In silico studies of anti-oxidative and hot temperament-based phytochemicals as natural inhibitors of SARS-CoV-2 Mpro

Ramin Naderi Beni; Parisa Elyasi-Ebli; Sajjad Gharaghani; Arefeh Seyedarabi

doi:10.1371/journal.pone.0295014

Abstract

Main protease (M^pro) of SARS-CoV-2 is considered one of the key targets due to its role in viral replication. The use of traditional phytochemicals is an important part of complementary/alternative medicine, which also accompany the concept of temperament, where it has been shown that hot medicines cure cold and cold medicines cure hot, with cold and hot pattern being associated with oxidative and anti-oxidative properties in medicine, respectively. Molecular docking in this study has demonstrated that a number of anti-oxidative and hot temperament-based phytochemicals have high binding affinities to SARS-CoV-2 M^pro, both in the monomeric and dimeric deposited states of the protein. The highest ranking phytochemicals identified in this study included savinin, betulinic acid and curcumin. Complexes of savinin, betulinic acid, curcumin as well as Nirmatrelvir (the only approved inhibitor, used for comparison) bound to SARS-CoV-2 M^pro were further subjected to molecular dynamics simulations. Subsequently, RMSD, RMSF, Rg, number of hydrogen bonds, binding free energies and residue contributions (using MM-PBSA) and buried surface area (BSA), were analysed. The computational results suggested high binding affinities of savinin, betulinic acid and curcumin to both the monomeric and dimeric deposited states of M^pro, while highlighting the lower binding energy of betulinic acid in comparison with savinin and curcumin and even Nirmatrelvir, leading to a greater stability of the betulinic acid-SARS-CoV-2 M^pro complex. Overall, based on the increasing mutation rate in the spike protein and the fact that the SARS-CoV-2 M^pro remains highly conserved, this study provides an insight into the use of phytochemicals against COVID-19 and other coronavirus diseases.

Citation: Naderi Beni R, Elyasi-Ebli P, Gharaghani S, Seyedarabi A (2023) In silico studies of anti-oxidative and hot temperament-based phytochemicals as natural inhibitors of SARS-CoV-2 M^pro. PLoS ONE 18(11): e0295014. https://doi.org/10.1371/journal.pone.0295014

Editor: Ahmed A. Al-Karmalawy, HUE: Horus University, EGYPT

Received: March 14, 2023; Accepted: November 13, 2023; Published: November 30, 2023

Copyright: © 2023 Naderi Beni et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

As stated by the World Health Organization (WHO), viral diseases display serious threats to the global health. There have been numerous viral pandemics over the past two decades. One example has been the Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV) in 2002 to 2003 in China and Hong Kong. Later in 2009, the world was challenged by the H1N1 influenza and afterwards in 2012, a new respiratory syndrome coronavirus, the Middle East Respiratory Syndrome Coronavirus (MERS-CoV), appeared in the middle east, Saudi Arabia [1]. In 2019, a new respiratory syndrome coronavirus, called the 2019 novel coronavirus, appeared initially in China and then spread all over the world. The International Committee on Taxonomy of Viruses (ICTV) named the virus Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and the WHO officially announced a new name for the disease; coronavirus disease 2019 (COVID-19) [2,3]. COVID-19 has since become a global pandemic and up until mid November 2023, over 697,000,000 confirmed infected cases, and over 6,900,000 million deaths have been reported, with over 669,000,000 recovered cases [4]. The number of confirmed cases and deaths reported are still on the rise.

SARS-CoV-2 is an enveloped, positive-sense, single-stranded RNA virus and the length of the genome is around 30 Kb. The replication of SARS-CoV-2 depends on polyproteins pp1a and pp1ab, which are encoded by about two-thirds of the 5’ end of the genome. These polyproteins are later processed into 16 non-structural proteins (NSPs), by two viral proteases, the main protease (M^pro, also called the 3C like protease, 3CLpro) and the papain like protease (PLpro) [5–7]. The M^pro, therefore, is one the proteins that has the potential of being a key target for antiviral drug development, since it is essential in the infection process and viral replication of SARS-CoV-2 [8,9].

SARS-CoV-2 M^pro consists of a dimer. Each monomer has three domains, with domains I & II consisting of residues 8−101 and 102−184, respectively, forming an antiparallel β-barrel structure, while domain III, consists of residues 201−303, mainly composed of α-helices. SARS-CoV-2 M^pro includes a Cys145-His41 catalytic dyad, and a cleft between domains I and II, where the inhibitor/substrate binding site is located. Each N-terminus residue, or N-finger, that is stuck between domains II and III of the monomer and domain II of the other monomer, plays a crucial role in forming an active site of M^pro. In comparison with the monomer, which is principally inactive, the M^pro dimer is extremely active [10–12].

There has been growing evidence that natural plants/herbs have been used in the form of teas, topical ointments and dietary supplements to relieve respiratory ailments including cough and bronchospasm [13–15]. In addition, there is a growing research interest on natural products with anti-inflammatory and anti-oxidative abilities to prevent and clarify the pathogenesis of various diseases that involve chronic inflammatory responses by the host to self-protect and remove the harmful effects such as those brought about by pathogens [16]. Interestingly, a comprehensive review published in 2018 [17], reported that traditional herbal medicines are important part of complementary/alternative medicine and accompany the concept of ‘temperament’, which is of a great deal of importance not only in the body but also in disease, foods and drugs. Accordingly, it was stated that the temperament is divided into two main categories, the hot and cold. The imbalance of temperament may it be body, medicine or food, can therefore cause disease [18,19]. Even more interesting, some studies refer to cold and hot pattern in terms of oxidative and anti-oxidative properties in medicine, respectively [20]. Traditional Chinese medicine, states that hot medicines are supposed to cure cold and cold medicines are supposed to cure hot diseases [20–23]. A study in 2014 [24], revealed that food with hot (e.g. thyme) and cold (e.g. sumac) properties affect people of different temperaments. Another study showed the effect of medicine with hot or cold properties to have varying effects on patients with different temperaments [19]. Interestingly, the amount of water content decreases with the hot temperament [25]. Additionally, it has been reported that the coldness (and wetness) of the brain i.e. having a cold temperament, was associated with the severity of the COVID-19 infection and hence regarded as a factor to be considered in preventing the disease or in predicting the course of the disease [26].

The initial step taken in this study was to assess M^pro structural conservation using a number of known ligand-bound coronovirus structures. This was then followed by analysis of five PBD structures of SARS-CoV-2 M^pro itself, consisting of 6lu7, 6y2f, 6wtt, 6m0k and 6lze, with promising IC₅₀ and EC₅₀ values, for further emphasis on structural conservation and identification of SARS-CoV-2 M^pro residues involved in binding interactions with ligands/inhibitors. The reliability of molecular docking versus experimentally derived structural information was assessed, followed by evaluation of six antiviral drugs also used for COVID-19 including Remdesivir, Darunavir, Oseltamivir, Lopinavir, Ritonavir, Ribavirin and in particular Nirmatrelvir, as an approved inhibitor [27]. Subsequently, 38 compounds from plants/herbs/trees were selected in this study, with emphasis on their therapeutics properties and antiviral effects, and assessed via molecular docking. The phytochemicals with the highest binding affinities were then subjected to computational studies including molecular dynamics (MD) simulation analysis involving calculations of root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg) and number of hydrogen bonds, binding free energy and residue contribution calculations by using molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) and buried surface area (BSA) analysis. The study compared the monomeric (inactive) versus dimeric (active) states of the SARS-CoV-2 M^pro deposited structure, to assess the binding interactions of the selected phytochemicals, using Nirmatrelvir as a reference. The results highlighted the presence of certain phytochemicals, with anti-oxidantive and hot-temperament properties, as potential inhibitors of SARS-CoV-2 M^pro.

Results

Bioinformatics analyses of coronavirus M^pro structures with bound ligands/inhibitors

In this study, initially the fasta format of the SARS-CoV-2 M^pro sequence was taken and a protein blast search performed in ncbi against all known coronavirus M^pro PDB structures, in order to assess M^pro structural conservation. The outcome of the search was assessed and the non-mutated, native M^pro structures with bound inhibitors were selected and assessed by multiple sequence alignment (S1 Fig in S1 File). As seen in S1 Table in S1 File, the sequence identity of various coronavirus M^pro compared to SARS-CoV-2 M^pro ranged from the highest at 96% for SARS-CoV to about 41% for human coronavirus.

The ligand binding mode in each structure was then analysed by Ligplot (S2 Fig in S1 File) in order to identify the residues and type(s) of binding interactions involved (whether covalent, hydrogen bonding or hydrophobic interaction), also shown in S1 Table in S1 File. The Ligplot analysis showed that Cys145 was involved in covalent bonding in most of the ligand bound SARS- and MERS-related M^pro structures and even in gastroenteritis virus and human coronavirus M^pro structures, except for PDB IDs 4mds, 2op9, 2qiq (SARS-related M^pro) and PDB ID 4ylu (MERS-related M^pro), representing non-covalent inhibitors. Subsequently, the coronavirus structures were superposed (using Chimera) showing the presence of a common inhibitor binding site (Fig 1). As reported in S1 Table in S1 File, common residues involved in ligand binding via hydrogen bonding included the following residues (or their equivalent in terms of residue position): Phe140, Gly143, His163, His164, Glu166 and Gln189. Residue Cys145 (or its equivalent) was also seen to be involved in hydrogen bonding with ligands, as well as His41, which was shown to be involved in hydrogen bonding to ligands in some of the structures. The common residues involved in hydrophobic interactions included but was not limited to Thr25/Thr26, His41, Met49, Leu141, Asn142, Met165, Asp187 and Arg188. Residues such as Phe140, Gln189, and Cys145 were also involved in hydrophobic interactions in some of the M^pro ligand-bound structures. Additionally, it was clear that the ligand-bound M^pro structures most closely related to the SARS-CoV-2 M^pro sequence, had 7–8 hydrogen bondings and 14–16 hydrophobic interactions as well as a covalent bonding involving the Cys145 residue. It should be added here that SARS-CoV-2 M^pro has 12 cysteines occupying positions 16, 22, 38, 44, 85, 117, 128, 145, 156, 160, 265 and 300. Out of those, cysteines at positions 16, 38, 117 and 145 were identified to be conserved (S3 Fig in S1 File), amongst the coronavirus sequences analysed and aligned in this study (S1 Fig in S1 File).

Download:

Fig 1. Ribbon representation of coronavirus M^pro structures in complex with ligands/inhibitors superposed, showing the main ligand binding site, using SARS-CoV-2 sequence as the query.

