key: cord-1044649-f2k9mn6q
authors: Khanal, Pukar; Dey, Yadu Nandan; Patil, Rajesh; Chikhale, Rupesh; Wanjari, Manish M.; Gurav, Shailendra S.; Patil, B. M.; Srivastava, Bhavana; Gaidhani, Sudesh N.
title: Combination of system biology to probe the anti-viral activity of andrographolide and its derivative against COVID-19
date: 2021-01-27
journal: RSC advances
DOI: 10.1039/d0ra10529e
sha: 4293f71da3df6810e7708bd82e9e58d151d47afa
doc_id: 1044649
cord_uid: f2k9mn6q

The present study aimed to investigate the binding affinity of andrographolide and its derivative i.e., 14-deoxy-11,12-didehydroandrographolide with targets related to COVID-19 and their probable role in regulating multiple pathways in COVID-19 infection. SMILES of both compounds were retrieved from the PubChem database and predicted for probably regulated proteins. The predicted proteins were queried in STRING to evaluate the protein–protein interaction, and modulated pathways were identified concerning the KEGG database. Drug-likeness and ADMET profile of each compound was evaluated using MolSoft and admetSAR 2.0, respectively. Molecular docking was carried using Autodock 4.0. Andrographolide and its derivative were predicted to have a high binding affinity with papain-like protease, coronavirus main proteinase, and spike protein. Molecular dynamics simulation studies were performed for each complex which suggested the strong binding affinities of both compounds with targets. Network pharmacology analysis revealed that both compounds modulated the immune system by regulating chemokine signaling, Rap1 signaling, cytokine–cytokine receptor interaction, MAPK signaling, NF-kappa B signaling, RAS signaling, p53 signaling, HIF-1 signaling, and natural killer cell-mediated cytotoxicity. The study suggests strong interaction of andrographolide and 14-deoxy-11,12-didehydroandrographolide against COVID-19 associated target proteins and exhibited different immunoregulatory pathways.

In December 2019, a severe acute respiratory syndrome caused by novel severe acute respiratory syndrome novel coronavirus 2 (SARS-CoV-2) 1 emerged as a global pandemic from Wuhan city, Hubei province, China. WHO designated this nSARS-CoV-2 infection as Coronavirus disease . The n-SARS-CoV-2 is a highly contagious virus that can be transmitted from person to person, 2 leading to community transmission. COVID-19 became a major global threat by inuencing around 212 countries, with almost half a million deaths worldwide. 3 Presently, it has majorly affected subjects with comorbidities and low immunity that are suffering from infectious and noninfectious diseases. 4 Patients with COVID-19, especially those with severe pneumonia, showed substantially lower lymphocyte counts, and severely ill patients exhibited a reduction in CD4+ T cells, CD8+ T cells, and natural killer cells. 5, 6 The higher plasma concentrations of several inammatory cytokines, such as IL-6 and tumor necrosis factor (TNF), were observed in COVID-19 patients. 7 The pathological ndings in patients with showed that immune-mediated lung injury was involved in acute respiratory distress syndrome (ARDS). 8 This evidence suggested an immune imbalance in COVID-19, and it was contemplated that the immune modulation can provide some prophylaxis and promising benet against COVID-19. 9 Also, there is a need to utilize the concept to identify the new therapeutic agent with immunomodulatory action and anti-viral property against the COVID-19 as reported. 10, 11 The effectiveness of treatment based on traditional medicinal plants has been reported during 2003 SARS. [12] [13] [14] [15] Therefore, the scientic community has already started studies on medicinal plants, based on their history and traditional uses, as plausible leads in the treatment of COVID-19. 3,16-20 For thousands of years, medicinal plants have played a vital role in managing multiple infectious and non-infectious diseases. [21] [22] [23] Among them, Andrographis paniculata (Family: Acanthaceae), also called known as 'King of bitters' and 'Indian Echinacea' reserves its importance in the management of various infectious and non-infectious diseases. 9, [24] [25] [26] Further, it has been studied well for its potency as a modulator of the immune system. 25, 27 In Andrographis paniculata, andrographolide 28 is a major bioactive that possesses benecial effects in multiple pathogenic conditions, including the immunity booster role. 29 Further, two important databases, i.e., ChEBI and PCIDB, also record andrographolide ( Fig. 1 ) as chief bioactive from Andrographis paniculata. Andrographolide and its derivative(s) also exhibited decisive immunomodulatory action 25, 27 and have broad-spectrum anti-viral properties. 30 Further, it was found to be effective against multiple viral infections like dengue, 31 swine u, 32 hepatitis C, 33 chikungunya, 24 inuenza, 34 Epstein-Barr virus (EBV) 35 and herpes simplex virus 1 (HSV-1) 36 in previous experimental studies. The andrographolide derivative, i.e., 14-deoxy-11,12-didehydroandrographolide is one of the major components/derivatives of A. paniculata reported for its antiviral properties. [37] [38] [39] Recently, andrographolide has been investigated as a potential inhibitor of SARS-CoV-2 main protease (3CL pro ) using an in-silico approach. 40 However, its potency to act over papain-like protease (PL pro ) and spike protein has not been investigated yet. Further, there are numerous reports wherein various in silico approaches 41 such as molecular docking, fast pulling of ligand (FPL), free energy perturbation (FEP), 42 density functional theory (DFT), 43 high throughput virtual screening, 44 and drug repurposing studies 45 have been exploited to investigate various target proteins of SARS-COV-2. Likewise, there are recent reports of in-silico investigations of murine natural products 46 and some diverse scaffolds of synthetic compounds designed through in silico insights. 47, 48 Since the risk of getting an infection with COVID-19 is reported to be higher in the subjects with compromised immunity, 10 it is important to consider in manipulating the immune system in them.

