Sage Journals: Discover world-class research

Abstract

Background

The COVID-19 pandemic has led to significant loss of life and economic disruption worldwide. Currently, there are limited effective treatments available for this disease. SARS-CoV-2 RNA-dependent RNA polymerase (SARS-CoV-2 RdRp) has been identified as a potential target for drug development against COVID-19. Natural products have been shown to possess antiviral properties, making them a promising source for developing drugs against SARS-CoV-2.

Objectives

The objective of this study is to identify the most effective natural inhibitors of SARS-CoV-2 RdRp among a set of 4924 African natural products using a multi-phase in silico approach.

Methods

The study utilized remdesivir (RTP), the co-crystallized ligand of RdRp, as a starting point to select compounds that have the most similar chemical structures among the examined set of compounds. Molecular fingerprints and structure similarity studies were carried out in the first part of the study. The second part of the study included molecular docking against SARS-CoV-2 RdRp (PDB ID: 7BV2) and Molecular Dynamics (MD) simulations including the calculation of RMSD, RMSF, Rg, SASA, hydrogen bonding, and PLIP. Moreover, the calculations of Molecular mechanics with generalised Born and surface area solvation (MM-GBSA) Lennard-Jones and Columbic electrostatic interaction energies have been conducted. Additionally, in silico ADMET and toxicity studies were performed to examine the drug likeness degrees of the selected compounds.

Results

Eight compounds were identified as the most effective natural inhibitors of SARS-CoV-2 RdRp. These compounds are kaempferol 3-galactoside, kaempferol 3-O-β-D-glucopyranoside, mangiferin methyl ether, luteolin 7-O-β-D-glucopyranoside, quercetin-O-β-D-3-glucopyranoside, 1-methoxy-3-indolylmethyl glucosinolate, naringenin, and asphodelin A 4’-O-β-D-glucopyranoside.

Conclusion

The results of this study provide valuable information for the development of natural product-based drugs against COVID-19. However, the elected compounds should be further studied in vitro and in vivo to confirm their efficacy in treating COVID-19.

Keywords

Natural products SARS-CoV-2 RdRp docking molecular dynamics simulations Lennard-Jones interaction energy Columbic electrostatic interaction energy MM-GBSA protein–ligand interaction profiler

Introduction

As of May 9th, 2023, the World Health Organization (WHO) reported a total of 765,903,278 confirmed COVID-19 infections worldwide, along with 6,927,378 recorded deaths due to the virus. Even though the number of administrated vaccine doses reached 13,350,530,518, the fatal virus still shows an ability to affect vastly.¹ These alarming statistics underscore the significant effort required from scientists worldwide to discover a cure.

Over the past two decades, Computer-Aided Drug Discovery (CADD) approaches have gained significant traction in pharmacology development and testing.^2,3 This widespread utilization can be attributed to several factors. Firstly, the elucidation of precise 3D structures of various protein targets within the human body.⁴ Secondly, the remarkable advancements in computer software and hardware capabilities.⁵ Lastly, the improved understanding of Structure–Activity Relationship (SAR) and Quantitative Structure–Activity Relationship (QSAR) principles.⁶ Consequently, computer-aided chemistry approaches have been employed to investigate various pharmacokinetic and pharmacodynamics properties, enabling the identification of connections between chemical structure and activity. These methods include structure similarity analysis,⁷ molecular fingerprints,⁸ QSAR modeling,⁹ pharmacophore analysis,¹⁰ homology modeling,¹¹ molecular docking,¹² molecular dynamics (MD) simulations,¹³ ADMET property predictions,¹⁴ and density functional theory (DFT) calculations.¹⁵

Throughout the annals of history, the nature has held an indispensable role in meeting fundamental human needs. Nature has not only provided solutions for medicinal treatments but has also offered sustenance and contributed to the formulation of various cosmetic products.^16,17 Throughout the past 30 years, natural products have consistently contributed to nearly one-third of newly authorized pharmaceuticals by the FDA. For instance, the cytosine arabinoside, Cytarabine, which was used to treat acute myeloid leukaemia.¹⁸ Eribulin, Halaven^®, the anti-breast cancer,¹⁹ the analgesic Ziconotide, Prialt^®,²⁰ and the ergot alkaloid, dihydroergocristine, for Alzheimer's disease.²¹ The continuity of this trend can be traced back to the 1940s, showcasing the enduring importance and worth of natural products within the realm of pharmaceutical research, development, and validation.

In targeting COVID-19, our team utilized various in silico methods to propose natural inhibitors. Four natural isoflavonoids were proposed as the most active inhibitors of hACE2 and viral main protease between fifty-nine isoflavonoids.²² Likewise, fifteen alkaloidal compounds were studied in silico against five crucial proteins of COVID-19.²³ In a like manner, we designed multi-stage CADD approaches to recommend the most fitting inhibitor against certain COVID-19 enzymes. Among a set of 310 antiviral natural metabolites, potential inhibitors for nsp10,²⁴ the main protease,^25,26 and the papain-like protease²⁷ have been determined. Furthermore, we investigated 3009 FDA-approved drugs and identified the most potent inhibitors against the SARS-CoV-2 nsp16-nsp10 2′-o-Methyltransferase Complex²⁸ and the RNA-Dependent RNA Polymerase.²⁹ Additionally, we explored a collection of 5956 compounds from traditional Chinese medicine to identify potential natural inhibitors for the SARS-CoV-2 Helicase.³⁰ Lastly, we discovered the most active semisynthetic inhibitor for the COVID-19 papain-like protease from a group of 69 molecules.³¹

For this study, a curated set of 4924 African natural products was meticulously selected.³² These compounds were isolated from diverse natural sources in Africa and obtained from the comprehensive African Natural Products Database (ANPDB). The dataset encompassed the period of 1962 to 2019, incorporating a wide range of scholarly sources such as international and local African journals, as well as MSc and Ph.D. theses archived in African university libraries. The selected compounds have been screened using different computational methods to detect the most potent SARS-CoV-2 RdRp inhibitors. Remdesivir, RTP, has been chosen as a lead compound due to its role as a co-crystallized ligand with the SARS-CoV-2 RdRp enzyme. Additionally, its robust potential to inhibit the SARS-CoV-2 RdRp has been demonstrated through both in silico computational predictions and in vitro experimental studies.³³ The utilized methods included molecular structure similarity and fingerprint studies against RTP, the molecular docking against SARS-CoV-2 RdRp, and several molecular dynamic (MD) simulation experiments.

