LitCovid-PD-HP

PMC:7784829 / 25475-25480 JSON TXT

Virtual screening of curcumin and its analogs against the spike surface glycoprotein of SARS-CoV-2 and SARS-CoV A. Patel et al. Journal of Biomolecular Structure and Dynamics Abstract Abstract COVID-19, a new pandemic caused by SARS-CoV-2, was first identified in 2019 in Wuhan, China. The novel corona virus SARS-CoV-2 and the 2002 SARS-CoV have 74% identity and use similar mechanisms to gain entry into the cell. Both the viruses enter the host cell by binding of the viral spike glycoprotein to the host receptor, angiotensin converting enzyme 2 (ACE2). Targeting entry of the virus has a better advantage than inhibiting the later stages of the viral life cycle. The crystal structure of the SARS-CoV (6CRV: full length S protein) and SARS-CoV-2 Spike proteins (6M0J: Receptor binding domain, RBD) was used to determine potential small molecule inhibitors. Curcumin, a naturally occurring phytochemical in Curcuma longa, is known to have broad pharmacological properties. In the present study, curcumin and its derivatives were docked, using Autodock 4.2, onto the 6CRV and 6M0J to study their capability to act as inhibitors of the spike protein and thereby, viral entry. The curcumin and its derivatives displayed binding energies, ΔG, ranging from −10.98 to −5.12 kcal/mol (6CRV) and −10.01 to −5.33 kcal/mol (6M0J). The least binding energy was seen in bis-demethoxycurcumin with: ΔG = −10.98 kcal/mol (6CRV) and −10.01 kcal/mol (6M0J). A good binding energy, drug likeness and efficient pharmacokinetic parameters suggest the potential of curcumin and few of its derivatives as SARS-CoV-2 spike protein inhibitors. However, further research is necessary to investigate the ability of these compounds as viral entry inhibitors. Communicated by Ramaswamy H. Sarma Introduction Coronaviruses are enveloped positive-sense single stranded RNA viruses that belong to the family coronaviridae. They usually infect birds and mammals and cause mild respiratory diseases (Kahn & McIntosh, 2005). However, in the recent past, these viruses have caused lethal endemics such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) endemics (Kahn & McIntosh, 2005). In December 2019, Wuhan city in China became the centre of a surge in cases of pneumonia by an unknown cause (Zhu et al., 2020). A novel corona virus, called SARS-CoV-2, was isolated from these pneumonia patients in January 2020 and the disease was called the Coronavirus disease 2019 (COVID-19) by the World Health Organization in February, 2020 (Gorbalenya et. al, 2020; Jiang et al., 2020; Who, 2020). Even though the epidemic may have started from a zoonotic transmission in a seafood market, that also sold wild animals, it became clear that this disease was transmitted from person to person (Li et al., 2020). The clinical characteristics of this disease are broad, constituting asymptomatic infections, mild respiratory disease, severe pneumonia with respiratory failure and even death (Chen et al., 2020; Huang et al., 2020; Wang et al., 2020). The COVID-19 pandemic poses a significant challenge to global public health (Phelan et al., 2020), appealing for the development of safe and successful prophylactics and therapies against infection of its causative agent, the SARS-CoV-2 virus. SARS-CoV-2 has a genome of ∼30 kilobases, which codes for multiple structural and non-structural proteins (Chan et al., 2020). The structural proteins, present on the surface of the mature virion, include the spike protein, the membrane protein, the envelope protein and the nucleocapsid protein (Chan et al., 2020). The Spike protein, of the beta coronaviruses SARS-CoV-2 and SARS-CoV, enables the attachment of the virus to the cells of the lower respiratory tract of humans to gain entry into the lung tissue (Hoffmann et al., 2020). Apart from attachment, the spike glycoprotein also appears to play a role in fusion and entry of the virus into the host (Chen et al., 2020; Hoffmann et al., 2020). The spike protein, of the novel coronavirus, utilizes the SARS-CoV receptor, ACE2 for entry (Chen et al., 2020) and is primed by the cellular protease, TMPRSS2 (Hoffmann et al., 2020). A serine protease inhibitor, which can act on TMPRSS2, has been shown to inhibit novel coronavirus entry (Hoffmann et al., 2020). Therefore, the spike protein of the novel coronavirus is a good drug target and identifying small molecules that bind to S protein would inhibit viral recognition of host cells and disrupt viral-host interactions. Curcumin, a naturally occurring phytochemical and principal component of Curcuma longa, has exhibited broad pharmacological properties including antioxidant, anti-inflammatory, anti-cancer and anti-viral effects (Khor et al., 2019; Kocaadam & Şanlier, 2017; Lal et al., 2016; Wiggers et al., 2017). Curcumin and its derivatives, due to its rich conventional medicinal interest, has undergone comprehensive in vitro and in vivo studies. It has, therefore, been associated with more than 100 cellular targets, including cytokines, proteins, transcription factors, and receptors. Previous studies have shown the potential of curcumin as a treatment against Influenza A virus infection, by an effect mediated by modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Curcumin has also been shown to have anti neuraminidase (NA) activity for the influenza virus NA protein (Richart et al., 2018). Therefore, in the present study, curcumin and its derivatives were docked onto the spike protein of the SARS-CoV and the SARS-CoV-2 to predict the binding interactions. We found that curcumin and few of its derivatives showed promising results to be potential spike protein inhibitors. One of the derivatives, bis-demethoxycurcumin, showed the best binding affinity to the spike protein of both the SARS-CoV and the novel corona virus, SARS-CoV-2. The possibility for few of the curcumin derivatives, that showed good binding affinity, could be tested for further therapeutic use against COVID-19. Experimental section Sequence analysis The sequences of the SARS-CoV-2 and SARS-CoV were downloaded from National Center for Biotechnology Information. Multiple sequence analysis and pairwise sequence identity was determined using the Clustal Omega server at the European Bioinformatics Institute (Sievers et al., 2011). Preparation of coordinate file The X-ray crystal structure of spike surface glycoprotein of SARS-CoV (PDB entry: 6CRV, resolution = 3.2 Å) and co-crystallized structure of SARS-CoV-2 RBD with human ACE2 Protein (PDB entry: 6M0J, resolution = 2.45 Å) was retrieved from Protein Data Bank (https:/www.rscb.org/pdb). The protein structure was prepared using the Discovery Studio Visualizer (version 3.1) and AutoDock Tools (ADT; version 1.5.4) through different steps viz. removal of water molecules and co-crystallized ligand, addition of missing hydrogen atoms, addition of Gasteiger-Marsili and Kollman charges, merging of non-polar hydrogens, and assignment of rotatable bonds. The file was then saved in pdbqt file format for further analysis. Preparation of ligands The chemical structures of curcumin derivatives were constructed using Chem3D 15.0 module of ChemOffice 15.0 and saved in PDB format. The structures were optimized using “Prepare Ligands” in the AutoDock 4.2, flexible torsions were assigned and the acyclic dihedral angles were allowed to rotate freely. The file was then saved as pdbqt file format for further analysis. Molecular docking Molecular docking simulations were conducted on the curcumin derivatives using the AutoDock 4.2 to get insight into their binding preferences within the active site of the receptor. The molecular docking simulations were performed on the PC based machines running on Windows 7 (x86) as operating system. The software included MGL tools 1.5.4 based AutoDock 4.2 (www.scripps.edu) which uses Python 2.7 language - Cygwin C:\program (www.cygwin.com) and Python 2.5 (www.python.com) (Morris et al., 2009). Docking methodology The flexible docking was performed using the refined spike protein of SARS-CoV (6CRV) and RBD domain of SARS-CoV-2 (6M0J). The grid maps of the interaction energies of various atom types were pre-calculated using AutoGrid 4.2. In each docking for spike surface glycoprotein and spike RBD, a grid box was created using a grid map of 45 × 45 × 45 points, 60 × 60 × 60 points with grid spacing of 0.375 Å and 0.420 Å respectively. The grid maps were centred on the corresponding ligand binding site within the protein structure. Lamarckian Genetic Algorithm (LGA) was adopted to perform docking simulations using the following default parameters, viz. 100 independent runs with step sizes of 0.2 Å for translations and 5 Å for orientations and torsions, an initial population of random individuals with a population size of 150 individuals, a maximum of 2.5 × 106 energy evaluations, maximum number of generations of 27,000; mutation and crossover rates of 0.02 and 0.8 respectively and an elitism value of 1. All the computations were carried out on Cygwin and was used to generate both grid parameter file (.gpf file) and docking parameter file (.dpf file) for each ligand. The docked conformations of each ligand were ranked into clusters based on the binding energy and the top ranked conformations were used for further study. The pose with the lowest ΔG-score was considered the best fitted one and was further analyzed for Ligand-receptor interactions. In-silico drug-likeness and pharmacokinetic property prediction The in-silico prediction studies were performed, using pkCSM online prediction platforms (Pires et al., 2015), to assess the theoretical pharmacokinetic parameters of the ligands to predict the drug-likeness of ligands. The software calculated pharmaceutically relevant properties such as H-bond donor, H-bond acceptor, octanol-water partition coefficient (LogP), surface area, and number of rotatable bonds. In addition the ADME parameters of the curcumin analogs such as water solubility, Caco2 permeability, human intestinal absorption, skin permeability, P-glycoprotein I and II inhibition, volume of distribution, fraction of unbound drug, Blood Brain Barrier and CNS permeability, cytochrome P450 (CYP3A4 and CYP2C9 inhibition) inhibition, total clearance, action as renal OCT2 (organic cation transporter 2) substrate were analysed. Results Sequence alignment of the SARS-CoV-2 and SARS-CoV spike protein sequences Prior to performing the docking studies the multiple sequence alignment was carried out for the spike proteins of the novel corona virus, SARS-CoV-2 and the 2002 SARS-CoV (Supplementary Figure 1 and Table 1). These results show that there is a very high percentage identity (greater than 99%) among the spike protein of the novel corona virus SARS-CoV-2 strains. There was around 74% identity of the SARS-CoV-2 spike proteins with the 2002 SARS-CoV. Therefore, the SARS-CoV-2 spike protein is very much closer to SARS-CoV than to the other corona viruses as shown in other studies (Ahmed et al., 2020). Figure 1. The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins. Table 1. Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV). Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100 Molecular docking studies The three-dimensional structure of the SARS-CoV spike protein and SARS-CoV-2 Spike protein RBD has been solved. The PDB files of the SARS-CoV spike protein (6CRV) and the Spike protein RBD of the SARS-CoV-2 (6M0J) was downloaded from the PDB databank. These structures were used to predict the ability of the binding of curcumin and its derivatives to the spike proteins of the two corona viruses. Curcumin and 24 of its derivatives (Figure 1 and Supplementary Figure 2) were docked on to 6CRV and 6M0J. Tables 2 and 3 gives the binding energies of curcumin and its derivatives with 6CRV (binding energies ranged from −10.98 to −5.12 kcal/mol) and with 6M0J (binding energies ranged −10.01 to −5.33 kcal/mol) respectively. Visual examination of the computationally docked optimal binding poses of curcumin and its derivatives on 6CRV and 6MOJ revealed the important role of various types of interactions viz. hydrogen bonding and hydrophobic interactions, including π–π stacking, π–cation, and π–σ interactions in the stability of the binding of the curcumin/derivatives to the spike protein (6CRV and 6M0J). All compounds showed one or more hydrogen bonds with 6CRV and 6M0J except a few. Figure 2. Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein. Table 2. Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –10.1 –0.14 –8.24 –10.9724 BHBC –8.28 –0.13 –7.87 –3.42239 MNC –8.01 –0.43 –6.98 –8.01452 BDMC –10.98 –0.18 –8.94 –14.0467 1 –5.12 –0.25 –4.17 0 2 –8.55 –0.41 –6.56 –18.4317 3 –6.56 –0.21 –4.92 –11.1148 4 –8.69 –0.13 –6.94 –19.1539 5 –7.03 –0.19 –5.93 –8.47153 6 –6.59 –0.36 –5.12 –7.41531 7 –7.96 –0.21 –6.14 –7.98548 8 –7.55 –0.19 –5.14 –6.14537 9 –6.91 –0.21 –5.18 0 10 –7.75 –0.21 –6.14 –6.73473 11 –8.02 –0.19 –6.19 –17.2465 12 –7.85 –0.38 –6.25 –14.6321 13 –5.94 –0.29 –4.12 –11.3576 14 –8.75 –0.25 –7.19 –21.1412 15 –7.87 –0.38 –7.26 –0.473228 16 –8.47 –0.25 –6.88 –8.458912 DMC –7.89 –0.21 –6.41 –10.47057 FAC –7.17 –0.36 –5.15 –9.14873 IBC –7.91 –0.23 –7.17 –6.90341 IVC –7.43 –0.44 –6.67 –19.9864 SYC –6.82 –0.26 –5.45 –9.44678 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Table 3. Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –9.81 –0.15 –8.35 –15.1512 BHBC −5.33 –0.13 –7.96 0 MNC –7.96 –0.27 –7.38 –4.24687 BDMC –10.01 –0.12 –9.18 –13.9824 1 –7.58 –0.36 –6.19 –18.3489 2 –8.81 –0.31 –8.38 1.2654 3 –6.14 –0.34 –5.45 –6.95102 4 –8.88 –0.18 –7.98 –14.0321 5 –6.23 –0.26 –5.88 –3.5891 6 –7.16 –0.37 –6.15 –10.1688 7 –6.89 –0.25 –5.98 –9.12515 8 –5.49 –0.12 –3.90 –5.92265 9 –6.92 –0.18 –6.22 –6.96719 10 –7.03 –0.24 –6.33 –7.0 11 –6.88 –0.27 –6.23 –6.44363 12 –7.95 –0.41 –5.70 –2.5581 13 –6.11 –0.39 –5.42 –7.34943 14 –7.45 –0.19 –6.40 –10.5367 15 –5.74 –0.39 –5.51 –2.36748 16 –7.83 –0.27 –7.48 –6.64321 DMC –5.98 –0.19 –5.38 –5.99913 FAC –5.85 –0.38 –5.25 –6.46613 IBC –6.65 –0.15 –6.20 –1.52841 IVC –8.96 –0.43 –7.73 –12.2707 SYC –5.87 –0.29 –5.43 –4.4411 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. In 6CRV, it was found that ARG99, GLY100, TRP 101, ILE116, ILE117, ASN118, VAL123, MET144, SER165, ASP166, PHE168, ARG183, PHE185, LYS198, ILE219, PHE220 were the major amino acids involved in hydrogen bonding, hydrophobic and van der Waals interaction with most of the curcumin compounds. However, majority of the curcumin compounds exhibited hydrogen bonding with ARG99, MET144, LYS198 except compound 1 and 9. In addition, all the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) interactions with ASP166, VAL123, PHE185 and van der Waals interaction with GLY100, TRP 101, ILE117, ASN118, ARG183, ILE116, PHE168, SER165, PHE220, ILE219. All compounds showed one or more hydrogen bonding interactions with spike RBD of the SARS-CoV-2 (6M0J) except 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin. It was found that ARG355, TYR396, PRO426, ASP428, THR430, GLY431, PHE429, PHE 464, GLU516, SER514, LEU517 and PHE515 were the major amino acids involved in hydrogen bonding hydrophobic and van der Waals interaction with the all of curcumin compounds. However, majority of the curcumin compounds exhibit hydrogen bonding with ARG355, LEU517 and THR430. In addition, the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) and van der Waals interactions with PHE464, GLY431, GLU516, SER514, PHE515, TYR396, PRO426, ASP428 and PHE429. The curcumin derivative with the best binding affinity was Bis-demethoxycurcumin for both 6CRV (ΔG = −10.98 kcal/mol) and 6M0J (ΔG = −10.01 kcal/mol). The native compound, curcumin, demonstrated a binding energy of −9.81 kcal/mol for 6M0J and −10.1 kcal/mol for 6CRV. Binding mode of Bis-demethoxycurcumin with SARS-CoV Spike protein and SARS-CoV-2 Spike protein RBD is shown in Figure 2 and 3, respectively. Figure 3. Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD. In-silico drug-likeness and pharmacokinetic property prediction All the structures used for docking were analysed for in silico drug-likeness based on the Lipinski’s rules using pkCSM server (Table 4). The lipophilicity (expressed as LogP) predicted for all the compounds were found to be well above the traditionally cut-off value of 5 used for drug design. Curcumin and its derivatives, used in this study, show suitable MW values (MW < 500) essential for a successful penetration through biological membranes. The surface area (SA) for all the compounds was observed to be in the range 115.89 − 240.65 Å2 which is well within the limit. All compounds, except 5-di-tert-butyl-4-hydroxybenzaldehyde curcumin (BHBC), 4-methoxy-1-naphthaldehyde curcumin (MNC), Syringaldehyde curcumin (SYC) and compound-16, fall into the appropriate range indicating good bioavailability of the candidate molecule. The number of hydrogen bond acceptors (HBA, ≤10) and donors (HBD, ≤5) for all the compounds were in accordance with the Lipinski’s rule of five. Table 4. In-silico prediction of drug-likeness for curcumin derivatives[a]. Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3 [a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Additionally, a variety of key ADMET (Absorption, Distribution, Metabolism and Excretion) properties have also been calculated with the aid of pkCSMserver. The results are listed in Table 5. All curcumin derivatives showed moderate to high water solubility ranging from −2.91 log mol/L (BHBC) to −6.99 log mol/L (compound-12), in addition to high Caco-2 permeability (permeation > 0.90) except compound BHBC and MNC (permeation = 0.67 and 0.43 respectively) which showed moderate permeability. Intestinal absorption (IA) has been found to be greater than 85% indicating good permeation across the intestinal membrane. Further, all curcumin compounds showed good permeation through skin (permeation > −2.5). Additionally, all curcumin compounds showed no inhibition towards P-glycoprotein I and P-glycoprotein II, except compounds BHBC, MNC, compound-13 and Ferrulic acid curcumin (FAC) which demonstrated inhibition towards P-glycoprotein I. Furthermore, all curcumin compounds showed poor BBB permeability and moderate CNS permeability except Isovanillin curcumin (IVC) and SYC. All showed inhibition towards the metabolizing enzyme CYP3A4 except BHBC, compounds-8, -11, -14, -15 and -16, while except compound-15, all showed inhibition towards CYP2C9. All curcumin derivatives were found to show the total clearance in the range 0.10 log mL/min/kg (compound-1 and -4) to 1.01 log mL/min/kg (compound-15). Further, all curcumin derivatives except compound 15 were found to act as OCT2 substrate, thus indicating that these compounds will not have any adverse interactions and no negative effect on renal clearance. Table 5. In-silico ADME prediction for curcumin derivatives[a]. Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No [a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Conclusion and discussion The spike proteins of corona viruses are essential for entry of the virus into the target cells. The spike protein exists as a trimer on the surface of the virus with one of the monomer in up conformation and the other two in down conformation (Wrapp et al., 2020). The N-terminal region (S1) of the S protein is important for binding to the cellular receptor ACE2 (Hoffmann et al., 2020; Tai et al., 2020). The S protein undergoes priming by cellular protease, TMPRSS2 and the S2 region of the protein is responsible for fusion of viral and cellular membrane (Hoffmann et al., 2020). Therefore, identifying therapeutics for the S protein of the novel corona virus could potentially target the critical process of entry and fusion of the virus. Curcumin and its derivatives are known for their many biological activities, one of them is its antiviral activity. Therefore, here, we examined the potential of curcumin, and its derivatives, to bind to the SARS-CoV and SARS-CoV-2 spike protein. From our computational molecular docking approach (using auto dock 4.2, PDB ID - 6CRV, 6M0J) and in-silico ADMET tool, we predicted that Bis-demethoxy curcumin, compound-4 and compound-2 were the most recommended curcumin compounds which bind to RBD domain of the SARS-CoV-2 Spike protein and SARS-CoV spike protein efficiently in in silico studies. Curcumin has earlier shown to have specific inhibitory effect on the NA activity in influenza virus (Chen et al., 2013; Richart et al., 2018) and modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Till date, Bis-demethoxy curcumin has shown to have anti-cancer and hepato-protective activities (Kumaravel et al., 2013; Rajagopalan et al., 2010). This is the first time; it is predicted to have a potential anti-viral activity. Therefore, these curcumin derivatives, which have been predicted to bind to the SARS-CoV-2 spike protein, could be explored as probable inhibitors of COVID-19 spike protein through experimental studies. Supplementary Material Supplementary_Figure.pdf Click here for additional data file. Acknowledgements We acknowledge the support rendered by the Department of Biochemistry, Indian Academy Degree College - Autonomous, Bengaluru and Ramanbhai Patel College of Pharmacy, Charotar University of Science and Technology, Gujarat for their support in carrying out this work. Disclosure statement The authors declare that they have no conflict of interests

Document structure show

article-title	Virtual screening of curcumin and its analogs against the spike surface glycoprotein of SARS-CoV-2 and SARS-CoV
alt-title	A. Patel et al.
alt-title	Journal of Biomolecular Structure and Dynamics
abstract	Abstract COVID-19, a new pandemic caused by SARS-CoV-2, was first identified in 2019 in Wuhan, China. The novel corona virus SARS-CoV-2 and the 2002 SARS-CoV have 74% identity and use similar mechanisms to gain entry into the cell. Both the viruses enter the host cell by binding of the viral spike glycoprotein to the host receptor, angiotensin converting enzyme 2 (ACE2). Targeting entry of the virus has a better advantage than inhibiting the later stages of the viral life cycle. The crystal structure of the SARS-CoV (6CRV: full length S protein) and SARS-CoV-2 Spike proteins (6M0J: Receptor binding domain, RBD) was used to determine potential small molecule inhibitors. Curcumin, a naturally occurring phytochemical in Curcuma longa, is known to have broad pharmacological properties. In the present study, curcumin and its derivatives were docked, using Autodock 4.2, onto the 6CRV and 6M0J to study their capability to act as inhibitors of the spike protein and thereby, viral entry. The curcumin and its derivatives displayed binding energies, ΔG, ranging from −10.98 to −5.12 kcal/mol (6CRV) and −10.01 to −5.33 kcal/mol (6M0J). The least binding energy was seen in bis-demethoxycurcumin with: ΔG = −10.98 kcal/mol (6CRV) and −10.01 kcal/mol (6M0J). A good binding energy, drug likeness and efficient pharmacokinetic parameters suggest the potential of curcumin and few of its derivatives as SARS-CoV-2 spike protein inhibitors. However, further research is necessary to investigate the ability of these compounds as viral entry inhibitors. Communicated by Ramaswamy H. Sarma
title	Abstract
p	COVID-19, a new pandemic caused by SARS-CoV-2, was first identified in 2019 in Wuhan, China. The novel corona virus SARS-CoV-2 and the 2002 SARS-CoV have 74% identity and use similar mechanisms to gain entry into the cell. Both the viruses enter the host cell by binding of the viral spike glycoprotein to the host receptor, angiotensin converting enzyme 2 (ACE2). Targeting entry of the virus has a better advantage than inhibiting the later stages of the viral life cycle. The crystal structure of the SARS-CoV (6CRV: full length S protein) and SARS-CoV-2 Spike proteins (6M0J: Receptor binding domain, RBD) was used to determine potential small molecule inhibitors. Curcumin, a naturally occurring phytochemical in Curcuma longa, is known to have broad pharmacological properties. In the present study, curcumin and its derivatives were docked, using Autodock 4.2, onto the 6CRV and 6M0J to study their capability to act as inhibitors of the spike protein and thereby, viral entry. The curcumin and its derivatives displayed binding energies, ΔG, ranging from −10.98 to −5.12 kcal/mol (6CRV) and −10.01 to −5.33 kcal/mol (6M0J). The least binding energy was seen in bis-demethoxycurcumin with: ΔG = −10.98 kcal/mol (6CRV) and −10.01 kcal/mol (6M0J). A good binding energy, drug likeness and efficient pharmacokinetic parameters suggest the potential of curcumin and few of its derivatives as SARS-CoV-2 spike protein inhibitors. However, further research is necessary to investigate the ability of these compounds as viral entry inhibitors.
p	Communicated by Ramaswamy H. Sarma
body	Introduction Coronaviruses are enveloped positive-sense single stranded RNA viruses that belong to the family coronaviridae. They usually infect birds and mammals and cause mild respiratory diseases (Kahn & McIntosh, 2005). However, in the recent past, these viruses have caused lethal endemics such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) endemics (Kahn & McIntosh, 2005). In December 2019, Wuhan city in China became the centre of a surge in cases of pneumonia by an unknown cause (Zhu et al., 2020). A novel corona virus, called SARS-CoV-2, was isolated from these pneumonia patients in January 2020 and the disease was called the Coronavirus disease 2019 (COVID-19) by the World Health Organization in February, 2020 (Gorbalenya et. al, 2020; Jiang et al., 2020; Who, 2020). Even though the epidemic may have started from a zoonotic transmission in a seafood market, that also sold wild animals, it became clear that this disease was transmitted from person to person (Li et al., 2020). The clinical characteristics of this disease are broad, constituting asymptomatic infections, mild respiratory disease, severe pneumonia with respiratory failure and even death (Chen et al., 2020; Huang et al., 2020; Wang et al., 2020). The COVID-19 pandemic poses a significant challenge to global public health (Phelan et al., 2020), appealing for the development of safe and successful prophylactics and therapies against infection of its causative agent, the SARS-CoV-2 virus. SARS-CoV-2 has a genome of ∼30 kilobases, which codes for multiple structural and non-structural proteins (Chan et al., 2020). The structural proteins, present on the surface of the mature virion, include the spike protein, the membrane protein, the envelope protein and the nucleocapsid protein (Chan et al., 2020). The Spike protein, of the beta coronaviruses SARS-CoV-2 and SARS-CoV, enables the attachment of the virus to the cells of the lower respiratory tract of humans to gain entry into the lung tissue (Hoffmann et al., 2020). Apart from attachment, the spike glycoprotein also appears to play a role in fusion and entry of the virus into the host (Chen et al., 2020; Hoffmann et al., 2020). The spike protein, of the novel coronavirus, utilizes the SARS-CoV receptor, ACE2 for entry (Chen et al., 2020) and is primed by the cellular protease, TMPRSS2 (Hoffmann et al., 2020). A serine protease inhibitor, which can act on TMPRSS2, has been shown to inhibit novel coronavirus entry (Hoffmann et al., 2020). Therefore, the spike protein of the novel coronavirus is a good drug target and identifying small molecules that bind to S protein would inhibit viral recognition of host cells and disrupt viral-host interactions. Curcumin, a naturally occurring phytochemical and principal component of Curcuma longa, has exhibited broad pharmacological properties including antioxidant, anti-inflammatory, anti-cancer and anti-viral effects (Khor et al., 2019; Kocaadam & Şanlier, 2017; Lal et al., 2016; Wiggers et al., 2017). Curcumin and its derivatives, due to its rich conventional medicinal interest, has undergone comprehensive in vitro and in vivo studies. It has, therefore, been associated with more than 100 cellular targets, including cytokines, proteins, transcription factors, and receptors. Previous studies have shown the potential of curcumin as a treatment against Influenza A virus infection, by an effect mediated by modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Curcumin has also been shown to have anti neuraminidase (NA) activity for the influenza virus NA protein (Richart et al., 2018). Therefore, in the present study, curcumin and its derivatives were docked onto the spike protein of the SARS-CoV and the SARS-CoV-2 to predict the binding interactions. We found that curcumin and few of its derivatives showed promising results to be potential spike protein inhibitors. One of the derivatives, bis-demethoxycurcumin, showed the best binding affinity to the spike protein of both the SARS-CoV and the novel corona virus, SARS-CoV-2. The possibility for few of the curcumin derivatives, that showed good binding affinity, could be tested for further therapeutic use against COVID-19. Experimental section Sequence analysis The sequences of the SARS-CoV-2 and SARS-CoV were downloaded from National Center for Biotechnology Information. Multiple sequence analysis and pairwise sequence identity was determined using the Clustal Omega server at the European Bioinformatics Institute (Sievers et al., 2011). Preparation of coordinate file The X-ray crystal structure of spike surface glycoprotein of SARS-CoV (PDB entry: 6CRV, resolution = 3.2 Å) and co-crystallized structure of SARS-CoV-2 RBD with human ACE2 Protein (PDB entry: 6M0J, resolution = 2.45 Å) was retrieved from Protein Data Bank (https:/www.rscb.org/pdb). The protein structure was prepared using the Discovery Studio Visualizer (version 3.1) and AutoDock Tools (ADT; version 1.5.4) through different steps viz. removal of water molecules and co-crystallized ligand, addition of missing hydrogen atoms, addition of Gasteiger-Marsili and Kollman charges, merging of non-polar hydrogens, and assignment of rotatable bonds. The file was then saved in pdbqt file format for further analysis. Preparation of ligands The chemical structures of curcumin derivatives were constructed using Chem3D 15.0 module of ChemOffice 15.0 and saved in PDB format. The structures were optimized using “Prepare Ligands” in the AutoDock 4.2, flexible torsions were assigned and the acyclic dihedral angles were allowed to rotate freely. The file was then saved as pdbqt file format for further analysis. Molecular docking Molecular docking simulations were conducted on the curcumin derivatives using the AutoDock 4.2 to get insight into their binding preferences within the active site of the receptor. The molecular docking simulations were performed on the PC based machines running on Windows 7 (x86) as operating system. The software included MGL tools 1.5.4 based AutoDock 4.2 (www.scripps.edu) which uses Python 2.7 language - Cygwin C:\program (www.cygwin.com) and Python 2.5 (www.python.com) (Morris et al., 2009). Docking methodology The flexible docking was performed using the refined spike protein of SARS-CoV (6CRV) and RBD domain of SARS-CoV-2 (6M0J). The grid maps of the interaction energies of various atom types were pre-calculated using AutoGrid 4.2. In each docking for spike surface glycoprotein and spike RBD, a grid box was created using a grid map of 45 × 45 × 45 points, 60 × 60 × 60 points with grid spacing of 0.375 Å and 0.420 Å respectively. The grid maps were centred on the corresponding ligand binding site within the protein structure. Lamarckian Genetic Algorithm (LGA) was adopted to perform docking simulations using the following default parameters, viz. 100 independent runs with step sizes of 0.2 Å for translations and 5 Å for orientations and torsions, an initial population of random individuals with a population size of 150 individuals, a maximum of 2.5 × 106 energy evaluations, maximum number of generations of 27,000; mutation and crossover rates of 0.02 and 0.8 respectively and an elitism value of 1. All the computations were carried out on Cygwin and was used to generate both grid parameter file (.gpf file) and docking parameter file (.dpf file) for each ligand. The docked conformations of each ligand were ranked into clusters based on the binding energy and the top ranked conformations were used for further study. The pose with the lowest ΔG-score was considered the best fitted one and was further analyzed for Ligand-receptor interactions. In-silico drug-likeness and pharmacokinetic property prediction The in-silico prediction studies were performed, using pkCSM online prediction platforms (Pires et al., 2015), to assess the theoretical pharmacokinetic parameters of the ligands to predict the drug-likeness of ligands. The software calculated pharmaceutically relevant properties such as H-bond donor, H-bond acceptor, octanol-water partition coefficient (LogP), surface area, and number of rotatable bonds. In addition the ADME parameters of the curcumin analogs such as water solubility, Caco2 permeability, human intestinal absorption, skin permeability, P-glycoprotein I and II inhibition, volume of distribution, fraction of unbound drug, Blood Brain Barrier and CNS permeability, cytochrome P450 (CYP3A4 and CYP2C9 inhibition) inhibition, total clearance, action as renal OCT2 (organic cation transporter 2) substrate were analysed. Results Sequence alignment of the SARS-CoV-2 and SARS-CoV spike protein sequences Prior to performing the docking studies the multiple sequence alignment was carried out for the spike proteins of the novel corona virus, SARS-CoV-2 and the 2002 SARS-CoV (Supplementary Figure 1 and Table 1). These results show that there is a very high percentage identity (greater than 99%) among the spike protein of the novel corona virus SARS-CoV-2 strains. There was around 74% identity of the SARS-CoV-2 spike proteins with the 2002 SARS-CoV. Therefore, the SARS-CoV-2 spike protein is very much closer to SARS-CoV than to the other corona viruses as shown in other studies (Ahmed et al., 2020). Figure 1. The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins. Table 1. Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV). Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100 Molecular docking studies The three-dimensional structure of the SARS-CoV spike protein and SARS-CoV-2 Spike protein RBD has been solved. The PDB files of the SARS-CoV spike protein (6CRV) and the Spike protein RBD of the SARS-CoV-2 (6M0J) was downloaded from the PDB databank. These structures were used to predict the ability of the binding of curcumin and its derivatives to the spike proteins of the two corona viruses. Curcumin and 24 of its derivatives (Figure 1 and Supplementary Figure 2) were docked on to 6CRV and 6M0J. Tables 2 and 3 gives the binding energies of curcumin and its derivatives with 6CRV (binding energies ranged from −10.98 to −5.12 kcal/mol) and with 6M0J (binding energies ranged −10.01 to −5.33 kcal/mol) respectively. Visual examination of the computationally docked optimal binding poses of curcumin and its derivatives on 6CRV and 6MOJ revealed the important role of various types of interactions viz. hydrogen bonding and hydrophobic interactions, including π–π stacking, π–cation, and π–σ interactions in the stability of the binding of the curcumin/derivatives to the spike protein (6CRV and 6M0J). All compounds showed one or more hydrogen bonds with 6CRV and 6M0J except a few. Figure 2. Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein. Table 2. Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –10.1 –0.14 –8.24 –10.9724 BHBC –8.28 –0.13 –7.87 –3.42239 MNC –8.01 –0.43 –6.98 –8.01452 BDMC –10.98 –0.18 –8.94 –14.0467 1 –5.12 –0.25 –4.17 0 2 –8.55 –0.41 –6.56 –18.4317 3 –6.56 –0.21 –4.92 –11.1148 4 –8.69 –0.13 –6.94 –19.1539 5 –7.03 –0.19 –5.93 –8.47153 6 –6.59 –0.36 –5.12 –7.41531 7 –7.96 –0.21 –6.14 –7.98548 8 –7.55 –0.19 –5.14 –6.14537 9 –6.91 –0.21 –5.18 0 10 –7.75 –0.21 –6.14 –6.73473 11 –8.02 –0.19 –6.19 –17.2465 12 –7.85 –0.38 –6.25 –14.6321 13 –5.94 –0.29 –4.12 –11.3576 14 –8.75 –0.25 –7.19 –21.1412 15 –7.87 –0.38 –7.26 –0.473228 16 –8.47 –0.25 –6.88 –8.458912 DMC –7.89 –0.21 –6.41 –10.47057 FAC –7.17 –0.36 –5.15 –9.14873 IBC –7.91 –0.23 –7.17 –6.90341 IVC –7.43 –0.44 –6.67 –19.9864 SYC –6.82 –0.26 –5.45 –9.44678 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Table 3. Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –9.81 –0.15 –8.35 –15.1512 BHBC −5.33 –0.13 –7.96 0 MNC –7.96 –0.27 –7.38 –4.24687 BDMC –10.01 –0.12 –9.18 –13.9824 1 –7.58 –0.36 –6.19 –18.3489 2 –8.81 –0.31 –8.38 1.2654 3 –6.14 –0.34 –5.45 –6.95102 4 –8.88 –0.18 –7.98 –14.0321 5 –6.23 –0.26 –5.88 –3.5891 6 –7.16 –0.37 –6.15 –10.1688 7 –6.89 –0.25 –5.98 –9.12515 8 –5.49 –0.12 –3.90 –5.92265 9 –6.92 –0.18 –6.22 –6.96719 10 –7.03 –0.24 –6.33 –7.0 11 –6.88 –0.27 –6.23 –6.44363 12 –7.95 –0.41 –5.70 –2.5581 13 –6.11 –0.39 –5.42 –7.34943 14 –7.45 –0.19 –6.40 –10.5367 15 –5.74 –0.39 –5.51 –2.36748 16 –7.83 –0.27 –7.48 –6.64321 DMC –5.98 –0.19 –5.38 –5.99913 FAC –5.85 –0.38 –5.25 –6.46613 IBC –6.65 –0.15 –6.20 –1.52841 IVC –8.96 –0.43 –7.73 –12.2707 SYC –5.87 –0.29 –5.43 –4.4411 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. In 6CRV, it was found that ARG99, GLY100, TRP 101, ILE116, ILE117, ASN118, VAL123, MET144, SER165, ASP166, PHE168, ARG183, PHE185, LYS198, ILE219, PHE220 were the major amino acids involved in hydrogen bonding, hydrophobic and van der Waals interaction with most of the curcumin compounds. However, majority of the curcumin compounds exhibited hydrogen bonding with ARG99, MET144, LYS198 except compound 1 and 9. In addition, all the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) interactions with ASP166, VAL123, PHE185 and van der Waals interaction with GLY100, TRP 101, ILE117, ASN118, ARG183, ILE116, PHE168, SER165, PHE220, ILE219. All compounds showed one or more hydrogen bonding interactions with spike RBD of the SARS-CoV-2 (6M0J) except 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin. It was found that ARG355, TYR396, PRO426, ASP428, THR430, GLY431, PHE429, PHE 464, GLU516, SER514, LEU517 and PHE515 were the major amino acids involved in hydrogen bonding hydrophobic and van der Waals interaction with the all of curcumin compounds. However, majority of the curcumin compounds exhibit hydrogen bonding with ARG355, LEU517 and THR430. In addition, the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) and van der Waals interactions with PHE464, GLY431, GLU516, SER514, PHE515, TYR396, PRO426, ASP428 and PHE429. The curcumin derivative with the best binding affinity was Bis-demethoxycurcumin for both 6CRV (ΔG = −10.98 kcal/mol) and 6M0J (ΔG = −10.01 kcal/mol). The native compound, curcumin, demonstrated a binding energy of −9.81 kcal/mol for 6M0J and −10.1 kcal/mol for 6CRV. Binding mode of Bis-demethoxycurcumin with SARS-CoV Spike protein and SARS-CoV-2 Spike protein RBD is shown in Figure 2 and 3, respectively. Figure 3. Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD. In-silico drug-likeness and pharmacokinetic property prediction All the structures used for docking were analysed for in silico drug-likeness based on the Lipinski’s rules using pkCSM server (Table 4). The lipophilicity (expressed as LogP) predicted for all the compounds were found to be well above the traditionally cut-off value of 5 used for drug design. Curcumin and its derivatives, used in this study, show suitable MW values (MW < 500) essential for a successful penetration through biological membranes. The surface area (SA) for all the compounds was observed to be in the range 115.89 − 240.65 Å2 which is well within the limit. All compounds, except 5-di-tert-butyl-4-hydroxybenzaldehyde curcumin (BHBC), 4-methoxy-1-naphthaldehyde curcumin (MNC), Syringaldehyde curcumin (SYC) and compound-16, fall into the appropriate range indicating good bioavailability of the candidate molecule. The number of hydrogen bond acceptors (HBA, ≤10) and donors (HBD, ≤5) for all the compounds were in accordance with the Lipinski’s rule of five. Table 4. In-silico prediction of drug-likeness for curcumin derivatives[a]. Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3 [a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Additionally, a variety of key ADMET (Absorption, Distribution, Metabolism and Excretion) properties have also been calculated with the aid of pkCSMserver. The results are listed in Table 5. All curcumin derivatives showed moderate to high water solubility ranging from −2.91 log mol/L (BHBC) to −6.99 log mol/L (compound-12), in addition to high Caco-2 permeability (permeation > 0.90) except compound BHBC and MNC (permeation = 0.67 and 0.43 respectively) which showed moderate permeability. Intestinal absorption (IA) has been found to be greater than 85% indicating good permeation across the intestinal membrane. Further, all curcumin compounds showed good permeation through skin (permeation > −2.5). Additionally, all curcumin compounds showed no inhibition towards P-glycoprotein I and P-glycoprotein II, except compounds BHBC, MNC, compound-13 and Ferrulic acid curcumin (FAC) which demonstrated inhibition towards P-glycoprotein I. Furthermore, all curcumin compounds showed poor BBB permeability and moderate CNS permeability except Isovanillin curcumin (IVC) and SYC. All showed inhibition towards the metabolizing enzyme CYP3A4 except BHBC, compounds-8, -11, -14, -15 and -16, while except compound-15, all showed inhibition towards CYP2C9. All curcumin derivatives were found to show the total clearance in the range 0.10 log mL/min/kg (compound-1 and -4) to 1.01 log mL/min/kg (compound-15). Further, all curcumin derivatives except compound 15 were found to act as OCT2 substrate, thus indicating that these compounds will not have any adverse interactions and no negative effect on renal clearance. Table 5. In-silico ADME prediction for curcumin derivatives[a]. Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No [a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Conclusion and discussion The spike proteins of corona viruses are essential for entry of the virus into the target cells. The spike protein exists as a trimer on the surface of the virus with one of the monomer in up conformation and the other two in down conformation (Wrapp et al., 2020). The N-terminal region (S1) of the S protein is important for binding to the cellular receptor ACE2 (Hoffmann et al., 2020; Tai et al., 2020). The S protein undergoes priming by cellular protease, TMPRSS2 and the S2 region of the protein is responsible for fusion of viral and cellular membrane (Hoffmann et al., 2020). Therefore, identifying therapeutics for the S protein of the novel corona virus could potentially target the critical process of entry and fusion of the virus. Curcumin and its derivatives are known for their many biological activities, one of them is its antiviral activity. Therefore, here, we examined the potential of curcumin, and its derivatives, to bind to the SARS-CoV and SARS-CoV-2 spike protein. From our computational molecular docking approach (using auto dock 4.2, PDB ID - 6CRV, 6M0J) and in-silico ADMET tool, we predicted that Bis-demethoxy curcumin, compound-4 and compound-2 were the most recommended curcumin compounds which bind to RBD domain of the SARS-CoV-2 Spike protein and SARS-CoV spike protein efficiently in in silico studies. Curcumin has earlier shown to have specific inhibitory effect on the NA activity in influenza virus (Chen et al., 2013; Richart et al., 2018) and modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Till date, Bis-demethoxy curcumin has shown to have anti-cancer and hepato-protective activities (Kumaravel et al., 2013; Rajagopalan et al., 2010). This is the first time; it is predicted to have a potential anti-viral activity. Therefore, these curcumin derivatives, which have been predicted to bind to the SARS-CoV-2 spike protein, could be explored as probable inhibitors of COVID-19 spike protein through experimental studies. Supplementary Material Supplementary_Figure.pdf Click here for additional data file.
sec	Introduction Coronaviruses are enveloped positive-sense single stranded RNA viruses that belong to the family coronaviridae. They usually infect birds and mammals and cause mild respiratory diseases (Kahn & McIntosh, 2005). However, in the recent past, these viruses have caused lethal endemics such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) endemics (Kahn & McIntosh, 2005). In December 2019, Wuhan city in China became the centre of a surge in cases of pneumonia by an unknown cause (Zhu et al., 2020). A novel corona virus, called SARS-CoV-2, was isolated from these pneumonia patients in January 2020 and the disease was called the Coronavirus disease 2019 (COVID-19) by the World Health Organization in February, 2020 (Gorbalenya et. al, 2020; Jiang et al., 2020; Who, 2020). Even though the epidemic may have started from a zoonotic transmission in a seafood market, that also sold wild animals, it became clear that this disease was transmitted from person to person (Li et al., 2020). The clinical characteristics of this disease are broad, constituting asymptomatic infections, mild respiratory disease, severe pneumonia with respiratory failure and even death (Chen et al., 2020; Huang et al., 2020; Wang et al., 2020). The COVID-19 pandemic poses a significant challenge to global public health (Phelan et al., 2020), appealing for the development of safe and successful prophylactics and therapies against infection of its causative agent, the SARS-CoV-2 virus. SARS-CoV-2 has a genome of ∼30 kilobases, which codes for multiple structural and non-structural proteins (Chan et al., 2020). The structural proteins, present on the surface of the mature virion, include the spike protein, the membrane protein, the envelope protein and the nucleocapsid protein (Chan et al., 2020). The Spike protein, of the beta coronaviruses SARS-CoV-2 and SARS-CoV, enables the attachment of the virus to the cells of the lower respiratory tract of humans to gain entry into the lung tissue (Hoffmann et al., 2020). Apart from attachment, the spike glycoprotein also appears to play a role in fusion and entry of the virus into the host (Chen et al., 2020; Hoffmann et al., 2020). The spike protein, of the novel coronavirus, utilizes the SARS-CoV receptor, ACE2 for entry (Chen et al., 2020) and is primed by the cellular protease, TMPRSS2 (Hoffmann et al., 2020). A serine protease inhibitor, which can act on TMPRSS2, has been shown to inhibit novel coronavirus entry (Hoffmann et al., 2020). Therefore, the spike protein of the novel coronavirus is a good drug target and identifying small molecules that bind to S protein would inhibit viral recognition of host cells and disrupt viral-host interactions. Curcumin, a naturally occurring phytochemical and principal component of Curcuma longa, has exhibited broad pharmacological properties including antioxidant, anti-inflammatory, anti-cancer and anti-viral effects (Khor et al., 2019; Kocaadam & Şanlier, 2017; Lal et al., 2016; Wiggers et al., 2017). Curcumin and its derivatives, due to its rich conventional medicinal interest, has undergone comprehensive in vitro and in vivo studies. It has, therefore, been associated with more than 100 cellular targets, including cytokines, proteins, transcription factors, and receptors. Previous studies have shown the potential of curcumin as a treatment against Influenza A virus infection, by an effect mediated by modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Curcumin has also been shown to have anti neuraminidase (NA) activity for the influenza virus NA protein (Richart et al., 2018). Therefore, in the present study, curcumin and its derivatives were docked onto the spike protein of the SARS-CoV and the SARS-CoV-2 to predict the binding interactions. We found that curcumin and few of its derivatives showed promising results to be potential spike protein inhibitors. One of the derivatives, bis-demethoxycurcumin, showed the best binding affinity to the spike protein of both the SARS-CoV and the novel corona virus, SARS-CoV-2. The possibility for few of the curcumin derivatives, that showed good binding affinity, could be tested for further therapeutic use against COVID-19.