Hence, the present study aimed to investigate the prospective potential of andrographolide and one of the major derivatives i.e. 14-deoxy-11,12-didehydroandrographolide as a potent anti-viral agent by targeting three proteins of COVID-19, i.e., 3CL pro , PL pro , and spike protein. Further, the study also evaluated the plausible pathways to be regulated in enhancing the immune system.

SMILES of 14-deoxy-11,12-didehydroandrographolide and andrographolide was retrieved from the PubChem (https:// pubchem.ncbi.nlm.nih.gov/) database 49 and queried for protein-based prediction in DIGEP-Pred 50 at a pharmacological activity (Pa) > pharmacological inactivity (Pi).

The list of up-and-down-regulated proteins was queried in the STRING database. 51 The biological process, cell component, and molecular function were recorded. The modulated protein and their associated pathways were identied using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database. The interaction between the compounds, their targets, and pathways was constructed using Cytoscape (https:// cytoscape.org/) and was analyzed using "edge count" of respective node.

3D structures of 14-deoxy-11,12-didehydroandrographolide and andrographolide were retrieved from the PubChem database in .sdf format and converted into .pdb format using Discovery studio 2019. The ligand of each molecule was minimized using mmff94 forceeld and converted into '.pdbqt' format. Structures of 3CL pro (PDB: 6LU7) and PL pro (PDB: 4M0W) were retrieved from the RCSB database, 52 which were complexed with water molecules and hetero atoms; removed using Discovery studio 2019 and saved in '.pdb' format. The spike protein of coronavirus was homology modeled target using accession number AVP78042.1 as query sequence and PDB: 6VSB as a template using SWISS-MODEL. 53 Docking was carried using Autodock 4.0. 54 Aer docking ten different ligand conformations were obtained in which ligand possessing minimum binding energy was chosen to visualize the ligand-protein interaction using Discovery studio 2019.

