MATERIALS AND METHODS

Study area: The study was carried out at the Department of Biotechnology, Malankara Catholic College, Mariagiri, Kaliakkavilai, Tamil Nadu, India from June, 2019 to June, 2021.

Ligand preparation: The compounds present in the Curcuma longa was identified from a literature search and Dr. Dukes Phytochemical and Ethnobotanical database (https://phytochem.nal.usda.gov/phytochem/search/list).

The identified compounds were searched in PubChem and ZINC databases for the retrieval of 3D structure in SDF format. The Open Babel software was used for the conversion of all the 3D structures in SDF format into PDB format. The structure of hydroxychloroquine (3652) was taken as the reference molecule for the comparison of the activity.

Protein preparation: The UniProt database was used to identify the structure of SARS-CoV-2 Mpro. The crystal structure of a protein was selected based on the lowest resolution and the amino acids in Ramachandran plot. The 3D structure of the selected protein was downloaded from the PDB (https://www.rcsb.org) database. The heterogeneous atoms and the water molecules present in the protein were removed by using MG Tools of AutoDock 4 and the PDB format of the protein was made for molecular docking analysis.

Molecular docking: All the docking analysis were performed by AutoDock 4.2 software with the optimized docking parameters³⁴. The polar hydrogen atoms and Kollman united atom type charges were added to the protein molecule and it was saved in PDBQT format. The energies of the 85 phytocompounds were minimized by adding the Gasteiger partial charges, merging the non-polar hydrogen atoms and defining the rotatable bonds. The grid maps were constructed with the spacing of 0.375 Å and the dimensions of 80×80×80 Å by using AutoGrid. Lamarckian genetic algorithm 4.2 was employed for molecular docking. The phytocompounds with the activity of less than -7.5 will be selected as the highly active compounds for the interaction analysis. Protein-ligand Interaction Profiler and Ligplot+ v.1.4.5 software were used to analyze the interactions between SARS-CoV-2 Mproprotein and the active compounds.

Drug-likeness properties and pharmacophore study: Since the Curcuma longa was always ingested orally, an in silico integrative model of Absorption, Distribution, Metabolism and Excretion (ADME) was used to screen for natural compounds that may be bioactive via oral administration. SwissADME³⁵ was used to analyze the drug-likeness and the ADME properties of the highly active compounds from the docking studies. PharmaGist webserver³⁶ was used to study the important pharmacophore features of the active compounds and the reference molecules (Hydroxychloroquine).

RESULTS AND DISCUSSION

Ligand and protein preparation: In Curcuma longa, there are 267 compounds available in the Dr. Dukes Phytochemical and Ethnobotanical Database. Among the 267 compounds, the overlapping compounds were removed and the remaining 170 compounds were subjected to retrieve the 3D structure from the PubChem and ZINC databases. The availability of 170 compounds in various parts of Curcuma longa was given in Table 1. The structure of 85 compounds is only available in the PubChem database which was retrieved and used for the docking analysis (Table 2). Since hydroxychloroquine was recommended to treat the COVID-19 patients through clinical trial³⁷, it was used as a reference molecule in this study.

The main protease of SARS-CoV2 (SARS-CoV-2 Mpro) is composed of 306 amino acids and is 33.83 kDa in size (theoretical). The crystal structure with the PDB ID of 6Y84 was selected as SARS-CoV-2 Mpro from the UniProt database (ID: PODTD1) based on the low resolution (1.35 A) and the amino acids in the core, allowed and disallowed regions of the Ramachandran plot. Since the SARS-CoV-2 Mpro protein becomes a potential target for the inhibition of SARS-CoV-2 replication, several researchers crystallized the protein and deposited their structure in the Protein Data Bank (PDB).

Fig. 1(a-t):

Molecular docking of the active compounds from Curcuma longa to the amino acid residues of main protease (PDB ID: 6Y84), (a) Beta-carotene, (b) Beta-sitosterol, (c) Stigmasterol, (d) Campesterol, (e) Cholesterol, (f) Phytosterols, (g) Ascorbic-acid, (h) Cyclocurcumin, (i) Quercetin, (j) Demethoxycurcumin, (k) Riboflavin, (l) Dicinnamoylmethane (m) Bisacurone, (n) Calebin-A, (o) Dihydrocurcumin, (p) Curcumenone, (q) Isoprocurcumenol, (r) Curcumenol, (s) Beta-turmerone and (t) Bis-demethoxycurcumin

The UniProt ID of PODTD1 showed more than 75 crystal structures of SARS-CoV-2 Mpro. The 6Y84 was the crystallized structure with an unliganded active site.