Results and discussion

Filter using fingerprints

A co-crystallized ligand is a compound that exhibits a notable affinity for a specific protein, resulting in the formation of a crystallizable complex between the ligand and protein.³⁴ This complex offers valuable information about the interactions between the ligand and protein, unveiling crucial structural and chemical characteristics responsible for the strong binding affinity. The chemical structure of the co-crystallized ligand serves as a valuable blueprint for the design and development of inhibitors that can efficiently bind to the target protein, aiding in the discovery of potential therapeutic agents.³⁵ By analyzing the structural characteristics and functional groups of the co-crystallized ligand, we can develop a more profound comprehension of the critical elements responsible for its robust binding affinity. With this knowledge, we employed a strategic approach to select compounds exhibiting similar structural features to RTP. The objective was to identify potent inhibitors that display a high affinity for the RdRp protein. This selection process is rooted in the fundamental principle of Structure–Activity Relationship (SAR), which asserts that compounds with comparable chemical structures are likely to exhibit similar biological activities.³⁶

Various computational methods that describe the similarities between different molecules have gained more interest in drug discovery.³⁷ One of the most helpful techniques in this approach is fingerprints.³⁸ The fingerprints study is binary strings that computes the existence or absence of vital sub-structural fragments to calculate the structural similarity between molecules. This technique is currently utilized in virtual screening and detection of similarities between hit compounds and a lead one. Fingerprints technique was applied using Discovery Studio and the following parameters were examined: H-bond acceptor and donor,³⁹ charge,⁴⁰ hybridization,⁴¹ positive and negative ionizable,⁴² halogen, aromatic, or none of them besides the ALogP category of atoms. The fingerprints technique represents those structural features (in the examined compounds and RTP) in a format that can be easily compared and analyzed. These fingerprints are often represented as binary strings (bits), where each bit represents the presence or absence of a specific structural feature. In the study, the software was adjusted to select the most similar 200 natural African compounds (Table 1) to RTP in their chemical structure.

Table 1.

Most similar 200 candidates to RTP according to fingerprints similarity.

Comp.	Similarity	SA	SB	SC	Comp.	Similarity	SA	SB	SC
RTP	1	206	0	0	2143	1	223	209	−17
9	0.588	164	73	42	2144	1	239	253	−33
11	1	184	128	22	2205	0.507389	206	200	0
13	0.601	167	72	39	2314	0.503968	127	46	79
14	1	245	235	−39	2341	0.509434	216	218	−10
15	0.516291	206	193	0	2355	0.601	155	52	51
23	0.605	164	65	42	2372	0.604	166	69	40
27	0.524229	238	248	−32	2379	0.591	153	53	53
28	1	162	73	44	2385	0.534545	147	69	59
30	0.509009	226	238	−20	2398	0.507042	216	220	−10
39	0.498866	220	235	−14	2423	0.562069	163	84	43
40	1	231	229	−25	2424	1	168	83	38
41	1	239	247	−33	2426	0.530909	146	69	60
49	0.509804	130	49	76	2430	0.502463	204	200	2
88	0.501272	197	187	9	2451	0.530909	146	69	60
150	0.506083	208	205	−2	2452	1	168	83	38
154	0.509434	216	218	−10	2465	1	137	52	69
156	0.506083	208	205	−2	2468	0.500921	272	337	−66
196	0.525	231	234	−25	2475	0.497596	207	210	−1
245	0.541096	158	86	48	2531	0.522901	137	56	69
263	0.595	147	41	59	2541	1	153	57	53
264	0.595	147	41	59	2616	0.506912	220	228	−14
275	0.506083	208	205	−2	2620	1	130	16	76
290	0.506427	197	183	9	2628	0.595	147	41	59
291	0.506427	197	183	9	2645	1	248	255	−42
295	0.498708	193	181	13	2646	0.535088	244	250	−38
299	0.503546	142	76	64	2647	0.545455	240	234	−34
319	0.515571	149	83	57	2654	0.589	146	42	60
320	0.503356	150	92	56	2670	1	214	161	−8
342	0.506083	208	205	−2	2679	0.507042	216	220	−10
344	0.508235	216	219	−10	2692	1	144	63	62
345	0.509434	216	218	−10	2709	0.547826	189	139	17
351	0.505025	201	192	5	2710	0.547826	189	139	17
367	0.507795	228	243	−22	3099	0.5	206	206	0
535	0.505025	201	192	5	3149	0.539267	206	176	0
553	0.503759	201	193	5	3150	0.542005	200	163	6
591	0.515892	211	203	−5	3164	0.516291	206	193	0
592	1	211	201	−5	3266	0.541096	158	86	48
593	0.5025	201	194	5	3267	1	168	96	38
606	0.5025	201	194	5	3299	0.508235	216	219	−10
609	0.50495	204	198	2	3300	0.505593	226	241	−20
624	0.511962	214	212	−8	3301	0.504717	214	218	−8
625	0.511962	214	212	−8	3372	0.575385	187	119	19
647	1	149	50	57	3382	1	203	176	3
648	0.611	168	69	38	3404	0.50495	204	198	2
649	0.604	151	44	55	3406	0.513189	214	211	−8
650	0.574519	239	210	−33	3408	0.511962	214	212	−8
671	0.5	201	196	5	3409	0.513189	214	211	−8
682	0.50495	204	198	2	3446	0.511002	209	203	−3
687	0.510896	211	207	−5	3447	0.511002	209	203	−3
690	0.506143	206	201	0	3490	0.557491	160	81	46
712	0.515	206	194	0	3491	0.557491	160	81	46
714	0.515012	223	227	−17	3642	0.512315	208	200	−2
743	0.501272	197	187	9	3643	0.5	213	220	−7
952	0.498807	209	213	−3	3644	0.5	213	220	−7
976	0.505774	219	227	−13	3645	0.6	159	59	47
980	1	136	40	70	3650	0.5	226	246	−20
997	0.515723	164	112	42	3655	0.5	213	220	−7
999	0.611	157	51	49	3656	0.515513	216	213	−10
1040	0.50495	204	198	2	3674	0.59	164	72	42
1042	0.506361	199	187	7	3739	0.524917	158	95	48
1043	0.508642	206	199	0	3740	0.541096	158	86	48
1120	1	208	193	−2	3741	0.526846	157	92	49
1121	0.503704	136	64	70	3743	1	168	96	38
1122	0.587	155	58	51	3751	1	206	190	0
1123	0.599	211	146	−5	3752	0.506818	223	234	−17
1126	0.509921	257	298	−51	3780	0.5	213	220	−7
1127	0.599	170	78	36	3819	0.576389	166	82	40
1128	1	242	246	−36	3848	0.5	213	220	−7
1129	0.605	173	80	33	3849	0.509524	214	214	−8
1130	0.588	163	71	43	3958	0.512748	181	147	25
1131	0.59	157	60	49	3969	0.615	176	80	30
1266	0.511848	216	216	−10	3988	0.504831	209	208	−3
1358	0.507042	216	220	−10	3989	0.6	159	59	47
1425	0.501193	210	213	−4	4038	0.497653	212	220	−6
1430	0.516291	206	193	0	4039	0.497653	212	220	−6
1433	0.511962	214	212	−8	4331	0.5	221	236	−15
1435	0.509524	214	214	−8	4332	0.516291	206	193	0
1440	0.506203	204	197	2	4333	0.516291	206	193	0
1441	0.507463	204	196	2	4352	0.587	162	70	44
1697	0.577406	138	33	68	4559	0.509524	214	214	−8
1703	1	206	193	0	4590	0.504255	237	264	−31
1790	0.506361	199	187	7	4651	1	244	252	−38
1800	0.503704	204	199	2	4658	0.605	159	57	47
1824	0.546468	147	63	59	4660	0.503006	251	293	−45
1825	0.506083	208	205	−2	4684	1	162	72	44
1840	0.593	200	131	6	4693	0.503704	204	199	2
1841	1	198	134	8	4694	0.503704	204	199	2
1842	1	149	50	57	4767	0.567358	219	180	−13
2059	0.566751	225	191	−19	4772	1	227	204	−21
2071	1	170	104	36	4773	0.511848	216	216	−10
2072	0.541096	158	86	48	4778	0.587	162	70	44
2075	0.545455	138	47	68	4779	0.587	162	70	44
2109	0.507042	216	220	−10	4781	0.539171	234	228	−28
2116	0.516291	206	193	0	4867	0.516291	206	193	0
2117	0.509524	214	214	−8	4869	0.507042	216	220	−10
2119	0.595	157	58	49	4870	0.51358	208	199	−2
2120	1	165	93	41	4882	0.503448	219	229	−13
2121	0.593	176	91	30	4883	0.507282	209	206	−3
2122	0.526961	215	202	−9	4908	0.516291	206	193	0
2124	0.50116	216	225	−10

SA: The number bits in both RTP and the target candidates.

