Potential inhibitors of coronavirus 3-chymotrypsin-like protease (3CLpro): an in silico screening of alkaloids and terpenoids from African medicinal plants.

The novel coronavirus disease 2019 (COVID-19) caused by SARS-COV-2 has raised myriad of global concerns. There is currently no FDA approved antiviral strategy to alleviate the disease burden. The conserved 3-chymotrypsin-like protease (3CLpro), which controls coronavirus replication is a promising drug target for combating the coronavirus infection. This study screens some African plants derived alkaloids and terpenoids as potential inhibitors of coronavirus 3CLpro using in silico approach. Bioactive alkaloids (62) and terpenoids (100) of plants native to Africa were docked to the 3CLpro of the novel SARS-CoV-2. The top twenty alkaloids and terpenoids with high binding affinities to the SARS-CoV-2 3CLpro were further docked to the 3CLpro of SARS-CoV and MERS-CoV. The docking scores were compared with 3CLpro-referenced inhibitors (Lopinavir and Ritonavir). The top docked compounds were further subjected to ADEM/Tox and Lipinski filtering analyses for drug-likeness prediction analysis. This ligand-protein interaction study revealed that more than half of the top twenty alkaloids and terpenoids interacted favourably with the coronaviruses 3CLpro, and had binding affinities that surpassed that of lopinavir and ritonavir. Also, a highly defined hit-list of seven compounds (10-Hydroxyusambarensine, Cryptoquindoline, 6-Oxoisoiguesterin, 22-Hydroxyhopan-3-one, Cryptospirolepine, Isoiguesterin and 20-Epibryonolic acid) were identified. Furthermore, four non-toxic, druggable plant derived alkaloids (10-Hydroxyusambarensine, and Cryptoquindoline) and terpenoids (6-Oxoisoiguesterin and 22-Hydroxyhopan-3-one), that bind to the receptor-binding site and catalytic dyad of SARS-CoV-2 3CLpro were identified from the predictive ADME/tox and Lipinski filter analysis. However, further experimental analyses are required for developing these possible leads into natural anti-COVID-19 therapeutic agents for combating the pandemic.Communicated by Ramaswamy H. Sarma.

Introduction

The novel coronavirus called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the pandemic officially called coronavirus disease 2019 (COVID-19), and was first identified in December 2019 in Wuhan, China from cluster of patients(Zhu et al., 2020). This is somewhat different from the world-wide endemic human coronaviruses HCoV-229E, HCoV-NL63, HCoV-HKU1, and HCoV-OC43; the zoonotic Middle East respiratory syndrome coronavirus (MERS-CoV); and the severe acute respiratory syndrome coronavirus (SARS-CoV) known with high mortality (Who, 2020). People infected with SARS-CoV-2 have shown symptoms including fever, cough, apnea, dyspnea and respiratory symptoms (Cascella et al., 2020). Unfortunately, there is currently no specific medicine or treatment for COVID-19 (Wu et al., 2020). From the first documented case, COVID-19 has since spread to over 190 countries and territories in the world including many African countries like Algeria, Egypt, Tunisia, Morocco, Nigeria, South Africa, and Senegal where hundreds of millions of people live with HIV/AIDS, tuberculosis or malaria (Dong et al., 2020). The genome of SARS-CoV-2 is about 80% identical and evolutionarily related to the beta-coronavirus responsible for the severe acute respiratory syndrome (SARS), which caused global outbreak in 2003 (Chen et al., 2020; Wang et al., 2020; Wu et al., 2020).

An overarching therapeutic enquiry is to rapidly identify and develop antiviral agents, as no drug nor vaccineis currently available to contain this on-going pandemic. In this direction, molecular docking and other computation techniques have proved valuable in the initial large-scale screening of several natural compounds and small molecules that directly inhibit important target proteins. Based on the available knowledge of the SARS-CoV-2 and closely related coronaviruses, reports on virtual screening of existing antiviral drugs (Boopathi et al., 2020 {Muralidharan et al., 2020), available databases (Khan et al., 2020) and natural agents (Aanouz et al., 2020; Elfiky & Azzam, 2020; Pant et al., 2020) against emerging targets such as viral spike proteins (Hasan et al., 2020), envelop protein (Gupta et al., 2020), protease (Khan et al., 2020), nucleocapsid protein (Sarma et al., 2020), 2′-O-ribose methyltransferase (Khan et al., 2020) and 3CL hydrolase (Elmezayen et al., 2020) is rapidly emerging.

The 3-chymotrypsin-like protease (3CLpro) also known as main protease (Mpro), and the papain-like protease (PLpro) of the virus are two proteases vital to transcription and replication of the proteins the viral genome encodes because of their peculiarity to split the two translated polyproteins (PP1a and PP1ab) into separate functional constituents (Chen et al., 2020). Therefore, the 3CLpro is considered to be a promising drug target and a lot of efforts have been committed to its study because of its key role in the replication cycle of the virus (Wang et al., 2020). The 3CLpro is responsible for the catalytic cleavage of 11 conserved sites in polyprotein 1ab (PP1ab) and 1a (PP1a) containing a large hydrophobic residue, a glutamine, and a small aliphatic amino acid residue (Anand et al., 2003). The structure and catalytic mechanism of 3CLpro makes it a selective target for drug development hence providing possible leads.