MD simulations were carried out with the AMBER18 soware package. 55 The ligands 14-deoxy-11,12-didehydroandrographolide and andrographolide were parameterized with ANTE-CHAMBER 56 employing GAFF force eld. The amino acid residues of each protein under study were parameterized with the FF14SB force eld. The xLEAP program was used to prepare the protein-ligand complexes of the target proteins 3CL pro , PL pro , and spike protein with ligands' docked poses. Each proteinligand complex was solvated in a truncated octahedron of the TIP3P box. Appropriate counter ions Na + and Cl À were added to neutralize the system. Thus prepared, the protein-ligand complexes were subjected to 100 ns MD simulations on Nvidia V100-SXM2-16GB GPU using the PMEMD.CUDA module. Initially, the system was subjected to energy minimization, where water and the complete system were minimized in two steps. Simulated annealing optimization, and the NVT and NPT equilibration steps of 5 ns, each was performed to equilibrate the system. The production phase MD simulations were performed at 1 atm constant pressure using Monte Carlo barostat and 300 K constant temperature by using Langevin thermostat. During the simulation, a collision frequency of 2 ps À1 and the volume exchange was attempted for every 100 fs. Hydrogen bonds were constrained by using the SHAKE algorithm, and the integration step of two fs was employed. Particle Mesh Ewald (PME) method was used to compute the long-range electrostatic interactions, while the cut-off of 8Å was used to compute the short-range interactions. The program CPPTRAJ was used to analyze the interactions at every four ps on the result from the full trajectory. The MD simulation results were analyzed in terms of RMSD and RMSF of protein-ligand complexes.

The compound's drug-likeness was calculated based on the rule of ve using Molso 57 by querying the SMILES of compounds. Further, absorption, distribution, metabolism, excretion and toxicity (ADMET) prole were calculated using admetSAR2.0. 58

Andrographolide was predicted to regulate 36 proteins, of which 17 were down-regulated, and 19 were upregulated. Likewise, 14deoxy-11,12-didehydroandrographolide regulated 48 proteins in which 21 were downregulated, and 27 were upregulated. The list of regulated proteins with their Pa and Pi of both compounds is summarized in Table 1 .

A total of seventy-two different pathways were identied to be regulated by the andrographolide, among which, pathways in cancer were primarily regulated by modulating nine genes, i.e., AR, ESR2, IL6R, MDM2, PRKCA, RAC1, RARA, RHOA, RXRA at the false discovery rate of 4.96 Â 10 À5 . Similarly, 14-deoxy-11,12didehydroandrographolide was predicted to regulate seventyseven different pathways by modulating the Estrogen signaling pathway via seven genes, i.e., ESR2, FKBP5, KRT16, KRT17, KRT18, PGR, RARA at the false discovery rate of 7.57 Â 10 À6 . Pathways modulated by andrographolide and 14-deoxy-11,12-didehydroandrographolide with their respective genes are summarized in Tables 2 and 3, respectively. Similarly, the interaction of both compounds with the proteins and regulated pathways is represented in Fig. 2 and 3 .

Further, the number of genes in multiple cellular components, biological process, and molecular function for andrographolide and 14-deoxy-11,12-didehydroandrographolide are represented in Fig. 4 and5, respectively. Similarly, network analysis of 14-deoxy-11,12-didehydroandrographolide identied prime regulation of PRKCA protein and estrogen signaling pathway. Further, andrographolide primarily modulated PRKCA protein and pathways in cancer. 

Among andrographolide and 14-deoxy-11,12didehydroandrographolide, 14-deoxy-11,12-didehydroandrographolide was predicted to have the highest binding affinity with PL pro , i.e., À6.7 kcal mol À1 ; however, it did not have any hydrogen bond interactions. Similarly, andrographolide showed À6.5 kcal mol À1 binding energy with PL pro with 1 hydrogen bond interaction, i.e., Tyr274. Although both molecules had equal binding energy with 3CL pro (À6.8 kcal mol À1 ), the number of hydrogen bond interactions were more in andrographolide due to interaction with Thr190, His163, and Cys145. Further, both molecules showed a binding affinity with spike protein, i.e., 6.9 kcal mol À1 ; however, andrographolide showed 1 hydrogen bond interaction with Lys807 ( Table 4 ).

The interaction of each compound with the respective proteins is represented in Fig. 6 .