SB: The number bits in the target candidates but not RTP.

SC: The number bits in RTP but not the target candidates.

Molecular similarity

The base of the different molecular similarity studies is the principle that molecules of similar structures are predicted to exhibit similar biological activities.⁴³ The practical applications of molecular similarity as an essential approach in drug design have increased effectively in the last few years.⁴⁴ The molecular descriptors involved in this kind of study include the physicochemical properties and more specific features in the chemical structure such as H-bond donors (HBA),⁴⁵ H-bond acceptors (HBD),⁴⁶ the octanol–water partition coefficient (ALog p),⁴⁷ molecular weight (M. Wt),⁴⁸ rotatable bonds,⁴⁹ rings and aromatic rings,⁵⁰ besides molecular fractional polar surface area (MFPSA).⁵¹ An investigation into molecular similarity, considering the aforementioned features, was conducted between the top 200 candidates showing the highest similarity (Figure 1) and RTP. This analysis was performed using Discovery Studio software to identify the most similar compounds. A subsequent filtration step by molecular docking was applied to a subset of thirty-four compounds (the selected most similar compounds).

Figure 1.

Similarity analysis between the tested candidates and RTP.

The software, Discovery Studio, was adjusted to compare the molecular descriptors that were mentioned before to select the most similar 40 candidates (Figure 2 and Table 2).

Figure 2.

Thirty-four compounds with good molecular similarity with the co-crystallized ligand (RTP) of RdRp (PDB ID: 7BV2).

Table 2.

Most similar 34 candidates to RTP according to molecular similarity study.

Comp.	ALog p	M. Wt	HBA	HBD	Rotatable bonds	Rings	Aromatic rings	MFPSA	Minimum distance
RTP	−1.5	371.24	11	5	4	3	2	0.612
13	−0.67	302.28	8	6	4	2	1	0.485	1.376
28	−1.56	327.29	8	5	2	3	1	0.474	1.345
150	−0.06	448.38	11	7	4	4	2	0.463	1.339
156	−0.06	448.38	11	7	4	4	2	0.463	1.339
245	−0.34	354.31	9	6	5	2	1	0.487	1.307
263	−0.95	316.26	9	6	4	2	1	0.532	1.109
264	−0.95	316.26	9	6	4	2	1	0.532	1.109
275	−0.06	448.38	11	7	4	4	2	0.463	1.339
295	−0.17	436.37	11	7	3	4	2	0.473	1.281
299	−1.05	344.27	10	7	4	2	1	0.58	1.039
342	−0.06	448.38	11	7	4	4	2	0.463	1.339
609	0.24	448.38	11	7	4	4	2	0.463	1.385
649	−0.71	300.26	8	5	4	2	1	0.482	1.356
682	0.24	448.38	11	7	4	4	2	0.463	1.385
712	−1	463.37	12	7	4	4	2	0.501	1.126
980	−0.83	288.25	8	6	3	2	1	0.518	1.297
999	−0.9	311.29	8	5	4	2	1	0.47	1.376
1040	0.24	448.38	11	7	4	4	2	0.463	1.385
1043	−0.06	448.38	11	7	4	4	2	0.463	1.339
1120	−0.76	447.37	11	6	4	4	2	0.466	1.144
1123	−2.3	478.49	12	5	8	3	2	0.493	1.343
1130	−0.49	342.3	9	6	5	2	1	0.482	1.316
1440	0.24	448.38	11	7	4	4	2	0.463	1.385
1441	0.24	448.38	11	7	4	4	2	0.463	1.385
1790	−0.93	447.37	11	7	3	4	2	0.494	1.132
1824	−1.2	332.26	10	7	4	2	1	0.579	1.044
1825	−0.06	448.38	11	7	4	4	2	0.463	1.339
2072	−0.34	354.31	9	6	5	2	1	0.487	1.307
2119	−0.73	330.29	9	5	5	2	1	0.457	1.378
2205	0.18	448.38	11	7	4	4	2	0.463	1.376
2372	−0.49	342.3	9	6	5	2	1	0.482	1.316
2430	−1	463.37	12	7	4	4	2	0.501	1.126
2620	−1.88	267.24	8	4	2	3	2	0.539	1.064
2628	−0.95	316.26	9	6	4	2	1	0.532	1.109
3099	−0.31	464.38	12	8	4	4	2	0.499	1.36
3266	−0.34	354.31	9	6	5	2	1	0.487	1.307
3404	0.24	448.38	11	7	4	4	2	0.463	1.385
3739	−0.34	354.31	9	6	5	2	1	0.487	1.307
3740	−0.34	354.31	9	6	5	2	1	0.487	1.307
3741	−0.34	354.31	9	6	5	2	1	0.487	1.307

Docking studies

Molecular docking analysis was conducted on forty compounds that had a high molecular resemblance to the co-crystallized ligand (RTP) of RdRp. The binding free energies of the approved metabolites against RdRp (PDB ID: 7BV2) were mentioned in Table 3, while the binding modes were investigated and discussed in detail for RTP and the most promising compounds. In detail, the co-crystallized ligand (RTP) was employed as a reference. Table 3 shows the computed binding free energies of the investigated drugs versus RdRp, as well as the reference drug (RTP). By re-docking the co-crystalized ligands into their complexed enzyme, the measured RMSD between the docked and co-crystallized ligands was less than 1.20 Å, suggesting that the docking method was valid. Discovery Studio 4.5 visualizer was used to visualize the obtained docking results (Figure 3).

Table 3.

Binding free energies (ΔG in Kcal/mol) of the studied candidates and RTP.

