Introduction
Coronaviruses are a family of viruses that cause a wide range of diseases including common cold, severe acute respiratory syndrome (SARS), etc. The novel coronavirus, severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), shares about 80% and 50% homology with severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV) respectively [
1,
2]. The virus, SARS-CoV-2, is an enveloped positive-sense, single-stranded RNA virus, with genome size of about 30 kb [
1]. It has crown-like spike protein on its membrane, and uses this to bind cell receptor, thereby facilitating virus–host cell membrane fusion and infection of target human cells [
3]. The genetic content of the virus is then emptied into the host cell where the translation machinery of the host cell is used to make viral polyproteins. The proteolytic processing of this polyproteins yields structural and non-structural proteins. The non-structural proteins include papain-like protease (PLpro), 3-chymotrypsin-like protease (3CLpro), helicase and RNA-dependent RNA polymerase, among others [
4]. The proteolytic processing of the polyproteins is performed by the viral cysteine proteases to yield 16 non-structural proteins; the PLpro cleaves first three sites at the N-terminus while the other protease, 3CLpro, cleaves and modify the viral polyproteins at 11 other sites [
5,
6]. In addition to this proteolytic activity, PLpro reverses cellular ISGylation and ubiquitination processes, and may deubiquinate some host cell proteins, like interferon factor 3 and NF-κB, to suppress host innate immune system [
7] and aid survival of the virus. Hence, both 3CLpro and PLpro are viable therapeutic targets.
Drugs such as remdesivir and chloroquine have been reported to be potent inhibitors of SARS-CoV-2 in vitro, and have been proposed for the treatment of the virus [
8]. Furthermore, remdesivir has been reported previously to inhibit zoonotic CoV prior to emergency of SARS-CoV-2 [
9]. Following remdesivir and chloroquine, several drugs and natural compounds have been repurposed for treatment of SARS-CoV-2, to facilitate accelerated drug discovery for the pandemic [
10‐
14].
Number of cases and death associated with coronavirus disease 2019 (COVID-19) remains a concern. Hence, the need for alternative therapy against the causative agent, SARS-CoV-2. The use of natural products for the treatment of diseases is a practice that has been in existence since ancient times and their activities are attributed to the secondary metabolites present in them [
15]. Flavonoids are secondary metabolites present in diverse plant species. They play a wide range of physiological roles in plants and many reports have indicated their pharmacological activities against infectious diseases, metabolic disorders and degenerative diseases. Hence, they are used as antibacterial, antifungal, anti-inflammatory and antiviral agents [
16‐
19]. Flavonoids have been reported to be active against bacteria, such as Mycobacterium tuberculosis, and viruses, such as Streptococcus pneumonia, influenza virus and zika virus [
19‐
22]. Among the benefits of antiviral activities of flavonoids is their several mechanisms by which they inhibit and act on the viruses. Flavonoids acts at multiple stages of viral infection, targeting their attachment, entrance, obstruct phases of viral DNA replication, translation of proteins, poly-protein processing and could inhibit the release of viruses from invasion other healthy host cells [
23,
24]. Flavonoids are naturally occurring, ubiquitously in plants and major secondary metabolites [
25]. Thus, the use of flavonoids against COVID-19 could prove to be a treatment option that is accessible, of low cost, and with little or no adverse effect on infected individuals. 5,7-dimethoxyflavanone-40-O-b-d-glucopyranoside, baicalin, Euchresta flavanone A, flemiflavanone D, hesperidin, kaempferol, luteolin, myricetin 3-rutinoside, naringen, quercetin-3-O-rhamnoside, rhoifolin and rutin are among several flavonoids identified through computational studies as potent inhibitors of SARS-CoV-2 [
26‐
30]. Further studies, through combinatorial molecular simulations, ADMET analysis, and hybrid QM/MM approaches has identified some mechanism of action of the flavonoids against SARS-CoV-2, and their amino acid interactions in the binding pockets of SARS-CoV-2 structural and none structural proteins. Compounds which are active against SARS-CoV-2 proteases are expected to cause inhibition of the enzymes. Compounds that inhibit the viral proteases PLpro and 3CLpro could terminate the replication process of SARS-CoV-2. This study was, therefore, embarked on to evaluate the interaction of selected flavonoids with SARS-CoV-2 PLpro and 3CLpro and their pharmacokinetic parameters in silico.
Discussion
Therapeutic compounds against COVID-19 could stimulate human cells such as receptors and immune system. Also, structural and non-structural proteins of virus could be targeted by compounds. Compounds that target the virus inhibit binding of virus to human cell receptors, prevents viral replication processes, inhibit viral protein modification and/or inhibit self-assembly process of the virus. Viral spike protein, 3CLpro, PLpro, RNA-dependent RNA polymerase, and host cell proteins (ACE2 and TMPRSS2) that facilitate viral entry to cell are important therapeutic targets.