Molecular docking analysis: The virtual screening of all the 85 compounds as well as the reference molecule (Hydroxychloroquine) was done against the receptor protein (SARS-CoV-2 Mpro 6Y84). The compounds with a binding energy of less than -7.5 kcal mol^–1 were considered as highly active compounds. Among the tested 85 compounds and 20 compounds such as Beta-Carotene, Beta-Sitosterol, Stigmasterol, Campesterol, Cholesterol, Phytosterols, Ascorbic-Acid, Cyclocurcumin, Quercetin, Demeth-oxycurcumin, Riboflavin, Dicinnamoylmethane, Bisacurone, Calebin-A, Dihydrocurcumin, Curcumenone, Isopro-curcumenol, Curcumenol, Beta-Turmerone, Bis-Demethoxycurcumin were showed the binding energy of less than -7.5 kcal mol^–1 which were considered as active compounds (Table 2). The molecular docking data of the active 20 compounds were given in Fig. 1a-t (a) Beta-Carotene, (b) Beta-Sitosterol, (c) Stigmasterol, (d) Campesterol, (e) Cholesterol, (f) Phytosterols, (g) Ascorbic-Acid, (h) Cyclocurcumin, (i) Quercetin, (j) Demethoxycurcumin, (k) Riboflavin, (l) Dicinnamoylmethane (m) Bisacurone, (n) Calebin-A, (o) Dihydrocurcumin, (p) Curcumenone, (q) Isoprocurcumenol, (r) Curcumenol, (s) Beta-Turmerone, (t) Bis-Demethoxycurcumin and the reference molecule (Hydroxychloroquine) was given in Fig. 2a. Among the docked 85 compounds and 20 compounds showed the binding energy in the range of -10.12-7.52 kcal mol^–1. The Hydroxychloroquine (reference molecule) has shown the binding energy of -7.23 kcal mol^–1 which was lesser than the binding energy of 30 compounds from Curcuma longa. The 20 compounds from Curcuma longa showed a binding energy of less than -7.5 kcal mol^–1 which were considered as highly active compounds against SARS-CoV-2. The seven compounds such as Alpha-Tocopherol, Zedoarondiol, Beta-Sesquiphellandrene, Caffeic-Acid, Alpha-Turmerone, Procurcumenol, Dehydrocurdione were also showing the binding energy in the range of -7.47 to -7.24 kcal mol^–1 which was lesser than the reference molecule hydroxychloroquine (Table 2).

Among the 20 compounds, Beta-Carotene was showing the highest activity with the binding energy of -10.12 kcal mol^–1 and the other compounds such as Beta-Sitosterol, Stigmasterol, Campesterol, Cholesterol, Phytosterols, Ascorbic-Acid and Cyclocurcumin were displaying the binding energy between -9.96 to -9.01 kcal mol^–1. The compounds such as Quercetin, Demetho-xycurcumin, Riboflavin, Di-Cinnamoyl Methane, Bisacurone, Calebin-A, Dihydrocurcumin, Curcumenone, Isopro-curcumenol, Curcumenol, Beta-Turmerone and Bisdemetho-xycurcumin were showing the binding energy between -8.84 to -7.52 kcal mol^–1. Most of these active compounds were present in the rhizome of the Curcuma longa (Table 1).

Among these compounds, ascorbic acid (vitamin C) was studied by the Zhongnan Hospital (NCT04264533) to check its clinical efficacy against COVID-19. Quite a few studies have proved that ascorbic acid can be used to treat severe viral respiratory tract infections and SARS-associated viral diseases³⁸ by boosting the immune system. Germacrone was one of the principal components of Curcuma longa rhizome essential oil which can inhibit the growth of pseudorabies and Influenza viruses^39,40. In this study, it was observed that the germacrone is having less activity (-6.82) against SARS-CoV-2.

Fig. 2(a-b):

(a) Molecular docking and (b) Interactions of hydroxychloroquine (reference molecule) with the amino acid residues of the main protease

Several researchers have suggested that the curcumin from Curcuma longa could be used as a potential drug to treat COVID-19 patients^41,33. But this study proved that 32 compounds (Table 2) from Curcuma longa had more activity than the curcumin (-7.16). The activity of curcumin was lesser than the activity of Hydroxychloroquine (-7.23 kcal mol^–1). It was also observed that the active compounds such as Bis-Demethoxycurcumin, Cyclocurcumin, Demethoxycurcumin, Dihydrocurcumin, Tetrahydrocurcumin are the derivatives/metabolites of curcumin⁴². Curcumin, Bisdemethoxycurcumin and Demethoxycurcumin are the Curcuminoids produced by Curcuminoid synthase^43-45 which are responsible for all the pharmacological activities of Curcuma longa^46-48. Moreover, Chakraborty et al.⁴⁹ proved that cyclocurcumin have higher neuronal protection activity than curcumin.