The exibility at the binding site and the desolvation mechanism is not considered in the rigid docking methodology. However, molecular dynamics simulations can provide deeper insights into the interaction between ligand and protein amino acid residues at the atomistic level. The integrated workow comprising molecular docking and molecular dynamics simulations is more suited in such situations as docking provides the most favourable bioactive poses of inhibitor molecules. In contrast, MD provides the insights of interactions and energetics in a biological environment. 59, 60 The extended time scale MD simulations allow exploring a vast space of conformational optimization and its stability. In the present work, 100 ns MD simulation of well-equilibrated systems was performed on the complexes of 3CL pro , PL pro , and modeled spike protein, each of which is bound to 14-deoxy-11,12-didehydroandrographolide and andrographolide, respectively. The analysis of resulting trajectories comprising of 10 000 frames provides insights into the binding modes of inhibitor molecules, the formation of hydrogen bonds, pi-pi interactions, van der Waals interactions, and the consequent stability of the system in terms of RMSD, RMSF, and ligand-RMSD.

The MD trajectories of PL pro with 14-deoxy-11,12didehydroandrographolide and andrographolide were analyzed for the protein RMSD, ligand RMSD and per residue uctuations as RMSF (Fig. 7a-c) . The RMSD analyses of PL pro bound with 14-deoxy-11,12-didehydroandrographolide showed a slight gradual increase in RMSD with initial equilibration at around 2Å for the rst 25 ns, aer which it had gradually risen to 2.75Å between 25 to 100 ns. These RMSD values point out the system's stability and strong binding affinity between the PL pro and the 14-deoxy-11,12-didehydroandrographolide molecule. The subtle but gradual increase in RMSD could be attributed to the binding site adaptation supported by the RMSF for the binding site residues aa150 to aa200 and aa220 to aa240. Interestingly, a similar trend in the RMSF was also observed in the complex with andrographolide. These residues are present at the binding cavity, and possibly they adopt the conformation suitable for both the ligands. These RMSD uctuations in the case of PL pro and andrographolide complex were observed, reaching a maximum RMSD of 3Å at around 85 ns. Aer that, they were gradually decreasing to around 2.5Å towards the end of the simulation. The uctuations in RMSD may be in part due to the C11-C12 rotatable bond, which may give rise to better conformational exibility in the andrographolide molecule. The Lig-RMSD of 14-deoxy-11,12-didehydroandrographolide remains stable at 2.5Å for about 75 ns and aer that uctuates and sharply rises to 15Å. However, the Lig-RMSD of andrographolide remains stable at RMSD of 2.75Å until 50 ns and further rises to a stable RMSD of 5Å until the end of the simulation. The MD trajectories were visually inspected to investigate the uctuations in Lig-RMSD (Fig. 7d and e) . Both the phytochemicals adopt a conformationally more stable position by binding at the shallow binding cavity. Possibly because the hydroxyl group at the 14th position in andrographolide allows it to adopt a conformationally stable form throughout the simulation; however, the lack of this hydroxyl group and restricted rotation around the C11-C12 bond in 14- deoxy-11,12-didehydroandrographolide may be responsible for larger uctuations in Lig-RMSD aer 75 ns. The residues Val164 and Tyr274 participate in hydrogen bond formation with a carbonyl oxygen atom at C16 position in both the ligands. These bonds break in 14-deoxy-11,12didehydroandrographolide more oen aer 75 ns due to restricted rotation around the C11-C12 bond. The ligand superposition also shows some structural conformational changes in both the ligands. The MDS studies on the PL proligand complexes suggest that these complexes are relatively stable.