title	Introduction
p	Coronaviruses are enveloped positive-sense single stranded RNA viruses that belong to the family coronaviridae. They usually infect birds and mammals and cause mild respiratory diseases (Kahn & McIntosh, 2005). However, in the recent past, these viruses have caused lethal endemics such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) endemics (Kahn & McIntosh, 2005). In December 2019, Wuhan city in China became the centre of a surge in cases of pneumonia by an unknown cause (Zhu et al., 2020). A novel corona virus, called SARS-CoV-2, was isolated from these pneumonia patients in January 2020 and the disease was called the Coronavirus disease 2019 (COVID-19) by the World Health Organization in February, 2020 (Gorbalenya et. al, 2020; Jiang et al., 2020; Who, 2020). Even though the epidemic may have started from a zoonotic transmission in a seafood market, that also sold wild animals, it became clear that this disease was transmitted from person to person (Li et al., 2020). The clinical characteristics of this disease are broad, constituting asymptomatic infections, mild respiratory disease, severe pneumonia with respiratory failure and even death (Chen et al., 2020; Huang et al., 2020; Wang et al., 2020). The COVID-19 pandemic poses a significant challenge to global public health (Phelan et al., 2020), appealing for the development of safe and successful prophylactics and therapies against infection of its causative agent, the SARS-CoV-2 virus.
p	SARS-CoV-2 has a genome of ∼30 kilobases, which codes for multiple structural and non-structural proteins (Chan et al., 2020). The structural proteins, present on the surface of the mature virion, include the spike protein, the membrane protein, the envelope protein and the nucleocapsid protein (Chan et al., 2020). The Spike protein, of the beta coronaviruses SARS-CoV-2 and SARS-CoV, enables the attachment of the virus to the cells of the lower respiratory tract of humans to gain entry into the lung tissue (Hoffmann et al., 2020). Apart from attachment, the spike glycoprotein also appears to play a role in fusion and entry of the virus into the host (Chen et al., 2020; Hoffmann et al., 2020). The spike protein, of the novel coronavirus, utilizes the SARS-CoV receptor, ACE2 for entry (Chen et al., 2020) and is primed by the cellular protease, TMPRSS2 (Hoffmann et al., 2020). A serine protease inhibitor, which can act on TMPRSS2, has been shown to inhibit novel coronavirus entry (Hoffmann et al., 2020). Therefore, the spike protein of the novel coronavirus is a good drug target and identifying small molecules that bind to S protein would inhibit viral recognition of host cells and disrupt viral-host interactions.
p	Curcumin, a naturally occurring phytochemical and principal component of Curcuma longa, has exhibited broad pharmacological properties including antioxidant, anti-inflammatory, anti-cancer and anti-viral effects (Khor et al., 2019; Kocaadam & Şanlier, 2017; Lal et al., 2016; Wiggers et al., 2017). Curcumin and its derivatives, due to its rich conventional medicinal interest, has undergone comprehensive in vitro and in vivo studies. It has, therefore, been associated with more than 100 cellular targets, including cytokines, proteins, transcription factors, and receptors. Previous studies have shown the potential of curcumin as a treatment against Influenza A virus infection, by an effect mediated by modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Curcumin has also been shown to have anti neuraminidase (NA) activity for the influenza virus NA protein (Richart et al., 2018). Therefore, in the present study, curcumin and its derivatives were docked onto the spike protein of the SARS-CoV and the SARS-CoV-2 to predict the binding interactions.
p	We found that curcumin and few of its derivatives showed promising results to be potential spike protein inhibitors. One of the derivatives, bis-demethoxycurcumin, showed the best binding affinity to the spike protein of both the SARS-CoV and the novel corona virus, SARS-CoV-2. The possibility for few of the curcumin derivatives, that showed good binding affinity, could be tested for further therapeutic use against COVID-19.
sec	Experimental section Sequence analysis The sequences of the SARS-CoV-2 and SARS-CoV were downloaded from National Center for Biotechnology Information. Multiple sequence analysis and pairwise sequence identity was determined using the Clustal Omega server at the European Bioinformatics Institute (Sievers et al., 2011). Preparation of coordinate file The X-ray crystal structure of spike surface glycoprotein of SARS-CoV (PDB entry: 6CRV, resolution = 3.2 Å) and co-crystallized structure of SARS-CoV-2 RBD with human ACE2 Protein (PDB entry: 6M0J, resolution = 2.45 Å) was retrieved from Protein Data Bank (https:/www.rscb.org/pdb). The protein structure was prepared using the Discovery Studio Visualizer (version 3.1) and AutoDock Tools (ADT; version 1.5.4) through different steps viz. removal of water molecules and co-crystallized ligand, addition of missing hydrogen atoms, addition of Gasteiger-Marsili and Kollman charges, merging of non-polar hydrogens, and assignment of rotatable bonds. The file was then saved in pdbqt file format for further analysis. Preparation of ligands The chemical structures of curcumin derivatives were constructed using Chem3D 15.0 module of ChemOffice 15.0 and saved in PDB format. The structures were optimized using “Prepare Ligands” in the AutoDock 4.2, flexible torsions were assigned and the acyclic dihedral angles were allowed to rotate freely. The file was then saved as pdbqt file format for further analysis. Molecular docking Molecular docking simulations were conducted on the curcumin derivatives using the AutoDock 4.2 to get insight into their binding preferences within the active site of the receptor. The molecular docking simulations were performed on the PC based machines running on Windows 7 (x86) as operating system. The software included MGL tools 1.5.4 based AutoDock 4.2 (www.scripps.edu) which uses Python 2.7 language - Cygwin C:\program (www.cygwin.com) and Python 2.5 (www.python.com) (Morris et al., 2009). Docking methodology The flexible docking was performed using the refined spike protein of SARS-CoV (6CRV) and RBD domain of SARS-CoV-2 (6M0J). The grid maps of the interaction energies of various atom types were pre-calculated using AutoGrid 4.2. In each docking for spike surface glycoprotein and spike RBD, a grid box was created using a grid map of 45 × 45 × 45 points, 60 × 60 × 60 points with grid spacing of 0.375 Å and 0.420 Å respectively. The grid maps were centred on the corresponding ligand binding site within the protein structure. Lamarckian Genetic Algorithm (LGA) was adopted to perform docking simulations using the following default parameters, viz. 100 independent runs with step sizes of 0.2 Å for translations and 5 Å for orientations and torsions, an initial population of random individuals with a population size of 150 individuals, a maximum of 2.5 × 106 energy evaluations, maximum number of generations of 27,000; mutation and crossover rates of 0.02 and 0.8 respectively and an elitism value of 1. All the computations were carried out on Cygwin and was used to generate both grid parameter file (.gpf file) and docking parameter file (.dpf file) for each ligand. The docked conformations of each ligand were ranked into clusters based on the binding energy and the top ranked conformations were used for further study. The pose with the lowest ΔG-score was considered the best fitted one and was further analyzed for Ligand-receptor interactions. In-silico drug-likeness and pharmacokinetic property prediction The in-silico prediction studies were performed, using pkCSM online prediction platforms (Pires et al., 2015), to assess the theoretical pharmacokinetic parameters of the ligands to predict the drug-likeness of ligands. The software calculated pharmaceutically relevant properties such as H-bond donor, H-bond acceptor, octanol-water partition coefficient (LogP), surface area, and number of rotatable bonds. In addition the ADME parameters of the curcumin analogs such as water solubility, Caco2 permeability, human intestinal absorption, skin permeability, P-glycoprotein I and II inhibition, volume of distribution, fraction of unbound drug, Blood Brain Barrier and CNS permeability, cytochrome P450 (CYP3A4 and CYP2C9 inhibition) inhibition, total clearance, action as renal OCT2 (organic cation transporter 2) substrate were analysed.
title	Experimental section
sec	Sequence analysis The sequences of the SARS-CoV-2 and SARS-CoV were downloaded from National Center for Biotechnology Information. Multiple sequence analysis and pairwise sequence identity was determined using the Clustal Omega server at the European Bioinformatics Institute (Sievers et al., 2011).
title	Sequence analysis
p	The sequences of the SARS-CoV-2 and SARS-CoV were downloaded from National Center for Biotechnology Information. Multiple sequence analysis and pairwise sequence identity was determined using the Clustal Omega server at the European Bioinformatics Institute (Sievers et al., 2011).
sec	Preparation of coordinate file The X-ray crystal structure of spike surface glycoprotein of SARS-CoV (PDB entry: 6CRV, resolution = 3.2 Å) and co-crystallized structure of SARS-CoV-2 RBD with human ACE2 Protein (PDB entry: 6M0J, resolution = 2.45 Å) was retrieved from Protein Data Bank (https:/www.rscb.org/pdb). The protein structure was prepared using the Discovery Studio Visualizer (version 3.1) and AutoDock Tools (ADT; version 1.5.4) through different steps viz. removal of water molecules and co-crystallized ligand, addition of missing hydrogen atoms, addition of Gasteiger-Marsili and Kollman charges, merging of non-polar hydrogens, and assignment of rotatable bonds. The file was then saved in pdbqt file format for further analysis.
title	Preparation of coordinate file
p	The X-ray crystal structure of spike surface glycoprotein of SARS-CoV (PDB entry: 6CRV, resolution = 3.2 Å) and co-crystallized structure of SARS-CoV-2 RBD with human ACE2 Protein (PDB entry: 6M0J, resolution = 2.45 Å) was retrieved from Protein Data Bank (https:/www.rscb.org/pdb). The protein structure was prepared using the Discovery Studio Visualizer (version 3.1) and AutoDock Tools (ADT; version 1.5.4) through different steps viz. removal of water molecules and co-crystallized ligand, addition of missing hydrogen atoms, addition of Gasteiger-Marsili and Kollman charges, merging of non-polar hydrogens, and assignment of rotatable bonds. The file was then saved in pdbqt file format for further analysis.
sec	Preparation of ligands The chemical structures of curcumin derivatives were constructed using Chem3D 15.0 module of ChemOffice 15.0 and saved in PDB format. The structures were optimized using “Prepare Ligands” in the AutoDock 4.2, flexible torsions were assigned and the acyclic dihedral angles were allowed to rotate freely. The file was then saved as pdbqt file format for further analysis.
title	Preparation of ligands
p	The chemical structures of curcumin derivatives were constructed using Chem3D 15.0 module of ChemOffice 15.0 and saved in PDB format. The structures were optimized using “Prepare Ligands” in the AutoDock 4.2, flexible torsions were assigned and the acyclic dihedral angles were allowed to rotate freely. The file was then saved as pdbqt file format for further analysis.
sec	Molecular docking Molecular docking simulations were conducted on the curcumin derivatives using the AutoDock 4.2 to get insight into their binding preferences within the active site of the receptor. The molecular docking simulations were performed on the PC based machines running on Windows 7 (x86) as operating system. The software included MGL tools 1.5.4 based AutoDock 4.2 (www.scripps.edu) which uses Python 2.7 language - Cygwin C:\program (www.cygwin.com) and Python 2.5 (www.python.com) (Morris et al., 2009).
title	Molecular docking
p	Molecular docking simulations were conducted on the curcumin derivatives using the AutoDock 4.2 to get insight into their binding preferences within the active site of the receptor. The molecular docking simulations were performed on the PC based machines running on Windows 7 (x86) as operating system. The software included MGL tools 1.5.4 based AutoDock 4.2 (www.scripps.edu) which uses Python 2.7 language - Cygwin C:\program (www.cygwin.com) and Python 2.5 (www.python.com) (Morris et al., 2009).
sec	Docking methodology The flexible docking was performed using the refined spike protein of SARS-CoV (6CRV) and RBD domain of SARS-CoV-2 (6M0J). The grid maps of the interaction energies of various atom types were pre-calculated using AutoGrid 4.2. In each docking for spike surface glycoprotein and spike RBD, a grid box was created using a grid map of 45 × 45 × 45 points, 60 × 60 × 60 points with grid spacing of 0.375 Å and 0.420 Å respectively. The grid maps were centred on the corresponding ligand binding site within the protein structure. Lamarckian Genetic Algorithm (LGA) was adopted to perform docking simulations using the following default parameters, viz. 100 independent runs with step sizes of 0.2 Å for translations and 5 Å for orientations and torsions, an initial population of random individuals with a population size of 150 individuals, a maximum of 2.5 × 106 energy evaluations, maximum number of generations of 27,000; mutation and crossover rates of 0.02 and 0.8 respectively and an elitism value of 1. All the computations were carried out on Cygwin and was used to generate both grid parameter file (.gpf file) and docking parameter file (.dpf file) for each ligand. The docked conformations of each ligand were ranked into clusters based on the binding energy and the top ranked conformations were used for further study. The pose with the lowest ΔG-score was considered the best fitted one and was further analyzed for Ligand-receptor interactions.
title	Docking methodology
p	The flexible docking was performed using the refined spike protein of SARS-CoV (6CRV) and RBD domain of SARS-CoV-2 (6M0J). The grid maps of the interaction energies of various atom types were pre-calculated using AutoGrid 4.2. In each docking for spike surface glycoprotein and spike RBD, a grid box was created using a grid map of 45 × 45 × 45 points, 60 × 60 × 60 points with grid spacing of 0.375 Å and 0.420 Å respectively. The grid maps were centred on the corresponding ligand binding site within the protein structure.
p	Lamarckian Genetic Algorithm (LGA) was adopted to perform docking simulations using the following default parameters, viz. 100 independent runs with step sizes of 0.2 Å for translations and 5 Å for orientations and torsions, an initial population of random individuals with a population size of 150 individuals, a maximum of 2.5 × 106 energy evaluations, maximum number of generations of 27,000; mutation and crossover rates of 0.02 and 0.8 respectively and an elitism value of 1. All the computations were carried out on Cygwin and was used to generate both grid parameter file (.gpf file) and docking parameter file (.dpf file) for each ligand. The docked conformations of each ligand were ranked into clusters based on the binding energy and the top ranked conformations were used for further study. The pose with the lowest ΔG-score was considered the best fitted one and was further analyzed for Ligand-receptor interactions.