Comp.	Name	ΔG (kcal/mol)
Reference	RTP	−18.33
13	7-o-β-glucopyranosyl benzyl alcohol	−18.51
28	(2r)-2-o-β-D-glucopyranosyl-4-hydroxy-2h-1,4-benzoxazin-3(4h)-one	−18.64
150	Kaempferol 3-galactoside	−23.46
156	Kaempferol 3-o-β-D-glucopyranoside	−23.72
245	Chlorogenic acid	−20.7
263	Gentisic acid 2-o-glucopyranoside	−18.31
264	Gentisic acid 5-o-glucopyranoside	−18.44
295	Mangiferin methyl ether	−23.06
299	3-o-galloylquinic acid	−17.67
609	Luteolin 7-o-β-4c1-D-glucopyranoside	−24.88
649	1-o-p-hydroxybenzoyl-β-D-glucopyranoside	−18.33
712	Quercetin-3-glucopyranoside	−22.62
980	Phloroglucinol-1-o-β-D-glucopyranoside	−17.6
999	Taxiphyllin	−20.17
1043	Kaempferol-7-o-glucopyranoside	−20.13
1123	1-methoxy-3-indolylmethyl glucosinolate	−23.08
1130	1-o-caffeoyl-β-D-glucopyranoside	−18.68
1440	Naringenin	−23.12
1441	(+)-Aromadendrin	−20.26
1790	Isoorientin	−21.07
1824	1-o-galloyl-β-d-glucose	−18.58
2072	5-o-caffeoylquinic acid	−20.21
2119	Vanillic acid 4-o-β-D-glucopyranoside	−19.96
2205	Asphodelin a 4'-o-β-D-glucopyranoside	−21.68
2372	6-o-trans-caffeoyl-β-D-glucopyranoside	−18.83
2430	Quercetin-7-o-glucopyranoside	−20.34
2620	Adenosine	−16.03
2628	2,5-dihydroxybenzoic acid-2-o-β-4c1-glucopyranoside	−18.83
3099	Nigrosphaerin A	−20.65
3266	5-o-caffeoylquinic acid	−20.35
3404	Luteolin-7-galactoside	−21.23
3739	1-o-caffeoylquinic acid	−19.94
3740	Neochlorogenic acid	−20.86
3741	Cryptochlorogenic acid	−18.99

Figure 3.

Superimposition of the co-crystallized pose (purple) and the re-docking pose (blue) of RTP indicate nearly the same binding mode.

RdRp enzyme consists of main three parts: (1) ATP binding site represented by a network of different amino acids including Asp760, Asn691, Ser759, Cys622, Lys545, and the key amino acid Arg555. (2) RNA primer is represented by many nucleotides including uridine 20 (U20), uridine 10 (U10), and adenine 11 (A11). (3) Pyrophosphate group (POP1003).

The mode of binding of RTP inside RdRp was illustrated in Figure 4. It was noticed that RTP interacted with the active site via the formation of four hydrogen bonds with U10 and U20, four hydrophobic interactions with U20 and A11 besides five electrostatic interactions with Arg555 and phosphate group.

Figure 4.

Three-dimension (A) and two-dimension (B) representations of the binding mode RTP in the active site of RdRp. A: adenine, U: uracil, and POP: pyrophosphate.

The binding of compound 156 (∆G = −23.72) demonstrated that it occupied the ATP binding site forming four hydrogen bonds with Asp760, Ser759, and Asn691 via its 3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran moiety. Also, the hydroxymethyl moiety formed a hydrogen bond with the pyrophosphate group. Furthermore, compound 156 interacted with the RNA primer to form one hydrogen bond, two pi–pi interactions, and two electrostatic interactions with U20, U10, and A11 via its 5,7-dihydroxy-4H-chromen-4-one moiety. Also, one hydrogen bond with U10 and two electrostatic interactions with Arg555 were observed (Figure 5).

Figure 5.

Mapping surface (A), three-dimension (B), and two-dimension (C) representations of compound 156 in the active site of RdRp. A: adenine, U: uracil, and POP: pyrophosphate.

Investigation of the top docking pose of compound 295 showed that it interacted with the pocket active site through the formation of nine hydrogen bonds and one hydrophobic besides three electrostatic interactions. The 1,6,7-trihydroxy-3-methoxy-9H-xanthen-9-one interacted with U20 nucleotide of the RNA primer through two hydrogen bonds, two electrostatic interactions, and one hydrophobic interaction. In addition, the 1-hydroxyl group of xanthene moiety formed one hydrogen bond with the pyrophosphate group. Moreover, the pyran part occupied the ATP binding site by the formation of two hydrogen bonds between its 3-hydroxy and 5-hydroxy groups with Arg555 and Asn691. Also, the 6-hydroxymethyl group of pyran ring formed three extra hydrogen bonds with Asn691, Asp760, and Ser759 (Figure 6).

Figure 6.

Mapping surface, (A), three-dimension (B), and two-dimension (C) representations of compound 295 in the active site of RdRp. U: uracil and POP: pyrophosphate.

Next, the proposed binding pattern of the chromen-4-one derivative 609 was illustrated in Figure 7. It was noticed that the titled compounds were able to interact with the RNA primer through the formation of many hydrogen bonds and hydrophobic interactions with the key nucleotides U20, U10, and A11. At the same time, the two derivatives formed hydrogen bonds with the pyrophosphate group through 3,4-dihydroxyphenyl moiety. Also, the pyran ring in both compounds binds with the active residues in the ATP binding site including Ser759, Asn691, Thr680, and the key amino acid Arg555.

Figure 7.

Mapping surface (A), three-dimension (B), and two-dimension (C) representations of compound 609 in the active site of RdRp. A: adenine, U: uracil, and POP: pyrophosphate.

The docking pose accomplished by compound 1123 (affinity value of −23.08) revealed that it occupied the RdRP active site. It interacted with the active site through seven hydrogen bonds with Asn691, Asp760, Arg555, and the pyrophosphate group. Likewise, the 1-methoxy-1H-indole moiety was involved in two electrostatic interactions with Arg555 besides three pi–pi interactions with U20 and A11 (Figure 8).

Figure 8.

Mapping surface (A), three-dimension (B), and two-dimension (C) representations of compound 1123 in the active site of RdRp. A: adenine, U: uracil, and POP: pyrophosphate.

Finally, the metabolite 1440 was stabilized in the RdRP active site through the formation of eight hydrogen bonds with U20, U10 (in RNA primer), Asn691, Asp760, Ser759, Arg555 (in ATP active site), and pyrophosphate group. Also, it formed four electrostatic (with U20 and Arg555) and two hydrophobic interactions (with U20 and A11) (Figure 9).

Figure 9.

Mapping surface, (A), three-dimension (B), and two-dimension (C) representations of compound 1440 in the active site of RdRp. A: adenine, U: uracil, and POP: pyrophosphate.

ADMET studies

Ten compounds with a higher energy score in the docking study were explored further utilizing computer-based ADMET experiments to determine their pharmacokinetic profiles. RTP was used as a reference. The calculated parameters include the potential to pass from the blood–brain barrier (BBB-L), the solubility level in aqueous media (S-L), the potential to be absorbed from the human gut (A-L) level, the potential to bind and inhibit the cytochrome P2D6 (CYP2D6-I), and the potential to bind to the plasma protein (PB-L).

The findings (Table 4) demonstrated that all the tested metabolites have extremely poor BBB penetration. As a result, no CNS side effects are anticipated. Furthermore, all members tested performed effectively at ideal water solubility levels. On the other hand, the candidates who were tested had very poor intestinal absorption. All of the metabolites studied were anticipated to be non-inhibitors of CYP2D6 except compound 1790. As a result, they were all classified as non-hepatotoxic. Finally, all of the metabolites studied were expected to bind to plasma protein at a rate of less than 90%. In general, we can deduce that all metabolites studied had a bitter profile of drug-likeness and passed ADMET filter properly except isoorientin, 1790, which was excluded in the further study.