Interaction analysis: The structure of the SARS-CoV-2 Mpro 6Y84 protein was considered a rigid molecule and the phytocompounds of Curcuma longa were considered as flexible molecules during the docking analysis. From the docking, it was found that the 20 compounds from Curcuma longa were highly active against SARS-CoV-2. All the active 20 compounds and the reference molecule (Hydroxychloroquine) were subjected to interaction analysis by a protein-ligand interaction profiler. The interaction analysis showed that all the active compounds and the reference molecule (Hydroxychloroquine) were interacted with the amino acid residues of SARS-CoV-2 Mpro by forming hydrogen bonds as well as hydrophobic interactions except Beta-Carotene (Fig. 2b and 3a-t). The LYS5 and TYR126 amino acid residues of SARS-CoV-2 Mpro were commonly interacted by all the active compounds against SARS-CoV-2 except Beta-Carotene (Table 3, 4).

The Beta-Carotene interacted with the target protein through nine hydrophobic interactions. The compounds such as Riboflavin and Dihydrocurcumin were also showed the TT-cation interaction in the LYS5 residue of protein (Table 5). The LYS5 residue was also involved in other kinds of interactions such as hydrophobic and hydrogen bond interactions (Table 3, 4). The hydroxychloroquine was showing three hydrophobic interactions with the amino acid residues such as LYS5 and TYR126 while the other tested molecules from Curcuma longa was showing a higher number of hydrophobic interactions except for riboflavin which was showing two i.e., LYS5 and TYR126. The TYR126 hydrophobic interaction was displayed by all the active molecules of Curcuma longa except Bis-Demethoxycurcumin. The LYS5 hydrophobic interaction was also observed in all the active molecules except Beta-Carotene.

Hydroxychloroquine is interacting with the protein by six hydrogen bond interactions with three amino acid residues such as LYS5, ARG4 and GLN127 (Fig. 2b). The Beta-Carotene was not showing any hydrogen bond interaction while all other active compounds were showing the hydrogen bond interactions. The LYS5 residue of the SARS-CoV-2 Mpro interacted through hydrogen bond interaction with all the active compounds except Isoprocurcumenol, Campesterol and Beta-Sitosterol. ARG4 residue of the SARS-CoV-2 Mpro was interacted by the compounds such as Stigmasterol, Cholesterol and Curcumenone through hydrogen bond interaction. The six active compounds such as Quercetin, Riboflavin, Bisacurone, Calebin-A, Dihydrocurcumin and Isoprocurcumenol interacted through hydrogen bond interaction with the GLN127 residue. LYS137 residue was interacted by the active molecules such as Campesterol, Cyclocurcumin, Quercetin, Demethoxycurcumin, Riboflavin and Bisacurone through hydrogen bond interaction (Table 4).

The common interaction analysis between the five best active compounds (Beta-Carotene, Beta-Sitosterol, Stigmasterol, Campesterol and Cholesterol) and the SARS-CoV-2 Mpro was done with the help of Ligplot+ v.1.4.5 software. It was observed that the amino acid residues such as TYR126, LYS137, GLY138 and Ser139 interacted with all the five best active compounds (Fig. 4). The common interaction analysis between the Beta-Carotene and Hydroxychloroquine showed that the TYR126, LYS137 and GLU290 amino acid residues were involved in a common interaction (Fig. 5). These results confirmed that the amino acids residues of SARS-CoV-2 Mpro were affected by the active compounds of Curcuma longa.

Drug-likeness properties: The ADME profiles of all the active phytocompounds of Curcuma longa and hydroxychloroquine was displayed in Table 6. The molecular weight of all the active compounds of Curcuma longa was under 500 and the bioavailability score was 0.55/0.56 except Beta-Carotene. All the active phytocompounds of Curcuma longa were agreeing with Lipinski’s rule of 5 except Beta-Carotene. Even though the Beta-Carotene was the highest active compound (-10.12 kcal mol^–1 of binding energy) in Curcuma longa against SARS-CoV-2, it could not be used as a drug for the treatment of COVID-19 because it is not obeying Lipinski’s rule of 5. But some studies proved that Beta-Carotene is not showing any cytotoxic effect against normal and cancer cells^50,51. The interaction between the Beta-Carotene and casein in milk was necessary for the stable conformation of casein⁵². The Beta-Carotene is also rich in pigmented fruits and vegetables (orange and carrot) and is responsible for their biological activities and colour^53-55.