Some natural compounds and their derivatives that possess anti-inflammatory and anti-virus effects exhibit high binding affinity to 3CLpro (Wu et al., 2020). Likewise, alkaloids and terpenoids, from African plants, with documented antiviral, antimicrobial, antimalarial, antifungal, antileishmanial (Amoa Onguéné et al., 2013; Ndhlala et al., 2013; Osafo et al., 2017; Setzer et al., 2001) properties may inhibit the 3CLpro of SARS-CoV-2. This could also buttress reports that some patients with COVID-19 showed improvement when treated with hydroxychloroquine, an antimalarial drug (Wang et al., 2020; Wu et al., 2020)

The 3C-like cleavage sites on the polyproteins of coronaviruses are highly conserved, and their sequence and substrate specificities are identical i.e. SARS-CoV-2, MERS-CoV and SARS-CoV (Chen et al., 2020). This sequential similarity provides the basis for comparing SARS-CoV-2 with its previous counterpart leading to the identification of compounds with potentials to control or inhibit the replication of SARS-CoV-2. Therefore, identifying compounds with inhibitory capacity against the replication mechanism of SARS-CoV-2 could serve as breakthrough in the fight against COVID-19. The aim of this study is to establish a molecular model to delineate the possible inhibitory role of alkaloids and terpenoids against the 3CLpro of SARS-CoV-2, thereby providing a possible lead target against the novel pandemic coronavirus disease (COVID-19).

Materials and methods

Protein preparation

The crystal structures of coronoviruses 3CLpro used in the docking analysis were retrieved from the Protein Databank (http://www.rcsb.org) with their various PDB identification codes [COVID-19 (6lu7) (Souers et al., 2013); SARS-CoV (2duc) (Lessene et al., 2013); MERS-CoV (2yna) (Pelz et al., 2016)]. All the crystal structures of the 3CLpro were processed by removing existing ligands and water molecules while missing hydrogen atoms were added according to the amino acid protonation state at pH 7.0 utilizing Autodock version 4.2 program (Scripps Research Institute, La Jolla, CA). Thereafter, non-polar hydrogens were merged while polar hydrogens were added to each protein. The process was repeated for each protein and subsequently saved into a dockable pdbqt format for molecular docking.

Ligand preparation

Bioactive alkaloids (62) and terpenoids (100) from African plants were used in this study.

The Structure Data Format (SDF) structures of the 3CLpro reference inhibitors (Lopinavir and Ritonavir), and some of the compounds were retrieved from the PubChem database (www.pubchem.ncbi.nlm.nih.gov), other compounds not present on the database were drawn with chemdraw version 19. All the compounds and inhibitors were converted to mol2 chemical format using Open babel (O'Boyle et al., 2011). The nonpolar hydrogen molecules were merged with the carbons, polar hydrogen charges of the Gasteiger-type were assigned and the internal degrees of freedom and torsions were set to zero. The protein and ligand molecules were further converted to the dockable pdbqt format using Autodock tools.

Molecular docking

Virtual screening of the coronaviruses 3CLpro active regions, and determination of binding affinities of the alkaloids, terpenoids and reference inhibitors were carried out using AutoDock vina 4.2 with full ligand flexibility (Trott & Olson, 2010). The results of the top docked compounds obtained from vina were further validated by BINDSURF (https://bio-hpc.ucam.edu/achilles/) (Sanchez-Linares et al., 2012). Pdbqt form of individual protein and ligands were uploaded, and exhaustive series of docking calculations across the whole protein surface to find the spots with the best binding affinities were performed with vina. After the affinities were calculated, vina clusters the results according to the spatial overlapping of the resulting poses. For each cluster, the pose with the best affinity was taken as the representation of this cluster. The binding affinities of compounds for the selected cluster were recorded. The compounds were then ranked by their affinity. The molecular interactions between proteins and respective ligands with higher binding affinity were viewed using Discovery Studio Visualizer version 16.

ADMET study

To evaluate the drug-likeness prediction of the top compounds that showed a significant binding affinity for the 3CLpro of the of SARS-CoV-2, they were subjected to Lipinski filter in which an orally bio-active drug is expected not to violate more than one of the criteria for drug-likeness namely: cLogP, hydrogen donor and acceptor molecular mass, and molar refractive index (Nickel et al., 2014). The predicted Absorption Distribution Metabolism, Excretion and toxicity (ADME/tox) study were analyzed using the SuperPred webserver (http://lmmd.ecust.edu.cn:8000/) which is reported as an important tool in drug discovery (Cheng et al., 2012). The SuperPred webserver was employed for predicting important descriptors of drug-likeness. The SDF file and canonical SMILES of the compounds were downloaded from PubChem Database or copied from ChemDraw to calculate ADMET properties using default parameters.

Results and discussion

The result of the binding energy from the docking analysis of bioactive alkaloids (62) and terpenoids (100) to 3-Chymotrypsin-like protease (3CLpro) of the novel SARS-CoV-2 is represented in (supplementary material).

The alkaloids and terpenoids with more negative binding energy values have higher binding affinities for the specific 3CLpro protein, and were ranked higher. The twenty alkaloids and terpenoids with best binding affinities to 3CLpro of SARS-CoV-2, along with their binding affinities to 3CLpro of SARS-CoV and MERS-CoV are presented in Tables 1 and 2.

Table 1.

Binding affinities of reference compounds (ritonavir and lopinavir) and top 20 bioactive alkaloids from African plants to the 3CLpro of coronaviruses.