sec	In-silico drug-likeness and pharmacokinetic property prediction The in-silico prediction studies were performed, using pkCSM online prediction platforms (Pires et al., 2015), to assess the theoretical pharmacokinetic parameters of the ligands to predict the drug-likeness of ligands. The software calculated pharmaceutically relevant properties such as H-bond donor, H-bond acceptor, octanol-water partition coefficient (LogP), surface area, and number of rotatable bonds. In addition the ADME parameters of the curcumin analogs such as water solubility, Caco2 permeability, human intestinal absorption, skin permeability, P-glycoprotein I and II inhibition, volume of distribution, fraction of unbound drug, Blood Brain Barrier and CNS permeability, cytochrome P450 (CYP3A4 and CYP2C9 inhibition) inhibition, total clearance, action as renal OCT2 (organic cation transporter 2) substrate were analysed.
title	In-silico drug-likeness and pharmacokinetic property prediction
p	The in-silico prediction studies were performed, using pkCSM online prediction platforms (Pires et al., 2015), to assess the theoretical pharmacokinetic parameters of the ligands to predict the drug-likeness of ligands. The software calculated pharmaceutically relevant properties such as H-bond donor, H-bond acceptor, octanol-water partition coefficient (LogP), surface area, and number of rotatable bonds. In addition the ADME parameters of the curcumin analogs such as water solubility, Caco2 permeability, human intestinal absorption, skin permeability, P-glycoprotein I and II inhibition, volume of distribution, fraction of unbound drug, Blood Brain Barrier and CNS permeability, cytochrome P450 (CYP3A4 and CYP2C9 inhibition) inhibition, total clearance, action as renal OCT2 (organic cation transporter 2) substrate were analysed.
sec	Results Sequence alignment of the SARS-CoV-2 and SARS-CoV spike protein sequences Prior to performing the docking studies the multiple sequence alignment was carried out for the spike proteins of the novel corona virus, SARS-CoV-2 and the 2002 SARS-CoV (Supplementary Figure 1 and Table 1). These results show that there is a very high percentage identity (greater than 99%) among the spike protein of the novel corona virus SARS-CoV-2 strains. There was around 74% identity of the SARS-CoV-2 spike proteins with the 2002 SARS-CoV. Therefore, the SARS-CoV-2 spike protein is very much closer to SARS-CoV than to the other corona viruses as shown in other studies (Ahmed et al., 2020). Figure 1. The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins. Table 1. Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV). Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100
title	Results
sec	Sequence alignment of the SARS-CoV-2 and SARS-CoV spike protein sequences Prior to performing the docking studies the multiple sequence alignment was carried out for the spike proteins of the novel corona virus, SARS-CoV-2 and the 2002 SARS-CoV (Supplementary Figure 1 and Table 1). These results show that there is a very high percentage identity (greater than 99%) among the spike protein of the novel corona virus SARS-CoV-2 strains. There was around 74% identity of the SARS-CoV-2 spike proteins with the 2002 SARS-CoV. Therefore, the SARS-CoV-2 spike protein is very much closer to SARS-CoV than to the other corona viruses as shown in other studies (Ahmed et al., 2020). Figure 1. The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins. Table 1. Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV). Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100
title	Sequence alignment of the SARS-CoV-2 and SARS-CoV spike protein sequences
p	Prior to performing the docking studies the multiple sequence alignment was carried out for the spike proteins of the novel corona virus, SARS-CoV-2 and the 2002 SARS-CoV (Supplementary Figure 1 and Table 1). These results show that there is a very high percentage identity (greater than 99%) among the spike protein of the novel corona virus SARS-CoV-2 strains. There was around 74% identity of the SARS-CoV-2 spike proteins with the 2002 SARS-CoV. Therefore, the SARS-CoV-2 spike protein is very much closer to SARS-CoV than to the other corona viruses as shown in other studies (Ahmed et al., 2020).
figure	Figure 1. The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins.
label	Figure 1.
caption	The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins.
p	The curcumin and its derivatives which displayed high binding affinity for SARS-CoV-2 and SARS-CoV spike proteins.
table-wrap	Table 1. Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV). Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100
label	Table 1.
caption	Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV).
p	Percentage identity Matrix of Spike glycoprotein of SARS-CoV-2 and SARS-CoV. Indian 1 and 2 (QHS34546.1 and QIA98583.1), Korea (QHZ00379.1), Brazil (QIG55994.1), USA (QHO60594.1), Wuhan (QHD43416.1), Pakistan (QIQ22760.1), Italy (QIA98554.1), Australia (QHR84449.1) are spike proteins from SARS-CoV-2. SARS-6CRV is the spike protein from the 2002 SARS-CoV spike protein (6CRV).
table	Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77 Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84 Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84 Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92 Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100
tr	Viral Strain SARS-CoV Indian1 Korea Indian2 Brazil USA Wuhan Pakistan Italy Australia
th	Viral Strain
th	SARS-CoV
th	Indian1
th	Korea
th	Indian2
th	Brazil
th	USA
th	Wuhan
th	Pakistan
th	Italy
th	Australia
tr	SARS-CoV 100 73.75 73.77 73.69 73.77 73.77 73.77 73.77 73.77 73.77
td	SARS-CoV
td	100
td	73.75
td	73.77
td	73.69
td	73.77
td	73.77
td	73.77
td	73.77
td	73.77
td	73.77
tr	Indian1 73.75 100 99.84 99.84 99.92 99.92 99.92 99.92 99.92 99.84
td	Indian1
td	73.75
td	100
td	99.84
td	99.84
td	99.92
td	99.92
td	99.92
td	99.92
td	99.92
td	99.84
tr	Korea 73.77 99.84 100 99.84 99.92 99.92 99.92 99.92 99.92 99.84
td	Korea
td	73.77
td	99.84
td	100
td	99.84
td	99.92
td	99.92
td	99.92
td	99.92
td	99.92
td	99.84
tr	Indian2 73.69 99.84 99.84 100 99.92 99.92 99.92 99.92 99.92 99.84
td	Indian2
td	73.69
td	99.84
td	99.84
td	100
td	99.92
td	99.92
td	99.92
td	99.92
td	99.92
td	99.84
tr	Brazil 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92
td	Brazil
td	73.77
td	99.92
td	99.92
td	99.92
td	100
td	100
td	100
td	100
td	100
td	99.92
tr	USA 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92
td	USA
td	73.77
td	99.92
td	99.92
td	99.92
td	100
td	100
td	100
td	100
td	100
td	99.92
tr	Wuhan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92
td	Wuhan
td	73.77
td	99.92
td	99.92
td	99.92
td	100
td	100
td	100
td	100
td	100
td	99.92
tr	Pakistan 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92
td	Pakistan
td	73.77
td	99.92
td	99.92
td	99.92
td	100
td	100
td	100
td	100
td	100
td	99.92
tr	Italy 73.77 99.92 99.92 99.92 100 100 100 100 100 99.92
td	Italy
td	73.77
td	99.92
td	99.92
td	99.92
td	100
td	100
td	100
td	100
td	100
td	99.92
tr	Australia 73.77 99.84 99.84 99.84 99.92 99.92 99.92 99.92 99.92 100
td	Australia
td	73.77
td	99.84
td	99.84
td	99.84
td	99.92
td	99.92
td	99.92
td	99.92
td	99.92
td	100
sec	Molecular docking studies The three-dimensional structure of the SARS-CoV spike protein and SARS-CoV-2 Spike protein RBD has been solved. The PDB files of the SARS-CoV spike protein (6CRV) and the Spike protein RBD of the SARS-CoV-2 (6M0J) was downloaded from the PDB databank. These structures were used to predict the ability of the binding of curcumin and its derivatives to the spike proteins of the two corona viruses. Curcumin and 24 of its derivatives (Figure 1 and Supplementary Figure 2) were docked on to 6CRV and 6M0J. Tables 2 and 3 gives the binding energies of curcumin and its derivatives with 6CRV (binding energies ranged from −10.98 to −5.12 kcal/mol) and with 6M0J (binding energies ranged −10.01 to −5.33 kcal/mol) respectively. Visual examination of the computationally docked optimal binding poses of curcumin and its derivatives on 6CRV and 6MOJ revealed the important role of various types of interactions viz. hydrogen bonding and hydrophobic interactions, including π–π stacking, π–cation, and π–σ interactions in the stability of the binding of the curcumin/derivatives to the spike protein (6CRV and 6M0J). All compounds showed one or more hydrogen bonds with 6CRV and 6M0J except a few. Figure 2. Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein. Table 2. Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –10.1 –0.14 –8.24 –10.9724 BHBC –8.28 –0.13 –7.87 –3.42239 MNC –8.01 –0.43 –6.98 –8.01452 BDMC –10.98 –0.18 –8.94 –14.0467 1 –5.12 –0.25 –4.17 0 2 –8.55 –0.41 –6.56 –18.4317 3 –6.56 –0.21 –4.92 –11.1148 4 –8.69 –0.13 –6.94 –19.1539 5 –7.03 –0.19 –5.93 –8.47153 6 –6.59 –0.36 –5.12 –7.41531 7 –7.96 –0.21 –6.14 –7.98548 8 –7.55 –0.19 –5.14 –6.14537 9 –6.91 –0.21 –5.18 0 10 –7.75 –0.21 –6.14 –6.73473 11 –8.02 –0.19 –6.19 –17.2465 12 –7.85 –0.38 –6.25 –14.6321 13 –5.94 –0.29 –4.12 –11.3576 14 –8.75 –0.25 –7.19 –21.1412 15 –7.87 –0.38 –7.26 –0.473228 16 –8.47 –0.25 –6.88 –8.458912 DMC –7.89 –0.21 –6.41 –10.47057 FAC –7.17 –0.36 –5.15 –9.14873 IBC –7.91 –0.23 –7.17 –6.90341 IVC –7.43 –0.44 –6.67 –19.9864 SYC –6.82 –0.26 –5.45 –9.44678 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Table 3. Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –9.81 –0.15 –8.35 –15.1512 BHBC −5.33 –0.13 –7.96 0 MNC –7.96 –0.27 –7.38 –4.24687 BDMC –10.01 –0.12 –9.18 –13.9824 1 –7.58 –0.36 –6.19 –18.3489 2 –8.81 –0.31 –8.38 1.2654 3 –6.14 –0.34 –5.45 –6.95102 4 –8.88 –0.18 –7.98 –14.0321 5 –6.23 –0.26 –5.88 –3.5891 6 –7.16 –0.37 –6.15 –10.1688 7 –6.89 –0.25 –5.98 –9.12515 8 –5.49 –0.12 –3.90 –5.92265 9 –6.92 –0.18 –6.22 –6.96719 10 –7.03 –0.24 –6.33 –7.0 11 –6.88 –0.27 –6.23 –6.44363 12 –7.95 –0.41 –5.70 –2.5581 13 –6.11 –0.39 –5.42 –7.34943 14 –7.45 –0.19 –6.40 –10.5367 15 –5.74 –0.39 –5.51 –2.36748 16 –7.83 –0.27 –7.48 –6.64321 DMC –5.98 –0.19 –5.38 –5.99913 FAC –5.85 –0.38 –5.25 –6.46613 IBC –6.65 –0.15 –6.20 –1.52841 IVC –8.96 –0.43 –7.73 –12.2707 SYC –5.87 –0.29 –5.43 –4.4411 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. In 6CRV, it was found that ARG99, GLY100, TRP 101, ILE116, ILE117, ASN118, VAL123, MET144, SER165, ASP166, PHE168, ARG183, PHE185, LYS198, ILE219, PHE220 were the major amino acids involved in hydrogen bonding, hydrophobic and van der Waals interaction with most of the curcumin compounds. However, majority of the curcumin compounds exhibited hydrogen bonding with ARG99, MET144, LYS198 except compound 1 and 9. In addition, all the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) interactions with ASP166, VAL123, PHE185 and van der Waals interaction with GLY100, TRP 101, ILE117, ASN118, ARG183, ILE116, PHE168, SER165, PHE220, ILE219. All compounds showed one or more hydrogen bonding interactions with spike RBD of the SARS-CoV-2 (6M0J) except 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin. It was found that ARG355, TYR396, PRO426, ASP428, THR430, GLY431, PHE429, PHE 464, GLU516, SER514, LEU517 and PHE515 were the major amino acids involved in hydrogen bonding hydrophobic and van der Waals interaction with the all of curcumin compounds. However, majority of the curcumin compounds exhibit hydrogen bonding with ARG355, LEU517 and THR430. In addition, the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) and van der Waals interactions with PHE464, GLY431, GLU516, SER514, PHE515, TYR396, PRO426, ASP428 and PHE429. The curcumin derivative with the best binding affinity was Bis-demethoxycurcumin for both 6CRV (ΔG = −10.98 kcal/mol) and 6M0J (ΔG = −10.01 kcal/mol). The native compound, curcumin, demonstrated a binding energy of −9.81 kcal/mol for 6M0J and −10.1 kcal/mol for 6CRV. Binding mode of Bis-demethoxycurcumin with SARS-CoV Spike protein and SARS-CoV-2 Spike protein RBD is shown in Figure 2 and 3, respectively. Figure 3. Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD. In-silico drug-likeness and pharmacokinetic property prediction All the structures used for docking were analysed for in silico drug-likeness based on the Lipinski’s rules using pkCSM server (Table 4). The lipophilicity (expressed as LogP) predicted for all the compounds were found to be well above the traditionally cut-off value of 5 used for drug design. Curcumin and its derivatives, used in this study, show suitable MW values (MW < 500) essential for a successful penetration through biological membranes. The surface area (SA) for all the compounds was observed to be in the range 115.89 − 240.65 Å2 which is well within the limit. All compounds, except 5-di-tert-butyl-4-hydroxybenzaldehyde curcumin (BHBC), 4-methoxy-1-naphthaldehyde curcumin (MNC), Syringaldehyde curcumin (SYC) and compound-16, fall into the appropriate range indicating good bioavailability of the candidate molecule. The number of hydrogen bond acceptors (HBA, ≤10) and donors (HBD, ≤5) for all the compounds were in accordance with the Lipinski’s rule of five. Table 4. In-silico prediction of drug-likeness for curcumin derivatives[a]. Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3 [a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Additionally, a variety of key ADMET (Absorption, Distribution, Metabolism and Excretion) properties have also been calculated with the aid of pkCSMserver. The results are listed in Table 5. All curcumin derivatives showed moderate to high water solubility ranging from −2.91 log mol/L (BHBC) to −6.99 log mol/L (compound-12), in addition to high Caco-2 permeability (permeation > 0.90) except compound BHBC and MNC (permeation = 0.67 and 0.43 respectively) which showed moderate permeability. Intestinal absorption (IA) has been found to be greater than 85% indicating good permeation across the intestinal membrane. Further, all curcumin compounds showed good permeation through skin (permeation > −2.5). Additionally, all curcumin compounds showed no inhibition towards P-glycoprotein I and P-glycoprotein II, except compounds BHBC, MNC, compound-13 and Ferrulic acid curcumin (FAC) which demonstrated inhibition towards P-glycoprotein I. Furthermore, all curcumin compounds showed poor BBB permeability and moderate CNS permeability except Isovanillin curcumin (IVC) and SYC. All showed inhibition towards the metabolizing enzyme CYP3A4 except BHBC, compounds-8, -11, -14, -15 and -16, while except compound-15, all showed inhibition towards CYP2C9. All curcumin derivatives were found to show the total clearance in the range 0.10 log mL/min/kg (compound-1 and -4) to 1.01 log mL/min/kg (compound-15). Further, all curcumin derivatives except compound 15 were found to act as OCT2 substrate, thus indicating that these compounds will not have any adverse interactions and no negative effect on renal clearance. Table 5. In-silico ADME prediction for curcumin derivatives[a]. Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No [a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
title	Molecular docking studies
p	The three-dimensional structure of the SARS-CoV spike protein and SARS-CoV-2 Spike protein RBD has been solved. The PDB files of the SARS-CoV spike protein (6CRV) and the Spike protein RBD of the SARS-CoV-2 (6M0J) was downloaded from the PDB databank. These structures were used to predict the ability of the binding of curcumin and its derivatives to the spike proteins of the two corona viruses. Curcumin and 24 of its derivatives (Figure 1 and Supplementary Figure 2) were docked on to 6CRV and 6M0J. Tables 2 and 3 gives the binding energies of curcumin and its derivatives with 6CRV (binding energies ranged from −10.98 to −5.12 kcal/mol) and with 6M0J (binding energies ranged −10.01 to −5.33 kcal/mol) respectively. Visual examination of the computationally docked optimal binding poses of curcumin and its derivatives on 6CRV and 6MOJ revealed the important role of various types of interactions viz. hydrogen bonding and hydrophobic interactions, including π–π stacking, π–cation, and π–σ interactions in the stability of the binding of the curcumin/derivatives to the spike protein (6CRV and 6M0J). All compounds showed one or more hydrogen bonds with 6CRV and 6M0J except a few.