Table 4.

Predicted ADMET descriptors for the examined compounds and RTP.

Comp. No.	BBB-L^a	A-L^b	S-L^c	CYP2D6-I^d	PB-L^e
295	4	3	3	N-I	L
1120	4	3	3	N-I	L
1790	4	3	3	I	L
712	4	3	3	N-I	L
609	4	3	3	N-I	L
1440	4	3	3	N-I	L
2205	4	3	3	N-I	L
1123	4	3	3	N-I	L
156	4	3	3	N-I	L
150	4	3	3	N-I	L
RTP	4	3	2	N-I	M

^a0 is very high, 1 is high, 2 is medium, 3 is low, and 4 is very low.

^b0 is good, 1 is moderate, 2 is poor, and 3 is very poor.

^c0 is extremely low, 1 is very low, 2 is low, 3 is good, and 4 is optimal.

^dN-I is a non-inhibitor and I is an inhibitor.

^eL means less than 90% and M means more than 90%.

Toxicity studies

The toxicity of any new compound is a critical parameter in the process of drug development. The optimization of toxicity is the most time and money-consuming part of the pre-clinical trials during the drug discovery process.⁵² Also, the accurate identification of the toxicological profile of a new compound is of great importance to avoid any side effects or late withdrawal.⁵³

Next, we looked at the toxicity profiles of fifteen metabolites that passed via the ADMET filter. This entails using the Discovery Studio 4.0 programme to create seven created toxicity models (Table 5). According to the findings, all members were found to be non-carcinogenic in the FDA rat male carcinogenicity model. Also, all the examined compounds had TD₅₀ values ranging from 10.159 to 53.265 g/kg body, which were higher than RTP (2.014 g/kg body).

Table 5.

Toxicity properties of tested compounds and the reference molecule.

Comp.	FDA carcinogenicity	TD₅₀	(RMTD-L)^a	LD₅₀^a	LOAEL-L^a	Ocular irritancy	Skin irritancy
RTP	Non-carcinogen	2.014	0.003	0.209	0.002	Mild	None
150		13.836	1.167	0.695	0.056	Moderate	None
156		13.836	1.167	0.695	0.056	Moderate	None
295		10.159	0.770	0.523	0.069	Mild	None
609		10.718	1.230	1.358	0.030	Moderate	None
712		14.619	0.868	0.336	0.059	Moderate	None
1123		53.265	0.108	0.208	0.025	Moderate	Mild
1440		13.641	1.230	1.704	0.037	Moderate	None
2205		16.158	1.574	2.610	0.046	Severe	None

^aUnit: g/kg body.

Additionally, all members had rat maximum tolerated doses (RMTD-L) ranging from 0.577 to 1.574 g/kg body weight, which were extremely greater than RTP (0.003 g/kg body). Moreover, all compounds, except for 1123, had rat oral LD₅₀ values ranging from 0.336 to 2.610 g/kg body weight, which were higher than that of RTP (0.209 g/kg body).

Furthermore, all compounds demonstrated rat chronic LOAEL (LOAEL-L) values ranging from 0.030 to 0.069 g/kg body which were higher than that of RTP (0.002 g/kg body). Finally, all the tested members revealed mild to moderate ocular irritancy against the ocular irritancy model except compound 2205 that displayed severe irritancy. Also, all members displayed non-irritancy against the skin irritancy model except compound 1123 which showed mild irritancy against this model (Table 5). To summarize, all compounds passed the toxicity filter properly.

Molecular dynamic simulation

The MD simulation studies can explore several essential biomolecular processes that happen to a specific protein upon binding with a certain ligand over a determined time. These processes include protein conformational changes, protein folding, and stability of ligand binding.⁵⁴ All these processes have been predicted at an atomic level through the revelation of any change in the atomic positions at a temporal resolution of femtosecond.⁵⁵ Interestingly, such studies predict biomolecules’ responding, at an atomic level, to any perturbation such as mutation, protonation, phosphorylation, binding, or unbinding to a ligand.

To evaluate the integrity as well as the stability of the kaempferol 3-O-β-D-glucopyranoside-RdRp complex, the root-mean square deviation (RMSD) of the backbone atoms, root-mean square fluctuation (RMSF) of the Cα atoms, solvent accessible surface area (SASA), and the radius of gyration (R_g) of RdRp were investigated. In addition, the RMSD of the ligand, the number of H-bonds formed with the RdRp along the duration of the simulation, and the binding energy between the ligand and the protein using MM-GBSA were calculated (with decomposition).

The RMSD values of the protein which measures the overall change of each frame from the reference frame across the simulation show an increasing trend in the first half of the simulation and then stabilizes around 2.7 Å (Figure 10). This indicates a stable conformation during the second half of the simulation. The RMSF values show a large fluctuation from Arg331 to Asn360, which corresponds to the loop motion during the simulation (Figure 11). The SASA values (Figure 12) identify how much of the protein surface is exposed to the solvent, and the radius of gyration (R_g) (Figure 13) measures the protein compactness. Both of them indicate stable values around 37,300 Å² and 28.8 Å for the SASA and R_g, respectively, indicating that the RdRp was stably folded, which was critical for the validity of the simulation analysis.

Figure 10.

RMSD of the protein in RdRp-kaempferol 3-O-β-D-glucopyranoside complex, calculated throughout a 100-ns MD simulation.

Figure 11.

RMSF of RdRp Cα atoms when complexed with kaempferol 3-O-β-D-glucopyranoside, calculated over the course of a 100-ns MD simulation.

Figure 12.

SASA values of RdRp when complexed with kaempferol 3-O-β-D-glucopyranoside, calculated for the 100-ns MD simulation.

Figure 13.

Radius of gyration of RdRp when complexed with kaempferol 3-O-β-D-glucopyranoside, calculated over the course of a 100-ns MD simulation.

Furthermore, the RMSD of kaempferol 3-O-β-D-glucopyranoside is a measure of how much its conformation has varied relative to the reference structure. kaempferol 3-O-β-D-glucopyranoside reached a relatively stable conformation after 25 ns, with an RMSD fluctuating around 0.48 nm (Figure 14), indicating small changes in the kaempferol 3-O-β-D-glucopyranoside’s position and conformation while preserving its binding to the RdRp’s active site. Additionally, the number of H-bonds formed between the compound and RdRp was calculated (Figure 15) and was shown to have two to three bonds, on average.

Figure 14.

RMSD of the ligand with reference to the starting configuration, calculated over the course of the 100-ns simulation.

Figure 15.

Change in the number of H-bonds formed between the ligand and the RdRp over the course of the 100-ns simulation.

Additionally, the RdRp-kaempferol 3-O-β-D-glucopyranoside interactions were analyzed to assess the kaempferol 3-O-β-D-glucopyranoside’s affinity towards the RdRp and the strength of their interactions. The energetics analysis revealed that the average Lennard-Jones energy was −92.5382 kJ/mol (−22.11 kcal/mol) (Figure 16), and the average Coulombic interaction energy was −206.63 kJ/mol (−49.42 kcal/mol) (Figure 17), indicating the ligand’s high affinity towards the RdRp.

