecules were removed from the protein structure, polar hydrogen was added followed by the addition of AD4 type atoms to the protein. In the same manner for the ligand's energy optimization was done. Lamarckian Genetic Algorithm embedded in AutoDock software was employed to evaluate the binding score of ligand protein interactions. The blind docking was performed, the protein and ligands was confined into the grid box in all axes which could accommodate the active site. The grid parameters are listed in table S1. Thereafter, a total of 100 runs were set in the software to carry out the docking. After successful completion of all the docking steps, the obtained conformers were ranked depending upon the highest binding energy value from molecular docking results. The obtained results from molecular docking were further analysed using visualisation software. Depending upon the highest binding affinity score the stable conformer was selected and was visualised using PyMol software. The 3D view of selected conformer and interaction involved (with bond length) between ligand and protein, UCFS Chimera 1.2 software was employed [5] . Additionally, the 2D view of most appropriate conformer was obtained using Discovery studio software (BIOVIA-2016) [15] . Pharmacokinetics properties helps in predicting the safety and effectiveness of therapeutic agent taken in the study in the initial stages of drug discovery or development. The pharmacokinetics properties of therapeutic agent involve adsorption, distribution, metabolism and excretion of compound in the body. Freely available software SwissADME at http://www.swissadme.ch/index.php was used to predicts the pharmacokinetics of designed molecules (as shown in Figure 1 ) [27] . Toxicity of therapeutic agent in drug discovery is always been a matter of concern for the science fraternity. The computational approach for the prediction of toxicity of compounds plays a key role in the studies done for drug discovery. The toxicity profiling includes various parameters such as hepatotoxicity, immunogenicity, mutagenicity, carcinogenicity, etc. The selected compounds (shown in Figure 1 ) were screened for their toxicity profiling using a freely available software named software at http://tox.charite.de/tox [28] . between ILs and AVDs. The best fit molecules were screened out on the basis of minimum binding energy obtained from molecular docking study and DFT calculations were done for the screened molecules. We used Gaussian 03W for performing the DFT calculation and the Gaussian view was used to for optimization + frequency of the molecules. The tools set for DFT was B3LYP, 6-31G (d) basic set, +. Various Physiochemical descriptors viz. global electrophilicity index (ω), electronegativity (χ), chemical potential (μ), softness (S), and chemical hardness (η) of IL, drug and their complex, IL-drug were determined using equations 1-5. (1) μ = (E HOMO + E LUMO )/2 Table 2 it can be seen that with increasing the alkyl chain length the value of binding energy decreases. CL12 showed minimum binding energy with target protein which depicts that with increasing carbon chain length at N-atom the antiviral activity of the IL increase and becomes more potent against novel coronavirus. Our results comply with the results we achieved where it was clearly seen that with increasing hydrophobicity the antimicrobial characteristic of the corresponding IL increases. choline-based ILs with M pro protease (target protein) are summarized in Table 2 h-bond donor should be less than 5, H-bond acceptor must be less than 10, log P value should be less than 5 for a compound to be a candidate for drug development [33] . These parameters for selected class of compounds were calculated using SwizzAdme software as described in section 2.5. The values obtained for the above said parameters are listed in Table S3 . The best fir molecules on the basis of minimum binding energy obtained from molecular docking was, ABA_CL12, ABA_CL10, ABA_CL8 and ABA. Amongst 23 compounds, the best fit molecule on the basis of binding energy was screened out and the results for best molecules are listed in Table 3 . From Table 3 , the binding energy of ABA_CL12, ABA_CL10, ABA_CL8 and ABA was found to be -8.13, -7.12, -7.01, -6.91 kcal/mol, respectively. The value binding energy of screened conjugates was found to be much lower than that of conventional antiviral drug, ABA which suggests the high potency of the conjugates than that of antiviral drug. The ADME property describing parameters were found to be in range Toxicity determination is a very crucial to be counted before designing any therapeutic agent. Therefore, the toxicity profile of all selected drugs/IL and designed conjugates were done using online available software as discussed in section 2.6. Various features such as blood brain barrier (BBB) penetration, GI adsorption, CYP inhibitory promiscuity, and