S/No	Bioactive Compounds	Class of compound	Plant species (Family)	SARS-Cov-2	SARS-CoV	MERS-CoV
	Ritonavir			−6.8	−6.6	−7.9
	Lopinavir			−8.3	−7.2	−5.6
1	10-Hydroxyusambarensine	Indole alkaloids	Strychnos usambarensis (Loganiaceae)	−10.0	−10.1	−8.5
2	Cryptoquindoline	Cryptolepines	Cryptolepis sanguinolenta (Periplocaceae)	−9.7	−9.3	−9.9
3	Cryptospirolepine	Cryptolepines	Cryptolepis sanguinolenta (Periplocaceae)	−9.2	−10.5	−10.9
4	Chrysopentamine	Indole alkaloids	Strychnos usambarensis (Loganiaceae)	−8.5	−8.9	−9.0
5	Isocryptolepine	Cryptolepines	Cryptolepis sanguinolenta (Periplocaceae)	−8.5	−6.7	−7.4
6	Strychnopentamine	Indole alkaloids	Strychnos usambarensis (Loganiaceae)	−8.2	−8.9	−9.6
7	Isostrychnopentamine	Indole alkaloids	Strychnos usambarensis (Loganiaceae)	−8.1	−9.0	−9.5
8	Normelicopicine	Acridones	Teclea trichocarpa (Rutaceae)	−8.1	−6.8	−6.7
9	Jozipeltine A	Naphthoisoquinolines	Triphyophyllum peltatum,	−8.0
10	5′-O-Demethyl-dioncophylline A	Naphthoisoquinolines	Triphyophyllum peltatum (Dioncophyllaceae)	−8.0	−7.8	−8.1
11	Dioncophylline C	Naphthoisoquinolines	Triphyophyllum peltatum (Dioncophyllaceae)	−7.9	−8.7	−7.7
12	Dioncopeltine A	Naphthoisoquinolines	Triphyophyllum peltatum (Dioncophyllaceae)	−7.8	−7.7	−7.9
13	Liriodenine	Indole alkaloids	Glossocalyx brevipes (Siparunaceae)	−7.6	−7.8	−8.1
14	5,6-Dihydronitidine	Furoquinolines	Toddalia asiatica (Rutaceae)	−7.6	−7.0	−8.2
15	Hydroxycryptolepine	Cryptolepines	Cryptolepis sanguinolenta (Periplocaceae)	−7.6	−7.0	−7.2
16	Cryptoheptine	Cryptolepines	Cryptolepis sanguinolenta (Periplocaceae)	−7.6	−7.1	−8.5
17	Annonidine F	Indole alkaloids	Monodora angolensis (Annonaceae)	−7.5	−7.9	−7.8
18	Ancistrotanzanine C	Naphthoisoquinolines	Ancistrocladus tanzaniensis (Acistrocladaceae)	−7.5	−7.6	−8.7
19	Fagaronine	Indole alkaloids	Fagara zanthoxyloides (Rutaceae)	−7.4	−7.7	−7.3
20	Alstonine	Indole alkaloids	Fagara zanthoxyloides (Rutaceae)	−7.4	−7.4	−8.1

Compounds having the highest binding affinity for the corresponding proteins are the ones indicated in bold values.

Table 2.

Binding affinities of top 20 bioactive terpenoids from African plants to the 3CLpro of coronaviruses.

S/No	Bioactive Compounds	Class of compound	Plant species (Family)	SARS-Cov-2	SARS-CoV	MERS-Cov
1	6-Oxoisoiguesterin	Bisnorterpenes	Bisnorterpenes	−9.1	−9.5	−9.3
2	22-Hydroxyhopan-3-one	Pentacyclic triterpenes	Cassia siamea (Fabaceae)	−8.6	−8.5	−9.1
3	Isoiguesterin	Bisnorterpenes	Bisnorterpenes	−8.1	−7.4	−9.4
4	20-Epi-isoiguesterinol	Bisnorterpenes	Bisnorterpenes	−8.1	−9.2	−9.3
5	20-Epibryonolic acid	Pentacyclic triterpenes	Cogniauxia podolaena (Cucurbitaceae)	−8.1	−8.6	−9.4
6	Oleanolic acid	Pentacyclic triterpenes	Nuxia sphaerocephala (Loganiaceae)	−8.5	−8.6	−8.2
7	3-Oxolupenal (3-oxolup-20(29)-en-30-al)	Pentacyclic triterpenes	Nuxia sphaerocephala (Loganiaceae)	−8.4	−7.8	−8.8
8	2 ,3 ,19 -Trihydroxy-urs-12-20-en-28-oic acid	Pentacyclic triterpenes	Kigelia africana (Bignoniaceae)	−8.4	−9.0	−8.7
9	3-Oxolupenol (30-hydroxylup-20(29)-en-3-one)	Pentacyclic triterpenes	Nuxia sphaerocephala (Loganiaceae)	−8.3	−8.1	−8.9
10	3-O-betulinic acid p-coumarate	Pentacyclic triterpenes	Baillonella toxisperma (Sapotaceae)	−8.3	−8.2	−8.8
11	Isoiguesterinol	Bisnorterpenes	Bisnorterpenes	−8.1	−8.9	−9.3
12	3- Benzoylhosloppone	Abietane diterpenes	Hoslundia opposita (Lamiaceae)	−8.1	−8.5	−8.7
13	7 -Acetoxy-6,12-dihydroxy-abieta-8, 12-Diene-11,14-dione	Abietane diterpenes	Plectranthus hadiensis (Lamiaceae)	−7.9	−7.7	−6.5
14	Cryptobeilic acid C	Beilshmiedic acid derivatives	Beilschmiedia cryptocaryoides (Lauraceae)	−7.9	−8.3	−7.8
15	3 -Hydroxylupenal (3 -hydroxylup-20(29)-en-30-al)	Pentacyclic triterpenes	Nuxia sphaerocephala (Loganiaceae)	−7.9	−7.8	−9.3
16	3-Friedelanone	Pentacyclic triterpenes	Hypericum lanceolatum (Hypericaceae)	−7.9	−8.7	−8.7
17	6-Acetylswietenolide	Limonoids	Khaya grandifoliola (Meliaceae)	−7.8	−7.6	−7.9
18	11-Hydroxy-19-(4-hydroxy-benzoyloxy)-abieta-5, 7,9(11),13-tetraene-12-one	Abietane diterpenes	Plectranthus purpuratus (Lamiaceae)	−7.8	−8.2	−8.6
19	11-Hydroxy-19-(3,4-dihydroxybenzoyloxy)-abieta-5, 7,9(11),13-tetraene-12-one	Abietane diterpenes	Plectranthus purpuratus (Lamiaceae)	−7.8	−8.5	−8.3
20	3-Hydroxy-20(29)-lupen-28-ol	Pentacyclic triterpenes	Schefflera umbellifera (Araliaceae)	−7.8	−8.3	−8.2

Compounds having the highest binding affinity for the corresponding proteins are the ones indicated in bold values.