The main ligands/inhibitors are circled in red. The water molecules are shown as small red balls. The figure was generated in Chimera. PDB information about the M^pro structures used for superposition in this figure are given in S1 Table in S1 File.

https://doi.org/10.1371/journal.pone.0295014.g001

Comparison between experimentally derived structures of five synthetic covalent inhibitors bound to SARS-CoV-2 M^pro and that of molecular docking results

Five experimentally derived structures of SARS-CoV-2 M^pro with PDB IDs 6lu7, 6y2f, 6wtt, 6m0k and 6lze and corresponding synthetic covalent inhibitors, N3, alpha-ketoamide inhibitor (13b), GC-376, and peptidomimetic inhibitors 11a and 11b, respectively, were chosen in this study, based on their promising IC₅₀ and EC₅₀ values (S2 Table in S1 File).

As reported in S2 Table in S1 File, common residues involved in ligand binding via hydrogen bonding in all of the five structures included Phe140, Gly143 (except for PDB ID 6wtt), His163, His164 (except for PDB ID 6lu7) and Glu166. Residue Cys145 was also seen to be involved in hydrogen bonding in all structures as well as associated with all the inhibitors via covalent bonding. The most potent of inhibitors with an EC₅₀ value of 0.53 ± 0.01 μM was related to PDB ID 6lze, where the following residues including His41, Met49, Tyr54, Leu141, Asn142, Ser144, Met165, His172, Asp187, Arg188 and Gln189, were involved in hydrophobic interactions with the ligand.

For the molecular docking analysis, the ligands from the five structures were taken out of the halo structures (which were all deposited in the monomeric state of SARS-CoV-2 M^pro in asymmetric unit, except for PDB ID 6wtt) and then docked using DockThor with the apo/unbound SARS-CoV-2 M^pro structure (PDB ID 6yb7; containing SARS-CoV-2 M^pro in the monomeric state of the protein in the asymmetric unit), and results presented in S3 Table in S1 File. N3 had the highest binding affinity to M^pro (-8.636 Kcal/mol) based on molecular docking analysis in this study. A comparative analysis of the results from the molecular docking and that of the actual solved structures revealed that residues involved in ligand binding in both experimentally solved structures and molecular docking results included His41, Asn142, Gly143, Cys145, Met165, Glu166 and Gln189. Covalent binding of Cys145 was not detected in the molecular docking results, but instead Cys145 was observed to be involved in hydrophobic interactions in all structures except for the M^pro-alpha-ketoamide inhibitor (13b) complex, where Cys145 was not involved in the binding interaction at all (S3 Table in S1 File), contrary to the experimental data.

Molecular docking of six antiviral drugs also used for COVID-19

Molecular docking analyses of six drugs used for treatment of COVID-19, targeting the SARS-CoV-2 M^pro, including Remdesivir [28,29], Darunavir [30], Oseltamivir [29], Lopinavir [30,31], Ritonavir [29,30] and Ribavirin [32], were also performed using the DockThor program. The results revealed that the highest binding affinity was -8.894 Kcal/mol for Remdesivir followed by -8.399 Kcal/mol for Darunavir (S4 Table in S1 File). Remdesivir was found to bind to SARS-CoV-2 M^pro via hydrogen bonding to Glu166 and 9 hydrophobic interactions involving His41, Cys44, Phe140, Leu141, Asn142, Cys145, His163, Pro168 and Gln189. The binding interaction of Remdesivir through molecular docking seemed quite plausible since it involved common residues that have been identified in experimentally derived structures of potent inhibitors such as that of the peptidomimetic inhibitor(11a)-bound to SARS-CoV-2 M^pro (with the lowest EC₅₀ value of 0.53 ± 0.01 μM; S2 Table in S1 File), including common residues His41, Phe140, Leu141, Asn142, Cys145, His163, Glu166 and Gln189.

Molecular docking of 38 phytochemicals with SARS-CoV-2 M^pro

The DockThor program was further used to evaluate the binding of 38 active constituents/phytochemicals from the following herbs/plants/trees including ginger, turmeric, elderberry, cinnamon, garlic, onion, black cumin, peppermint, broccoli, dill, black pepper, Acanthopanax henryi plant, Boswellia plant, Birch and Camphor tree (S5 and S6 Tables in S1 File), with the SARS-CoV-2 M^pro. The active constituents/phytochemicals of the herbs/plants/trees were selected based on their anti-oxidative properties and antiviral effectiveness and benefits (using almost 300 supporting references; please see S5 and S6 Tables in S1 File).

Looking at the deposited structures of SARS-CoV-2 M^pro, most had reported the monomeric state (i.e. the asymmetric unit of the crystal structures mainly contained a monomer of the SARS-CoV-2 M^pro), with the homodimeric (dimeric) state deposited in some structures, which actually represents the biologically active form of SARS-CoV-2 M^pro. Therefore, in this study, the apo/unbound SARS-CoV-2 M^pro structures in the monomeric and dimeric states, with PDB IDs 6yb7 and 7ali, respectively, were both used for molecular docking. The emphasis on the DockThor outputs and the ranking of the compounds was based on affinity prediction values (Kcal/mol) of different ligands in virtual screening experiments considering the top-energy pose (according to Total Energy) of each compound, in both the monomeric and dimeric states of SARS-CoV-2 M^pro (Table 1A and 1B and S7 and S8 Figs in S1 File).

Download:

Table 1. Molecular docking results of top ranking compounds savinin, betulinic acid, and curcumin with (A) SARS-CoV-2 M^pro monomer and (B) SARS-CoV-2 M^pro dimer, respectively, based on affinity of binding.

The top-energy pose (according to Total Energy) of each compound is presented (with an RMSD of 0.0). The DockThor molecular docking program was used for analysis of interactions between SARS-CoV-2 M^pro monomer and dimer and the three compounds, using default settings. Ligplot was used to reveal the residues involved in binding and the mode of interaction with the ligands. The SARS-CoV-2 M^pro monomer and SARS-CoV-2 M^pro dimer (chain A) used in this study were from PDB IDs 6yb7 and 7ali, respectively.

https://doi.org/10.1371/journal.pone.0295014.t001

Molecular docking using PDB ID 6y7b, containing the deposited monomeric state of SARS-CoV-2 M^pro in the asymmetric unit

The results of docking using PDB ID 6yb7, containing the monomeric state of SARS-CoV-2 M^pro in the asymmetric unit, showed that the phytochemical savinin from the Acanthopanax henryi plant and betulinic acid from birch tree had the highest two scores in terms of affinity (-8.864 and -8.644 Kcal/mol, respectively), followed by curcumin from turmeric (-8.373 Kcal/mol) (Table 1A and Figs 2 and 3). Referring to the savinin-SARS-CoV-2 M^pro molecular docking results analysed by LigPlot (Fig 3A) and Chimera’s hydrophobic and electrostatic interaction analyses (Fig 3B and 3C, respectively), savinin was bound to SARS-CoV-2 M^pro through two hydrogen bondings involving Gly143 and His163, and 11 hydrophobic interactions involving residues Thr25, His41, Met49, Phe140, Leu141, Asn142, Ser144, Cys145, Met165, Glu166 and Gln189. This seems a plausible interaction as savinin and the most potent synthetic SARS-CoV-2 inhibitor listed in S2 Table in S1 File, interacted with the following common residues (as shown through experimental data) including His41, Met49, Phe140, Leu141, Asn142, Gly143, Cys145, His163, Met165, Glu166 and Gln189. Additionally, the drugs mentioned in S4 Table in S1 File, and in particular Remdesivir, interacted with common residues similar to savinin, including His41, Phe140, Leu141, Asn142, Cys145, His163; Glu166 and Gln189. Additionally, according to a previous in silico docking study [33], where savinin was shown to bind to SARS-CoV-2 M^pro with a binding affinity of -8.3 kcal/mol (weaker binding affinity than that reported in this study), His163 (in addition to Ser144) was involved via hydrogen bonding (Gly143 was not reported in that study). Furthermore, similar residues including those identified in this study were involved in hydrophobic interactions with savinin, including His41, Phe140, Cys145, Met165 and Glu166 [33]. Out of the residues involved in binding interaction with savinin in our study, His41, Phe140, Gly143, Cys145, His163 and Glu166 were conserved amongst the native M^pro sequences analysed in this study (S1 Fig in S1 File).

Download:

Fig 2. Hydrophobic surface representation of the highest scoring herbal compound-SARS-CoV-2 M^pro monomer complexes.

(a-c) Savinin, betulinic acid and curcumin, respectively, in complex with SARS-CoV-2 M^pro monomer using DockThor. The phytochemicals are shown in ball and stick representation. The figure was generated using Chimera.

https://doi.org/10.1371/journal.pone.0295014.g002

Download:

Fig 3. Structural analysis of savinin-, betulinic acid- and curcumin-SARS-CoV-2 M^pro monomer complexes as revealed by molecular docking, with the highest scores in terms of affinity.

(a, d and g) Ligplot analyses of the savinin-, betulinic acid- and curcumin-SARS-CoV-2 M^pro monomer complexes, respectively, showing M^pro residues involved in hydrogen bonding and hydrophobic interactions. (b, e and h) Hydrophobic representation of the savinin-, betulinic acid- and curcumin-SARS-CoV-2 M^pro monomer complexes, respectively. (c, f and i) Electrostatic potential representation of the savinin-, betulinic acid- and curcumin-SARS-CoV-2 M^pro monomer complexes, respectively. The residues involved in interaction with savinin, betulinic acid and curcumin are labelled. The figure was generated using the Ligplot and Chimera programs.

https://doi.org/10.1371/journal.pone.0295014.g003

As for the binding interaction of betulinic acid to SARS-CoV-2 M^pro analysed by LigPlot and Chimera (Fig 3D–3F), Cys44 was involved in hydrogen bonding and a total of 10 residues were involved in hydrophobic interactions including Thr24, Thr25, Thr26, Thr45, Ser46, Met49, Asn142, Gly143, Glu166 and Glu189. According to a previous in silico docking study [34], betulinic acid was found to inhibit SARS-CoV-2 M^pro with an IC₅₀ value of 14.55 as well as a binding energy of −8.1 kcal/mol, and had exhibited hydrogen bonding with His41 and Phe140 (this is different to what we have seen in this current study). Out of the residues involved in binding interaction with betulinic acid, Cys44 and Glu166 were the two residues conserved amongst the native M^pro sequences analysed in this study (S1 Fig in S1 File).

Additionally, the analysis of binding interactions of curcumin to SARS-CoV-2 M^pro by LigPlot and Chimera (Fig 3G–3I), revealed that Thr26 was involved in hydrogen bonding and a total of 10 residues were involved in hydrophobic interactions including Thr25, His41, Met49, His164, Met165, Glu166, Pro168, Gln189, Thr190 and Gln192. Out of the residues involved in binding interaction with curcumin, His41, Glu166 and Gln192 were the three residues conserved amongst the native M^pro sequences analysed in this study (S1 Fig in S1 File).