The MDS trajectories of 3CL pro bound to each ligand were analyzed, and the protein RMSD, ligand RMSD and per amino acid residue uctuations, RMSF were recorded (Fig. 8a-c) . The protein RMSD for the 14-deoxy-11,12-didehydroandrographolide rise to about 3Å during the rst 25 ns and then equilibrated at around 3Å until 75 ns, aer which it uctuates and rose to around 4.5Å towards the end of MDS. On the other hand, the andrographolide bound to 3CL pro equilibrates quickly, and the RMSD remains in 2 to 3Å throughout the simulation. It indicates the fair stability of 3CL pro bound to andrographolide. This nding is also clearly reected in the RMSF and ligand RMSD. The ligand RMSD of andrographolide Larger uctuations in the RMSD were observed aer 70 ns with an increase in RMSD to an average of around 7.5Å. The results of ligand RMSD indicates better conformational stability of andrographolide than 14-deoxy-11,12-didehydroandrographolide (Fig. 8b) . The per residue RMSF for both the complexes has a similar pattern of uctuating residues involvement with the uctuations ranging between 0.5 to 4Å; however, the RMSF values for 14-deoxy-11,12didehydroandrographolide are slightly higher than andrographolide. The residues aa48-aa52 and aa150-aa200 clearly show larger deviations in RMSF with 14-deoxy-11,12didehydroandrographolide (Fig. 8c) . A visual analysis of the MDS trajectories was performed to ascertain these observations, as shown in Fig. 8d and e. In the initial conformation of 14deoxy-11,12-didehydroandrographolide bound to 3CL pro before MDS, a hydrogen bond between the carbonyl oxygen at C14 and Arg131 residue of the active site was observed. However, this hydrogen bond breaks and new hydrogen bonds were formed with other residues such as Gln109 and Thr190. Probably due to conformationally restricted bond rotation around C11-C12, these hydrogen bonds are formed less frequently, which is evident in RMSF values in these residues and ligand RMSD (Fig. 8d) . In the case of andrographolide bound to 3CL pro , the initial conformation has three hydrogen bonds between C16- . Interaction analysis of the PL pro bound to ligands during the molecular dynamics simulation; (d) equilibrated structure of 14-deoxy-11,12didehydroandrographolide bound to the PL pro before MDS production phase (green) and post-MDS production phase (red); (e) equilibrated structure of andrographolide bound to the PL pro before MDS production phase (green) and post-MDS production phase (red). . Interaction analysis of the 3CL pro bound to ligands during the molecular dynamics simulation; (d) equilibrated structure of 14-deoxy-11,12-didehydroandrographolide bound to the spike protein before MDS production phase (green) and post-MDS production phase (red); (e) equilibrated structure of andrographolide bound to the spike protein before MDS production phase (green) and post-MDS production phase (red). . Interaction analysis of the 3CL pro bound to ligands during the molecular dynamics simulation; (d) equilibrated structure of 14-deoxy-11,12-didehydroandrographolide bound to the 3CL pro before MDS production phase (green) and post-MDS production phase (red); (e) equilibrated structure of andrographolide bound to the 3CL pro before MDS production phase (green) and post-MDS production phase (red).

carbonyl oxygen-His163, C14-hydroxyl group oxygen-Cys145, and C19-hydroxyl oxygen-Thr190. During MDS's progress, some of these hydrogen bonds break, and new hydrogen bonds were formed with adjacent residues such as Ala191 and His164. However, due to conformational exibility in andrographolide around the C11-C12 bond, the ligand stabilizes and quickly gains an energetically lower conformation (Fig. 8e) . These observations suggest conformationally better stabilization of the andrographolide at the binding site of 3CL pro .

The MDS trajectories of modelled spike protein bound to both the ligands were analyzed. The protein RMSD, ligand RMSD and per amino acid residue uctuations, and RMSF were recorded (Fig. 9a-c) . The RMSD in spike protein bound to each ligand shows uctuations in the range 5 to 12Å, and it is acceptable due to the amino acid composition of the protein comprising more than 600 residues. In spike protein bound to 14-deoxy-11,12-didehydroandrographolide, an initial increase in RMSD to 10Å till 25 ns simulation was observed, which remained stable with minor deviations, thereaer till the end of simulation with RMSD of 10Å. This suggests the conformational stability of 14-deoxy-11,12-didehydroandrographolide at the binding cavity, which resulted in the system stability.