The results from this study revealed that lopinavir and ritonavir, the reference inhibitors, had a binding affinity of −8.3 and −6.8 Kcal/mol, respectively, for 3CLpro of SARS-CoV-2 (Table 1). The binding affinity of lopinavir and ritonavir for 3CLpro of SARS-CoV was −7.2 and −6.6 Kcal/mol, respectively, while for 3CLpro of MERS-CoV was −5.6 and −7.9 Kcal/mol, respectively (Table 1). It was observed that more than half of the selected top 20 alkaloids and terpenoids had a binding affinity for the 3CLpro of the SARS-coronaviruses that surpassed that of the reference inhibitors (Tables 1 and 2).

The 2 top docked alkaloids to SARS-CoV-2 3CLpro are 10-hydroxyusambarensine (-10.0 kcal mol−1) and cryptoquindoline (−9.7 kcal.mol−1) (Table 1). It was observed that while 10-hydroxyusambarensine was the second top docked compound to the 3CLpro of SARS-CoV, cryptospirolepine had the highest binding affinity to that of SARS-CoV and MERS-CoV (Table 1). The result further showed that 10-hydroxyusambarensine was more selective for SARS-CoV-2 though interacted strongly with the target protein of the other coronavirus, while cryptospirolepine was more selective for the 3CLpro of the MERS-CoV and SARS-CoV respectively (Table 1).

The terpenoids, 6-oxoisoiguesterin (−9.1 kcal.mol−1) a bisnorterpenes, and 22-hydroxyhopan-3-one (−8.6 kcal.mol−1) a pentacyclic triterpenes are the 2 top-docked compounds base on the binding affinities (Table 2). 6-oxoisoiguesterin had the highest binding affinity to the 3CLpro of SARS-CoV-2 while 20-epi-isoiguesterinol, isoiguesterin 20-epibryonolic acid were the top docked compounds to 3CLpro of SARS-CoV and MERS-CoV (Table 2). The binding energies of the terpenoids revelaed that 6-oxoisoiguesterin was more selective for the 3CLpro of SARS-CoV and SARS-Cov-2, while isoiguesterin and 20-epibryonolic with the same binding energy (−9.4 kcal.mol−1) interacted more strongly with the 3CLpro of MERS-CoV than that of other coronaviruses.

Amino acid interaction of selected bioactive alkaloids and terpenoids with 3CLpro of coronaviruses

The interactions of reference inhibitors, and top ranked alkaloids and terpenoids with the amino acids of 3CLpro of coronaviruses are represented in Table 3.

Table 3:

Interacting amino acid residues of 3CLpro of coronaviruses with the top binding alkaloids and terpenoids from African plants.

Bioactive compound	Coronavirus	Interacted residues	Protein atom involved in H-bonding (BOND DISTANCE)
Ritonavir	SARS-Cov-2	GLU166 GLY143 MET49 MET165 PRO168	GLY143 (2.97) GLU166 (2.97)
Lopinavir		GLN110 ASP153 SER158 ILE106 VAL104 PHE294 VAL297 PRO293 VAL202 ILE249	GLN110 (2.11) ASP153 (2.80) SER158(3.09)
10 -Hydroxyusambarensine		GLN189 TYR54 MET49 MET165 HIS163 CYS145 GLU166 PRO168	GLN189 (2.97)
Cryptoquindoline		CYS148 MET49 MET165
6-Oxoisoiguesterin		GLN189 MET49 MET165 HIS41 CYS145	GLN189 (2.75)
22-Hydroxyhopan-3-one		LYS137 LEU275 LEU287 LEU286 TYR239	LYS137 (3.16)
10-Hydroxyusambarensine	SARS-CoV	PHE294 LEU202 PRO293 VAL104 ASP153
Cryptospirolepine		MET49 GLU47 CYS145
6-Oxoisoiguesterin		THR292 THR111 PRO252 PRO293 ILE294 PHE294 VAL297	THR292 (3.30) THR111 (2.01)
20-Epi-isoiguesterinol		THR24 THR25 ALA46 CYS145 HIS41 MET165	THR24 (2.97) THR25(2.92)
Cryptospirolepine	MERS-CoV	ASP294 SER114 ALA113 THR154 ASP295 MET298	ASP294
Cryptoquindoline		ASP294 ASP295 MET298SER114 ALA113 THR154
Isoiguesterin		ASP294 THR292 ALA113 PRO293 LYS110HIS135 VAL246 PRO111 CYS203 ILE205	ASP294 (2.35)THR292 (3.08)
20-Epibryonolic acid		ASP294 CYS203 SER250 PRO293 ILE205VAL246	ASP294 (2.94) CYS203 (2.56) SER250 (2.99)

The ligands majorly interacted with the residues through hydrophobic interactions, with few H-bonding above 3.40 Å.