Molecular docking using PDB ID 7ali, containing the dimeric and biologically active form of SARS-CoV-2 M^pro

The results of molecular docking using PDB ID 7ali, containing the dimeric state of SARS-CoV-2 M^pro, showed that in contrary to the monomeric state of M^pro where savinin, betulinic acid and curcumin had the highest consecutive scores in terms of affinity, instead curcumin ranked first with the highest score in terms of affinity, followed by betulinic acid and savinin (-8.554, -8.548 and -8.518 Kcal/mol, respectively) (Table 1B and Figs 4 and 5). Having said that, the binding affinities were very close to each other.

Download:

Fig 4. Hydrophobic surface representation of the highest scoring herbal ligand-SARS-CoV-2 M^pro dimer complexes.

(a-c) Curcumin, betulinic acid and savinin, respectively, in complex with SARs-CoV-2 M^pro dimer using DockThor. The herbal ligands are shown in ball and stick representation. The figure was generated using Chimera.

https://doi.org/10.1371/journal.pone.0295014.g004

Download:

Fig 5. Structural analysis of the curcumin- and betulinic acid- and savinin-SARS-CoV-2 M^pro dimer complexes as revealed by molecular docking, with the highest scores in terms of affinity.

(a, d and g) Ligplot analyses of the curcumin- and betulinic acid- and savinin-SARS-CoV-2 M^pro dimer complexes, respectively, showing M^pro residues involved in hydrogen bonding and hydrophobic interactions. (b, e and h) Hydrophobic representation of the curcumin-, betulinic acid- and savinin-SARS-CoV-2 M^pro dimer complexes, respectively. (c, f and i) Electrostatic potential representation of curcumin-, betulinic acid- and savinin-SARS-CoV-2 M^pro dimer complexes, respectively. The residues involved in interaction with either curcumin, betulinic acid or savinin are labelled. The figure was generated using the Ligplot and Chimera programs.

https://doi.org/10.1371/journal.pone.0295014.g005

Curcumin was bound to SARS-CoV-2 M^pro through a hydrogen bonding with Thr26 and 9 hydrophobic interactions involving residues Thr25, Met49, Tyr118, Asn119, Asn142, Gly143, His164, Met165 and Gln189 (Fig 5A–5C). Compared to curcumin bound to the monomeric state of SARS-CoV-2 M^pro, the common residues included Thr26 (via hydrogen bonding), Thr25, Met49, Asn142, His164, Met165 and Gln189. Residue Cys145 was not seen in the binding interaction with curcumin, neither in the monomeric nor the dimeric states of SARS-CoV-2 M^pro.

There were no hydrogen bondings for docking results of betulinic acid and savinin bound to SARS-CoV-2 M^pro in the dimeric state. Both ligands appeared to bind via 9 and 10 hydrophobic interactions, respectively. Residues involved in binding to betulinic acid included Thr24, Thr25, Thr26, His41, Ser46, Met49, Asn119, Asn142 and Gly143 (Fig 5D–5F). Compared to the docking results using the monomeric state of SARS-CoV-2 M^pro, the common residues involved in hydrophobic interactions included Thr24, Thr25, Thr26, Ser46, Met49, Asn142 and Gly143. Consistent with curcumin-bound to SARS-CoV-2 M^pro, Cys145 was not involved in binding to betulinic acid, neither in the monomeric nor dimeric states of SARS-CoV-2 M^pro. As for savinin, the following residues were involved in the hydrophobic interactions: Thr25, Cys44, Met49, Leu141, Asn142, Ser144, Cys145, His163, Glu166 and Gln189 (Fig 5G–5I). Compared to the molecular docking results using the monomeric state of SARS-CoV-2 M^pro, the common residues involved in hydrophobic interactions included Thr25, Met49, Asn142, Ser144, Cys145, Glu166 and Gln189. Residue Cys145 was present in hydrophobic bonding interactions to savinin in both the monomeric and dimeric states of SARS-CoV-2 M^pro.

Referring to S9 Table in S1 File, all of the three phytochemicals including savinin, betulinic acid and curcumin, with the highest scores in terms of binding affinity, using both the monomeric and dimeric states of SARS-CoV-2 M^pro, were also listed to have antioxidant/hot-temperament based properties, as well as antiviral effects; savinin and betulinic acid, in particular, were reported to have antiviral effects against SARS-CoV-2 [33,34].

Re-docking of savinin, betulinic acid and curcumin

As a means of gaining more confidence in our results from molecular docking, re-docking of the best pose of savinin, betulinic acid and curcumin with SARS-CoV-2 M^pro were performed in DockThor and results revealed the following RMSD values including 0.233 Å, 0.041 Å and 0.433 Å, respectively. As can be clearly observed, the betulinic acid pose for binding to SARS-CoV-2 M^pro had the lowest and hence the best re-docking RMSD value, followed by savinin and curcumin.

Covalent versus non-covalent SARS-CoV-2 M^pro inhibitors

As stated previously, SARS-CoV-2 M^pro includes a Cys145-His41 catalytic dyad (where the ionization of sulfhydryl group of Cys145 and imidazole group of His41 play an important role in the catalytic processing of substrate hydrolysis), and a cleft between domains I and II, where the inhibitor/substrate binding site is located. His41 and Cys145 were shown to be involved in savinin binding to the SARS-CoV-2 M^pro (in the monomeric state) and helps to further recommend savinin as an effective covalent inhibitor. Furthermore, although DockThor failed to reveal the covalent binding of Cys145 to the ligands of confirmed covalent inhibitors, there is a high probability that Cys145 is involved in covalent interaction with savinin. Having said that, non-covalent inhibitors also exist such as that seen in PDB ID 7l0d, with the inhibitor ML188 bound to SARS-CoV-2 M^pro (deposited in the monomeric state). Non-covalent inhibitors, like ML188, may actually be more potent than covalent inhibitors [35]. Therefore, it is not necessary to have a covalent interaction between Cys145 and the inhibitor in order to have a strong inhibition against SARS-CoV-2 M^pro. S10 Table in S1 File shows that the non-covalent inhibitor ML188 binds to SARS-CoV-2 M^pro via four hydrogen bondings involving Asn142, Gly143 (with two hydrogen bondings) and His163. Additionally, 14 residues are involved in hydrophobic interactions including Thr25, Thr26, Leu27, His41, Met49, Leu141, Phe140, Ser144, Cys145, His164, Glu166, Asp187, Arg188 and Gln189. As can be clearly seen, both the His41 and Cys145 residues are involved in the binding but via hydrophobic interactions. Molecular docking of ML188 with PDB ID 6yb7 (monomeric state of SARS-CoV-2 M^pro, same as that of the experimentally observed deposited state of SARS-CoV-2 M^pro in PDB ID 7l0d), for comparison, revealed a high binding affinity of -8.883 Kcal/mol (S11 Table in S1 File), showing two hydrogen bondings involving His163 and Glu166 and 9 hydrophobic interactions involving residues including His41, Ser46, Phe140, Asn142, Gly143, Cys145, His164, Met165 and Gln189. The common residues between the experimentally derived structure of the ML188-SARS-CoV-2 M^pro complex and that the of the molecular docking result, included His41, Asn142, Gly143, His163, Cys145, His146 and Gln189 (through either hydrogen bonding/hydrophobic interactions).

Nirmatrelvir as an approved specific inhibitor of SARS-CoV-2 M^pro

To make the results of this study more meaningful, the structure of Nirmatrelvir bound to SARS-CoV-2 M^pro (PDB ID 7te0), as an approved inhibitor of the only FDA approved drug against COVID-19, PAXLOVID, regarded as a combination therapy of nirmatrelvir and ritonavir, was used in this study [27].

Nirmatrelvir has been reported (S10 Table in S1 File) to bind to Cys145, His163, Glu166 and Gln192 through hydrogen bonding. Nirmatrelvir also binds to Cys145 via a covalent interaction. Additionally, 8 residues are involved in hydrophobic interactions with Nirmatrelvir including His41, Met49, Phe140, Asn142, Gly143, Met165, His172 and Gln189. Molecular docking of Nirmatrelvir with PDB ID 6yb7 (monomeric state of SARS-CoV-2 M^pro, same as that of the experimentally observed deposited state of SARS-CoV-2 M^pro in PDB ID 7te0), for comparison, revealed a binding affinity of -7.793 Kcal/mol (S11 Table in S1 File), showing a hydrogen bonding with Gly143 and 9 hydrophobic interactions involving residues Ser46, Phe140, Asn142, Cys145, His163, Glu166, Leu167, Pro168 and Gln189. The common residues between the experimentally derived structure of the Nirmatrelvir-SARS-CoV-2 M^pro complex and that the of the molecular docking result (using the monomeric state of SARS-CoV-2 M^pro), included residues Phe140, Gly143, Asn142, Cys145, His163, Glu166 and Gln189 (through either hydrogen bonding/hydrophobic interactions). Additional molecular docking using the dimeric state of SARS-CoV-2 M^pro (PDB ID 7ali) revealed a stronger binding affinity of -8.404 Kcal/mol and the involvement of Glu166 in hydrogen bonding as well as 11 residues involved in hydrophobic interactions, including Thr25, Thr26, Leu27, His41, Ser46, Met49, Asn119, Asn142, Gly143, Cys145 and Gln189 (S11 Table in S1 File). The common residues between the experimentally derived structure of the Nirmatrelvir-SARS-CoV-2 M^pro complex and that the of the molecular docking result (using the dimeric state of SARS-CoV-2 M^pro), included residues His41, Met49, Asn142, Gly143, Cys145, Glu166 and Gln189 (through either hydrogen bonding/hydrophobic interactions).

ADMET analysis

The four ligands including savinin, betulinic acid, curcumin and Nirmatrelvir (as the reference ligand in this study) were all evaluated using the ADMETlab2.0 online program [36]. A summary of some of the calculated parameters for each ligand are given in S12 Table in S1 File. Three rules under the category of medicinal chemistry, described as Lipinski, Pfizer and GSK rules, were assessed in addition to the parameters shown in S12 Table in S1 File. Nirmatrelvir, satisfied many of the factors relating to the analysis in addition to the Lipinski and Pfizer rules. Curcumin satisfied all the three rules and had the highest favourable calculated values for the parameters evaluated compared to savinin and betulinic acid (S12 Table in S1 File). Savinin was accepted by the Lipinski and GSK rules, while betulinic acid was accepted by the Lipinski rule only. The golden triangle (where a molecular weight of between 50 or less and 200 or more, with a logD of less than or equal to 5 indicated a more favourable ADMET profile) revealed that only Nirmatrelvir, savinin and curcumin satisfied the criteria.