figure	Figure 2. Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein.
label	Figure 2.
caption	Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein.
p	Molecular docking of Bis-demethoxy curcumin on SARS-CoV spike protein (6CRV). (A) 3 D structure of SARS-CoV Spike Glycoprotein Macromolecule (6CRV). (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residues of SARS-CoV Spike glycoprotein (PDB ID-6CRV). (C) Superimposed binding mode of Bis-demethoxy curcumin with SARS-CoV Spike protein. (D) 2 D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV Spike protein.
table-wrap	Table 2. Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –10.1 –0.14 –8.24 –10.9724 BHBC –8.28 –0.13 –7.87 –3.42239 MNC –8.01 –0.43 –6.98 –8.01452 BDMC –10.98 –0.18 –8.94 –14.0467 1 –5.12 –0.25 –4.17 0 2 –8.55 –0.41 –6.56 –18.4317 3 –6.56 –0.21 –4.92 –11.1148 4 –8.69 –0.13 –6.94 –19.1539 5 –7.03 –0.19 –5.93 –8.47153 6 –6.59 –0.36 –5.12 –7.41531 7 –7.96 –0.21 –6.14 –7.98548 8 –7.55 –0.19 –5.14 –6.14537 9 –6.91 –0.21 –5.18 0 10 –7.75 –0.21 –6.14 –6.73473 11 –8.02 –0.19 –6.19 –17.2465 12 –7.85 –0.38 –6.25 –14.6321 13 –5.94 –0.29 –4.12 –11.3576 14 –8.75 –0.25 –7.19 –21.1412 15 –7.87 –0.38 –7.26 –0.473228 16 –8.47 –0.25 –6.88 –8.458912 DMC –7.89 –0.21 –6.41 –10.47057 FAC –7.17 –0.36 –5.15 –9.14873 IBC –7.91 –0.23 –7.17 –6.90341 IVC –7.43 –0.44 –6.67 –19.9864 SYC –6.82 –0.26 –5.45 –9.44678 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	Table 2.
caption	Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold.
p	Molecular docking results of curcumin and its derivatives with SARS-CoV Spike protein (6CRV). The top hits with the binding energy lesser than -8.4 are in bold.
table	Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –10.1 –0.14 –8.24 –10.9724 BHBC –8.28 –0.13 –7.87 –3.42239 MNC –8.01 –0.43 –6.98 –8.01452 BDMC –10.98 –0.18 –8.94 –14.0467 1 –5.12 –0.25 –4.17 0 2 –8.55 –0.41 –6.56 –18.4317 3 –6.56 –0.21 –4.92 –11.1148 4 –8.69 –0.13 –6.94 –19.1539 5 –7.03 –0.19 –5.93 –8.47153 6 –6.59 –0.36 –5.12 –7.41531 7 –7.96 –0.21 –6.14 –7.98548 8 –7.55 –0.19 –5.14 –6.14537 9 –6.91 –0.21 –5.18 0 10 –7.75 –0.21 –6.14 –6.73473 11 –8.02 –0.19 –6.19 –17.2465 12 –7.85 –0.38 –6.25 –14.6321 13 –5.94 –0.29 –4.12 –11.3576 14 –8.75 –0.25 –7.19 –21.1412 15 –7.87 –0.38 –7.26 –0.473228 16 –8.47 –0.25 –6.88 –8.458912 DMC –7.89 –0.21 –6.41 –10.47057 FAC –7.17 –0.36 –5.15 –9.14873 IBC –7.91 –0.23 –7.17 –6.90341 IVC –7.43 –0.44 –6.67 –19.9864 SYC –6.82 –0.26 –5.45 –9.44678
tr	Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond
th	Compound
th	Binding Energy (ΔG)
th	Ligand Efficiency
th	Intermolecular energy
th	XPH-Bond
tr	Curcumin –10.1 –0.14 –8.24 –10.9724
td	Curcumin
td	–10.1
td	–0.14
td	–8.24
td	–10.9724
tr	BHBC –8.28 –0.13 –7.87 –3.42239
td	BHBC
td	–8.28
td	–0.13
td	–7.87
td	–3.42239
tr	MNC –8.01 –0.43 –6.98 –8.01452
td	MNC
td	–8.01
td	–0.43
td	–6.98
td	–8.01452
tr	BDMC –10.98 –0.18 –8.94 –14.0467
td	BDMC
td	–10.98
td	–0.18
td	–8.94
td	–14.0467
tr	1 –5.12 –0.25 –4.17 0
td	1
td	–5.12
td	–0.25
td	–4.17
td	0
tr	2 –8.55 –0.41 –6.56 –18.4317
td	2
td	–8.55
td	–0.41
td	–6.56
td	–18.4317
tr	3 –6.56 –0.21 –4.92 –11.1148
td	3
td	–6.56
td	–0.21
td	–4.92
td	–11.1148
tr	4 –8.69 –0.13 –6.94 –19.1539
td	4
td	–8.69
td	–0.13
td	–6.94
td	–19.1539
tr	5 –7.03 –0.19 –5.93 –8.47153
td	5
td	–7.03
td	–0.19
td	–5.93
td	–8.47153
tr	6 –6.59 –0.36 –5.12 –7.41531
td	6
td	–6.59
td	–0.36
td	–5.12
td	–7.41531
tr	7 –7.96 –0.21 –6.14 –7.98548
td	7
td	–7.96
td	–0.21
td	–6.14
td	–7.98548
tr	8 –7.55 –0.19 –5.14 –6.14537
td	8
td	–7.55
td	–0.19
td	–5.14
td	–6.14537
tr	9 –6.91 –0.21 –5.18 0
td	9
td	–6.91
td	–0.21
td	–5.18
td	0
tr	10 –7.75 –0.21 –6.14 –6.73473
td	10
td	–7.75
td	–0.21
td	–6.14
td	–6.73473
tr	11 –8.02 –0.19 –6.19 –17.2465
td	11
td	–8.02
td	–0.19
td	–6.19
td	–17.2465
tr	12 –7.85 –0.38 –6.25 –14.6321
td	12
td	–7.85
td	–0.38
td	–6.25
td	–14.6321
tr	13 –5.94 –0.29 –4.12 –11.3576
td	13
td	–5.94
td	–0.29
td	–4.12
td	–11.3576
tr	14 –8.75 –0.25 –7.19 –21.1412
td	14
td	–8.75
td	–0.25
td	–7.19
td	–21.1412
tr	15 –7.87 –0.38 –7.26 –0.473228
td	15
td	–7.87
td	–0.38
td	–7.26
td	–0.473228
tr	16 –8.47 –0.25 –6.88 –8.458912
td	16
td	–8.47
td	–0.25
td	–6.88
td	–8.458912
tr	DMC –7.89 –0.21 –6.41 –10.47057
td	DMC
td	–7.89
td	–0.21
td	–6.41
td	–10.47057
tr	FAC –7.17 –0.36 –5.15 –9.14873
td	FAC
td	–7.17
td	–0.36
td	–5.15
td	–9.14873
tr	IBC –7.91 –0.23 –7.17 –6.90341
td	IBC
td	–7.91
td	–0.23
td	–7.17
td	–6.90341
tr	IVC –7.43 –0.44 –6.67 –19.9864
td	IVC
td	–7.43
td	–0.44
td	–6.67
td	–19.9864
tr	SYC –6.82 –0.26 –5.45 –9.44678
td	SYC
td	–6.82
td	–0.26
td	–5.45
td	–9.44678
table-wrap-foot	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
footnote	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
p	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
table-wrap	Table 3. Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold. Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –9.81 –0.15 –8.35 –15.1512 BHBC −5.33 –0.13 –7.96 0 MNC –7.96 –0.27 –7.38 –4.24687 BDMC –10.01 –0.12 –9.18 –13.9824 1 –7.58 –0.36 –6.19 –18.3489 2 –8.81 –0.31 –8.38 1.2654 3 –6.14 –0.34 –5.45 –6.95102 4 –8.88 –0.18 –7.98 –14.0321 5 –6.23 –0.26 –5.88 –3.5891 6 –7.16 –0.37 –6.15 –10.1688 7 –6.89 –0.25 –5.98 –9.12515 8 –5.49 –0.12 –3.90 –5.92265 9 –6.92 –0.18 –6.22 –6.96719 10 –7.03 –0.24 –6.33 –7.0 11 –6.88 –0.27 –6.23 –6.44363 12 –7.95 –0.41 –5.70 –2.5581 13 –6.11 –0.39 –5.42 –7.34943 14 –7.45 –0.19 –6.40 –10.5367 15 –5.74 –0.39 –5.51 –2.36748 16 –7.83 –0.27 –7.48 –6.64321 DMC –5.98 –0.19 –5.38 –5.99913 FAC –5.85 –0.38 –5.25 –6.46613 IBC –6.65 –0.15 –6.20 –1.52841 IVC –8.96 –0.43 –7.73 –12.2707 SYC –5.87 –0.29 –5.43 –4.4411 Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	Table 3.
caption	Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold.
p	Molecular docking analysis of curcumin and its derivatives with SARS-CoV-2 Spike protein Receptor binding domain (6M0J). The top hits with the binding energy lesser than -8.4 are in bold.
table	Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond Curcumin –9.81 –0.15 –8.35 –15.1512 BHBC −5.33 –0.13 –7.96 0 MNC –7.96 –0.27 –7.38 –4.24687 BDMC –10.01 –0.12 –9.18 –13.9824 1 –7.58 –0.36 –6.19 –18.3489 2 –8.81 –0.31 –8.38 1.2654 3 –6.14 –0.34 –5.45 –6.95102 4 –8.88 –0.18 –7.98 –14.0321 5 –6.23 –0.26 –5.88 –3.5891 6 –7.16 –0.37 –6.15 –10.1688 7 –6.89 –0.25 –5.98 –9.12515 8 –5.49 –0.12 –3.90 –5.92265 9 –6.92 –0.18 –6.22 –6.96719 10 –7.03 –0.24 –6.33 –7.0 11 –6.88 –0.27 –6.23 –6.44363 12 –7.95 –0.41 –5.70 –2.5581 13 –6.11 –0.39 –5.42 –7.34943 14 –7.45 –0.19 –6.40 –10.5367 15 –5.74 –0.39 –5.51 –2.36748 16 –7.83 –0.27 –7.48 –6.64321 DMC –5.98 –0.19 –5.38 –5.99913 FAC –5.85 –0.38 –5.25 –6.46613 IBC –6.65 –0.15 –6.20 –1.52841 IVC –8.96 –0.43 –7.73 –12.2707 SYC –5.87 –0.29 –5.43 –4.4411
tr	Compound Binding Energy (ΔG) Ligand Efficiency Intermolecular energy XPH-Bond
th	Compound
th	Binding Energy (ΔG)
th	Ligand Efficiency
th	Intermolecular energy
th	XPH-Bond
tr	Curcumin –9.81 –0.15 –8.35 –15.1512
td	Curcumin
td	–9.81
td	–0.15
td	–8.35
td	–15.1512
tr	BHBC −5.33 –0.13 –7.96 0
td	BHBC
td	−5.33
td	–0.13
td	–7.96
td	0
tr	MNC –7.96 –0.27 –7.38 –4.24687
td	MNC
td	–7.96
td	–0.27
td	–7.38
td	–4.24687
tr	BDMC –10.01 –0.12 –9.18 –13.9824
td	BDMC
td	–10.01
td	–0.12
td	–9.18
td	–13.9824
tr	1 –7.58 –0.36 –6.19 –18.3489
td	1
td	–7.58
td	–0.36
td	–6.19
td	–18.3489
tr	2 –8.81 –0.31 –8.38 1.2654
td	2
td	–8.81
td	–0.31
td	–8.38
td	1.2654
tr	3 –6.14 –0.34 –5.45 –6.95102
td	3
td	–6.14
td	–0.34
td	–5.45
td	–6.95102
tr	4 –8.88 –0.18 –7.98 –14.0321
td	4
td	–8.88
td	–0.18
td	–7.98
td	–14.0321
tr	5 –6.23 –0.26 –5.88 –3.5891
td	5
td	–6.23
td	–0.26
td	–5.88
td	–3.5891
tr	6 –7.16 –0.37 –6.15 –10.1688
td	6
td	–7.16
td	–0.37
td	–6.15
td	–10.1688
tr	7 –6.89 –0.25 –5.98 –9.12515
td	7
td	–6.89
td	–0.25
td	–5.98
td	–9.12515
tr	8 –5.49 –0.12 –3.90 –5.92265
td	8
td	–5.49
td	–0.12
td	–3.90
td	–5.92265
tr	9 –6.92 –0.18 –6.22 –6.96719
td	9
td	–6.92
td	–0.18
td	–6.22
td	–6.96719
tr	10 –7.03 –0.24 –6.33 –7.0
td	10
td	–7.03
td	–0.24
td	–6.33
td	–7.0
tr	11 –6.88 –0.27 –6.23 –6.44363
td	11
td	–6.88
td	–0.27
td	–6.23
td	–6.44363
tr	12 –7.95 –0.41 –5.70 –2.5581
td	12
td	–7.95
td	–0.41
td	–5.70
td	–2.5581
tr	13 –6.11 –0.39 –5.42 –7.34943
td	13
td	–6.11
td	–0.39
td	–5.42
td	–7.34943
tr	14 –7.45 –0.19 –6.40 –10.5367
td	14
td	–7.45
td	–0.19
td	–6.40
td	–10.5367
tr	15 –5.74 –0.39 –5.51 –2.36748
td	15
td	–5.74
td	–0.39
td	–5.51
td	–2.36748
tr	16 –7.83 –0.27 –7.48 –6.64321
td	16
td	–7.83
td	–0.27
td	–7.48
td	–6.64321
tr	DMC –5.98 –0.19 –5.38 –5.99913
td	DMC
td	–5.98
td	–0.19
td	–5.38
td	–5.99913
tr	FAC –5.85 –0.38 –5.25 –6.46613
td	FAC
td	–5.85
td	–0.38
td	–5.25
td	–6.46613
tr	IBC –6.65 –0.15 –6.20 –1.52841
td	IBC
td	–6.65
td	–0.15
td	–6.20
td	–1.52841
tr	IVC –8.96 –0.43 –7.73 –12.2707
td	IVC
td	–8.96
td	–0.43
td	–7.73
td	–12.2707
tr	SYC –5.87 –0.29 –5.43 –4.4411
td	SYC
td	–5.87
td	–0.29
td	–5.43
td	–4.4411
table-wrap-foot	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
footnote	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
p	Abbreviations: BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
p	In 6CRV, it was found that ARG99, GLY100, TRP 101, ILE116, ILE117, ASN118, VAL123, MET144, SER165, ASP166, PHE168, ARG183, PHE185, LYS198, ILE219, PHE220 were the major amino acids involved in hydrogen bonding, hydrophobic and van der Waals interaction with most of the curcumin compounds. However, majority of the curcumin compounds exhibited hydrogen bonding with ARG99, MET144, LYS198 except compound 1 and 9. In addition, all the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) interactions with ASP166, VAL123, PHE185 and van der Waals interaction with GLY100, TRP 101, ILE117, ASN118, ARG183, ILE116, PHE168, SER165, PHE220, ILE219.