Cell entry by virus is followed by translation of the two-thirds of the genome, from the 5' end to yield two large replicase polyproteins (pp1a and pp1ab), which are cleaved by PLpro and 3CLpro to yield up to 16 non-structural proteins (nsp1 – nsp16). The assembly of the proteins give rise to membrane-bound replicase complex, which facilitate replication and structural gene translation of viral genome [
7,
48]. In addition to the N-terminus proteolytic cleavage activities of PLpro to yield nsp1, nsp2 and nsp3, the protease also antagonizes the host’s innate immunity via its deubiquitinating/deISGylating activity [
49‐
51]. Further downstream, proteolytic cleavage of polyproteins by 3CLpro yields nsp4 –nsp16 [
52].
The amino acid residues Cys111, His272 and Asp286 forms the catalytic triad of SARS-CoV-2 PLpro [
7,
48], while Trp106, Gly256, and Lys274 are amino acid residues at the catalytic region [
48]. Also, residues Glu167, Leu162, Asp164 and Tyr264 have been reported to be essential or required for deubiquitinating activity of PLpro [
53]. The response of host innate immune system is critical to controlling SARS-CoV-2 infection. The PLpro of SARS-CoV-2 reverses post-translational modification of immune proteins, like interferon factor 3 and NF-κB, which is achieved by ubiquitin and interferon-stimulated gene product 15 (ISG15). The reversal of ubiquitination and ISGylation suppresses host innate immune responses [
7,
54,
55], and assists SARS-CoV-2 escape from host innate immune responses. The results revealed that licorice, the flavonoid with the lowest binding energy, interacted with HIS89, ASN109B, GLY160A, ASN109C, GLY160C HIS89, VAL159, GLY160, and did not interact with amino acids required for post-translational and deubiquitination activities of PLpro, thus it may not directly inhibit these activities. However, licorice, through its strong interactions with neighbouring amino acid residues may cause some conformational changes at the active site which may indirectly inhibit PLpro. However, lopinavir (a reference protease inhibitor used in this study) interacted with Leu162 of PLpro via Pi-alkyl interaction, suggesting that it may inhibit the deubiquitinating activity of PLpro. The interaction of procyanidin with PLpro, via its Pi-Sigma interaction with Trp106, Pi-alkyl interaction with Leu162 and Pi–Pi interaction with Tyr264 (amino acids required for the deubiquitinating activity of the enzyme), suggests that procyanidin may alter the catalytic conformation of PLpro and inhibit its ability to reverse ubiquitination. The interactions of silymarin via Pi-donor hydrogen bond, ugonin M via alkyl interaction, gallocatechin gallate and isonymphaeol B via Pi-alkyl interaction with Leu162 of PLpro suggests that the flavonoids may possess inhibitory activities against the deubiquitinating function of SARS-CoV-2 PLpro, thereby limiting the ability of the virus to suppress the host immunity. The results suggest that procyanidin may be more potent in inhibiting the deubiquitinating activity of SARS-CoV-2 PLpro compared to other flavonoids and lopinavir (the reference compound).
The catalytic dyad (His41 and Cys145) of 3CLpro is domiciled between its domain I (residues 8–101) and domain II (residues 102–184) [
56]. A long loop (residues 185–200) that connects domain II and domain III (residues 201–303) completes the 3CLpro monomer [
52]. The 3CLpro recognises Leu-Gln*Ser, Leu-Gln*Ala and Leu-Gln*Gly sequence at most sites for cleavage (cleavage site asterisked) [
56]. 3CLpro is a good therapeutic target as no human protease appears to have such recognition sequence [
56]. The interaction of lopinavir and ritonavir with Gln110 of 3CLpro occurred via hydrogen bonding while baicalin and apigetrin interacted with the amino acid residue via a Pi-donor hydrogen bond. Both isonymphaeol B and abyssinone II interacted with a number of the same amino acids of 3CLpro (Asp295, Phe294, Pro293, Ile249, Val202, and His246) as did lopinavir. In addition to these, isonymphaeol B also interacted with Asp153 while abyssinone II interacted with Ile200 (a residue that ritonavir also interacted with), both amino acids being points of interaction for lopinavir. While lopinavir interacted with ASP295, a domain III amino acid residue of 3CLpro, via a Pi-anion bond, isonymphaeol B and abyssinone II interacted via hydrogen bonding. The large number of hydrogen bonds involved in the interaction of flavonoids with the least binding energies (isonymphaeol B, baicalin, abyssinone II, apigetrin, and tomentin A) with 3CLpro seems to be responsible for the higher binding affinities of the protein for them. Despite that these flavonoids did not interact with the catalytic dyad of 3CLpro, their strong interactions with neighbouring amino acid residues may cause some conformational changes at the active site which may inhibit the catalytic activity of 3CLpro. The strong hydrogen bonding and hydrophobic interaction exhibited by isonymphaeol B suggest its potential as a potent 3CLpro inhibitor. The anti-SARS-CoV-2 activities of baicalin and silymarin, have similarly being reported by Akhter et al. [
29]. The flavonoids were identified as potentially active compounds against SARS-CoV-2 Mpro.