Validation of the molecular docking results

To validate the results of molecular docking, MD simulation analyses including calculations of RMSD, RMSF, Rg and number of hydrogen bonds,BSA analysis and MM-PBSA, were performed for savinin-, betulinic acid- and curcumin-SARS-CoV-2 M^pro complexes as well as Nirmatrelvir-SARS-CoV-2 M^pro complex, as a means to make a comparison with the only approved inhibitor of SARS-CoV-2 M^pro. The validation was performed using both the monomeric and dimeric states of SARS-CoV-2 M^pro, both in the apo and halo forms.

MD simulation analysis of the complexes

MD simulation, a method to investigate the movements of atoms and molecules over a given period of time, was performed to discover the stability and intermolecular interactions of the phytochemicals and Nirmatrelvir with the SARS-CoV-2 M^pro (in both the monomeric and dimeric states).

RMSD anlaysis

RMSD, an indicator of the stability of the structures during simulation, was calculated for the backbone atoms of apo- and ligand bound-SARS-CoV-2 M^pro structures. The RMSD values of the ligands from the ligand-monomeric SARS-CoV-2 M^pro complexes converged over a range of 0.10–0.45 nm (Fig 6A), while, the RMSD of the ligands from the ligand-dimeric SARS-CoV-2 M^pro complexes converged over a range of 0.15–0.3 nm (Fig 6B). As seen in Fig 6A and 6B, the RMSD plots of the ligands through MD simulation of both monomeric and dimeric states of SARS-CoV-2 M^pro showed betulinic acid and curcumin with the lowest and highest RMSD values, respectively. The RMSD values of Nirmatrelvir bound to SARS-CoV-2 M^pro in the dimeric state exhibited higher RMSD values than curcumin. Overall, the presented results indicated the stability of the complexes and ligands during the MD simulations. Despite the stable RMSD values of the betulinic acid-, savinin- and Nirmatrelvir-SARS-CoV-2 M^pro dimer complexes and the apo dimer, the curcumin-SARS-CoV-2 M^pro dimer experienced a 10 nm rise, at the mid-point of the 45 ns to 55 ns period (Fig 6D). As for the RMSD of monomeric complexes and the apo monomer, a greater rise was observed in the RMSD value for the betulinic-SARS-CoV-2 M^pro monomer complex (Fig 6C).

Download:

Fig 6. RMSD plots of ligands from both monomeric and dimeric states of SARS-CoV-2 M^pro and the RMSD plots of the apo and ligand-bound SARS-CoV-2 M^pro complexes in the dimeric state.

(A) RMSD of the ligands from the ligand-monomeric SARS-CoV-2 M^pro complexes. Betulinic acid (green) showed the highest stability followed by Nirmatrelvir (red), Savinin (yellow) and Curcumin (navy blue). (B) RMSD of the ligands from the ligand-dimeric SARS-CoV-2 M^pro complexes. Both chains A and B of Betulinic acid, Savinin, Curcumin and Nirmatrelvir are presented. (C) RMSD of the ligand-monomeric SARS-CoV-2 M^pro complexes. (D) RMSD of the ligand-dimeric SARS-CoV-2 M^pro complexes. Apo SARS-CoV-2 M^pro (green); Betulinic acid-SARS-CoV-2 M^pro (yellow), Savinin-SARS-CoV-2 M^pro (red), Nirmatrelvir-SARS-CoV-2 M^pro (dark blue) and curcumin-SARS-CoV-2 M^pro (light blue). Apo-SARS-CoV-2 M^pro (Apo), Betulininc acid-SARS-CoV-2 M^pro complex (BET), Savinin-SARS-CoV-2 M^pro complex (SAV), Curcumin-SARS-CoV-2 M^pro complex (CUR) and Nirmatrelvir-SARS-CoV-2 M^pro complex (NIR).

https://doi.org/10.1371/journal.pone.0295014.g006

Fig 7 represents the overlay of the ligand-dimeric SARS-CoV-2 M^pro complexes at 0, 50 and 100 ns intervals of MD simulations. The overlayed structures of the dimers revealed that the positions of the monomers remained consistent relative to each other.

Download:

Fig 7. Overlay of ligand-dimeric SARS-CoV-2 M^pro complexes at 0, 50 and 100 ns intervals of MD simulations.

(a) Savinin- SARS-CoV-2 M^pro complex, (b) Betulinic acid-SARS-CoV-2 M^pro complex, (c) Curcumin-SARS-CoV-2 M^pro complex and (d) Nirmatrelvir-SARS-CoV-2 M^pro at 0, 50 and 100 ns (in grey, pink and green, respectively) of the MD simulation.

https://doi.org/10.1371/journal.pone.0295014.g007

RMSF analysis

The RMSF analysis was used to investigate fluctuation in the residues during the simulation. The overall RMSF values for apo- and ligand-bound SARS-CoV-2 M^pro structures, in the dimeric state were relatively similar, however, Met49 showed lower fluctuation when stabilised by the ligands (Fig 8). RMSF values of betulinic acid was higher than other complexes denoting the higher fluctuation of the conformation. ΔRMSF of apo- and Nirmatrelvir-SARS-CoV-2 M^pro complex for Met49 was 0.495 nm, indicating the higher stability of this residue in the presence of Nirmatrelvir. ΔRMSF of Met49 in apo- and betulinic acid-SARS-CoV-2 M^pro complex was 0.344 nm. This suggests that Nirmatrelvir reduced the fluctuation of Met49, more than betulinic acid. Similarly, Tyr154 was highly unstable in the betulinic acid-SARS-CoV-2 M^pro complex, while other complexes experienced lower fluctuations. Overall, RMSF values indicated that all complexes had a similar behaviour during the MD simulation, in both the monomeric and dimeric states of SARS-CoV-2 M^pro (Fig 8A and 8B).

Download:

Fig 8. RMSF analysis of apo- and ligand-bound SARS-CoV-2 M^pro structures, in both the monomeric and dimeric states.

(A) RMSF values of apo- and ligand- bound SARS-CoV-2 M^pro structures, in the monomeric state. (B) RMSF values of apo- and ligand-bound SARS-CoV-2 M^pro structures, in the dimeric state. Chains A and B of SARS-CoV-2 M^pro in the dimeric state were analysed separately.

https://doi.org/10.1371/journal.pone.0295014.g008

Rg analysis

The Rg analysis measures the degree of structural compactness against time, where conformational stability of the protein maintains a steady value of Rg and misfolding increases the Rg. Fluctuation of the Rg values, between 2.56 and 2.63 nm, indicated that the conformation of all the apo- and ligand-SARS-CoV-2 M^pro complex structures, in the dimeric state, were reserved during the simulation (Fig 9). The curcumin-SARS-CoV-2 M^pro complex, in the dimeric state, had experienced a 0.1 nm rise, between 40 and 50 ns, which was consistent with the rise observed in RMSD values (Fig 6C). The apo- and ligand-SARS-CoV-2 M^pro complex structures, in the monomeric state, presented a lower fluctuation range, between 2 to 2.30 nm, confirming the conformational stability of the structures (Fig 9).

Download:

Fig 9. Radius of Gyration of apo- and ligand-bound SARS-CoV-2 M^pro structures, both in the monomeric and dimeric states.

Apo-SARS-CoV-2 M^pro (Mpro), Betulininc acid-SARS-CoV-2 M^pro complex (BET), Savinin-SARS-CoV-2 M^pro complex (SAV), Curcumin-SARS-CoV-2 M^pro complex (CUR) and Nirmatrelvir-SARS-CoV-2 M^pro complex (NIR). Dimer denotes the dimeric form of the SARS-CoV-2 M^pro protein, either in apo or halo forms.

https://doi.org/10.1371/journal.pone.0295014.g009

Number of hydrogen bonds

As shown in Figs 10 and 11, multiple hydrogen bondings were formed between both the monomeric and dimeric states of SARS-CoV-2 M^pro and the ligands, respectively. In MD simulation of betulinic acid-SARS-CoV-2 M^pro complex (using the monomeric state of SARS-CoV-2 M^pro), hydrogen bonds with Thr25, His41, Thr190 and Gln192 were observed (Fig 10A–10C). Additionally, the MD simulation of betulinic acid-SARS-CoV-2 M^pro using the dimeric state of SARS-CoV-2 M^pro suggested that His164 participated in the formation of a hydrogen bond and double hydrogen bonds were formed with Thr25 side chain and an additional hydrogen bond with Thr24 main chain (Fig 11B and 11C). The MD simulation results of savinin-SARS-CoV-2 M^pro complex in the monomeric state presented hydrogen bonding with side chains of Thr25, Thr26, Ser46, Asn142, Cys145 and main chain of Gln189 (Fig 10D–10F). The savinin-SARS-CoV-2 M^pro complex in the dimeric state revealed a hydrogen bonding between Glu166 and savinin (Fig 11A). In MD simulation of the curcumin-SARS-CoV-2 M^pro complex in the monomeric state, hydrogen bonds were observed between His41 (Fig 10G), main chain of Asp187, Gln189 and main chain of Thr26 (Fig 10H), and main chain of Gly143 and Cys145 (Fig 10I). In the curcumin-SARS-CoV-2 M^pro complex in the dimeric state, hydrogen bonds were formed between curcumin and residues Thr25, Thr26, Ser46, Cys145, Gln189 and Ser301 of the adjacent monomer (Fig 11D–11F). As for the MD results of Nirmatrelvir-monomeric SARS-CoV-2 M^pro complex, hydrogen bondings were formed between Nirmatrelvir and residues His41, Cys44 and Asn119 (Fig 10J–10L). Nirmatrelvir-dimeric SARS-CoV-2 M^pro complex exhibited hydrogen bonds formed between Nirmatrelvir and residues Thr26, Thr45, Ser46, Asn142 and Gln189, (Fig 11G–11I).

Download:

Fig 10. Hydrogen bonding analysis between the ligands and the monomeric state of SARS-CoV-2 M^pro as revealed by MD.