Figure 16.

Lennard-Jones interaction energy during the MD simulation, showing an average value of −92.5382 kJ/mol.

Figure 17.

Columbic electrostatic interaction energy during MD simulation, showing an average value of −206.63 kJ/mol.

Moreover, the RdRp _kaempferol 3-O-β-D-glucopyranoside interactions were analyzed to assess the kaempferol 3-O-β-D-glucopyranoside’s affinity towards the RdRp and the contribution of amino acids around 1 nm of the compound. The binding energy analysis using MM-GBSA revealed that the average binding energy is −25 kcal/mol (Figure 18) with electrostatic interactions as the most effective component (−67.33 Kcal/Mol) compared to the hydrophobic interactions (−25.63 Kcal/Mol). In addition, decomposition was performed to obtain the contribution of each amino acid towards the binding energy (Figure 19). There are three amino acids (Arg553 [−1.38 Kcal/Mol], Arg555 [−4.1 Kcal/Mol], and Asp760 [−1.39 Kcal/Mol]) with a contribution of less than −1 Kcal/Mol which are consistent with the docking and PLIP results.

Figure 18.

Different energetic components of MM-GBSA and their values. Bars represent the standard deviations.

Figure 19.

Binding free energy decomposition of the RdRp_kaempferol 3-O-β-D-glucopyranoside complex.

Protein–ligand interaction profiler experiment

PLIP stands as a valuable computational tool employed to meticulously dissect and visualize the intricate interactions between a protein and a ligand, those diminutive molecules of significance.⁵⁶ This tool affords a closer, atomic-level look into the profound engagement of a ligand with a protein. Its prowess lies in unraveling binding sites, exposing the nuances of hydrogen bonding, shedding light on hydrophobic interactions, and unearthing an array of pertinent information. Such insights are tantamount to unraveling the complex behavior exhibited by the intricate interplay of protein–ligand complexes.⁵⁷

In this study, the MD trajectory was clustered into 5 clusters, and for each cluster, a representative frame was obtained and used with PLIP to get the interacting amino acids. Table 6 shows the interacting amino acids in each cluster representative. As can be shown, several common amino acids appear in each representative (marked in bold) such as Arg555 (forms a salt bridge and H-bond), Val557 (forms a hydrophobic interaction), and Asp760 (forms a hydrogen bond). In addition, Arg555, Asp760, Asn691, and Ser759 are persistent amino acids that are common between the docking and MD simulation (Figure 20).

Table 6.

Interacting amino acids obtained from PLIP for each representative cluster. Bold amino acids are the amino acids that are common between the docking step and the MD simulation.

Cluster number	Number of salt bridges	Amino acids in RDRP	Number of hydrophobic interactions	Amino acids in RDRP	Number of hydrogen bonds	Amino acids in RDRP
C1	1	Arg555	1	Val557	7	Arg555 - Asp623 - Ser682 - Asn691 - Ser759 - Asp760 (2)
C2	1	Arg555	1	Val557	10	Arg553 - Arg555 (2) - Asp623 - Thr680 - Ser682 - Asn691 - Ser759 - Asp760 (2)
C3	1	Arg555	3	Lys545 - Arg555 - Val557	7	Ala547 - Arg553 - Arg555 (3) - Asp623 - Asp760
C4	1	Arg555	1	Val557	8	Ala547 - Arg553 - Arg555 (2) - Asp623 - Ser759 - Asp760 (2)
C5	1	Arg555	2	Lys545 - Val557	7	Ala547 - Arg553 - Arg555 (2) - Ser759 - Asp760 (2)

Figure 20.

Clustering of the trajectory using TTClust and the interactions obtained from PLIP on the representative frames.

Conclusion

Eight compounds among 4924 African natural products (kaempferol 3-galactoside, kaempferol 3-O-β-D-glucopyranoside, mangiferin methyl ether, luteolin 7-O-β-4C1-D-glucopyranoside, quercetin-3-glucopyranoside, 1-methoxy-3-indolylmethyl glucosinolate, naringenin, and asphodelin A 4’-O-β-D-glucopyranoside) were decided as the most potent inhibitors against SARS-CoV-2 RdRp (PDB ID: 7BV2). The study depended on a multi-phase in silico approach to select the most structurally similar compounds to RTP, the co-crystallized ligand of the RdRp employing a molecular fingerprints and structure similarity studies. Then, molecular docking for the chosen 34 compounds preferred 10 candidates. Followingly, ADMET study excluded a compound. Finally, MD simulation studies verified the correct binding of kaempferol 3-O-β-D-glucopyranoside against RdRp for 100 ns. The outputted results are a solid base to conduct further in vitro and in vivo studies on the preferred compounds to find a cure against COVID-19.

Method

Fingerprint studies

For the fingerprints analysis of 4924 African natural products to select the most similar 200 compounds to RTP, we utilized the Discovery Studio 4.0 software to compare the number of charges, H. bond acceptors, H. bond donors, f-positive ionizable atoms, negative ionizable atoms, halogen atoms, and aromatic rings in addition to hybridization and ALogP as described in the Supplementary data section.

Molecular similarity detection

For the molecular similarity analysis of 200 African natural products, to select the most similar compounds (34 compounds) to RTP, we utilized the Discovery Studio 4.0 software to compare the number of rotatable bonds, rings, aromatic rings, H. bond donors, and H. bond acceptors, in addition to ALog p, molecular weight, and the MFPSA as described in the Supplementary data section. A maximum distance of 1.385 was selected.

Docking studies

Docking studies were performed against SARS-CoV-2 RdRp (PDB ID: 7BV2) using Discovery Studio software⁵⁸ and depending on RTP as a reference (see method part in Supplementary data).

ADMET analysis

Discovery Studio 4.0 was used to predict human intestinal absorption, aqueous solubility, blood–brain barrier penetration, plasma protein binding, cytochrome P450 2D6 inhibition, and hepatotoxicity (see method part in Supplementary data).^59,60

Toxicity studies

Discovery Studio 4.0 software was used⁶¹ to predict the FDA rat carcinogenicity, TD50, MTD, oral LD50 in rats, LOAEL, ocular, and dermal irritancy (see method part in Supplementary data).

Molecular dynamics simulations

A 100 ns molecular dynamics (MD) simulation was performed for the kaempferol 3-O-β-D-glucopyranoside–RdR complex using GROMACS v.2021.4. Once the simulation concluded, the periodic boundary conditions (PBCs) were eliminated from the system using the trjconv command, restoring the RdRp protein to its complete form. Subsequently, VMD TK scripts were applied to analyze the trajectory. Calculations were conducted for the root-mean square deviation (RMSD) of both RdRp and kaempferol 3-O-β-D-glucopyranoside. Essential metrics such as hydrogen bond count, radius of gyration (RoG), solvent-accessible surface area (SASA), and root-mean square fluctuation (RMSF) were also determined. Following this, the trajectory of the kaempferol 3-O-β-D-glucopyranoside-RdRp complex underwent clustering through TTClust, generating representative frames for each cluster. To comprehensively examine these frames for the spectrum and types of kaempferol 3-O-β-D-glucopyranoside–RdRp interactions, the Protein–Ligand Interaction Profiler (PLIP) was enlisted. The outcomes were compiled in a .pse file format, readily viewable through PyMOL. In order to ascertain the binding free energy, the MM-GBSA method was employed using the gmx_MMPBSA programme. Furthermore, decomposition analysis was executed to quantify the contributions of interacting amino acids within 1 nm of kaempferol 3-O-β-D-glucopyranoside. Notably, the ionic strength was set at 0.154 M, and the solvation method (igb) was designated as 5 for the calculations (see method part in Supplementary data).