The result obtained from the ligand-protein binding interaction showed that ritonavir was docked into the receptor-binding site and catalytic dyad (Cys-145 and His-41) of SARS-CoV-2 (Figure 2e). Ritonavir interacted via a conventional hydrogen bond to GLY143 and GLU166. It further interacted with MET165 via a Pi-Sulfur bond and via Pi-Alkyl interaction to PRO168 and MET49 (Figure 2e). Lopinavir with a considerable higher binding energy (−8.3 kcal.mol−1) than ritonavir did not show significant binding to the catalytic dyad (Cys-145 and His-41) of SARS-CoV-2. It interacted via Hydrogen bond to GLN110, ASP153, and SER158, Pi-Sigma bond to ILE106 of Domain II; Pi–Pi Stacking to PHE294 of Domain III; Amide-Pi Stacking to PRO293 of Domain III; Alkyl and Pi-Alkyl to the other residues (Table 3, Figure 2f).

Figure 2.

Visualization of SARS-Cov-2 3CLpro amino acids interactions with ligands (a) 10-Hydroxyusambarensine (b) Cryptoquindoline (c) 6-Oxoisoiguesterin (d) 22-Hydroxyhopan-3-one (e) Ritonavir (f) Lopinavir; (i) Cartoon representation, showing binding conformation (ii) interactions view with important residues.

10-Hydroxyusambarensine the topmost docked compound to the 3CLpro of SARS-CoV-2 interacted via a conventional hydrogen bond to the GLN189 residue a carbon-hydrogen bond to GLN166 and CYS145, a Pi-Sulfor to MET165, a Pi-Cation to HIS41, a Pi-Alkyl and Pi-Pi Stacking to other residues (Figure 2a and Table 3) Cryptoquindoline interacted to CYS145 in a similar pattern as 10-Hydroxyusambarensine, and via Pi-Alkyl to MET49 (Figure 2b). A conventional hydrogen bond was observed between 6-Oxoisoiguesterin and ARG189. 6-Oxoisoiguesterin exhibited similar binding patterns, creating an alkyl and Pi-Alkyl stacking with MET49, MET165, and CYS145 (Figure 2c). 22-Hydroxyhopan-3-one interacted via conventional hydrogen bond to LYS137 and via Alkyl and Pi-Alkyl to LEU275 LEU287 LEU286 and TYR239 (Figure 2d).

Unlike the ligand-protein binding interaction of 10-Hydroxyusambarensine to SARS-CoV-2 3CLpro that targeted the Cys-His catalytic dyad (Cys-145 and His-41) along with the other binding residues, the docking analysis revealed that the SARS-CoV 3CLpro interacted with the same ligand differently. A Pi-Pi and Amide- Pi Stacking was observed between 10-Hydroxyusambarensine, PHE294, and PRO293 respectively, it further interacted with VAL104 and ASP153 of the Domain II (102-184 residues) region of the protease via Pi-Alkyl and Pi-Anion bonds (Figure 4a). Cryptospirolepine interacted via a Carbon Hydrogen, Pi-Donor Hydrogen, and Pi-Alkyl bond to CYS145, GLU47, and MET49 respectively (Figure 4b), these residues are implicated in the binding of a substrate that targeted the Cys-His catalytic dyad (Cys-145 and His-41) of the protein. THR292 and THR111 of the protein interacted with 6-oxoisoiguesterin via hydrogen bond, while the other residues (Table 3) interacted via hydrophobic bonds to 6-oxoisoiguesterin (figure 4c). The interaction of 20-epi-isoiguesterinol to the SARS-CoV 3CLpro revealed a hydrogen bonding to THR24 and THR25 while hydrophobic interaction was observed between CYS145 HIS41 MET165 and the ligand (Figure 3d). These residues are also conserved for the Cys-His catalytic dyad binding hotspot.

Figure 4.

Visualization of SARS-Cov 3CLpro amino acids interactions with ligands (a) 10-Hydroxyusambarensine (b) Cryptospirolepine (c) 6-Oxoisoiguesterin (d) 20-Epi-isoiguesterinol.

Figure 3.

Surface view of ligand in binding cavity of SARS-Cov-2 3CLpro (a) 10-Hydroxyusambarensine (b) Cryptoquindoline (c) 6-Oxoisoiguesterin (d) 22-Hydroxyhopan-3-one (e) Ritonavir (f) Lopinavir.

Cryptospirolepine, cryptoquindoline, isoiguesterin and 20-epibryonolic acid interacted with the MERS-CoV 3CLpro in a conservative manner that is different from the other two coronaviruses. The predominate residue involved in the binding processes was outside the Domain I and II of the protein. A hydrogen bond was observed between cryptospirolepine/isoiguesterin/20-epibryonolic acid and ASP294 while an extra hydrogen bond interaction with SER250 in the case of the latter was also observed (Figure 5 a–c). Cryptospirolepine/cryptoquindoline interacted via hydrophobic (Pi-Sulfur, Pi-Anaion, Pi-Alkyl, and Pi-Donor Hydrogen bonds) to MET298ASP294 ALA113, SER114, and THR154, respectively (Figure 5d).

Figure 5.

Visualization of MERS-CoV 3CLpro amino acids interactions with ligands (a) Cryptospirolepine (b) Cryptoquindoline (c) Isoiguesterin (d) 20-Epi-isoiguesterinol.

Summary of phamacokinetic properties of top binding alkaloids and terpenoids from African plants to the 3CLpro of SARS-Cov-2: (a) 10-hydroxyusambarensine, (b) cryptoquindoline, (c) 6-oxoisoiguesterin, (d) 22-hydroxyhopan-3-one.

The colour space is the suitable physiochememical space for oral bioavailability.

LIPO Lipophility: -0.7 < XLOGP3 < +5.0.

SIZE: 150g/mol : < MW < 500g/mol.

POLAR (Polarity): 20Å2 < TPSA < 130 Å 2.