(a, b and c) Hydrogen bonds formed between betulinic acid and SARS-CoV-2 M^pro. ⁽d, e and f) Hydrogen bonds formed between savinin and SARS-CoV-2 M^pro. (g, h and i) Hydrogen bonds formed between curcumin and SARS-CoV-2 M^pro. (j, k and l) Hydrogen bonds formed between Nirmatrelvir and SARS-CoV-2 M^pro. (a) Thr25 and His41, (b) Thr190 and His41 and (c) Gln192 and His41 participated in hydrogen bonding with betulinic acid. (d) Ser46 and main chain of Thr26, (e) Thr25, Cys145 and main chain of Gln189 and (f) Asn142, Cys145 and main chain of Gln189 formed hydrogen bonded with savinin. (g) His41, (h) Gln189, Asp187, Thr26 and (i) Gly143 and Cys145 formed hydrogen bonds with curcumin. (j) Cys44, (k) Asn119 and (l) His41 hydrogen bonded with Nirmatrelvir.

https://doi.org/10.1371/journal.pone.0295014.g010

Download:

Fig 11. Hydrogen bonding analysis between the ligands and the dimeric state of SARS-CoV-2 M^pro as revealed by MD.

(a) Hydrogen bonding formed between savinin and Glu166 of SARS-CoV-2 M^pro. (b and c) Hydrogen bondings formed between betulinic acid and residues Thr25, Thr26 and His164 of SARS-CoV-2 M^pro. (d, e and f) Hydrogen bondings formed between curcumin and residues Thr26 and Gln189 of SARS-CoV-2 M^pro, as well as Ser301 of the adjacent monomer. (g, h and i) Hydrogen bondings formed between Nirmatrelvir and residues Thr25, Thr45, Ser46, Asn142 and Gln189 of SARS-CoV-2 M^pro.

https://doi.org/10.1371/journal.pone.0295014.g011

Binding free energies and residue contributions

To investigate the overall contributions of the residues to the binding energy of the ligands to the dimeric state of SARS-CoV-2 M^pro, binding free energy calculations were performed. Using g_mmpbsa package [37,38], Van der Waal, electrostatic, polar, solvent-accessible surface area (SASA) and the total binding free energies, were calculated for the dimeric complexes for 100 ns MD simulations (Table 2). The highest binding energy belonged to betulinic acid at -83.579 kJ/mol, while savinin had the lowest binding energy. Nirmatrelvir with a total binding energy of -60.521 kJ/mol, had the highest binding energy after betulinic acid. Curcumin, with a total binding energy of -50.588 kj/mol, higher than the binding energy for savinin (-48.284 kJ/mol), had a lower total binding energy compared to Nirmatrelvir. Despite the unfavoured effect of polar interactions for the binding of the ligands, hydrophobic non-polar interactions played the main role in ligand binding. These data suggested that betulinic acid might be a more effective inhibitor for SARS-CoV-2 M^pro.

Download:

Table 2. Summary of the MM/PBSA results of the ligand-bound SARS-CoV-2 M^pro dimeric complexes.

The values are given in kJ/mol.

https://doi.org/10.1371/journal.pone.0295014.t002

In order to inspect the contribution of residues in binding to ligands, contribution energies were also calculated for each residue of ligand-bound SARS-CoV-2 M^pro compelxes, in the dimeric state of the protein (Fig 12). As can be observed from the results, in the Nirmatrelvir-SARS-CoV-2 M^pro complex, the following residues contributed positively to the energy of binding including His41, Val42, Gln142, His164 and Asp187. As for the curcumin-SARS-CoV-2 M^pro complex, Thr26, Gln142, Gly143 and Gly302 (from the other monomer) contributed to the binding. His41, had overall destabilised the binding of curcumin despite the hydrogen bond formed (Fig 11D). This suggests a low formation rate of this hydrogen bond and probably a more disrupting role of this residue in the binding pocket occupied by curcumin. Fig 12 further shows that the contribution energy calculated for betulinic acid-SARS-CoV-2 M^pro complex involved residues Thr25, His41, Gln189, Thr190, and Gln192 and ensured the stability of the betulinic acid during the MD simulation, while Val186 and Arg188 destabilised the ligand binding. Even though Glu166 led to higher stability of curcumin, in complex with betulinic acid, it destabilised the interaction. Similarly, Thr25, Thr26, Ser46, Gln142, Cys145, His164, Glu166 and Gln189 contributed to the energy of savinin binding to SARS-CoV-2 M^pro.

Download:

Fig 12. Contribution energy of residues from the ligand-bound SARS-CoV-2 M^pro complexes, in the dimeric state of the protein.

The contribution energies of residues from SARS-CoV-2 M^pro in the betulinic acid-, Savinin-, Curcumin- and Nirmatrelvir-bound SARS-CoV-2 M^pro complexes are given. The positive values in kJ/mol indicates the negative contribution of the residue for ligand binding and the negative values represent the probable positive role of the residues in stabilisation of the ligand binding to the SARS-CoV-2 M^pro.

https://doi.org/10.1371/journal.pone.0295014.g012

Buried surface area analysis

Buried Surface Area (BSA) analysis was performed based on a previous study [39]. The overlayed structures of the dimers (Fig 7) revealed that the positions of the monomers remained consistent relative to each other. Based on the consistency of conformation, for the calculation of SASA, each ligand and its corresponding monomer were selected for this calculation. As depicted in Fig 13, throughout the MD simulation, savinin followed by curcumin had the lowest BSA, allowing more solvent molecules to access the binding pocket. Conversely, Nirmatrelvir exhibited the highest BSA, indicating a greater occupation of the binding site, which is consistent with the bulky structure of this molecule. Despite the smaller structure of betulinic acid in comparison with other ligands, it covered the binding pocket more than savinin and curcumin, which might be due to the hydrophobic interactions in the binding pocket. Furthermore, the absence of abrupt changes in the BSA values suggested that the ligands occupied the binding pocket steadily, as demonstrated by the RMSD plot of each ligand in the dimeric form (Fig 6D).

Download:

Fig 13. Buried Surface Area analysis of ligand-bound SARS-CoV-2 M^pro complexes, in the dimeric state of the protein.

This figure compared the BSA value of each ligand in relation to the chain it was bound to. Nirmatrelvir (NIR) exhibited the highest BSA values indicating more area covered by this ligand. Savinin (SAV) presented the lowest values and hence higher access of solvent molecules to the binding pocket. Betulinic acid (BET) and curcumin (CUR) stood in middle, with betulinic acid covering more buried surface area than curcumin.

https://doi.org/10.1371/journal.pone.0295014.g013

Conclusions

One of the essential proteins needed for the infection process and viral replication of SARS-CoV-2 is the 3C-like cysteine protease, also known as the Main protease (M^pro), with a potential of being a key target for antiviral drug development with a conserved substrate binding site.

Using molecular docking it was possible to evaluate the binding affinities of 38 phytochemicals with SARS-CoV-2 M^pro, in both the monomeric and dimeric states of the protein (as deposited asymmetric states of SARS-CoV-2 M^pro in the protein data bank). Three top scoring phytochemicals in terms of binding affinity, were identified to bind SARS-CoV-2 M^pro in the monomeric and dimeric states of the protein, respectively, including savinin from the Acanthopanax henryi plant (-8.864 and -8.518 Kcal/mol), betulinic acid from birch tree (-8.644 and -8.548 Kcal/mol) and curcumin from turmeric (-8.373 and -8.554 Kcal/mol). Savinin, betulinic acid and curcumin were reported to have therapeutic properties with antioxidant properties and hence hot temperaments [20], as well as antiviral effects [33,34]. To make the results more meaningful, the only approved inhibitor Nirmatrelvir [27] was used for comparison. Molecular docking was performed on Nirmatrelvir with SARS-CoV-2 M^pro, in both the monomeric and dimeric states of the protein with the following binding affinities of -7.798 and -8.404 kcal/mol, respectively. ADMET analysis was conducted on all four ligands, showing that curcumin followed by savinin and betulinic acid, could satisfy the calculated rules as drugs (in terms of favourable criteria), compared to Nirmatrelvir.

In order to validate the molecular docking results, MD simulation analyses including calculations of RMSD, RMSF, Rg and number of hydrogen bonds, MM-PBSA, and BSA analysis, were performed on apo and savinin-, betulinic acid-, curcumin-, and Nirmatrelvir-bound SARS-CoV-2 M^pro, in both the monomeric and dimeric states of the protein. The molecular behaviour and conformational dynamic differences between the monomeric and dimeric states of SARS-CoV-2 M^pro, were assessed. The results showed that the active site residues interacted with the ligands via hydrogen bonding. For example, savinin, curcumin and Nirmatrelvir formed hydrogen bonds with Gln189. Cys145 also hydrogen bonded with curcumin and savinin. It should be noted here, that while Cys145 is commonly seen to be associated in covalent interactions with inhibitors, potent non-covalent inhibitors such as ML188 also exist [35], where Cys145 is non-covalently involved through either hydrogen bonding or hydrophobic interactions giving an even higher affinity of binding as assessed by molecular docking in this study (-8.518 Kcal/mol).

Moreover, His41 interacted with betulinic acid, curcumin and Nirmatrelvir via hydrogen bonds. Out of the ligands analysed, only curcumin formed hydrogen bonding with a residues from the neighbouring monomer (Ser301), which may emphasise on the effect of certain ligands in the stabilisation of the dimeric/active state of SARS-CoV-2 M^pro. The MM-PBSA tool was used to calculate the binding energies and found that betulinic acid had the highest affinity for SARS-CoV-2 M^pro active site. Nirmatrelvir also had a high affinity but lower than betulinic acid. The analysis of the contribution energies of the residues for binding to the ligands revealed that Gln189 and His164 had positive effects. Buried surface area (BSA) analysis evaluated the ligand coverage in the binding pocket and showed that betulinic acid had a lower coverage than Nirmatrelvir, meaning that more water molecules could access the binding pocket. It was also concluded that hydrophobic non-polar interactions played the main role in ligand binding.

Overall, the findings from this study seem important, since the use of traditional medicine is advantageous over synthetic drugs, which can help overcome the toxicity and side effects of synthetic drugs. It is therefore highly recommended that experimental data for the binding of betulinic acid, savinin and curcumin bound to SARS-CoV-2 M^pro should be pursued. The study of phytochemicals against SARS-CoV-2 M^pro is most important at the time where there is still an ongoing number of infected cases and deaths reported due to increasing rate of mutations in the spike protein of SARS-CoV-2 [40], which renders vaccination less effective [41]. Additionally, the concept of temperament is highly important and can affect the severity of a disease as reported [26].

Methods

Bioinformatics analyses

Initially the fasta format of the SARS-CoV-2 M^pro sequence was taken and a protein blast search was performed in ncbi against all known coronavirus M^pro PDB structures. Subsequently, the outcome of the search was assessed and the non-mutated, native M^pro structures with bound inhibitors were selected and assessed by multiple sequence alignment. The ligand binding mode (whether bound via hydrogen bonding or hydrophobic interactions) in each structure was then analysed by Ligplot [42] and the common residues of coronavirus M^pro involved in binding to the ligands assessed.