rat acute toxicity (LD50) value were obtained from software. The parameters obtained are listed in Table S4 . Amongst 23 compounds, the best fit molecule on the basis of binding energy was screened out and the results for best molecules are listed in Table 4 . From Table 4 , the data suggests that the conjugates, ABA_CL12, ABA_CL10, ABA_CL8 showed high GI absorption but showed negative results for BBB which implies that the conjugates can't cross the blood brain barrier. CYP2D6, CYP3A4 is an importantenzymes found in liver and intestine. It helps in the oxidation of foreign organic molecule and supports in excretion [34] . Our results showed the inhibition of these enzymes in the presence of the conjugate whereas no inhibition of the important enzymes such as CYP1A2, CYP2C19, and CYP2C9 which further helps in the excretion of foreign molecule from the body. The LD50 value of all the studied compounds is summarized in Table S4 . The value of LD50 decides the toxicity factor of any therapeutic agent [34] . According to the literature the LD50 range is classified into 4 categories. First in which LD50 ≤ 50 mg/kg; second in which 50 mg/kg ˂ LD50 ≤500 mg/kg and are considered as toxic; third category in which 500 mg/kg ˂ LD50 ≤ 5000 mg/kg) and fourth category where, LD50 ˃ 5000 mg/kg are considered to be non-toxic [5, 35] . In our study the screen molecule falls under the category 3 and are considered to be non-toxic. Table 5 were used to calculate different physiochemical parameters viz. global electrophilicity index (ω), electronegativity (χ), chemical potential (μ), softness (S), and chemical hardness (η), respectively using Equation 1-5. The popularity of DFT applications to elucidate the broad range of problems in pharmaceutical and biochemical interest has been growing rapidly, especially the interactions of candidate molecules such as drug-drug [36] , drug-ionic liquids [37] , To understand the reactivity of the molecules, the physiochemical parameters obtained from DFT calculation plays an important role. As per the reported literature, HOMO behaves as the electron donor while LUMO behaves as acceptor of electron [38] . The difference between the HOMO and LUMO energy level (ΔE H-L ) reveals the reactivity of the studied molecule. On moving from CL8 to CL12 with ABA, the negative value of E H-L increases and hence the stability of the complex increases. Therefore, system ABA_CL8, ABA_CL10, ABA_CL12 comes out to be most stable as depicted from the Table 5 . 3.8 Molecular dynamic simulation. MD simulation was performed to evaluate the stability of target protein and screened conjugate. It also helped in determining any induced change in the protein by conjugate [39, 40] . The RMSD profile of docked complexes for the best obtained conjugate and target protein was obtained using Webgro server as shown in Figure 7 . The trajectories of generated simulation run were analysed using RMSF and RMSD calculations. From MD simulations results, RMSF of screened best complex with coronavirus protease was analysed using CPPTRAJ module at 100ns time period [41] . The RMSD of the protein backbone was consistent after 50 ns. In case of ABA_CL12-protease complex, the protein backbone RMSD stabilized after 80ns as shown in Figure 7a . The overall result suggests that after 80ns the complex was stable. The amino acids contribute for the formation of the complex between main protease with the ABA_CL12 and showed structural fluctuations [42] . RMSF helps in determining the flexibility of each amino acid or the residue in the protease of the complex. The amino acids surrounding the ABA_CL12, showed fluctuations but was less as compared to the complex as in Figure 7b . The RMSF value for complex of main protease of novel coronavirus was found to be 0.75-3.75 at 300 K, while RMSF value of complex with ABA_CL12 was found to be 0.51-3.25 at 300 K. The maximum fluctuation was seen at 80-120 positions. The lower fluctuations thereafter with RMSF value indicating more stability with a pronounced role of these residues in interaction. Rg results indicated that the compactness of the complex increases as Rg value found to be decreasing with time from 2.2ns to 2.1ns as shown in Figure 7c . This confirms that point mutation in the conserved site caused structural stability leading to the increasing protein compactness [43] . India. The second wave is much devastating as compared to first wave [44] . The delta variant is much more dangerous as that of ß variant which was found responsible for first wave. Delta variant transmission was in fact much faster as that of ß variant [45] . Though we have number of vaccine available to combat Covid-19 infection, but no such evidence has been seen showing the complete cure [10, 12, 46] . Therefore, we need to develop new effective drugs on urgent bases against novel SARS-CoV2. The combination therapy is always been a