INSOLU (insolubility): 0 < Log S (ESOL) < 6.

INSATU (insaturation): 0.25 < Fraction Csp3 < 1.

FLEX (Flexibity): 0 < Num. rotatable bonds < 9.

From the results obtained in this study, 7 compounds (10-hydroxyusambarensine, cryptoquindoline, 6-oxoisoiguesterin, 22-hydroxyhopan-3-one, cryptospirolepine, isoiguesterin, 20-epibryonolic acid) with a remarkable inhibitory tendency towards the SARS-coronavirus were identified from plant-derived alkaloids and terpenoids. 10-hydroxyusambarensine, cryptoquindoline, 6-oxoisoiguesterin, cryptospirolepine, and 20-epi-isoiguesterinol were docked into the Cys-His catalytic dyad of the coronaviruses 3CLpro in a similar pattern as ritonavir, while the rest compounds interacted with the amino residues in the cleft of Domain III of the protein.

Revalidation of docking scores

The result of the top docked compounds that were obtained from the docking analysis using vina was further revalidated by the docking scores from BINDSURF analysis. The results from Bindsurf analysis was similar to that of vina in most cases the ligand had the same binding affinities with respective proteins as presented, only few differ with the range of ± 2 Kcal/mol. All the Ligands were docked into cavities as revealed by Vina. The ligands exhibited a similar binding pattern and interacted with similar amino acids as presented by Vina. The additional interactions as presented by BINDSURF analysis are listed in Table 4.

Table 4:

Revalidation of the binding and docking properties of top docked compounds to respective proteins.

Bioactive compound	Interacted residues	Interacted residues involved in H-bonding (BOND DISTANCE)	Binding energy Kcal/mol	Poses in Cluster	Best Pose	Binding site coordinate
SARS-CoV-2
Ritonavir	GLN189 GLU166 GLY143 MET49 MET165 PRO168	GLY143 (2.61) GLU166 (2.81)	−7.0	95	181	−10.94, 12.98, 68.43
Lopinavir	GLN110 ASP153 SER158 ILE106 VAL104 PHE294 VAL297 PRO293 VAL202 ILE249	GLN110 (2.11) ASP153 (2.80) SER158(3.09)	−8.4	81	133	−38.75, 11.96, 5111
10 -Hydroxyusambarensine	GLN189 TYR54 MET49 MET165 HIS163 CYS145 GLU166 PRO168 HIS41 PHE140	GLN189 (2.97) TYR54 (2.51)	−10.1	64	147	−11.34, 12.13, 69.13
Cryptoquindoline	CYS148 MET49 MET165 HIS41 PHE140 ASN142 GLN189		"−9.70	56	26	−9.89, 13.66, 67.80
6-Oxoisoiguesterin	GLN189 MET49 MET165 ASN142HIS41		−9.10	67	142	−9.76, 15.40, 68.75
22-Hydroxyhopan-3-one	LYS137 LEU275 LEU287 LEU286 TYR239 THR199 TYR273	LYS137 (3.16)	−8.70	65	168	−26.41, −6.36, 55.37
SARS-CoV
10 -Hydroxyusambarensine	VAL104 ILE106 GLN110 LEU202 ILE249 PHE294 PRO252 THR292		−10.10	74	106	56.09, −7.81, 27.36
Cryptospirolepine	MET49 GLU47 GLU166 THR25	HIS41 (2.21)	−10.50	58	24	42.31, 12.60, 3.68
6-Oxoisoiguesterin	THR292 THR111 PRO252 PRO293 ILE294 PHE294 VAL297 GLN110 ASN151	THR292 (3.30) THR111 (2.01) GLN110 (2.46) ASN151 (2.81)	−9.50	64	110	55.62, −8.30, 27.53
20-Epi-isoiguesterinol	THR24 THR25 ALA46 CYS145 HIS41 MET165	THR24 (2.97) THR25(2.92)	−9.3	58	24	42.99, 13.10, 3.81
MERS-CoV
Cryptospirolepine	ASP294 THR154 THR292	ASP294 (2.49) THR292 (3.13)	−10.90	78	308	9.50, 58.23, 30.81
Cryptoquindoline	MET6 THR154		−9.70	68	116	8.53, 57.22, 28.19
Isoiguesterin	ASP294 THR292 ALA113 PRO293 LYS110HIS135 VAL246 PRO111 CYS203 ILE205	ASP294 (2.35)THR292 (3.08)	−9.40	85	138	12.97, 55.41, 39.38
20-Epibryonolic acid	ASP294 CYS203 SER250 PRO293 ILE205VAL246	ASP294 (2.94) CYS203 (2.56) SER250 (2.99)	−9.5	78	178	12.51, 56.11, 38.22

Figure 3 shows that from the 4 top docked compounds to the 3CLpro of SARS-CoV, 3 compounds namely: 10-Hydroxyusambarensine, Cryptoquindoline, 6-Oxoisoiguesterin where further revealed to be docked into the substrate binding cavity (Cys-His catalytic dyad (Cys-145 and His-41)). The binding site coordinate further showed that the 3 compounds were docked into the same cavity as the ritonavir.

Pharmacokinetic properties of selected alkaloids and terpenoids

The result generated from the Lipinski and ADME/tox filtering analyses are represented in Figure 5 and Table 5.

Table 5.

Physicochemical properties of the top binding alkaloids and terpenoids from African plants.