Supplemental Material

Supplemental Material - Computer-assisted drug discovery of potential natural inhibitors of the SARS-CoV-2 RNA-dependent RNA polymerase through a multi-phase in silico approach

Supplemental Material for Computer-assisted drug discovery of potential natural inhibitors of the SARS-CoV-2 RNA-dependent RNA polymerase through a multi-phase in silico approach by Eslam B Elkaeed, Bshra A Alsfouk, Tuqa H Ibrahim, Reem K Arafa, Hazem Elkady, Ibrahim M Ibrahim, Ibrahim H Eissa, and Ahmed M Metwaly in Antiviral Therapy.

Footnotes

Author contributions

The study was conceptualized, designed, and supervised by IHE and AMM. EE and HE carried out the computational studies. TI and RA carried out the MD studies. HE was responsible for analyzing and interpreting the data. The fund accusation was provided by EBE and BAA, who were also included in the writing of the manuscript. All authors have thoroughly reviewed and agreed upon the final version of the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2023R142), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia. The authors extend their appreciation to the Research Center at AlMaarefa University for funding this work.

ORCID iDs

Eslam B Elkaeed

Hazem Elkady

Ahmed M Metwaly

Supplemental Material

Supplemental material for this article is available online.

References

WHO . WHO coronavirus (COVID-19) dashboard, 2023. [cited 2023 July, 10]; Available from: https://covid19.who.int/

J and

Hagler

. Chemoinformatics and drug discovery. Molecules 2002; 7(8): 566–600.

Engel

. Basic overview of chemoinformatics. J Chem Inf Model 2006; 46(6): 2267–2277.

Westbrook

JD and

Burley

. How structural biologists and the protein data bank contributed to recent FDA new drug approvals. Structure 2019; 27(2): 211–217.

Grimme

S and

Schreiner

. Computational chemistry: the fate of current methods and future challenges. Angew Chem Int Ed 2018; 57(16): 4170–4176.

Amin

Banerjee

Singh

, et al. First structure-activity relationship analysis of SARS-CoV-2 virus main protease (Mpro) inhibitors: an endeavor on COVID-19 drug discovery. Mol Divers 2021; 25(3): 1827–1838.

Ranjan

Devarapalli

Kundu

, et al. Isomorphism: 'molecular similarity to crystal structure similarity’ in multicomponent forms of analgesic drugs tolfenamic and mefenamic acid. IUCrJ 2020; 7(Pt 2): 173–183.

Baidya

Ghosh

Amin

, et al. In silico modelling, identification of crucial molecular fingerprints, and prediction of new possible substrates of human organic cationic transporters 1 and 2. New J Chem 2020; 44(10): 4129–4143.

Shi

. Support vector regression-based QSAR models for prediction of antioxidant activity of phenolic compounds. Sci Rep 2021; 11(1): 8806.

10.

Idris

Yekeen

Alakanse

, et al. Computer-aided screening for potential TMPRSS2 inhibitors: a combination of pharmacophore modeling, molecular docking and molecular dynamics simulation approaches. J Biomol Struct Dyn 2021; 39(15): 5638–5656.

11.

Y and

. A new computer model for evaluating the selective binding affinity of phenylalkylamines to T-Type Ca2+ channels. Pharmaceuticals 2021; 14(2): 141.

12.

Fan

J and

Zhang

LJQB

. Progress in molecular docking. Quant Biol 2019; 7: 83–89.

13.

Hospital

Goñi

Orozco

, et al. Molecular dynamics simulations: advances and applications. Comput Biol Chem Adv Appl 2015; 8: 37–47.

14.

Cheng

Liu

, et al. In silico ADMET prediction: recent advances, current challenges and future trends. Curr Top Med Chem 2013; 13(11): 1273–1289.

15.

Obot

Macdonald

D and

Gasem

ZJCS

. Density functional theory (DFT) as a powerful tool for designing new organic corrosion inhibitors. Part 1: an overview. Corrosion Sci 2015; 99: 1–30.

16.

Metwaly

Lianlian

Luqi

, et al. Black ginseng and its saponins: preparation, phytochemistry and pharmacological effects. Molecules 2019; 24(10): 1856.

17.

Metwaly

Ghoneim

Eissa

, et al. Traditional ancient Egyptian medicine: a review. Saudi J Biol Sci 2021; 28(10): 5823–5832.

18.

Löwenberg

Pabst

Vellenga

, et al. Cytarabine dose for acute myeloid leukemia. N Engl J Med 2011; 364(11): 1027–1036.

19.

Shetty

N and

Gupta

. Eribulin drug review. South Asian J Cancer 2014; 3(1): 57–59.

20.

Pope

JE and

Deer

. Ziconotide: a clinical update and pharmacologic review. Expet Opin Pharmacother 2013; 14(7): 957–966.

21.

Lei

Niu

, et al. The FDA-approved natural product dihydroergocristine reduces the production of the Alzheime’s disease amyloid-β peptides. Sci Rep 2015; 5(1): 16541.

22.

Alesawy

Abdallah

Taghour

, et al.

In Silico studies of some isoflavonoids as potential candidates against COVID-19 targeting human ACE2 (hACE2) and viral main protease (Mpro)

. Molecules 2021; 26(9): 2806.

23.

El-Demerdash

Metwaly

Hassan

, et al. Comprehensive virtual screening of the antiviral potentialities of marine polycyclic guanidine alkaloids against SARS-CoV-2 (COVID-19). Biomolecules 2021; 11(3): 460.

24.

Eissa

Khalifa

Elkaeed

, et al.

In Silico exploration of potential natural inhibitors against SARS-Cov-2 nsp10

. Molecules 2021; 26(20): 6151.

25.

Elkaeed

Youssef

Eissa

, et al. Multi-step In Silico discovery of natural drugs against COVID-19 targeting main protease. Int J Mol Sci 2022; 23(13): 6912.

26.

Elkaeed

Eissa

Elkady

, et al. A multistage in silico study of natural potential inhibitors targeting SARS-CoV-2 main protease. Int J Mol Sci 2022; 23(15): 8407.

27.

Elkaeed

Metwaly

Alesawy

, et al. Discovery of potential SARS-CoV-2 papain-like protease natural inhibitors employing a multi-phase In Silico approach. Life 2022; 12(9): 1407.

28.

Eissa

Alesawy

Saleh

, et al. Ligand and structure-based In Silico determination of the most promising SARS-CoV-2 nsp16-nsp10 2′-o-Methyltransferase complex inhibitors among 3009 FDA approved drugs. Molecules 2022; 27(7): 2287.

29.