The predictive physicochemical and pharmacokinetic analyses of flavonoids with the least binding energies revealed variations in the properties of these compounds. The results revealed that the gastrointestinal absorption was high for isonymphaeol B, abyssinone II and tomentin A but was low for other flavonoids with low binding energies for PLpro and 3CLpro (FLBEPCs), suggesting that the bioavailability of these three compounds is high compared to others. Permeability glycoprotein (P-gp) is extensively expressed in the intestinal epithelium, liver cells, proximal tubular cells of the kidney and capillary endothelial cells comprising the blood–brain barrier and blood-testis barrier, where it pumps xenobiotics back into the intestinal lumen, bile ducts, urine-conducting ducts and capillaries respectively [
57]. From the results, licorice, ugonin M, baicalin, apigetrin and tomentin A were substrates for P-gp, suggesting that their absorption into the earlier mentioned tissues will be low, thereby affecting their bioavailability and increasing their excretion, which in turn, will shorten their half-lives. SARS-CoV-2 has been reported to infect the brain, thus indicating its ability to cross the blood brain barrier (BBB) [
58]. Of all the FLBEPCs, only abyssinone II has the potential of crossing the BBB. Thus, it may be able to clear the viral load in the brain. Lipinski’s rule of five has always been used to evaluate the drug-likeness of any compound; the more the violation of the rule, the less the drug-likeness of the compound [
40]. Of all the FLBEPCs, only isonymphaeol B, ugonin M, silymarin, abyssinone II, apigetrin and tomentin A did not violate any of the five rules, thus suggesting that they had higher drug-likeness compared to others.
The toxicities of the FLBEPCs were also evaluated in silico. hERG channel plays a vital role in the repolarization and termination stages of action potential in cardiac cells [
59,
60]. hERG channel blockers cause cardiotoxicity [
61]. The potentials of these compounds as hERG channel blockers were evaluated. The results revealed that only ugonin M, isonymphaeol B, apigetrin and tomentin A did not exhibit the potential of being hERG channel blockers, suggesting that they may not cause hERG channel-related cardiotoxicity. Also, the mutagenicities of the FLBEPCs were evaluated in silico. The results revealed that gallocatechin gallate, baicalin and apigetrin exhibited mutagenicity in silico. Thus, they are probable mutagens, which can cause genetic mutations, which, in turn, may initiate the pathophysiology of other diseases, such as cancer. The liver is exposed to higher concentration of drugs, being the primary organ responsible for drug metabolism. In this study the effects of the compounds on the liver were evaluated. The results indicated that only ugonin M, abyssinone II and apigetrin were hepatotoxic. Of all the FLBEPCs, only ugonin M, silymarin, isonymphaeol B and abyssinone II exhibited the potential to be inhibitors of different variants of cytochrome P450, thus they may adversely affect phase I drug metabolism in the liver. Thus, of all the flavonoids studied, isonymphaeol B may be predicted as the most effective inhibitor of 3CLpro with favourable pharmacokinetic parameters and no toxicity while procyanidin may be predicted as the most effective inhibitor of PLpro with less favourable pharmacokinetic parameters, drug-likeness and low toxicity. However, the low drugability of procyanidin (violating 3 out of the Lipinski’s rule of five) suggests that further chemical modification of the structure of the compound is required in order to increase this parameter. Based on this, silymarin may be a better alternative, though it is more toxic than procyanidin.
The MDS was performed on 3CLpro and PLpro in apo form and in complex with isonymphaeol B and silymarin, respectively for 100 ns in NVT ensemble. Furthermore, the SASA, RMSD, RoG, and RMSF were calculated from the trajectory of MD simulation. The RMSD parameter demonstrate insight into the structural conformations of proteins through molecular dynamics simulation [
62]. Through this parameters, the stability of protein backbone can be analysed when bound with a ligand or small molecule. Low values during RMSD runs indicates high stability of protein ligand system, while high values during RMSD refers to comparatively low stability of the system [
62]. While lower RMSD values are considered ideally acceptable for protein systems [
63]. The 3CLpro-Isonymphaeol B and PLpro-Silymarin complexes demonstrated appreciable degree of stability throughout the period of the 100 ns MDS run. For each of the representative conformers for the selected clusters from the clustering analysis, it was observed that the interactions were maintained at different time frames compared to the initial interactions, indicating that the interaction can be maintained in a dynamic environment, thus can be well adapted for experimental procedures.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.