(a) Lipinski filter analysis
Lipinski filters	10 -Hydroxyusambarensine	Cryptoquindoline	6-Oxoisoiguesterin	22-Hydroxyhopan-3-one
Molecular weight (g/mol)	448.56	448.52	420.58	442.72
Num. heavy atoms	34	35	31	32
Num. rotatable bonds	2	1	0	1
Num. H-bond acceptors	3	2	3	2
Hydrogen bond donor	3	0	2	1
cLogP	3.31	4.02	4.80	4.41
Molar Refractivity	142.46	145.65	126.84	135.30
Lipinski violation	0	0	0	0
(b) admet SAR
Absorption (Probability)
Blood-Brain Barrier	BBB+ (0.83)	BBB+ (0.95)	BBB+ (0.53)	BBB+ (0.97)
Human Intestinal Absorption	HIA+ (0.98)	HIA+ (0.99)	HIA+ (0.99)	HIA+ (1.00)
Bioavailability Score	0.55	0.55	0.55	0.55
Caco-2 Permeability	Caco2+ (0.53)	Caco2+ (0.76)	Caco2+ (0.79)	Caco2+ (0.85)
P-glycoprotein Substrate	Substrate (0.91)	Non-inhibitor (0.69)	Non-inhibitor (0.77)	Substrate (0.54)
P-glycoprotein Inhibitor	Non-inhibitor (0.60)	Non-inhibitor (0.72)	Non-inhibitor (0.71)	Non-inhibitor (0.61)
Renal Organic Cation Transporter	Inhibitor (0.80)	Non-inhibitor (0.67)	Non-inhibitor (0.83)	Non-inhibitor (0.77)
Distribution (Probability)
Subcellular localization	Mitochondria (0.65)	Mitochondria (0.55)	Mitochondria (0.89)	Mitochondria (0.53)
Metabolism
CYP450 Substrate	Substrate (0.53) Non-inhibitor (0.83)	substrate Non-inhibitor (0.76)	Non-substrate (0.84) inhibitor (0.67)	Inhibitor (0.73) Non-substrate (0.63)
Toxicity
AMES Toxicity	Non AMES toxic (0.75)	AMES toxic (0.89)	Non AMES toxic (0.91)	Non AMES toxic (0.87)
Carcinogens	Non-carcinogens (0.97)	Non-carcinogens (0.92)	Non-carcinogens (0.90)	Non-carcinogens (0.88)
Acute Oral Toxicity	III (0.51)	III (0.67)	III (0.63)	III (0.75)
Rat Acute Toxicity LD50, mol/kg	2.7896	2.4420	2.0255	2.7443
Aqueous solubility (LogS)	−2.7626	−3.1120	−4.7201	−4.1004
Pharmacokinetics
GI absorption	High	Low	High	low
Log Kp (skin permeation) cm/s	−5.70	−3.97	−3.93	−3.29

Four (4) compounds fulfilled the requirement for Lipinski analysis of the rule of-five with corresponding favourable predicted ADME/tox parameters. This includes 2 alkaliods (10-hydroxyusambarensine and cryptoquindoline), and 2 terpenoids (6-oxoisoiguesterin and 22-hydroxyhopan-3-one) (Figure 1). The predicted physiochemical properties for bioavailability of the lead compounds was further represented in Figure 5.

Figure 1.

Structure of alkaloids and terpenoids with remarkable binding energy to SARS-CoV-2 3CLpro (1a) 10-Hydroxyusambarensine (1b) Cryptoquindoline (2a) 6-Oxoisoiguesterin (2b) 22-Hydroxyhopan-3-one.

The ADME/tox and pharmacokinetic properties from the filtering analyses suggested four compounds with a high probability of absorption, subcellular distribution, except for AMES toxicity parameter which indicated cryptoquindoline to be toxic at a probability of 0.89. All the four compounds were indicated to be non-carcinogenic, with very low acute toxicity and aqueous solubility of < 0. The gastrointestinal absorption index was indicated to be high for 10-hydroxyusambarensine and 6-oxoisoiguesterin but low for the other two compounds (Table 5).

Discussion

Potential anti-coronavirus therapies are classified based on their targets: action of the lead agent on the host immune system, host cells, or on the coronavirus itself. This study is focused on the inhibition of 3CLpro as a therapeutic approach of inhibiting the replicative potential of coronaviruses; the protein is specific for the virus and its inhibition may not pose health threats to the host.

The identification and development of potent and efficacious chemical agents against coronavirus could be achieved via three strategies (Wu et al., 2020; Zumla et al., 2016): testing existing broad-spectrum of anti-virals (Chan et al., 2013); the use of known molecular databases to screen for agents that may have a therapeutic effect on coronavirus (de Wilde et al., 2014; Dyall et al., 2014); and the direct-based genomic information and pathological characteristics of different coronaviruses to develop new targeted drugs. Some peptides and small molecules have been reported so far, as inhibitors that target SARS-CoV 3CLpro (Pillaiyar et al., 2016). Natural products such as alkaloids and terpenoids with antiviral, antimalarial, antibacterial, anti-inflammatory, and immunomodulatory activities could effectively interact with target such as 3CLpro of SARS-CoV-2 (Chen et al., 2020), and be vital for the prevention and treatment of COVID-19.

A monomer of 3CLpro has three domains: domain I (residues 8–101), domain II (residues 102–184), and domain III (residues 201–303); and a long loop (residues 185–200) connects domains II and III. The catalytic site of 3CLpro occupies the gap between domains I and II, and has a CysHis catalytic dyad (Cys145 and His41).The enzymatic activity of the 3CLpro resides in the catalytic dyad of Cys145 and His41 (Yang et al., 2003). Lopinavir and Ritonavir, the reference compounds used in this study are antiretroviral, protease inhibitors recommended for the treatment of SARS and MERS, which have similar mechanisms of action as HIV (Li et al., 2020).