Compound selection

The selection of compounds from plants/herbs/trees in this study were based on almost 300 references (please see supplementary data, S5 and S6 Tables in S1 File), with emphasis on the therapeutic and in particular antioxidant and antiviral properties of the active constituents.

Molecular docking

The molecular docking program ‘DockThor’ [43–45] was used to assess the binding of herbal compounds/phytochemicals, considering the initial results of docking five synthetic inhibitors, including N3, α-ketoamide inhibitor, GC-376, and peptidomimetic inhibitors 11a and 11b [10,11,46–51], from experimentally solved structures of SARS-CoV-2 M^pro in complex with the inhibitors. This was done in order to better understand how the DockThor program worked and to gain confidence in the results. In continuation, the docking program was used and the best solutions (with RMSD of 0.0) from the DockThor outputs were used and analysed by Ligplot and Chimera [52]. The emphasis on the DockThor output and the ranking of the compounds was based on affinity prediction values (Kcal/mol) of different ligands in virtual screening experiments considering the top-energy pose (according to Total Energy) of each compound. User defined docking in DockThor included grid centre values of 12, 0, 24 for axes x, y and z, respectively, for SARS-CoV-2 M^pro in the monomeric state (pdb ID 6yb7) and -10, -10, 40 for axes x, y, z, respectively, in the dimeric state (pdb ID 7ali). These regions covered the substrate binding area in SARS-CoV-2 M^pro (deposited either in the monomeric/dimeric states) as identified via superposition of previously ligand-bound structures of coronavirus M^pro. Default settings were used for the DockThor runs. Additionally, six drugs used for COVID-19 including Remdesivir, Darunavir, Oseltamivir, Lopinavir, Ritonavir and Ribavirin; a non-covalent inhibitor ML188; and an approved covalent inhibitor Nirmatrelvir, were all docked with SARS-CoV-2 M^pro, using the same methods described here and results tabulated.

Re-docking

Validation of the docking protocol was performed for the savinin-, betulinic acid- and curcumin-bound SARS-CoV-2 M^pro complexes in DockThor [43–45]. Initially the ligands were omitted from the files containing the complex information and used as the ‘protein file’. The best ranking ligand pose with RMSD of 0.0 obtained in the initial docking was then used as the ‘ligand’ and re-docking performed. The final result containing the best ranking ligand pose was compared with the reference conformation (the pose of the ligand upon initial docking), leading to the calculated RMSD values.

ADMET calculations

The ADMET calculations were performed using the ADMETlab2.0 [36].

MD simulation

The top scoring savinin-, betulinic acid-, curcumin- and Nirmatrelvir-bound SARS-CoV-2 M^pro complexes, as well as the apo structure of SARS-CoV-2 M^pro, were subjected to MD simulation (in both the monomeric and dimeric states), using GROMACS 2019.1 simulation package [53]. The MD simulations were conducted using CHARMM36 forcefield [54]. Structural parameters for savinin, betulinic acid, curcumin and Nirmatrelvir, based on CHARMM36 force field, were obtained from CGenFF webserver [55,56]. Structures were solvated with TIP3P water model [57] in a cubic box, a 1 nm distance between protein and box edges. The system was neutralized by adding sufficient counter charge ions. Energy minimisation of the system was performed by steepest descent integrator [58]. Subsequently, the simulations were performed with Berendsen [59] and Parrinello-Rahman [60] coupling algorithms and following SHAKE integrator [61]. The simulation results from the GROMACS software were analysed as RMSD, RMSF, Rg and number of Hydrogen bonds. Occupancy of the hydrogen bonds, during the MD simulations, were calculated from the number of hydrogen bonds formed between the SARS-CoV-2 M^pro and the ligand in each frame using Microsoft Excel (available at https://office.microsoft.com/excel.). UCSF Chimera was used for the visualisation of the hydrogen bonds and generation of figures. For MD simulation of either the apo or ligand-bound SARS-CoV-2 M^pro structures, molecular docking results using PDB IDs 6yb7 and 7ali for monomeric and dimeric states, respectively, were used.

Binding free energies and residue contributions

Binding free energies (ΔG_Bind) of protein-ligand (or protein-DNA and protein-protein) complexes in a solvent can be calculated from theMM-PBSA. Binding free energies and residue contributions for savinin-,betulinic acid-, curcumin- and Nirmatrelvir-SARS-CoV-2 M^pro complexes were calculated using the g_mmpbsa package [38]. The MM-PBSA calculation was performed based on the protocol available at https://rashmikumari.github.io/g_mmpbsa.

BSA analysis

BSA analysis was performed based on a previous study [37]. The SASA for each ligand was computed using Visual Molecular Dynamics (VMD) [62] and a TCL script, following the Richards and Lee method with a water probe size of 1.4 Å [63]. Subsequently, the BSA was calculated using the formula BSA = 0.5*(SASA ligand + SASA protein–SASA complex).

Supporting information

S1 File. S1-S3 Figs and S1-S12 Tables are provided in the file entitled “Supporting Information”.

The Figure legends and Table titles are also included in the “Supporting Information” file.

https://doi.org/10.1371/journal.pone.0295014.s001

(PDF)