On the other hand, spike protein bound with andrographolide showed a similar trend in RMSD initially till 25 ns, which rises to around 12.5Å during 25 ns to 100 ns. The binding site residues undergo conformational change during this simulation period. Possibly, the conformational change in the residues is due to conformational exibility in the andrographolide molecule. The ligand RMSD and per residue RMSF supports this observation for andrographolide (Fig. 9b and c) . The ligand RMSD of andrographolide increases sharply during initial MDS to around 15Å until 25 ns and decreases to around 5Å until 50 ns. However, it is unable to converge to a stable RMSD aer that which suggests the major conformational changes in andrographolide and, consequently, the conformational changes in the binding site.

In contrast, the RMSD uctuations in 14-deoxy-11,12didehydroandrographolide are very subtle, with an initial rise to around 5Å, and remain stable at this RMSD with minor deviations through the rest of the simulation period, suggests a stable complex and strong binding between the protein and ligand. The RMSF in spike protein residues also supports these observations. The residues aa300-aa550 clearly shows larger deviations in RMSF of around 7 to 12Å with andrographolide (Fig. 9b) . Most of these residues belong to the binding cavity. The corresponding RMSF values in the case of 14-deoxy-11,12didehydroandrographolide for these residues range from 5 to 7Å. A visual analysis of the MDS trajectories was also performed to ascertain these observations. The initial equilibrated conformation of 14-deoxy-11,12-didehydroandrographolide bound to spike protein has a hydrogen bond between the carbonyl oxygen at C14 and Tyr585 and C19-hydroxyl group oxygen and Pro263. However, these hydrogen bond breaks and a new hydrogen bond were formed between the C14 carbonyl oxygen and Trp582. In the case of andrographolide bound to spike protein, the initial equilibrated conformation shows a hydrogen bond between C3 hydroxyl group hydrogen and Asn285 and Gln282 residues. Due to conformational exibility in andrographolide, during the production phase of MDS, these hydrogen bonds break. However, no new hydrogen bond formation was observed towards the end of the simulation.

14-Deoxy-11,12-didehydroandrographolide scored higher druglikeness score, i.e., À0.52 compared to andrographolide, which was computed based on molecular weight, number of hydrogen bond donor, number of hydrogen bond acceptor, and log P value (Table 5 ). It has inuenced both compounds' pharmacokinetic characters by affecting absorption, distribution, metabolism, excretion, and toxicity (Fig. 10 ).

The present study investigated one of the active biomolecules, andrographolide, and its derivative i.e., 14-deoxy-11,12didehydroandrographolide from Andrographis paniculate, for the regulation of the proteins/immunomodulatory pathways and also assess their binding affinity with three targets, i.e., 3CL pro , PL pro , and spike protein involved in the COVID infection. Further, we investigated the drug-likeness property of both molecules in which andrographolide scored a lower druglikeness character compared to 14-deoxy-11,12-didehydroandrographolide. However, on looking at the binding affinity and number of hydrogen bond interactions, andrographolide showed a higher interaction towards the selected targets. It suggests fewer modications could be made in the andrographolide moiety to enhance its drug-likeness property without altering the binding affinity towards the targeted proteins. Further, some modications could be made in 14-deoxy-11,12didehydroandrographolide to amplify the hydrogen bond interaction and eventually increase the higher binding affinity with targeted proteins. Subjects with lower immunity system are more prone to infection with COVID 19 due to compromised immunity system, 61 which is well proven in subjects suffering from an infectious and non-infectious disease(s). In this case, it is crucial to enhance the subjects' immunity to minimize the probability of viral infection. In the present study, via the enrichment analysis, we identied multiple pathways involved in boosting the immune system, which is modulated by andrographolide and 14deoxy-11,12-didehydroandrographolide.

In the present study, we identied potential modulation of few pathways, directly or indirectly linked with the modulation of the immune system, i.e., chemokine signaling pathway, Rap1 signaling pathway, cytokine-cytokine receptor interaction, MAPK signaling pathway, NF-kappa B signaling pathway, RAS signaling pathway, p53 signaling pathway, HIF-1 signaling pathway, and natural killer cell-mediated cytotoxicity. Among the above pathways, chemokine signaling pathways, Rap1 signaling pathway, and cytokine-cytokine receptor interaction are the choice of interest pathways as they are directly linked with the immune system's regulation, and they scored minimum false discovery rate compared to the rest of the pathways. The chemokine signaling pathway was modulated by andrographolide and 14deoxy-11,12-didehydroandrographolide, which could control the migration of immune cells in tissues. 62 Further, the Rap1 signaling pathway is involved in activating three secondary messengers, i.e., cAMP, calcium, and diacylglycerol, 63 which are needed in the signaling of cell position during viral infections; modulated by andrographolide by regulating ID1, PRKCA, RAC1, RAP1A, and RHOA and by 14-deoxy-11,12didehydroandrographolide by regulating FLT1, ID1, PRKCA, RAC1, RAP1A, and RHOA. Similarly, the KEGG database has recorded cytokine-cytokine receptor interaction as an entry (hsa04060) in various auto-immune disorders. Since COVID-19 has a more risk over the infections on the altered immune system of subjects, modulation of this pathway could be benecial in them, modulated by andrographolide and 14-deoxy-11,12didehydroandrographolide. Further, the MAPK signaling pathway has been identied to play an essential role in the functioning of T lymphocytes, 64 was observed to be modulated by andrographolide and 14-deoxy-11,12-didehydroandrographolide. Additionally, other pathways like NF-kappa B signaling pathway, ras signaling pathway, p53 signaling pathway, HIF-1 signaling pathway, and natural killer cell-mediated cytotoxicity are also regulated which has been well reported to be involved in the modulation of the immune system.

In COVID-19 infection, the n-CoV-2 binds to ACE-2, enters into the cell, and starts deregulating the intracellular functions by altering the normal homeostatic stimulus. 65 Hence, it is needed to control the components by binding over them or responding towards the stimulus 3CL pro , or at least to minimize its effect by controlling the intracellular cascade initiated by the viral infection. Further, gene ontology enrichment analysis identied andrographolide and 14-deoxy-11,12didehydroandrographolide to target the intracellular components, binding capacity towards various proteins as a molecular function and responder towards stimulus which could be the possible action of these two agents over the viral infection.

A concept of modulation of multiple proteins by a single molecule is the choice of research interest in identifying the lead hit towards respective targets. Further, andrographolide has been previously reported to possess anti-viral properties. 29 Hence, based on the same concept, andrographolide and 14deoxy-11,12-didehydroandrographolide may also possess the anti-viral efficacy over COVID-19, which kindled us evaluating the binding affinity of these bioactives over PL pro , 3CL pro , and spike protein. Although the drug-likeness score model predicted 14-deoxy-11,12-didehydroandrographolide to behave like a drug based on "Rule of Five", the binding affinity and number of hydrogen bond interactions reected andrographolide to act more on three proteins of COVID-19 i.e. PL pro , 3CL pro , and spike protein.

The present study utilized the system biology approach to investigate the andrographolide and 14-deoxy-11,12-didehydroandrographolide against COVID-19 by modulating the multiple pathways in which the chemokine signaling pathway could be a choice of interest as it is directly linked in modulating the immune response and scored the lowest false discovery rate. Further, andrographolide could possess higher importance than 14-deoxy-11,12-didehydroandrographolide as it scored higher interaction with the targeted proteins of COVID-19. However, the present ndings are completely based on computer simulations and database query; the outcome may vary based on processing units and database updates; suggests the necessity to conrm the present ndings via well-designed experimental protocols and is the future scope of present ndings.

MMW and BMP: conceptualization, supervision, investigation; RVC and RBP: molecular docking and dynamics studies; PK and YDD: network pharmacology and analysis; SNG and BV: methodology, soware data analysis; SSG and RBP: writing-original dra preparation; MMW and SSG: formal analysis, visualization, reviewing and editing.

This study doesn't include any animal or human study.

This research did not receive any specic grant from funding agencies in the public, commercial, or not-for-prot sectors.

The authors declare no competing interests.

of the International

MAP Kinase Signaling Protocols

Authors are thankful to the Director-General, Central Council for Research in Ayurvedic Sciences (Ministry of AYUSH, Govt. of India) New Delhi and KLE College of Pharmacy, Belagavi, for providing necessary facilities.