p	All compounds showed one or more hydrogen bonding interactions with spike RBD of the SARS-CoV-2 (6M0J) except 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin. It was found that ARG355, TYR396, PRO426, ASP428, THR430, GLY431, PHE429, PHE 464, GLU516, SER514, LEU517 and PHE515 were the major amino acids involved in hydrogen bonding hydrophobic and van der Waals interaction with the all of curcumin compounds. However, majority of the curcumin compounds exhibit hydrogen bonding with ARG355, LEU517 and THR430. In addition, the compounds exhibited hydrophobic (π–π stacking, π–cation, and π–σ) and van der Waals interactions with PHE464, GLY431, GLU516, SER514, PHE515, TYR396, PRO426, ASP428 and PHE429.
p	The curcumin derivative with the best binding affinity was Bis-demethoxycurcumin for both 6CRV (ΔG = −10.98 kcal/mol) and 6M0J (ΔG = −10.01 kcal/mol). The native compound, curcumin, demonstrated a binding energy of −9.81 kcal/mol for 6M0J and −10.1 kcal/mol for 6CRV. Binding mode of Bis-demethoxycurcumin with SARS-CoV Spike protein and SARS-CoV-2 Spike protein RBD is shown in Figure 2 and 3, respectively.
figure	Figure 3. Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD.
label	Figure 3.
caption	Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD.
p	Molecular docking of Bis-demethoxy curcumin on SARS-CoV-2 spike protein receptor binding domain (6M0J). (A) 3D structure of SARS-CoV-2 spike protein RBD bound with ACE2 protein Macromolecule. (B) Binding Interaction of Bis-demethoxy curcumin with amino acid residue of spike protein RBD of SARS-CoV-2 (PDB ID-6M0J). (C) Superimposed binding mode of Bis-demethoxy curcumin, docked with SARS-CoV-2 spike protein RBD. (D) 2D representation of the binding interaction of Bis-demethoxy curcumin with SARS-CoV-2 spike RBD.
sec	In-silico drug-likeness and pharmacokinetic property prediction All the structures used for docking were analysed for in silico drug-likeness based on the Lipinski’s rules using pkCSM server (Table 4). The lipophilicity (expressed as LogP) predicted for all the compounds were found to be well above the traditionally cut-off value of 5 used for drug design. Curcumin and its derivatives, used in this study, show suitable MW values (MW < 500) essential for a successful penetration through biological membranes. The surface area (SA) for all the compounds was observed to be in the range 115.89 − 240.65 Å2 which is well within the limit. All compounds, except 5-di-tert-butyl-4-hydroxybenzaldehyde curcumin (BHBC), 4-methoxy-1-naphthaldehyde curcumin (MNC), Syringaldehyde curcumin (SYC) and compound-16, fall into the appropriate range indicating good bioavailability of the candidate molecule. The number of hydrogen bond acceptors (HBA, ≤10) and donors (HBD, ≤5) for all the compounds were in accordance with the Lipinski’s rule of five. Table 4. In-silico prediction of drug-likeness for curcumin derivatives[a]. Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3 [a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin. Additionally, a variety of key ADMET (Absorption, Distribution, Metabolism and Excretion) properties have also been calculated with the aid of pkCSMserver. The results are listed in Table 5. All curcumin derivatives showed moderate to high water solubility ranging from −2.91 log mol/L (BHBC) to −6.99 log mol/L (compound-12), in addition to high Caco-2 permeability (permeation > 0.90) except compound BHBC and MNC (permeation = 0.67 and 0.43 respectively) which showed moderate permeability. Intestinal absorption (IA) has been found to be greater than 85% indicating good permeation across the intestinal membrane. Further, all curcumin compounds showed good permeation through skin (permeation > −2.5). Additionally, all curcumin compounds showed no inhibition towards P-glycoprotein I and P-glycoprotein II, except compounds BHBC, MNC, compound-13 and Ferrulic acid curcumin (FAC) which demonstrated inhibition towards P-glycoprotein I. Furthermore, all curcumin compounds showed poor BBB permeability and moderate CNS permeability except Isovanillin curcumin (IVC) and SYC. All showed inhibition towards the metabolizing enzyme CYP3A4 except BHBC, compounds-8, -11, -14, -15 and -16, while except compound-15, all showed inhibition towards CYP2C9. All curcumin derivatives were found to show the total clearance in the range 0.10 log mL/min/kg (compound-1 and -4) to 1.01 log mL/min/kg (compound-15). Further, all curcumin derivatives except compound 15 were found to act as OCT2 substrate, thus indicating that these compounds will not have any adverse interactions and no negative effect on renal clearance. Table 5. In-silico ADME prediction for curcumin derivatives[a]. Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No [a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
title	In-silico drug-likeness and pharmacokinetic property prediction
p	All the structures used for docking were analysed for in silico drug-likeness based on the Lipinski’s rules using pkCSM server (Table 4). The lipophilicity (expressed as LogP) predicted for all the compounds were found to be well above the traditionally cut-off value of 5 used for drug design. Curcumin and its derivatives, used in this study, show suitable MW values (MW < 500) essential for a successful penetration through biological membranes. The surface area (SA) for all the compounds was observed to be in the range 115.89 − 240.65 Å2 which is well within the limit. All compounds, except 5-di-tert-butyl-4-hydroxybenzaldehyde curcumin (BHBC), 4-methoxy-1-naphthaldehyde curcumin (MNC), Syringaldehyde curcumin (SYC) and compound-16, fall into the appropriate range indicating good bioavailability of the candidate molecule. The number of hydrogen bond acceptors (HBA, ≤10) and donors (HBD, ≤5) for all the compounds were in accordance with the Lipinski’s rule of five.
table-wrap	Table 4. In-silico prediction of drug-likeness for curcumin derivatives[a]. Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3 [a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	Table 4.
caption	In-silico prediction of drug-likeness for curcumin derivatives[a].
p	In-silico prediction of drug-likeness for curcumin derivatives[a].
table	Compound MW LogP SA HBA HBD nviolations Rotatable bonds BHBC 534.78 8.93 236.0 4 3 1 6 MNC 538.59 6.14 232.11 7 2 1 11 BDMC 308.33 3.83 133.51 4 3 5 1 270.353 4.21 115.89 3 1 4 2 332.739 3.79 136.57 5 4 4 3 314.337 3.77 133.64 5 4 4 4 322.36 4.36 139.56 3 3 4 5 364.441 4.93 159.62 4 0 5 6 320.388 4.95 141.45 4 3 1 7 325.32 3.60 138.02 5 1 6 8 329.193 4.45 125.75 2 1 4 9 462.181 5.24 164.71 7 2 0 10 351.358 4.13 149.74 4 5 1 11 360.375 5.07 149.89 2 1 3 12 419.565 6.24 185.40 9 4 1 13 389.495 5.22 171.67 4 4 1 14 342.778 3.81 143.82 4 2 2 15 469.629 4.38 208.20 6 5 1 16 486.564 6.19 211.58 11 6 1 0 DMC 342.391 3.20 146.43 5 2 9 FAC 550.604 4.7 232.81 9 2 15 IBC 556.655 5.24 240.62 7 1 15 IVC 504.535 4.69 214.22 8 3 11 SYC 430.453 3.77 180.12 10 8 1 3
tr	Compound MW LogP SA HBA HBD nviolations Rotatable bonds
th	Compound
th	MW
th	LogP
th	SA
th	HBA
th	HBD
th	nviolations
th	Rotatable bonds
tr	BHBC 534.78 8.93 236.0 4 3 1 6
td	BHBC
td	534.78
td	8.93
td	236.0
td	4
td	3
td	1
td	6
tr	MNC 538.59 6.14 232.11 7 2 1 11
td	MNC
td	538.59
td	6.14
td	232.11
td	7
td	2
td	1
td	11
tr	BDMC 308.33 3.83 133.51 4 3 5
td	BDMC
td	308.33
td	3.83
td	133.51
td	4
td	3
td
td	5
tr	1 270.353 4.21 115.89 3 1 4
td	1
td	270.353
td	4.21
td	115.89
td	3
td	1
td
td	4
tr	2 332.739 3.79 136.57 5 4 4
td	2
td	332.739
td	3.79
td	136.57
td	5
td	4
td
td	4
tr	3 314.337 3.77 133.64 5 4 4
td	3
td	314.337
td	3.77
td	133.64
td	5
td	4
td
td	4
tr	4 322.36 4.36 139.56 3 3 4
td	4
td	322.36
td	4.36
td	139.56
td	3
td	3
td
td	4
tr	5 364.441 4.93 159.62 4 0 5
td	5
td	364.441
td	4.93
td	159.62
td	4
td	0
td
td	5
tr	6 320.388 4.95 141.45 4 3 1
td	6
td	320.388
td	4.95
td	141.45
td	4
td	3
td
td	1
tr	7 325.32 3.60 138.02 5 1 6
td	7
td	325.32
td	3.60
td	138.02
td	5
td	1
td
td	6
tr	8 329.193 4.45 125.75 2 1 4
td	8
td	329.193
td	4.45
td	125.75
td	2
td	1
td
td	4
tr	9 462.181 5.24 164.71 7 2 0
td	9
td	462.181
td	5.24
td	164.71
td	7
td	2
td
td	0
tr	10 351.358 4.13 149.74 4 5 1
td	10
td	351.358
td	4.13
td	149.74
td	4
td	5
td
td	1
tr	11 360.375 5.07 149.89 2 1 3
td	11
td	360.375
td	5.07
td	149.89
td	2
td	1
td
td	3
tr	12 419.565 6.24 185.40 9 4 1
td	12
td	419.565
td	6.24
td	185.40
td	9
td	4
td
td	1
tr	13 389.495 5.22 171.67 4 4 1
td	13
td	389.495
td	5.22
td	171.67
td	4
td	4
td
td	1
tr	14 342.778 3.81 143.82 4 2 2
td	14
td	342.778
td	3.81
td	143.82
td	4
td	2
td
td	2
tr	15 469.629 4.38 208.20 6 5 1
td	15
td	469.629
td	4.38
td	208.20
td	6
td	5
td
td	1
tr	16 486.564 6.19 211.58 11 6 1 0
td	16
td	486.564
td	6.19
td	211.58
td	11
td	6
td	1
td	0
tr	DMC 342.391 3.20 146.43 5 2 9
td	DMC
td	342.391
td	3.20
td	146.43
td	5
td	2
td
td	9
tr	FAC 550.604 4.7 232.81 9 2 15
td	FAC
td	550.604
td	4.7
td	232.81
td	9
td	2
td
td	15
tr	IBC 556.655 5.24 240.62 7 1 15
td	IBC
td	556.655
td	5.24
td	240.62
td	7
td	1
td
td	15
tr	IVC 504.535 4.69 214.22 8 3 11
td	IVC
td	504.535
td	4.69
td	214.22
td	8
td	3
td
td	11
tr	SYC 430.453 3.77 180.12 10 8 1 3
td	SYC
td	430.453
td	3.77
td	180.12
td	10
td	8
td	1
td	3
table-wrap-foot	[a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
footnote	[a] MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	[a]
p	MW = Molecular weight, LogP = octanol-water partition coefficient, SA = Surface Area, HBA = Number of hydrogen bond acceptor, HBD = Number of hydrogen bond donor, nviolations = violations from Lipinski’s rule. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
p	Additionally, a variety of key ADMET (Absorption, Distribution, Metabolism and Excretion) properties have also been calculated with the aid of pkCSMserver. The results are listed in Table 5. All curcumin derivatives showed moderate to high water solubility ranging from −2.91 log mol/L (BHBC) to −6.99 log mol/L (compound-12), in addition to high Caco-2 permeability (permeation > 0.90) except compound BHBC and MNC (permeation = 0.67 and 0.43 respectively) which showed moderate permeability. Intestinal absorption (IA) has been found to be greater than 85% indicating good permeation across the intestinal membrane. Further, all curcumin compounds showed good permeation through skin (permeation > −2.5). Additionally, all curcumin compounds showed no inhibition towards P-glycoprotein I and P-glycoprotein II, except compounds BHBC, MNC, compound-13 and Ferrulic acid curcumin (FAC) which demonstrated inhibition towards P-glycoprotein I. Furthermore, all curcumin compounds showed poor BBB permeability and moderate CNS permeability except Isovanillin curcumin (IVC) and SYC. All showed inhibition towards the metabolizing enzyme CYP3A4 except BHBC, compounds-8, -11, -14, -15 and -16, while except compound-15, all showed inhibition towards CYP2C9. All curcumin derivatives were found to show the total clearance in the range 0.10 log mL/min/kg (compound-1 and -4) to 1.01 log mL/min/kg (compound-15). Further, all curcumin derivatives except compound 15 were found to act as OCT2 substrate, thus indicating that these compounds will not have any adverse interactions and no negative effect on renal clearance.
table-wrap	Table 5. In-silico ADME prediction for curcumin derivatives[a]. Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No [a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	Table 5.
caption	In-silico ADME prediction for curcumin derivatives[a].
p	In-silico ADME prediction for curcumin derivatives[a].