References

1. Cascella M, Rajnik M, Cuomo A, et al. Features, Evaluation, and Treatment of Coronavirus (COVID-19) [Updated 2021 Mar 1]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2021 Jan.
2. Ramphul K. and Mejias S.G., Coronavirus disease: a review of a new threat to public health. Cureus, 2020. 12(3). pmid:32300496
- View Article
- PubMed/NCBI
- Google Scholar
3. Wang L., et al., Review of the 2019 novel coronavirus (SARS-CoV-2) based on current evidence. International journal of antimicrobial agents, 2020. 55(6): p. 105948. pmid:32201353
- View Article
- PubMed/NCBI
- Google Scholar
4. Available from: https://www.worldometers.info/coronavirus/.
5. Khailany R.A., Safdar M., and Ozaslan M., Genomic characterization of a novel SARS-CoV-2. Gene reports, 2020. 19: p. 100682. pmid:32300673
- View Article
- PubMed/NCBI
- Google Scholar
6. Zhao Y., et al., Crystal structure of SARS-CoV-2 main protease in complex with protease inhibitor PF-07321332. Protein & cell, 2022. 13(9): p. 689–693. pmid:34687004
- View Article
- PubMed/NCBI
- Google Scholar
7. Ullrich S. and Nitsche C., The SARS-CoV-2 main protease as drug target. Bioorganic & medicinal chemistry letters, 2020. 30(17): p. 127377. pmid:32738988
- View Article
- PubMed/NCBI
- Google Scholar
8. Anand K., et al., Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs. Science, 2003. 300(5626): p. 1763–1767. pmid:12746549
- View Article
- PubMed/NCBI
- Google Scholar
9. La Monica G., et al., Targeting SARS-CoV-2 main protease for treatment of COVID-19: covalent inhibitors structure–activity relationship insights and evolution perspectives. Journal of medicinal chemistry, 2022. 65(19): p. 12500–12534. pmid:36169610
- View Article
- PubMed/NCBI
- Google Scholar
10. Jin Z., et al., Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. Nature, 2020. 582(7811): p. 289–293. pmid:32272481
- View Article
- PubMed/NCBI
- Google Scholar
11. Zhang L., et al., Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved α-ketoamide inhibitors. Science, 2020. 368(6489): p. 409–412.
- View Article
- Google Scholar
12. Silvestrini L., et al., The dimer-monomer equilibrium of SARS-CoV-2 main protease is affected by small molecule inhibitors. Scientific Reports, 2021. 11(1): p. 9283. pmid:33927258
- View Article
- PubMed/NCBI
- Google Scholar
13. Kao S.T., et al., Effects of Ding‐Chuan‐Tang on bronchoconstriction and airway leucocyte infiltration in sensitized guinea pigs. Immunopharmacology and immunotoxicology, 2004. 26(1): p. 113–124. pmid:15106736
- View Article
- PubMed/NCBI
- Google Scholar
14. Li A., et al., Anti-virus effect of traditional Chinese medicine Yi-Fu-Qing granule on acute respiratory tract infections. Bioscience trends, 2009. 3(4): p. 119–123 pmid:20103834
- View Article
- PubMed/NCBI
- Google Scholar
15. Park H.S., et al., The roles of phytochemicals in bronchial asthma. Molecules, 2010. 15(10): p. 6810–6834. pmid:20924320
- View Article
- PubMed/NCBI
- Google Scholar
16. Bazzano L.A., et al., Fruit and vegetable intake and risk of cardiovascular disease in US adults: the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. The American journal of clinical nutrition, 2002. 76(1): p. 93–99. pmid:12081821
- View Article
- PubMed/NCBI
- Google Scholar
17. Roodaki M., et al., Hot and cold: An old theory with modern applications. Trends in Pharmaceutical Sciences, 2018. 4(1).
- View Article
- Google Scholar
18. Mojahedi M., et al., Reliability and validity assessment of Mizaj questionnaire: a novel self-report scale in Iranian traditional medicine. 2014.
- View Article
- Google Scholar
19. Parvinroo S., et al., The effects of seeds with hot and cold temperaments on serum thyroid hormones, corticosterone and urine vanillylmandelic acid concentrations of healthy rats. Journal of Ethnopharmacology, 2014. 156: p. 216–221. pmid:25193009
- View Article
- PubMed/NCBI
- Google Scholar
20. Chao D.-P., et al., Effects of hot and cold foods on signals of heart rate variability and nail fold microcirculation of healthy young humans: a pilot study. Chinese Journal of Physiology, 2011. 54(3): p. 145–152. pmid:21789896
- View Article
- PubMed/NCBI
- Google Scholar
21. Lin Y.S., Differentiation of cold and heat patterns: ambiguities in diagnosis. Clinical Acupuncture and Oriental Medicine, 2001. 2(2): p. 102–110.
- View Article
- Google Scholar
22. Ma T., et al., Bridging the gap between traditional Chinese medicine and systems biology: the connection of Cold Syndrome and NEI network. Molecular BioSystems, 2010. 6(4): p. 613–619. pmid:20237638
- View Article
- PubMed/NCBI
- Google Scholar
23. Ryu H., et al., Reliability and validity of a cold–heat pattern questionnaire for traditional Chinese medicine. The Journal of Alternative and Complementary Medicine, 2010. 16(6): p. 663–667. pmid:20569034
- View Article
- PubMed/NCBI
- Google Scholar
24. Farsani G.M., et al., Is the Iranian Traditional Medicine warm and cold temperament related to Basal Metabolic Rate and activity of the sympathetic-parasympathetic system? Study protocol. Journal of Diabetes & Metabolic Disorders, 2014. 13: p. 1–6.
- View Article
- Google Scholar
25. Liu C., et al., The relationship between cold‐hot nature and nutrient contents of foods. Nutrition & Dietetics, 2012. 69(1): p. 64–68.
- View Article
- Google Scholar
26. Afzali A., et al., The Relationship between the Severity of the COVID-19 Disease, Temperament and Psychological Factors. Iranian Journal of Psychiatry, 2022. 17(4): p. 462–468.
- View Article
- Google Scholar
27. Yang K.S., et al., Evolutionary and Structural Insights about Potential SARS-CoV-2 Evasion of Nirmatrelvir. J. Med. Chem., 2022. 65(13): p. 8686–8698 pmid:35731933
- View Article
- PubMed/NCBI
- Google Scholar
28. Hall D.C. Jr and Ji H.-F., A search for medications to treat COVID-19 via in silico molecular docking models of the SARS-CoV-2 spike glycoprotein and 3CL protease. Travel medicine and infectious disease, 2020. 35: p. 101646. pmid:32294562
- View Article
- PubMed/NCBI
- Google Scholar
29. Al-Khafaji K., Al-Duhaidahawi D., and Taskin Tok T., Using integrated computational approaches to identify safe and rapid treatment for SARS-CoV-2. Journal of Biomolecular Structure and Dynamics, 2021. 39(9): p. 3387–3395. pmid:32364041
- View Article
- PubMed/NCBI
- Google Scholar
30. Pant S., et al., Peptide-like and small-molecule inhibitors against Covid-19. Journal of Biomolecular Structure and Dynamics, 2021. 39(8): p. 2904–2913. pmid:32306822
- View Article
- PubMed/NCBI
- Google Scholar
31. Wang J., Fast identification of possible drug treatment of coronavirus disease-19 (COVID-19) through computational drug repurposing study. Journal of chemical information and modeling, 2020. 60(6): p. 3277–3286. pmid:32315171
- View Article
- PubMed/NCBI
- Google Scholar
32. Kandeel M. and Al-Nazawi M., Virtual screening and repurposing of FDA approved drugs against COVID-19 main protease. Life sciences, 2020. 251: p. 117627. pmid:32251634
- View Article
- PubMed/NCBI
- Google Scholar
33. Ebenezer O. and Shapi M., Promising inhibitors against main protease of SARS CoV-2 from medicinal plants: In silico identification. Acta Pharmaceutica, 2022. 72(2): p. 159–169. pmid:36651513
- View Article
- PubMed/NCBI
- Google Scholar
34. Alhadrami H.A., et al., Olive-derived triterpenes suppress SARS COV-2 main protease: a promising scaffold for future therapeutics. Molecules, 2021. 26(9): p. 2654. pmid:34062737
- View Article
- PubMed/NCBI
- Google Scholar
35. Lockbaum G.J., et al., Crystal Structure of SARS-CoV-2 Main Protease in Complex with the Non-Covalent Inhibitor ML188. Viruses, 2021. 13(2): p. 174. pmid:33503819
- View Article
- PubMed/NCBI
- Google Scholar
36. Xiong G., et al., ADMETlab 2.0: an integrated online platform for accurate and comprehensive predictions of ADMET properties. Nucleic Acids Res., 2021. 49(W1):W5–W14. pmid:33893803
- View Article
- PubMed/NCBI
- Google Scholar
37. Baker N. Et al., Electrostatics of nanosystems: Application to microtubules and the ribosome. Proceedings of the National Academy of Sciences, 2001. 98(18):p. 10037–10041. pmid:11517324
- View Article
- PubMed/NCBI
- Google Scholar
38. Kumari R., et al., g_mmpbsa—A GROMACS tool for high-throughput MM-PBSA calculations. Journal of chemical information and modeling, 2014. 54(7): p. 1951–1962. pmid:24850022
- View Article
- PubMed/NCBI
- Google Scholar
39. Zhang L., et al., HBD-2 binds SARS-CoV-2 RBD and blocks viral entry: Strategy to combat COVID-19. iScience. 2022. 25(3): p. 103856. pmid:35128350
- View Article
- PubMed/NCBI
- Google Scholar
40. Callaway E., The mutation that helps Delta spread like wildfire. Nature, 2021. 596(7873): p. 472–473. pmid:34417582
- View Article
- PubMed/NCBI
- Google Scholar
41. Vaidyanathan G., News in focus. Nature, 2021. 593: p. 321.
- View Article
- Google Scholar
42. Wallace A.C., Laskowski R.A., and Thornton J.M., LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein engineering, design and selection, 1995. 8(2): p. 127–134. pmid:7630882
- View Article
- PubMed/NCBI
- Google Scholar
43. Guedes I.A., et al., New machine learning and physics-based scoring functions for drug discovery. Scientific reports, 2021. 11(1): p. 1–19.
- View Article
- Google Scholar
44. Santos K.B., et al., Highly flexible ligand docking: Benchmarking of the DockThor program on the LEADS-PEP protein–peptide data set. Journal of Chemical Information and Modeling, 2020. 60(2): p. 667–683. pmid:31922754
- View Article
- PubMed/NCBI
- Google Scholar
45. de Magalhães C.S., et al., A dynamic niching genetic algorithm strategy for docking highly flexible ligands. Information Sciences, 2014. 289: p. 206–224.
- View Article
- Google Scholar
46. Yang H., et al., Design of wide-spectrum inhibitors targeting coronavirus main proteases. PLoS biology, 2005. 3(10): p. e324. pmid:16128623
- View Article
- PubMed/NCBI
- Google Scholar
47. Xue X., et al., Structures of two coronavirus main proteases: implications for substrate binding and antiviral drug design. Journal of virology, 2008. 82(5): p. 2515–2527. pmid:18094151
- View Article
- PubMed/NCBI
- Google Scholar
48. Ren Z., et al., The newly emerged SARS-like coronavirus HCoV-EMC also has an" Achilles’ heel": current effective inhibitor targeting a 3C-like protease. Protein & cell, 2013. 4(4): p. 248. pmid:23549610
- View Article
- PubMed/NCBI
- Google Scholar
49. Wang F., et al., Structure of main protease from human coronavirus NL63: insights for wide spectrum anti-coronavirus drug design. Scientific reports, 2016. 6(1): p. 22677. pmid:26948040
- View Article
- PubMed/NCBI
- Google Scholar
50. Ma C., et al., Boceprevir, GC-376, and calpain inhibitors II, XII inhibit SARS-CoV-2 viral replication by targeting the viral main protease. Cell research, 2020. 30(8): p. 678–692. pmid:32541865
- View Article
- PubMed/NCBI
- Google Scholar
51. Dai W., et al., Structure-based design of antiviral drug candidates targeting the SARS-CoV-2 main protease. Science, 2020. 368(6497): p. 1331–1335. pmid:32321856
- View Article
- PubMed/NCBI
- Google Scholar
52. Pettersen E.F., et al., UCSF Chimera—a visualization system for exploratory research and analysis. Journal of computational chemistry, 2004. 25(13): p. 1605–1612. pmid:15264254
- View Article
- PubMed/NCBI
- Google Scholar
53. Abraham M.J., et al., GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX, 2015. 1: p. 19–25.
- View Article
- Google Scholar
54. Huang J. and MacKerell A.D. Jr, CHARMM36 all‐atom additive protein force field: Validation based on comparison to NMR data. Journal of computational chemistry, 2013. 34(25): p. 2135–2145. pmid:23832629
- View Article
- PubMed/NCBI
- Google Scholar
55. Vanommeslaeghe K. and MacKerell A.D. Jr, Automation of the CHARMM General Force Field (CGenFF) I: bond perception and atom typing. Journal of chemical information and modeling, 2012. 52(12): p. 3144–3154.
- View Article
- Google Scholar
56. Vanommeslaeghe K., Raman E.P., and MacKerell A.D. Jr, Automation of the CHARMM General Force Field (CGenFF) II: assignment of bonded parameters and partial atomic charges. Journal of chemical information and modeling, 2012. 52(12): p. 3155–3168. pmid:23145473
- View Article
- PubMed/NCBI
- Google Scholar
57. Jorgensen W.L., et al., Comparison of simple potential functions for simulating liquid water. The Journal of chemical physics, 1983. 79(2): p. 926–935.
- View Article
- Google Scholar
58. Haug E., Arora J., and Matsui K., A steepest-descent method for optimization of mechanical systems. Journal of Optimization Theory and Applications, 1976. 19: p. 401–424.
- View Article
- Google Scholar
59. Berendsen H.J., et al., Molecular dynamics with coupling to an external bath. The Journal of chemical physics, 1984. 81(8): p. 3684–3690.
- View Article
- Google Scholar
60. Parrinello M. and Rahman A., Polymorphic transitions in single crystals: A new molecular dynamics method. Journal of Applied physics, 1981. 52(12): p. 7182–7190.
- View Article
- Google Scholar
61. Ryckaert J.-P., Ciccotti G., and Berendsen H.J., Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. Journal of computational physics, 1977. 23(3): p. 327–341.
- View Article
- Google Scholar
62. Humphrey W., Dalke A. and Schulten K., "VMD—Visual Molecular Dynamics", J. Molec. Graphics, 1996. 14: p. 33–38. pmid:8744570
- View Article
- PubMed/NCBI
- Google Scholar
63. Lee B., and Richards F.M. The interpretation of protein structures: estimation of static accessibility. J Mol Biol., 1971. 55(3): p. 379–400. pmid:5551392
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Cascella M, Rajnik M, Cuomo A, et al. Features, Evaluation, and Treatment of Coronavirus (COVID-19) [Updated 2021 Mar 1]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2021 Jan.

[ref2] 2. Ramphul K. and Mejias S.G., Coronavirus disease: a review of a new threat to public health. Cureus, 2020. 12(3). pmid:32300496
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Wang L., et al., Review of the 2019 novel coronavirus (SARS-CoV-2) based on current evidence. International journal of antimicrobial agents, 2020. 55(6): p. 105948. pmid:32201353
View Article
PubMed/NCBI
Google Scholar

[7] View Article

[8] PubMed/NCBI

[9] Google Scholar

[ref4] 4. Available from: https://www.worldometers.info/coronavirus/.