successful attempt in treating various threatful diseases [47, 48] . issues related with the high mortality rate and irregular heartbeat [50] . In our earlier publication, we explored the application of ILs as an antiviral agent computationally where we concluded the application of ILs as an antiviral agent [5] . The reason of being concern while working with ILs is their toxicity. As earlier reported, imidazolium, pyrrolidinium, pyridinium-based ILs are toxic [51] . Therefore, in the present study we focused on the known greener, non-toxic choline based ILs. Also, ILs hold many fascinating features with them such as tunability, liquid at room temperature, low vapour pressure, biological activity, etc [52, 53] . The tunable characteristics of ILs make them a perfect class of compound where it can be designed on the basis of the application needed for [52] . We planned the repurposing of available antiviral drug by designing conjugates originating from (AVDs, ABA, ACY, and GAL) and choline-based ILs. The ABA, ACY and GAL are the known antiviral drugs active against novel coronavirus (details given in Table 1 ). The results showed the antiviral activity of choline based ILs against SARS-CoV2 virus. From Table 2 , it was observed that with increasing the carbon chain length the binding energy with target protein increased suggesting the increasing antiviral activity with increasing hydrophobicity likewise the antibacterial activity as reported in our previous publications [15, 16, 54] . Further, the molecular docking results showed GAL>ABA>ACY order of antiviral activity of the drugs. The designed conjugates showed much increased activity as compared to AVDs and ILs alone. Best results in terms of least binding energy with target protein was shown by ABA_CL12>ABA_CL10>ABA_CL8. Rest all other conjugates also showed much improved results as compared to corresponding drugs and ILs (alone). Also, with increasing hydrophobicity the binding energy increased in case of conjugates also suggesting a higher potential of the conjugates against SARS-CoV2 virus. Though lot of computational work has been done in screening of active drug molecule towards inhibiting the SARS-CoV2 viral strain, our present approach is not yet known. The ADME and toxicity results revealed that the screened molecules are non-toxic and biocompatible. Additionally, the DFT study was also performed to investigate the interaction and specifically the reactivity of the conjugates formed between drugs and ILs which revealed the stability of the screened complexes as reported in the literature [55] . In the end, amongst the screened complexed; the complex with highest binding affinity was tested for the stability after binding with the target protein using molecular dynamic simulation. MD simulation results revealed that ABA_CL12 forms the stable complex once it gets bound to the target protein and can therefore, act as an inhibitor of the Mpro protein leading to the prevention of entering of virus protein into the host cell [43] . In the present study, we have explored the new conjugation of antiviral drugs (abacavir, acyclovir, and Galdesivir) with choline-based ILs using computational approach. conjugates might be used as a potential antiviral agent but before any clinical application of these molecules detailed in-vivo and in-vitro studies would be needed in future to validate the findings. JS performed the whole work and wrote the manuscript. UR performed the DFT calculation in the study. RP is the overall corresponding author. The authors declare no competing financial interest. Table 4 : Toxicity Prediction of all studied compounds the potential inhibitors of Sars-CoV-2 obtained using ProTox-II prediction software. 4R)-4-(2-amino-9H-purin-9-yl)cyclopent-2-enyl)methoxy)ethyl)-N 4R)-4-(2-amino-9H-purin-9-yl)cyclopent-2-enyl)methoxy)ethyl)-N 4R)-4-(2-amino-9H-purin-9-yl)cyclopent-2-enyl)methoxy)ethyl)-N 4R)-4-(2-amino-9H-purin-9-yl)cyclopent-2-enyl)methoxy)ethyl)-N -hydroxyethoxy)methyl)-1H-purin-6(7H)-one -5.72 13 methoxy)ethoxy)ethyl)-N,N-dimethylbutan-1-aminium methoxy)ethoxy)ethyl)-N,N-dimethylhexan-1-aminium methoxy)ethoxy)ethyl)-N,N-dimethyloctan-1-aminium methoxy)ethoxy)ethyl)-N,N-dimethyldecan-1-aminium 5S)-5-(4-amino-5H-pyrrolo[3,2-d]pyrimidin-7-yl)-3,4-dihydroxypyrrolidin-2-yl)methoxy)ethyl)-N 5S)-5-(4-amino-5H-pyrrolo[3,2-d]pyrimidin-7-yl)-3,4-dihydroxypyrrolidin-2-yl)methoxy)ethyl)-N 5S)-5-(4-amino-5H-pyrrolo[3,2-d]pyrimidin-7-yl)-3,4-dihydroxypyrrolidin-2-yl)methoxy)ethyl)-N 5S)-5-(4-amino-5H-pyrrolo[3,2-d]pyrimidin-7-yl)-3,4-dihydroxypyrrolidin-2-yl)methoxy)ethyl)-N 5-(4-amino-5H-pyrrolo[3,2-d]pyrimidin-7-yl)-3,4-dihydroxypyrrolidin-2-yl)methoxy)ethyl)-N Dr. Rajan Patel greatly acknowledges the financial support from Science and EngineeringResearch Board (EEQ/2020/000437) New Delhi, India. The authors declare no competing financial interest.Graphical Abstract: