Mushroom-derived bioactive compounds potentially serve as the inhibitors of SARS-CoV-2 main protease: An in silico approach.

BACKGROUND AND AIM: Coronavirus Disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has now become the world pandemic. There is a race to develop suitable drugs and vaccines for the disease. The anti-HIV protease drugs are currently repurposed for the potential treatment of COVID-19. The drugs were primarily screened against the SARS-CoV-2 main protease. With an urgent need for safe and effective drugs to treat the virus, we have explored natural products isolated from edible and medicinal mushrooms that have been reported to possess anti-HIV protease. EXPERIMENTAL PROCEDURES: We have examined 36 compounds for their potential to be SARS-CoV-2 main protease inhibitors using molecular docking study. Moreover, drug-likeness properties including absorption, distribution, metabolism, excretion and toxicity were evaluated by in silico ADMET analysis.
RESULTS: Our AutoDock study showed that 25 of 36 candidate compounds have the potential to inhibit the main viral protease based on their binding affinity against the enzyme's active site when compared to the standard drugs. Interestingly, ADMET analysis and toxicity prediction revealed that 6 out of 25 compounds are the best drug-like property candidates, including colossolactone VIII, colossolactone E, colossolactone G, ergosterol, heliantriol F and velutin.
CONCLUSION: Our study highlights the potential of existing mushroom-derived natural compounds for further investigation and possibly can be used to fight against SARS-CoV-2 infection. TAXONOMY CLASSIFICATION BY EVISE: Disease, Infectious Disease, Respiratory System Disease, Covid-19, Traditional Medicine, Traditional Herbal Medicine, Phamaceutical Analysis.

Severe acute respiratory syndrome coronavirus-2

Coronavirus Disease 2019

Human immunodeficiency virus

Absorption Distribution Metabolism and Excretion

Protein Data Bank

Introduction

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), also known as 2019 novel coronavirus (2019-nCoV), has quickly spread worldwide since late 2019. Coronavirus Disease 2019 (COVID-19) has emerged as a severe global pandemic. As per the recent situational report released by the World Health Organization (WHO) on 27 December 2020, the disease has caused 1.7 million deaths and 78 million infections worldwide. The disease’s clinical characteristics can range from asymptomatic infections, mild respiratory disease, severe pneumonia with respiratory failure, and even fatal respiratory diseases such as acute respiratory distress syndrome.,

SARS-CoV-2 belongs to the family of the coronavirus. It is an enveloped virus with a positive-sense single-stranded RNA and a single linear RNA segment. SARS-CoV-2 has been linked to two other strains, severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV), in the zoonotic origins and the cause of fatal pneumonia., The main viral protease has been proposed as a key therapeutic target for drug development against coronavirus and COVID-19 treatment.6, 7, 8, 9, 10, 11 The enzyme plays an essential role in the viral life cycle, mostly involving maturation cleavage events within several precursor proteins., The active sites of this enzyme are highly conserved among all coronaviruses. Furthermore, the main viral protease is a non-human homologue, making it an ideal antiviral target.

Unfortunately, there is no anti-SARS-CoV-2 drug or vaccine sufficiently available to date. The urgent need for drugs to treat COVID-19 has led scientists to focus on protease inhibitors as potential drugs to cure the disease. Repurposing drugs currently used to treat HIV, HCV and SARS, taking advantage of drug safety, is one of the current anti-SARS-CoV-2 drug searching approaches. Repurposing HIV-drugs have been tested against SARS-CoV-2 main protease and shown effectiveness against the viral enzyme., Moreover, the combination of anti-HIV protease drugs, lopinavir and ritonavir, was currently employed in a clinical trial against COVID-19 in patients with mild and moderate COVID-19.

Lopinavir and Ritonavir, the approved drugs for HIV protease inhibitor, have been reported to inhibit the main protease of both SARS-CoV and MERS-CoV.17, 18, 19, 20, 21 With the urgent need to reduce the impact of COVID-19, a combination of these drugs was currently investigated to treat SARS-CoV-2-infected patients.,22, 23, 24, 25 However, these drugs’ curative effect remains minimal, with potentially toxic side effects that might be harmful to the patients. Identifying bioactive compounds from natural sources that can inhibit SARS-CoV-2 main protease is considered an alternative approach to combat COVID-19. 26, 27, 28, 29 In silico technique, a computational approach, is the promising preliminary evidence for drug discovery.30, 31, 32, 33, 34 Several molecular docking studies have identified bioactive compounds from natural products as the potential SARS-CoV-2 main protease inhibitors derived from natural sources.35, 36, 37, 38, 39, 40

Mushrooms are a rich source of bioactive compounds with antiviral activity., Some of the compounds showed anti-inflammatory activity.43, 44, 45 These dual activities would be an effective treatment of COVID-19. A number of bioactive compounds have been shown to inhibit HIV protease, suggesting their potential activity against the main proteases of coronaviruses. In this work, we carried out an in silico investigation of bioactive compounds found in edible and medicinal mushrooms. These compounds have been shown to display anti-HIV protease activity. We examined 36 compounds for their potential candidates as SARS-CoV-2 main protease inhibitors and investigated their drug-like properties using molecular docking study.

Materials and methods

Ligand preparation

A total of 36 mushroom-isolated compounds were selected from the literature, including those listed by Suwannarach et al. (Table 1). Compounds No. 1–32 were reported for their anti-HIV protease property, whereas compound No. 33–36 showed anti-HIV activity via HIV reverse transcriptase inhibition. All chemical structures were obtained from the PubChem database. Prior to docking studies, each compound was loaded into Discovery Studio (DS) Visualizer (BIOVIA, CA, USA) and optimized using "Clean Geometry" function in DS. All compounds’ files were converted to PDBQT format using AutoDockTools-1.5.6 software (The Scripps Research Institute, USA) for docking studies.

Table 1

List of anti- HIV bioactive compounds derived from mushrooms.

No.	Compound	Structure	Source	Reference
1	20(21)-dehydrolucidenic acid N	Image 2	Ganoderma sinense	46
2	20-hydroxylucidenic acid N	Image 3	Ganoderma sinense	46
3	Agrocybin	Image 4	Agrocybe cylindracea	47
4	Betulinic acid	Image 5	Trametes versicolor	48,49
5	Colossolactone A	Image 6	Ganoderma colosum	50
6	Colossolactone E	Image 7	Ganoderma colosum	50
7	Colossolactone G	Image 8	Ganoderma colosum	50
8	Colossolactone V	Image 9	Ganoderma colosum	50
9	Colossolactone VII	Image 10	Ganoderma colosum	50
10	Colossolactone VIII	Image 11	Ganoderma colosum	50
11	Ellagic acid	Image 12	Flammulina velutipes,Phellinus linteus,Pleurotus eryngii	51, 52, 53, 54
12	Ergosterol	Image 13	Auricularia polytricha,Flammulina velutipes,Lentinula edodes	55, 56, 57
13	Gallic acid	Image 14	Agaricus bisporus,Flammulina velutipes,Ganoderma lucidum,Laetiporus sulphureus,Lentinus lepideus,Leucoagaricus leucothites,Macrocybe gigantea,Pleurotus ostreatus	51,58, 59, 60, 61, 62
14	Ganoderic acid alpha	Image 15	Ganoderma lucidum	63
15	Ganoderic acid B	Image 16	Ganoderma lucidum	63,64
16	Ganoderic acid beta	Image 17	Ganoderma lucidum	65
17	Ganoderic acid C1	Image 18	Ganoderma lucidum	63
18	Ganoderic acid GS-2	Image 19	Ganoderma sinense	46
19	Ganoderic acid H	Image 20	Ganoderma lucidum	63
20	Ganoderiol A	Image 21	Ganoderma lucidum	63
21	Ganoderiol B	Image 22	Ganoderma lucidum	63
22	Ganoderiol F	Image 23	Ganoderma lucidum, Ganoderma sinense	46,63
23	Ganodermanondiol	Image 24	Ganoderma lucidum	65
24	Ganodermanontriol	Image 25	Ganoderma lucidum	63,65
25	Ganolucidic acid A	Image 26	Ganoderma lucidum	65
26	Ganomycin B	Image 27	Ganoderma colosum	50,66
27	Ganomycin I	Image 28	Ganoderma colosum	50,66
28	Heliantriol F	Image 29	Lignosus rhinocerus	67
29	Linoleic acid	Image 30	Auricularia polytricha,Lentinula edodes	55,57
30	Lucidumol B	Image 31	Ganoderma lucidum	65
31	Methyl gallate	Image 32	Pholiota adiposa	68
32	Oleanolic acid	Image 33	Hericium erinaceus,Lactarius deliciosus,Lactarius sanguifluus,Lactarius semisanguifluusRussula delica, Suillus bellinii	69, 70, 71
33	Semicochliodinol A	Image 34	Chrysosporium merdarium	72
34	Semicochliodinol B	Image 35	Chrysosporium merdarium	72
35	Ursolic acid	Image 36	Cynomorium songaricum,Hericium erinaceus,Russula delica, Suillus bellinii	70,71,73
36	Velutin	Image 37	Flammulina velutipes	74

Protein preparation

The crystal structure of COVID-19 main protease in complex with an inhibitor N3 (PDB ID: 6LU7) was obtained from RCSB Protein Data Bank. Before the docking studies, the protein structure was first prepared using the Prepare protein set up in AutoDock. The protein preparation is an optimization step that corrects structure, atomic and bond length and charges anomalies. Water molecules and the original inhibitor were removed from the protein structure. Then any missing hydrogen atoms were added. The optimization step was then employed to provide the stable conformation before converting to PDBQT format for docking analysis.

Molecular docking

Molecular docking studies were performed using the default protocol in Autodock tools 1.5.6 (AutoDock 4.2 software, The Scripps Research Institute, USA). The active site of SARS-CoV-2 main protease 6LU7 was set as a grid box. The grid box was set to 50 x 50 x 50 points in xyz-dimension that equated to a grid box spacing of 0.375 Å3, and the coordinate of the x, y and z centers of the box are at −9.1, 11.0 and 68.0, respectively. The docking simulations were performed using the Lamarckian Genetic Algorithm with default parameters, including 10 Genetic Algorithm runs. The docking results of 10 runs were ranked by energy. The compound with the best energy ranking was further closely analyzed for its protein-ligand interactions using the DS Visualizer (BIOVIA, San Diego, CA, USA).

Drug-likeness prediction

The structures of compounds presented in mushrooms (Table 1) were obtained from PubChem database in SMILE files. The compounds were then evaluated for drug-likeness using the Lipinski’s rule of five to predict their pharmacokinetic properties such as the Absorption, Distribution, Metabolism and Excretion (ADME) of the compounds using SwissADME (http://www.swissadme.ch/).

Toxicity, carcinogenicity and mutagenicity prediction

The canonical SMILES of the list of compounds in Table 1 was further submitted to the pkCSM database and Toxtree software to predict their toxicity and carcinogenicity, respectively. The toxicity was predicted via toxicity mode in pkCSM database, while carcinogenicity and mutagenicity prediction was analyzed using Toxtree 3.1.0 software based on the Benigni-Bossa rule.,

Results and discussion

Investigation of mushroom-derived bioactive compounds against SARS-CoV-2 main protease by molecular docking

The crystal structure of COVID-19 main protease in complex with the inhibitor N3 (PDB ID: 6LU7 obtained on 21 June 2020) was obtained from RCSB Protein Data Bank. The protein structure was prepared for the docking studies by removing water molecules and the inhibitor before optimization step to provide the stable conformation. The docking protocol was validated using the standard set up in AutoDock. The minimized structure of the inhibitor N3 was re-docked into the original binding site of the protein. The results of 10 Genetic Algorithm runs of the re-docking were ranked and shown in Table 2. The best docking score of N3 was re-docked structure 1 which showed binding energy at −7.97 kcal/mol. The conformation, orientation, position and interactions of the redocked N3 at the binding site were analyzed and compared to that found in the original X-ray structure (PDB ID: 6LU7). Superimposing of the redocked protein-N3 over the original X-ray structure indicated that N3 was able to seek out the location of the receptor site. It is noteworthy that the re-docked N3 has adopted a slightly different conformation, specifically at the 5-methylisoxazole-3-carboxamide part of the molecule compared to that found in the X-ray structure, as shown in Fig. 1B. The variation in conformation, perhaps due to the stable conformation of N3 adopted in silico.The best re-docked N3 was found to interact with only two active site amino acid residues (GLY143 and GLU166) out of nine different residues observed in the X-ray structures. The common key interactions are shown in Table 2. This indicates that N3 was able to recognize its original receptor site in the X-ray structure of 6LU7, using this docking protocol. The results above validated the docking protocol as a suitable method for this study.

Table 2

Hydrogen bond interaction between N3 and 6LU7.

6LU7-N3 interaction	Binding energy (kcal/mol)	Hydrogen bonding		3D diagrams
		Number	Amino acid interaction
X-ray structure	–	10	GLY143, THR190, GLN189, PHE140, HIS163, HIS164, MET165, HIS172, GLU166 (2)
Re-docked structure 1	−7.97	3	GLY143, LEU141, GLU166	Image 38
Re-docked structure 2	−6.28	2	GLN189, THR190	Image 39
Re-docked structure 3	−5.89	5	ASN142, HIS163, GLU166, GLY143, CYS145	Image 40
Re-docked structure 4	−5.86	2	CYS145, THR190	Image 41
Re-docked structure 5	−5.85	2	GLU166, GLN189	Image 42
Re-docked structure 6	−5.65	5	HIS41, CYS145, GLU166, GLN189,MET165	Image 43
Re-docked structure 7	−5.03	3	GLN189, PRO168, GLY143	Image 44
Re-docked structure 8	−4.36	6	ASN142, CYS145, HIS163, SER46, MET165, HIS172	Image 45
Re-docked structure 9	−4.30	4	GLU166, PRO168, GLU166(2)	Image 46
Re-docked structure 10	−3.55	3	ASN142, GLU166(2)	Image 47

The 10 Genetic Algorithm runs on the re-docking of 6LU7-N3 using the validated method were shown. The results were ranked by binding energy. The lowest binding energy showed the best overlapping between N3 ligand from crystal (violet) and re-docking via AutoDock (yellow).

Fig. 1

The results from validation method for molecular docking study of SARS-CoV-2 main protease (6LU7). (A) The 3D diagrams of interaction between N3 and 6LU7 at the active site demonstrated overlapping between N3 ligand from crystal (magenta) and re-docking via AutoDock (yellow). (B) and (C) displayed the 2D ligands-receptor interactions of N3 from crystal and re-docking ligand via AutoDock structure with 6LU7, respectively.

The validated docking protocol was subsequently used for in silico screening of potential compounds as inhibitors against SARS-CoV-2 main protease. The in silico inhibition of SARS-CoV-2 main protease is gauged on the potential binding affinity of the compounds. The binding energy expressed the binding affinity as Gibbs free energy by which the compounds displayed the higher negative binding energy were considered betters. The in silico binding affinities of Lopinavir and Ritonavir to 6LU7 were determined. Since Lopinavir and Ritonavir are anti-HIV protease drugs that have been currently used in hospitals for COVID-19 treatment.22, 23, 24, 25 Therefore, it is appropriate to use them as standard drugs in addition to N3 for comparative study. Lopinavir and Ritonavir showed binding energy at −7.41 and −6.19 kcal/mol against 6LU7, respectively. Andrographolide and kaempferol have been previously demonstrated the potential inhibition against 6LU7 via docking studies., They were also docked in this study to confirm our validated method. Our docking results agreed with recent reports that both andrographolide and kaempferol had lower binding energy (−7.93 and −7.99 kcal/mol, respectively) than that of anti-HIV agents, indicating their potency to inhibit SARS-CoV-2 main protease.

The docking scores of all compounds are presented in Table 3. Out of 36 compounds, seventeen compounds (No.1–17) showed lower binding energy (<−7.97 kcal/mol) binding to 6LU7 than that of original inhibitor (N3), Lopinavir and Ritonavir. Notably, 25 out of 36 mushroom-derived compounds (No.1–25) showed lower binding energy against 6LU7 than that of lopinavir, while 30 out of 36 compounds (No.1–30) showed lower binding energy than that of Ritonavir. From these results, colossolactone VIII exhibited the lowest binding energy against SARS-CoV-2 protease (−10.55 kcal/mol) followed by colossolactone E, colossolactone G, ergosterol, semicochliodinol A and heliantriol F (−10.10, −9.86, −9.49, −8.91 and −8.91 kcal/mol, respectively). Consistent with previous reports,81, 82, 83 our results from molecular docking found that ursolic acid, oleanolic acid and betulinic acid also exhibited the potential to inhibit SARS-CoV-2 main protease.

Table 3

Docking results of all mushroom-derived compounds with the SARS-CoV-2 main protease.

No	Compound	Binding energy (kcal/mol)	LigandEfficiency	Inhibition constant (Kpi)
	N3 (original inhibitor)	−7.97	−0.16	1.44 μM
	Lopinavir	−7.41	−0.16	3.32 μM
	Ritonavir	−6.19	−0.12	28.94 μM
	Andrographolide	−7.93	−0.32	1.55 μM
	Kaempferol	−7.99	−0.38	1.38 μM
1	Colossolactone VIII	−10.55	−0.27	18.58 nM
2	Colossolactone E	−10.10	−0.27	39.58 nM
3	Colossolactone G	−9.86	−0.25	58.86 nM
4	Ergosterol	−9.49	−0.33	111.07 nM
5	Semicochliodinol A	−8.91	−0.27	294.8 nM
6	Heliantriol F	−8.91	−0.27	292.54 nM
7	Semicochliodinol B	−8.71	−0.26	412.95 nM
8	Ganoderic acid GS-2	−8.68	−0.24	433.02 nM
9	Lucidumol B	−8.64	−0.26	464.56 nM
10	Ganodermanondiol	−8.60	−0.26	500.55 nM
11	Ursolic acid	−8.50	−0.26	598.68 nM
12	Colossolactone A	−8.49	−0.23	594.72 nM
13	Oleanolic acid	−8.35	−0.25	756.26 nM
14	Ganodermanontriol	−8.30	−0.24	818.74 nM
15	Ganolucidic acid A	−8.27	−0.23	865.36 nM
16	Ganoderic acid beta	−8.17	−0.23	1.02 μM
17	Ganoderic acid B	−7.99	−0.22	1.39 μM
18	Velutin	−7.97	−0.35	1.45 μM
19	Ganoderiol F	−7.95	−0.24	1.48 μM
20	20(21)-dehydrolucidenic acid N	−7.88	−0.24	1.67 μM
21	Ganoderic acid C1	−7.88	−0.21	1.66 μM
22	Ganoderiol A	−7.87	−0.23	1.72 μM
23	Ganoderiol B	−7.68	−0.23	2.34 μM
24	Ganoderic acid H	−7.65	−0.19	2.48 μM
25	Betulinic acid	−7.62	−0.23	2.61 μM
26	Ganomycin I	−7.37	−0.29	3.96 μM
27	Colossolactone VII	−7.33	−0.18	4.24 μM
28	20-hydroxylucidenic acid N	−6.98	−0.21	7.71 μM
29	Ganoderic acid alpha	−6.79	−0.17	10.55 μM
30	Ellagic acid	−6.76	−0.31	11.04 μM
31	Ganomycin B	−6.02	−0.24	38.57 μM
32	Linoleic acid	−5.37	−0.27	116.76 μM
33	Colossolactone V	−4.86	−0.11	274.96 μM
34	Agrocybin	−4.81	−0.44	300.41 μM
35	Methyl gallate	−4.79	−0.37	308.49 μM
36	Gallic acid	−4.26	−0.36	757.15 μM

ADME analysis

The ADME (Adsorption, Distribution, Metabolism and Excretion) analysis was performed to evaluate the drug-likeness of 36 compounds found in mushrooms. The prediction was performed using SwissADME database. Drug-likeness was indicated by Lipinski’s rule of five. With respect to the criteria, molecular weight (MW) ≤ 500, the number of hydrogen bond acceptor ≤ 10, the number of hydrogen bond donors ≤ 5 and the Log Po/w ≤ 5 , no more than 1 violation is allowed. We found compounds listed 1–25 with lower binding energy than those of both drug standards as ranked in Table 4. The drug-like property’s prediction was then evaluated.

Table 4

The physiochemical properties of the top 25 potential compounds derived from mushrooms as SARS-CoV-2 main protease inhibitors based on Lipinski’s rule of five.

No	Compound	MW (g/mol)(≤ 500 g/mol)	Num. H-bond acceptors (≤ 10)	Num. H-bond donors (≤ 5)	Log Po/w (≤ 5)	Violation (≤ 1)
1	Colossolactone VIIIa	558.75	7	1	5.49	2
2	Colossolactone Ea	522.67	4	0	5.27	2
3	Colossolactone G	538.67	7	1	4.57	1
4	Ergosterol	396.65	1	1	6.49	1
5	Heliantriol F	458.72	3	3	5.28	1
6	Semicochliodinol A	438.47	4	4	4.15	0
7	Semicochliodinol B	438.47	4	4	4.13	0
8	Ganoderic acid GS-2	500.67	6	3	4.07	1
9	Lucidumol B	458.72	3	3	5.63	1
10	Ganodermanondiol	456.70	5	2	5.71	1
11	Ursolic acid	456.70	3	2	5.94	1
12	Colossolactone Aa	516.75	5	3	5.46	2
13	Oleanolic acid	456.70	3	2	6.06	1
14	Ganodermanontriol	472.70	4	3	4.90	0
15	Ganolucidic acid A	500.67	6	2	4.10	1
16	Ganoderic acid beta	500.67	6	3	4.02	1
17	Ganoderic acid B	516.67	7	3	3.34	1
18	Velutin	314.29	6	2	2.60	0
19	Ganoderiol F	454.68	3	2	5.63	1
20	Ganoderic acid C1	514.65	7	2	3.27	1
21	20(21)-dehydrolucidenic acid N	476.60	7	4	2.41	0
22	Ganoderiol A	474.72	4	4	4.89	0
23	Ganoderiol B	470.68	4	3	4.90	0
24	Ganoderic acid H	572.69	9	2	3.20	1
25	Betulinic acid	456.70	3	2	6.11	1

Note.

Non-drug-likeness compounds based on Lipinski’s rule of five.

ADME is an essential tool for analyzing the proposed molecule’s oral bioavailability as possible drugs. As shown in Table 4, the prediction results showed that compounds list 1–22 passed the Lipinski’s rule of five. The best five compounds based on docking score and Lipinski’s rule of five for SARS-CoV-2 main protease inhibition are colossolactone G, ergosterol, heliantriol F, semicochliodinol A and semicochliodinol B. Although three compounds, i.e. colossolactone VIII, colossolactone E and colossolactone A, violated more than one rule, there is an exception to Lipinski’s rule for natural products. Therefore, these three are acceptable as druglikness compounds.

The toxicity of the potential compounds was predicted via toxicity mode in pkCSM database. The toxicity prediction was described in Supplementary data 1. The top 25 compounds’ carcinogenicity and mutagenicity were determined using Toxtree 3.1.0 software based on the Benigni-Bossa rule. The rule determines the compound’s carcinogenic and mutagenic potency based on the compound’s functional groups. The functional groups confer as either carcinogenic or mutagenic are acyl halide, haloalkene, epoxide, aliphatic halogen, alkyl nitrate, aldehyde, hydrazine, isocyanate, polyaromatic hydrocarbon, azide, alkyl/aromatic nitro, coumarin, diazo aromatic, benzyl sulfinyl ether, alkyl halide and thiocarbonyl.,, Predicted genotoxic carcinogenicity, nongenotoxic carcinogenicity and mutagenic potential of the compounds were shown in Supplementary data 2. Overall, six compounds are shown in Fig. 2, including colossolactone VIII, colossolactone E, colossolactone G, ergosterol, heliantriol F and velutin were predicted as non-toxic, non-carcinogenic and non-mutagenic compounds derived from mushrooms.

Fig. 2

Venn diagram describes the number of compounds predicted no toxicity, no carcinogenicity and no mutagenicity. There are six compounds predicted non-toxicity to skin and liver as well as non-mutagenicity and non-carcinogenicity.

Hydrogen bond analysis

The complex interactions between the SARS-CoV-2 protease and compounds listed 1–25 were visualized, and hydrogen bond analysis was conducted using Discovery Studio Visualizer. The number of hydrogen bonds and the residues involved in hydrogen bond interactions are summarized in Table 6. Colossolactone G, ergosterol, heliantriol F and velutin are the potential candidate compounds for SARS-CoV-2 inhibitors with drug-likeness property, no toxicity, no carcinogenicity and no mutagenicity. Colossolactone G has four hydrogen bonds interacting with the active site of SARS-CoV-2 main protease at the residue HIS41, ASN142, GLN189 and HIS172 (Fig. 3A). Ergosterol shows a hydrogen bond binding to viral protease at the residue ASP187 (Fig. 3B). Hydrogen bond interaction of heliantriol F against the main protease are two interactions at the residue THR190 and GLN189, while velutin exhibited two interactions at the residue THR190 and MET49 (Fig. 3D-E). It is noteworthy that colossolactone G, heliantriol F and velutin exhibited the common interactions when compared to X-ray structure of 6LU7-N3. The common interaction of colossolactone G and velutin is GLN189 residue, while heliantriol F is THR190 residue. Even though colossolactone VIII and colossolactone E were predicted as non-druglikness compounds; it was acceptable to include them as the potential anti-SARS-CoV-2 main protease candidates as we previously mentioned. Colossolactone VIII has two hydrogen bonds interacting with the viral protease’s active site at the residue HIS41 and GLN189, while colossolactone E has a hydrogen bond interaction at the residue HIS41 (Fig. 3E-F). Semicochliodinol A was predicted as a hepatotoxicity agent; it has shown the significant six interactions at the receptor site. Those involved residues are SER144, MET165, GLY143, CYS145, GLU166 and GLU166 (Fig. 3G), in which MET165, GLY143 and GLU166 are the common interactions when compared to X-ray structure of 6LU7-N3. Compared to the original inhibitor, the studied compounds’ common interactions could demonstrate mushroom-derived compounds as the potential inhibitors against SARS-CoV-2 main protease.

Table 6

H-bond interactions of top 25 potential compounds derived from mushrooms against SARS-CoV-2 main protease inhibitor.

No	Compound	Hydrogen bonding
		Number	Amino acid interaction: bond length
	X-ray structure (N3)	10	GLY143: 2.86529, THR190:2.84976 , GLN189: 2.88951, PHE140: 3.18679, HIS163: 2.36734, HIS164: 2.80304, MET165: 3.69210, HIS172: 3.72440, GLU166: 2.97927, GLU166: 2.83433
1	Colossolactone VIII	2	HIS41: 2.72763, GLN189: 1.96169
2	Colossolactone E	1	HIS41: 3.02314
3	Colossolactone G	4	HIS41: 2.58995, ASN142: 1.88729, GLN189: 1.65515, HIS172: 2.84417
4	Ergosterol	1	ASP187: 2.18813
5	Heliantriol F	2	THR190: 1.85702, GLN189: 2.18737
6	Semicochliodinol A	6	SER144: 1.82069, MET165: 3.03681, GLY143: 2.7737, CYS145: 3.69761, GLU166: 2.01863, GLU166: 3.36082
7	Semicochliodinol B	3	ARG188: 1.86701, CYS145: 3.18962, CYS145: 3.89699
8	Ganoderic acid GS-2	4	HIS41: 2.83185, GLU166: 2.24025, GLN189: 2.41463, ASN142: 2.11084
9	Lucidumol B	3	GLN189: 1.96259, GLN189: 2.14332, GLU166: 2.83998
10	Ganodermanondiol	8	GLY143: 2.49923, CYS145: 2.76771, HIS163: 2.08786, SER144: 1.90278, LEU141: 1.64009, SER144: 2.43463, LEU167: 3.51115, PRO168: 2.91688
11	Ursolic acid	3	SER144: 2.87305, CYS145: 2.02309, CYS145: 2.97091
12	Colossolactone A	3	HIS163: 1.91024, CYS145: 2.52781, THR190: 1.85473
13	Oleanolic acid	3	SER144: 2.97806, CYS145: 1.97439, CYS145: 2.65435
14	Ganodermanontriol	8	GLY143: 2.6619, HIS163: 2.25078, LEU141: 2.11468, SER144: 2.86257, LEU141: 1.79129, ASN142: 2.96571, LEU167: 3.47167, PRO168: 2.72296
15	Ganolucidic acid A	3	GLU166: 2.24559, GLN189: 2.44479, ASN142: 1.94471
16	Ganoderic acid beta	4	TYR54: 2.24534, GLU166: 2.24309, MET49: 2.05161, ARG188: 2.99000
17	Ganoderic acid B	6	HIS41: 2.52071, GLY143: 2.65843, CYS145: 3.63501, GLU166: 2.42919, MET49: 2.0714, ASN142: 1.77428
18	Velutin	2	THR190: 2.15553, MET49: 3.56078
19	Ganoderiol F	1	ASP187: 1.81627
20	Ganoderic acid C1	4	TYR54: 2.29094, GLN189: 2.96608, ASN142: 2.04706, ARG188: 3.33477
21	20(21)-dehydrolucidenic acid N	5	GLY143: 2.52618, GLU166: 2.2808, CYS145: 2.70312, MET49: 2.20979, ASN142: 1.81534
22	Ganoderiol A	4	MET49: 2.2339, GLU166: 1.94189, GLU166: 1.89131, GLU166: 3.10886
23	Ganoderiol B	6	CYS145: 2.73888, CYS145: 3.60588, MET165: 2.75963, LEU141: 2.23062, LEU141: 2.48645, SER144: 2.06885
24	Ganoderic acid H	5	ASN142: 1.83347, ASN142: 2.34095, GLU166: 2.10428, GLU166: 2.68895, THR190: 1.78362
25	Betulinic acid	2	PRO168: 2.29376, GLN189: 1.67449

Fig. 3

Hydrogen bond interactions between 6LU7 and (A) colossolactone G, (B) ergosterol, (C) heliantriol F, (D) velutin, (E) colossolactone VIII, (F) colossolactone E and (G) semicochliodinol A.

Our project plan’s next step is to confirm the anti- SARS-CoV-2 of the identified compounds using SARS-CoV-2 assay kit (BPS Bioscience, CA, USA). The compounds that show inhibitory activity will be screened for their toxicity against normal human cell models such as human peripheral blood mononuclear cells (PBMCs).

Conclusions

The current therapeutic drugs against COVID-19 is a group of anti-HIV protease drugs. However, the compound with more efficiency and less toxicity is needed. Moreover, the most severe cases of COVID-19 are caused by infection-induced hyper-inflammation of the lungs and followed by acute respiratory distress, resulting in death. Therefore, effective treatment of COVID-19 would likely be a combination of therapies, using both antiviral and anti-inflammatory drugs. Several bioactive compounds derived from mushrooms have been promisingly demonstrated to exhibit anti-HIV protease and anti-inflammatory activities. Those compounds may have the potential to inhibit SARS-CoV-2 main protease and become useful for the treatment of COVID-19 patients. In this study, mushroom-derived compounds were docked against the main viral protease. In addition, the in silico ADME analysis and toxicity of these compounds were conducted to predict their drug-like properties, toxicity, carcinogenicity and mutagenicity. Our study revealed six compounds found in mushrooms, namely colossolactone VIII, colossolactone E colossolactone G, ergosterol, heliantriol F and velutin the best potential candidates for anti-SAR-CoV-2 agents. Those potential compounds could be used and developed as an alternative or complementary medicine for COVID-19 treatment. Moreover, ergosterol has been reported as an anti-inflammatory agent., These could support the idea of ergosterol being a useful compound for COVID-19 treatment due to its dual activities. Strategies are in place to confirm theirs in vitro inhibitory activity against SARS-CoV-2 protease and toxicity in healthy mammalian cells such as human PBMCs. The discovery of such inhibitors with low or no toxicity would provide us with further opportunity to develop them into anti-COVID-19 in monotherapy or combination drugs.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.