table	Comp Absorption Distribution Metabolism Excretion WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No 1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No 2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No 3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No 4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No 5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No 6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No 7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No 8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No 9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No 10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No 11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No 12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No 13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No 14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No 15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes 16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No
tr	Comp Absorption Distribution Metabolism Excretion
th	Comp
th	Absorption
th	Distribution
th	Metabolism
th	Excretion
tr	WS CP IA SP PI-1 PI-2 VD FU BBB CNS CI-1 CI-2 TC RS
th
th	WS
th	CP
th	IA
th	SP
th	PI-1
th	PI-2
th	VD
th	FU
th	BBB
th	CNS
th	CI-1
th	CI-2
th	TC
th	RS
tr	BHBC –2.91 0.67 88.62 –2.73 Yes Yes –0.39 0.31 –0.67 –0.85 No Yes 0.15 No
td	BHBC
td	–2.91
td	0.67
td	88.62
td	–2.73
td	Yes
td	Yes
td	–0.39
td	0.31
td	–0.67
td	–0.85
td	No
td	Yes
td	0.15
td	No
tr	MNC –3.64 0.43 100 –2.73 Yes Yes –1.45 0.33 –0.39 –2.73 Yes Yes 0.26 No
td	MNC
td	–3.64
td	0.43
td	100
td	–2.73
td	Yes
td	Yes
td	–1.45
td	0.33
td	–0.39
td	–2.73
td	Yes
td	Yes
td	0.26
td	No
tr	BDMC –4.43 1.01 95.92 –2.99 No No –0.22 0.08 –0.72 –2.11 Yes Yes 0.12 No
td	BDMC
td	–4.43
td	1.01
td	95.92
td	–2.99
td	No
td	No
td	–0.22
td	0.08
td	–0.72
td	–2.11
td	Yes
td	Yes
td	0.12
td	No
tr	1 –4.66 1.25 91.80 –2.51 No No 0.18 0.06 0.01 –1.45 Yes Yes 0.10 No
td	1
td	–4.66
td	1.25
td	91.80
td	–2.51
td	No
td	No
td	0.18
td	0.06
td	0.01
td	–1.45
td	Yes
td	Yes
td	0.10
td	No
tr	2 –3.23 0.93 89.16 –2.76 No No –0.07 0.12 –1.07 –2.29 Yes Yes 0.05 No
td	2
td	–3.23
td	0.93
td	89.16
td	–2.76
td	No
td	No
td	–0.07
td	0.12
td	–1.07
td	–2.29
td	Yes
td	Yes
td	0.05
td	No
tr	3 –3.19 0.94 89.70 –2.76 No No –0.02 0.13 –1.06 –2.27 Yes Yes 0.06 No
td	3
td	–3.19
td	0.94
td	89.70
td	–2.76
td	No
td	No
td	–0.02
td	0.13
td	–1.06
td	–2.27
td	Yes
td	Yes
td	0.06
td	No
tr	4 –4.11 0.92 87.64 –3.07 No No 0.09 0 –0.74 –1.85 Yes Yes 0.10 No
td	4
td	–4.11
td	0.92
td	87.64
td	–3.07
td	No
td	No
td	0.09
td	0
td	–0.74
td	–1.85
td	Yes
td	Yes
td	0.10
td	No
tr	5 –6.39 1.06 94.85 –2.49 No No 0.25 0 –0.03 –1.30 Yes Yes 0.22 No
td	5
td	–6.39
td	1.06
td	94.85
td	–2.49
td	No
td	No
td	0.25
td	0
td	–0.03
td	–1.30
td	Yes
td	Yes
td	0.22
td	No
tr	6 –5.42 1.43 90.49 –2.73 No No 0.05 0 –0.33 –1.29 Yes Yes 0.23 No
td	6
td	–5.42
td	1.43
td	90.49
td	–2.73
td	No
td	No
td	0.05
td	0
td	–0.33
td	–1.29
td	Yes
td	Yes
td	0.23
td	No
tr	7 –4.54 1.04 91.97 –2.73 No No –0.37 0 0.28 –2.20 Yes Yes 0.13 No
td	7
td	–4.54
td	1.04
td	91.97
td	–2.73
td	No
td	No
td	–0.37
td	0
td	0.28
td	–2.20
td	Yes
td	Yes
td	0.13
td	No
tr	8 –5.09 1.66 90.44 –2.29 No No 0.27 0 0.16 –1.47 No Yes 0.14 No
td	8
td	–5.09
td	1.66
td	90.44
td	–2.29
td	No
td	No
td	0.27
td	0
td	0.16
td	–1.47
td	No
td	Yes
td	0.14
td	No
tr	9 –7.03 1.13 93.34 –2.51 No No 0.37 0 0.46 –1.26 Yes Yes 0.26 No
td	9
td	–7.03
td	1.13
td	93.34
td	–2.51
td	No
td	No
td	0.37
td	0
td	0.46
td	–1.26
td	Yes
td	Yes
td	0.26
td	No
tr	10 –4.89 0.94 91.50 –2.74 No No –0.04 0 –0.24 –1.93 Yes Yes 0.11 No
td	10
td	–4.89
td	0.94
td	91.50
td	–2.74
td	No
td	No
td	–0.04
td	0
td	–0.24
td	–1.93
td	Yes
td	Yes
td	0.11
td	No
tr	11 –5.93 1.28 89.16 –2.59 No No 0.36 0 0.66 –1.40 No Yes 0.13 No
td	11
td	–5.93
td	1.28
td	89.16
td	–2.59
td	No
td	No
td	0.36
td	0
td	0.66
td	–1.40
td	No
td	Yes
td	0.13
td	No
tr	12 –6.99 1.02 91.30 –2.75 No No 0.66 0 –0.32 –1.85 Yes Yes 1.04 No
td	12
td	–6.99
td	1.02
td	91.30
td	–2.75
td	No
td	No
td	0.66
td	0
td	–0.32
td	–1.85
td	Yes
td	Yes
td	1.04
td	No
tr	13 –5.42 1.27 92.76 –2.80 Yes Yes 0.54 0 –0.02 –1.58 Yes Yes 0.26 No
td	13
td	–5.42
td	1.27
td	92.76
td	–2.80
td	Yes
td	Yes
td	0.54
td	0
td	–0.02
td	–1.58
td	Yes
td	Yes
td	0.26
td	No
tr	14 –4.20 1.12 90.31 –2.92 No No –0.04 0.01 –0.31 –1.33 No Yes 0.08 No
td	14
td	–4.20
td	1.12
td	90.31
td	–2.92
td	No
td	No
td	–0.04
td	0.01
td	–0.31
td	–1.33
td	No
td	Yes
td	0.08
td	No
tr	15 –3.76 0.97 88.86 –2.79 No No 1.90 0.16 –0.20 –1.26 No No 1.01 Yes
td	15
td	–3.76
td	0.97
td	88.86
td	–2.79
td	No
td	No
td	1.90
td	0.16
td	–0.20
td	–1.26
td	No
td	No
td	1.01
td	Yes
tr	16 –6.92 0.90 94.67 –2.73 No No –0.59 0.07 –0.72 –2.62 No Yes 0.41 No
td	16
td	–6.92
td	0.90
td	94.67
td	–2.73
td	No
td	No
td	–0.59
td	0.07
td	–0.72
td	–2.62
td	No
td	Yes
td	0.41
td	No
tr	DMC –3.95 0.99 92.03 –2.74 No No –0.06 0.12 –0.23 –2.66 Yes Yes 0.32 No
td	DMC
td	–3.95
td	0.99
td	92.03
td	–2.74
td	No
td	No
td	–0.06
td	0.12
td	–0.23
td	–2.66
td	Yes
td	Yes
td	0.32
td	No
tr	FAC –4.02 1.41 83.68 –2.76 Yes Yes –0.58 0.24 –0.91 –3.26 Yes Yes 0.55 No
td	FAC
td	–4.02
td	1.41
td	83.68
td	–2.76
td	Yes
td	Yes
td	–0.58
td	0.24
td	–0.91
td	–3.26
td	Yes
td	Yes
td	0.55
td	No
tr	IBC –4.36 0.91 91.72 –2.73 No No –0.88 0.18 –0.53 –2.77 Yes Yes 0.62 No
td	IBC
td	–4.36
td	0.91
td	91.72
td	–2.73
td	No
td	No
td	–0.88
td	0.18
td	–0.53
td	–2.77
td	Yes
td	Yes
td	0.62
td	No
tr	IVC –3.95 0.96 89.31 –2.73 No No –0.60 0.24 –1.37 –3.02 Yes Yes 0.16 No
td	IVC
td	–3.95
td	0.96
td	89.31
td	–2.73
td	No
td	No
td	–0.60
td	0.24
td	–1.37
td	–3.02
td	Yes
td	Yes
td	0.16
td	No
tr	SYC –3.89 1.01 89.15 –2.73 No No 0.28 0.17 –1.47 –3.14 Yes Yes 0.23 No
td	SYC
td	–3.89
td	1.01
td	89.15
td	–2.73
td	No
td	No
td	0.28
td	0.17
td	–1.47
td	–3.14
td	Yes
td	Yes
td	0.23
td	No
table-wrap-foot	[a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
footnote	[a] Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
label	[a]
p	Abbreviations: WS—Water solubility (log mol/L), CP—Caco2 permeability (log Papp in 10-6 cm/s), IA—Human intestinal absorption (% Absorbed), SP—Skin permeability (log Kp),PI-1—P-glycoprotein I inhibitor, PI-2—P-glycoprotein II inhibitor, VD—Human volume ofdistribution (log L/kg), FU—Fraction unbound (human) (Fu), BBB—BBB permeability (logBB), CNS—CNS permeability (log PS), CI-1—CYP3A4 inhibitor, CI-2—CYP2C9 inhibitor,TC—Total clearance (log mL/min/kg), RS—Renal OCT2 (organic cation transporter 2) substrate. BHBC: 3-5-di-tert-butyl-4-hydroxybenzaldehyde curcumin, MNC: 4-methoxy-1-naphthaldehyde curcumin, BDMC: Bis-demethoxy curcumin, DMC: Demethoxy curcumin, FAC: Ferulic-acid curcumin, IBC: Ibuprofen curcumin, IVC: Isovanillin curcumin, SYC: Syringaldehyde curcumin.
sec	Conclusion and discussion The spike proteins of corona viruses are essential for entry of the virus into the target cells. The spike protein exists as a trimer on the surface of the virus with one of the monomer in up conformation and the other two in down conformation (Wrapp et al., 2020). The N-terminal region (S1) of the S protein is important for binding to the cellular receptor ACE2 (Hoffmann et al., 2020; Tai et al., 2020). The S protein undergoes priming by cellular protease, TMPRSS2 and the S2 region of the protein is responsible for fusion of viral and cellular membrane (Hoffmann et al., 2020). Therefore, identifying therapeutics for the S protein of the novel corona virus could potentially target the critical process of entry and fusion of the virus. Curcumin and its derivatives are known for their many biological activities, one of them is its antiviral activity. Therefore, here, we examined the potential of curcumin, and its derivatives, to bind to the SARS-CoV and SARS-CoV-2 spike protein. From our computational molecular docking approach (using auto dock 4.2, PDB ID - 6CRV, 6M0J) and in-silico ADMET tool, we predicted that Bis-demethoxy curcumin, compound-4 and compound-2 were the most recommended curcumin compounds which bind to RBD domain of the SARS-CoV-2 Spike protein and SARS-CoV spike protein efficiently in in silico studies. Curcumin has earlier shown to have specific inhibitory effect on the NA activity in influenza virus (Chen et al., 2013; Richart et al., 2018) and modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Till date, Bis-demethoxy curcumin has shown to have anti-cancer and hepato-protective activities (Kumaravel et al., 2013; Rajagopalan et al., 2010). This is the first time; it is predicted to have a potential anti-viral activity. Therefore, these curcumin derivatives, which have been predicted to bind to the SARS-CoV-2 spike protein, could be explored as probable inhibitors of COVID-19 spike protein through experimental studies.
title	Conclusion and discussion
p	The spike proteins of corona viruses are essential for entry of the virus into the target cells. The spike protein exists as a trimer on the surface of the virus with one of the monomer in up conformation and the other two in down conformation (Wrapp et al., 2020). The N-terminal region (S1) of the S protein is important for binding to the cellular receptor ACE2 (Hoffmann et al., 2020; Tai et al., 2020). The S protein undergoes priming by cellular protease, TMPRSS2 and the S2 region of the protein is responsible for fusion of viral and cellular membrane (Hoffmann et al., 2020). Therefore, identifying therapeutics for the S protein of the novel corona virus could potentially target the critical process of entry and fusion of the virus.
p	Curcumin and its derivatives are known for their many biological activities, one of them is its antiviral activity. Therefore, here, we examined the potential of curcumin, and its derivatives, to bind to the SARS-CoV and SARS-CoV-2 spike protein. From our computational molecular docking approach (using auto dock 4.2, PDB ID - 6CRV, 6M0J) and in-silico ADMET tool, we predicted that Bis-demethoxy curcumin, compound-4 and compound-2 were the most recommended curcumin compounds which bind to RBD domain of the SARS-CoV-2 Spike protein and SARS-CoV spike protein efficiently in in silico studies.
p	Curcumin has earlier shown to have specific inhibitory effect on the NA activity in influenza virus (Chen et al., 2013; Richart et al., 2018) and modulating immune response to prevent injury to the lung tissue (Han et al., 2018). Till date, Bis-demethoxy curcumin has shown to have anti-cancer and hepato-protective activities (Kumaravel et al., 2013; Rajagopalan et al., 2010). This is the first time; it is predicted to have a potential anti-viral activity. Therefore, these curcumin derivatives, which have been predicted to bind to the SARS-CoV-2 spike protein, could be explored as probable inhibitors of COVID-19 spike protein through experimental studies.
sec	Supplementary Material Supplementary_Figure.pdf Click here for additional data file.
title	Supplementary Material
caption	Supplementary_Figure.pdf
title	Supplementary_Figure.pdf
caption	Click here for additional data file.
p	Click here for additional data file.
back	Acknowledgements We acknowledge the support rendered by the Department of Biochemistry, Indian Academy Degree College - Autonomous, Bengaluru and Ramanbhai Patel College of Pharmacy, Charotar University of Science and Technology, Gujarat for their support in carrying out this work. Disclosure statement The authors declare that they have no conflict of interests
ack	Acknowledgements We acknowledge the support rendered by the Department of Biochemistry, Indian Academy Degree College - Autonomous, Bengaluru and Ramanbhai Patel College of Pharmacy, Charotar University of Science and Technology, Gujarat for their support in carrying out this work.
title	Acknowledgements
p	We acknowledge the support rendered by the Department of Biochemistry, Indian Academy Degree College - Autonomous, Bengaluru and Ramanbhai Patel College of Pharmacy, Charotar University of Science and Technology, Gujarat for their support in carrying out this work.
sec	Disclosure statement The authors declare that they have no conflict of interests
title	Disclosure statement
p	The authors declare that they have no conflict of interests

Annnotations TAB TSV DIC JSON TextAE

last updated at 2021-06-18 21:11:25 UTC

Denotations: 0
Blocks: 0
Relations: 0

PMC:7784829 / 25475-25480 JSONTXT

Document structure show

Annnotations TAB TSV DIC JSON TextAE

PMC:7784829 / 25475-25480 JSON TXT