Elkaeed

Elkady

Belal

, et al. Multi-phase In Silico discovery of potential SARS-CoV-2 RNA-dependent RNA polymerase inhibitors among 3009 clinical and FDA-approved related drugs. Processes 2022; 10(3): 530.

30.

Metwaly

Elwan

El-Attar

A-AMM

, et al. Structure-based virtual screening, docking, ADMET, molecular dynamics, and MM-PBSA calculations for the discovery of potential natural SARS-CoV-2 helicase inhibitors from the traditional Chinese medicine. J Chem 2022; 2022: 1–23.

31.

Alesawy

Elkaeed

Alsfouk

, et al.

In Silico screening of semi-synthesized compounds as potential inhibitors for SARS-CoV-2 papain-like protease: pharmacophoric features, molecular docking, ADMET, toxicity and DFT studies

. Molecules 2021; 26(21): 6593.

32.

ANPDB. African Natural Products Database

2021 [cited 2021 11-11-2021] An amalgamation of NP databases from various regions of Africa. Currently, ANPDB comprises of the northern African natural products Database (NANPDB) and the Eastern African natural products Database (EANPDB)]. Available from: http://african-compounds.org/anpdb/compounds_list/

33.

Wang

Stephen

Tao

, et al. Human endeavor for anti-SARS-CoV-2 pharmacotherapy: a major strategy to fight the pandemic. Biomed Pharmacother 2021; 137: 111232.

34.

Carvalho

Trincão

and Romão MJ. X-ray crystallography in drug discovery. Methods Mol Biol 2010; 572: 31–56.

35.

Hassell

G and

Bledsoe

, et al. Crystallization of protein-ligand complexes. Acta Crystallogr Sect D Biol Crystallogr 2007; 63(1): 72–79.

36.

Vidal

Garcia-Serna

Mestres

. Ligand-based approaches to in silico pharmacology. In: Chemoinformatics and computational chemical biology. Berlin: Springer, 2011, 489–502.

37.

Burke

. Developments in molecular shape analysis to establish spatial similarity among flexible molecules. Chicago, IL: University of Illinois at Chicago, Health Sciences Center, 1993.

38.

Briem

H and

Kuntz

. Molecular similarity based on DOCK-generated fingerprints. J Med Chem 1996; 39(17): 3401–3408.

39.

Chu

Q-x.

Wang

, et al. In silico design of novel benzohydroxamate-based compounds as inhibitors of histone deacetylase 6 based on 3D-QSAR, molecular docking, and molecular dynamics simulations. New J Chem 2020; 44(48): 21201–21210.

40.

Ieritano

Campbell

JL and

Hopkins

. Predicting differential ion mobility behaviour in silico using machine learning. The Analyst 2021; 146(15): 4737–4743.

41.

Taha

Ismail

Ali

, et al. Molecular hybridization conceded exceptionally potent quinolinyl-oxadiazole hybrids through phenyl linked thiosemicarbazide antileishmanial scaffolds: In silico validation and SAR studies. Bioorg Chem 2017; 71: 192–200.

42.

Opo

FADM

Rahman

Ahammad

, et al. Structure based pharmacophore modeling, virtual screening, molecular docking and ADMET approaches for identification of natural anti-cancer agents targeting XIAP protein. Sci Rep 2021; 11(1): 1–17.

43.

Maggiora

Vogt

Stumpfe

, et al. Molecular similarity in medicinal chemistry. J Med Chem 2014; 57(8): 3186–3204.

44.

Bender

A and

Glen

. Molecular similarity: a key technique in molecular informatics. Org Biomol Chem 2004; 2(22): 3204–3218.

45.

Altamash

Amhamed

Aparicio

, et al. Effect of hydrogen bond donors and acceptors on CO2 absorption by deep eutectic solvents. Processes 2020; 8(12): 1533.

46.

Wan

Tian

Wang

, et al. In silico studies of diarylpyridine derivatives as novel HIV-1 NNRTIs using docking-based 3D-QSAR, molecular dynamics, and pharmacophore modeling approaches. RSC Adv 2018; 8(71): 40529–40543.

47.

Turchi

Cai

Q and

Lian

. An evaluation of in-silico methods for predicting solute partition in multiphase complex fluids - A case study of octanol/water partition coefficient. Chem Eng Sci 2019; 197: 150–158.

48.

Sullivan

Enoch

Ezendam

, et al. An adverse outcome pathway for sensitization of the respiratory tract by low-molecular-weight chemicals: building evidence to support the utility ofIn VitroandIn SilicoMethods in a regulatory context. Appl In Vitro Toxicol 2017; 3(3): 213–226.

49.

Escamilla-Gutiérrez

Ribas-Aparicio

Córdova-Espinoza

, et al. In silico strategies for modeling RNA aptamers and predicting binding sites of their molecular targets.Nucleosides, Nucleotides Nucleic Acids 2021; 40(8): 798–807.

50.

Kaushik

Kumar

Bharadwaj

, et al. Ligand-based approach for in-silico drug designing. In: Bioinformatics Techniques for Drug Discovery. Berlin: Springer, 2018, 11–19.

51.

Zhang

Ren

J-X

, et al. Development of an in silico prediction model for chemical-induced urinary tract toxicity by using naïve Bayes classifier. Mol Divers 2019; 23(2): 381–392.

52.

Blomme

EA and

Will

. Toxicology strategies for drug discovery: present and future. Chem Res Toxicol 2016; 29(4): 473–504.

53.

Krewski

Acosta

Andersen

, et al. Toxicity testing in the 21st century: a vision and a strategy. J Toxicol Environ Health. Part B, Critical reviews 2010; 13(2–4): 51–138.

54.

Sousa

Fernandes

PA and

Ramos

. Protein-ligand docking: current status and future challenges. Proteins: Struct, Funct, Bioinf 2006; 65(1): 15–26.

55.

Hollingsworth

SA and

Dror

. Molecular dynamics simulation for all. Neuron 2018; 99(6): 1129–1143.

56.

Salentin

Schreiber

Haupt

, et al. PLIP: fully automated protein-ligand interaction profiler. Nucleic Acids Res 2015; 43(W1): W443–W447.

57.

Salentin

Adasme

Heinrich

, et al. From malaria to cancer: computational drug repositioning of amodiaquine using PLIP interaction patterns. Sci Rep 2017; 7(1): 11401.

58.

bank, P.d., and R.P.D. Bank , 2020, 2021: https://www.rcsb.org/structure/4OW0

59.

Alanazi

Elwan

Alsaif

, et al. Discovery of new 3-methylquinoxalines as potential anti-cancer agents and apoptosis inducers targeting VEGFR-2: design, synthesis, and in silico studies. J Enzym Inhib Med Chem 2021; 36(1): 1732–1750.

60.

Yousef

Sakr

Eissa

, et al. New quinoxaline-2 (1H)-ones as potential VEGFR-2 inhibitors: design, synthesis, molecular docking, ADMET profile and anti-proliferative evaluations. New J Chem. 2021; 45(36).

61.

Alesawy

Al‐Karmalawy

Elkaeed

, et al. Design and discovery of new 1,2,4‐triazolo[4,3‐c]quinazolines as potential DNA intercalators and topoisomerase II inhibitors. Arch Pharmazie 2021; 354(3): 2000237.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.38 MB