Fig. 3(a-t):

Interactions of the active compounds from Curcuma longa with the amino acid residues of the main protease, (a) Beta-carotene, (b) Beta-sitosterol, (c) Stigmasterol, (d) Campesterol, (e) Cholesterol, (f) Phytosterols, (g) Ascorbic-acid, (h) Cyclocurcumin, (i) Quercetin, (j) Demethoxycurcumin, (k) Riboflavin, (l) Dicinnamoylmethane, (m) Bisacurone, (n) Calebin-a, (o) Dihydrocurcumin, (p) Curcumenone, (q) Isoprocurcumenol, (r) Curcumenol, (s) Beta-turmerone and (t) Bis-demethoxycurcumin

Fig. 4:

Common interaction between the five active compounds from Curcuma longa Rounded amino acids indicated the common interaction, five active compounds (beta-carotene, beta-sitosterol, stigmasterol, campesterol, cholesterol)

Fig. 5:

Common interaction between the hydroxychloroquine and beta-carotene Rounded amino acids indicated the common interaction

In Curcuma longa also the Beta-Carotene is contributing to its biological activities. So, Beta-Carotene rich fruits and vegetables can be recommended as the natural remedy for COVID-19 patients. The other 19 molecules Curcuma longa are obey Lipinski’s rule of 5. So, they can be used as drug candidates against COVID-19 with clinical trials.

Pharmacophore study: PharmaGist webserver was used to identify the important special and structural features of active phytochemicals from Curcuma longa. Pharmacophore features in the screened active phytochemicals were represented by a set of special and structural features which were responsible for their biological activity in the active site of the target protein (Table 7). These pharmacophore features are helpful for the interaction with the target proteins. The normal pharmacophore of a compound is consisting of aromatic, hydrophobic elements, hydrogen bonds donors and acceptors, positive and negative functional groups⁵⁶. Among the active phytochemicals, Beta-Carotene is having the highest number of structural and special features which may be responsible for the highest binding activity (-10.12 kcal mol^–1). The number of donors and acceptors present in the molecules were not responsible for the binding activity because Beta-Carotene did not have any donors or acceptors whereas the riboflavin is having 6 donors and 7 acceptors.

The pharmacophoric features of the 20 active compounds were compared with the reference compound (Hydroxychloroquine). The pharmacophore analysis revealed that all the active 20 compounds from Curcuma longa have essential features comparable to the reference molecule. So, they can be used as a drug candidate against SARS-CoV-2. Besides, ADME results revealed that all the active compounds are non-toxic to humans except Beta-Carotene. All the 20 active compounds were present in the rhizome of the Curcuma longa (Table 1). Numerous studies have reported that the rhizome of Curcuma longa was used as herbal medicine to treat a wide variety of viral diseases⁵⁷. The present study also proved that the rhizome was having all the active substances against SARS-CoV-2. So, the Curcuma longa rhizome itself can be recommended as herbal medicine against COVID-19. The findings of this study are preliminary and further molecular dynamics and simulation studies are required for more precise confirmations. The dose-dependent activity of Curcuma longa against COVID-19 should be carried out to get a deeper insight into the mode of inhibition.

CONCLUSION

The results of this study proved that twenty active compounds from Curcuma longa were capable to bind to the main protease of SARS-CoV2. The phytochemicals such as Beta-Carotene, Beta-Sitosterol, Stigmasterol, Campesterol, Cholesterol, Phytosterols, Ascorbic-Acid and Cyclocurcumin had remarkably high binding energies and were present in the rhizome of Curcuma longa. So, the rhizome of Curcuma longa is a potential candidate for clinical research against COVID-19.

SIGNIFICANCE STATEMENT

This study discovers the active twenty phytocompounds from Curcuma longa to act against SARS-CoV2. All the twenty active compounds were present in the rhizome of Curcuma longa. It is possible to synergistically act all these compounds against COVID-19. So, the rhizome of Curcuma longa is a potential candidate for clinical research against COVID-19. This study will help the researcher to identify a suitable drug candidate against COVID-19.

ACKNOWLEDGMENTS

The authors wish to thank the management of the Malankara Catholic College and Hindustan Institute of Technology and Science for giving support to this study. The authors also wish to thank Dr. K. Nandakumar (Director, Research, Hindustan Institute of Technology and Science, Chennai), Rev. Fr. Prem Kumar, Rev. Fr. Jose Bright and Mr. K. Suresh for their encouragement in this study.