The top docked alkaloids and terpenoids with docking scores that surpassed that of the reference inhibitors suggests their promising efficacy as potential inhibitors of SARS-CoV-2 and the probable role of structural flexibility of interactions with the 3CLpro of other coronaviruses (Sang et al., 2020).This binding is probably achieved by covalent alteration of the thiolate anion of the active Cys residue (Berry et al., 2015) because most of the interactions involved the Cys residue. The antiviral activity of the alkaloids against coronaviruses is associated with the direct inhibition the 3CLpro (Jo et al., 2020).

The binding affinity exhibited by 10-hydroxyusambarensine, cryptoquindoline, and cryptospirolepine (the top docked alkaloids) portrays the alkaloids to have strong binding tendency with inhibitory potential against 3CLpro of SARS-CoV-2, SARS-CoV. While 10-hydroxyusambarensine exhibited the strongest interactions with 3CLpro of SARS-CoV-2, cryptospirolepine exhibited highest binding affinity and selectivity for 3CLpro of SARS-CoV and MERS-CoV. These compounds had higher docking scores, good binding affinities, and more interactions with the conserved catalytic dyad residues (Cys-145 and His-41) that are important for the catalytic function of the 3CLpro. These compounds also interacted with Asp187 know to enhance the catalytic efficiency of 3CLpro (Zhao et al., 2008).

The comparison between the revalidated and vina result showed a consistency in binding affinities and binding pattern, also the binding site coordinate of the compounds in selected cluster as indicated by BINDSURF docking analysis further revalidate the docking of compounds onto the Cys-His catalytic dyad (Cys-145 and His-41) of the SARS-CoV-2 3CLOpro.

The binding pattern of the top docked terpenoids was conserved between SARS-CoV-2 and SARS-CoV. The binding of 6-oxoisoiguesterin to the 3CLpro of SARS-CoV-2 and SARS-CoV makes it a valuable compound against the viral activities of these coronaviruses. However, Isoiguesterin and 20-epibryonolic acid may be better inhibitors of the viral function of MERS-CoV 3CLpro since they interacted with amino acid residue of the domain III. Domain III of 3CLpro is also proposed as a new target for specific inhibitors because of its critical importance in the function of the viral enzyme (Anand et al., 2003).

The better interactions of these terpenoids with 3CLpro compared with that of reference compounds, suggest the terpenoids may alter the viral protease function essential for processing the viral replicase polyproteins. A series of diterpenoids have been identified as moderate competitive inhibitor of the 3CLpro. A structural activity relationship suggested the quinine-methide and hydrophobic E ring assists in producing the inhibitory activity (Pillaiyar et al., 2016). The results obtained in this study corroborate findings from similar studies in which the selected compounds were docked into the conserved catalytic dyad residues (Cys-145 and His-41), and interacted with receptor-binding residues Thr24, Thr25, Thr26, Met49, Asn142, Cys44, Thr45, Ser46, Gly143, His164, Glu166 and Gln189 (Ul Qamar et al., 2020).

The strikingly similar ligand-binding pattern exhibited by ritonavir, 10-hydroxyusambarensine, cryptoquindoline, 6-oxoisoiguesterin and 20-Epi-isoiguesterinol to the CysHis catalytic dyad (Cys145 and His41) and HIS163/HIS172/GLU166 indicates the conservation of the catalytic dyad across the two. The latter residues are believed to provide the opening gate for the substrate in the active site of the protome (Yang et al., 2003) of SARS-Cov-2 and SARS-CoV 3CLpro. This further reveals that SARS-Cov-2 receptor-binding pocket conformation resembles that of the SARS-CoV 3CLpro and raises the possibility that inhibitors intended for SARS-Cov-2 3CLpro may also inhibit the activity of SARS-CoV 3CLpro. The different binding pattern exhibited by 3CLpro of MERS-CoV to its top docked ligands, in which most of them were docked into the cleft of domain III, is consistent with the findings that SARS-Cov-2 is more similar to SARS-CoV than MERS-CoV (Xu et al., 2020; Zhou et al., 2020; Zhu et al., 2020).

From the Lipinski and predicted ADME/tox filtering analyses, we identified four non-toxic, druggable natural compounds that bind to the receptor-binding site and catalytic dyad (Cys-145 and His-41) of SARS-CoV-2 3CLpro. The four compounds (Table 4) which are potential drug candidates had no violation of Lipinski’s rule. This accepted Rule-of-five defines the relationship physicochemical between and pharmacokinetics parameters (Lipinski, 2000). Lipinski’s rule states that, generally, an orally active drug will not have more than one violation of the following criteria: Not >5 hydrogen bond donors (oxygen or nitrogen atoms with one or more hydrogen atoms); Not >10 hydrogen bond acceptors (nitrogen or oxygen atoms); A molecular mass <500 Daltons; and an octanol-water partition coefficient (log P) not greater than 5. The result from the predicted filtering analyses of the four compounds showed parameters that suggests a favourable ADME/tox and pharmacokinetic properties. This further indicates the druggability potential of the best docked alkaloids and terpenoids. This may be a pointer to a structure-based design of drugs that targets the 3CLpro of SARS-CoV-2, and useful against COVID-19.

Conclusion

In conclusion, our study revealed that natural agents from the alkaloids and terpnoids class of compounds are capable of inhibiting the 3CLpro with a highly conserved inhibitory pattern to both SARS-CoV-2 and SARS-CoV. Taken into account the site of inhibition, the strength of the binding affinity, the interaction with the conserved catalytic dyad residues (Cys-145 and His-41) and the favorable predicted ADME/tox parameters; 10-Hydroxyusambarensine, Cryptoquindoline, 6-Oxoisoiguesterin and 22-Hydroxyhopan-3-one may be potent against SARS-CoV-2 3CLpro. Further experimental studies are suggested to check the possible preclinical and clinical efficacy of these agents for the prevention and treatment of COVID-19.