[ref5] 5. Khailany R.A., Safdar M., and Ozaslan M., Genomic characterization of a novel SARS-CoV-2. Gene reports, 2020. 19: p. 100682. pmid:32300673
View Article
PubMed/NCBI
Google Scholar

[12] View Article

[13] PubMed/NCBI

[14] Google Scholar

[ref6] 6. Zhao Y., et al., Crystal structure of SARS-CoV-2 main protease in complex with protease inhibitor PF-07321332. Protein & cell, 2022. 13(9): p. 689–693. pmid:34687004
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref7] 7. Ullrich S. and Nitsche C., The SARS-CoV-2 main protease as drug target. Bioorganic & medicinal chemistry letters, 2020. 30(17): p. 127377. pmid:32738988
View Article
PubMed/NCBI
Google Scholar

[20] View Article

[21] PubMed/NCBI

[22] Google Scholar

[ref8] 8. Anand K., et al., Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs. Science, 2003. 300(5626): p. 1763–1767. pmid:12746549
View Article
PubMed/NCBI
Google Scholar

[24] View Article

[25] PubMed/NCBI

[26] Google Scholar

[ref9] 9. La Monica G., et al., Targeting SARS-CoV-2 main protease for treatment of COVID-19: covalent inhibitors structure–activity relationship insights and evolution perspectives. Journal of medicinal chemistry, 2022. 65(19): p. 12500–12534. pmid:36169610
View Article
PubMed/NCBI
Google Scholar

[28] View Article

[29] PubMed/NCBI

[30] Google Scholar

[ref10] 10. Jin Z., et al., Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. Nature, 2020. 582(7811): p. 289–293. pmid:32272481
View Article
PubMed/NCBI
Google Scholar

[32] View Article

[33] PubMed/NCBI

[34] Google Scholar

[ref11] 11. Zhang L., et al., Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved α-ketoamide inhibitors. Science, 2020. 368(6489): p. 409–412.
View Article
Google Scholar

[36] View Article

[37] Google Scholar

[ref12] 12. Silvestrini L., et al., The dimer-monomer equilibrium of SARS-CoV-2 main protease is affected by small molecule inhibitors. Scientific Reports, 2021. 11(1): p. 9283. pmid:33927258
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref13] 13. Kao S.T., et al., Effects of Ding‐Chuan‐Tang on bronchoconstriction and airway leucocyte infiltration in sensitized guinea pigs. Immunopharmacology and immunotoxicology, 2004. 26(1): p. 113–124. pmid:15106736
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref14] 14. Li A., et al., Anti-virus effect of traditional Chinese medicine Yi-Fu-Qing granule on acute respiratory tract infections. Bioscience trends, 2009. 3(4): p. 119–123 pmid:20103834
View Article
PubMed/NCBI
Google Scholar

[47] View Article

[48] PubMed/NCBI

[49] Google Scholar

[ref15] 15. Park H.S., et al., The roles of phytochemicals in bronchial asthma. Molecules, 2010. 15(10): p. 6810–6834. pmid:20924320
View Article
PubMed/NCBI
Google Scholar

[51] View Article

[52] PubMed/NCBI

[53] Google Scholar

[ref16] 16. Bazzano L.A., et al., Fruit and vegetable intake and risk of cardiovascular disease in US adults: the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. The American journal of clinical nutrition, 2002. 76(1): p. 93–99. pmid:12081821
View Article
PubMed/NCBI
Google Scholar

[55] View Article

[56] PubMed/NCBI

[57] Google Scholar

[ref17] 17. Roodaki M., et al., Hot and cold: An old theory with modern applications. Trends in Pharmaceutical Sciences, 2018. 4(1).
View Article
Google Scholar

[59] View Article

[60] Google Scholar

[ref18] 18. Mojahedi M., et al., Reliability and validity assessment of Mizaj questionnaire: a novel self-report scale in Iranian traditional medicine. 2014.
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref19] 19. Parvinroo S., et al., The effects of seeds with hot and cold temperaments on serum thyroid hormones, corticosterone and urine vanillylmandelic acid concentrations of healthy rats. Journal of Ethnopharmacology, 2014. 156: p. 216–221. pmid:25193009
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref20] 20. Chao D.-P., et al., Effects of hot and cold foods on signals of heart rate variability and nail fold microcirculation of healthy young humans: a pilot study. Chinese Journal of Physiology, 2011. 54(3): p. 145–152. pmid:21789896
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref21] 21. Lin Y.S., Differentiation of cold and heat patterns: ambiguities in diagnosis. Clinical Acupuncture and Oriental Medicine, 2001. 2(2): p. 102–110.
View Article
Google Scholar

[73] View Article

[74] Google Scholar

[ref22] 22. Ma T., et al., Bridging the gap between traditional Chinese medicine and systems biology: the connection of Cold Syndrome and NEI network. Molecular BioSystems, 2010. 6(4): p. 613–619. pmid:20237638
View Article
PubMed/NCBI
Google Scholar

[76] View Article

[77] PubMed/NCBI

[78] Google Scholar

[ref23] 23. Ryu H., et al., Reliability and validity of a cold–heat pattern questionnaire for traditional Chinese medicine. The Journal of Alternative and Complementary Medicine, 2010. 16(6): p. 663–667. pmid:20569034
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref24] 24. Farsani G.M., et al., Is the Iranian Traditional Medicine warm and cold temperament related to Basal Metabolic Rate and activity of the sympathetic-parasympathetic system? Study protocol. Journal of Diabetes & Metabolic Disorders, 2014. 13: p. 1–6.
View Article
Google Scholar

[84] View Article

[85] Google Scholar

[ref25] 25. Liu C., et al., The relationship between cold‐hot nature and nutrient contents of foods. Nutrition & Dietetics, 2012. 69(1): p. 64–68.
View Article
Google Scholar

[87] View Article

[88] Google Scholar

[ref26] 26. Afzali A., et al., The Relationship between the Severity of the COVID-19 Disease, Temperament and Psychological Factors. Iranian Journal of Psychiatry, 2022. 17(4): p. 462–468.
View Article
Google Scholar

[90] View Article

[91] Google Scholar

[ref27] 27. Yang K.S., et al., Evolutionary and Structural Insights about Potential SARS-CoV-2 Evasion of Nirmatrelvir. J. Med. Chem., 2022. 65(13): p. 8686–8698 pmid:35731933
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref28] 28. Hall D.C. Jr and Ji H.-F., A search for medications to treat COVID-19 via in silico molecular docking models of the SARS-CoV-2 spike glycoprotein and 3CL protease. Travel medicine and infectious disease, 2020. 35: p. 101646. pmid:32294562
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref29] 29. Al-Khafaji K., Al-Duhaidahawi D., and Taskin Tok T., Using integrated computational approaches to identify safe and rapid treatment for SARS-CoV-2. Journal of Biomolecular Structure and Dynamics, 2021. 39(9): p. 3387–3395. pmid:32364041
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref30] 30. Pant S., et al., Peptide-like and small-molecule inhibitors against Covid-19. Journal of Biomolecular Structure and Dynamics, 2021. 39(8): p. 2904–2913. pmid:32306822
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref31] 31. Wang J., Fast identification of possible drug treatment of coronavirus disease-19 (COVID-19) through computational drug repurposing study. Journal of chemical information and modeling, 2020. 60(6): p. 3277–3286. pmid:32315171
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref32] 32. Kandeel M. and Al-Nazawi M., Virtual screening and repurposing of FDA approved drugs against COVID-19 main protease. Life sciences, 2020. 251: p. 117627. pmid:32251634
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref33] 33. Ebenezer O. and Shapi M., Promising inhibitors against main protease of SARS CoV-2 from medicinal plants: In silico identification. Acta Pharmaceutica, 2022. 72(2): p. 159–169. pmid:36651513
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref34] 34. Alhadrami H.A., et al., Olive-derived triterpenes suppress SARS COV-2 main protease: a promising scaffold for future therapeutics. Molecules, 2021. 26(9): p. 2654. pmid:34062737
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref35] 35. Lockbaum G.J., et al., Crystal Structure of SARS-CoV-2 Main Protease in Complex with the Non-Covalent Inhibitor ML188. Viruses, 2021. 13(2): p. 174. pmid:33503819
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref36] 36. Xiong G., et al., ADMETlab 2.0: an integrated online platform for accurate and comprehensive predictions of ADMET properties. Nucleic Acids Res., 2021. 49(W1):W5–W14. pmid:33893803
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref37] 37. Baker N. Et al., Electrostatics of nanosystems: Application to microtubules and the ribosome. Proceedings of the National Academy of Sciences, 2001. 98(18):p. 10037–10041. pmid:11517324
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref38] 38. Kumari R., et al., g_mmpbsa—A GROMACS tool for high-throughput MM-PBSA calculations. Journal of chemical information and modeling, 2014. 54(7): p. 1951–1962. pmid:24850022
View Article
PubMed/NCBI
Google Scholar

[137] View Article

[138] PubMed/NCBI

[139] Google Scholar

[ref39] 39. Zhang L., et al., HBD-2 binds SARS-CoV-2 RBD and blocks viral entry: Strategy to combat COVID-19. iScience. 2022. 25(3): p. 103856. pmid:35128350
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref40] 40. Callaway E., The mutation that helps Delta spread like wildfire. Nature, 2021. 596(7873): p. 472–473. pmid:34417582
View Article
PubMed/NCBI
Google Scholar

[145] View Article

[146] PubMed/NCBI

[147] Google Scholar

[ref41] 41. Vaidyanathan G., News in focus. Nature, 2021. 593: p. 321.
View Article
Google Scholar

[149] View Article

[150] Google Scholar

[ref42] 42. Wallace A.C., Laskowski R.A., and Thornton J.M., LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein engineering, design and selection, 1995. 8(2): p. 127–134. pmid:7630882
View Article
PubMed/NCBI
Google Scholar

[152] View Article

[153] PubMed/NCBI

[154] Google Scholar

[ref43] 43. Guedes I.A., et al., New machine learning and physics-based scoring functions for drug discovery. Scientific reports, 2021. 11(1): p. 1–19.
View Article
Google Scholar

[156] View Article

[157] Google Scholar

[ref44] 44. Santos K.B., et al., Highly flexible ligand docking: Benchmarking of the DockThor program on the LEADS-PEP protein–peptide data set. Journal of Chemical Information and Modeling, 2020. 60(2): p. 667–683. pmid:31922754
View Article
PubMed/NCBI
Google Scholar

[159] View Article

[160] PubMed/NCBI

[161] Google Scholar

[ref45] 45. de Magalhães C.S., et al., A dynamic niching genetic algorithm strategy for docking highly flexible ligands. Information Sciences, 2014. 289: p. 206–224.
View Article
Google Scholar

[163] View Article

[164] Google Scholar

[ref46] 46. Yang H., et al., Design of wide-spectrum inhibitors targeting coronavirus main proteases. PLoS biology, 2005. 3(10): p. e324. pmid:16128623
View Article
PubMed/NCBI
Google Scholar

[166] View Article

[167] PubMed/NCBI

[168] Google Scholar

[ref47] 47. Xue X., et al., Structures of two coronavirus main proteases: implications for substrate binding and antiviral drug design. Journal of virology, 2008. 82(5): p. 2515–2527. pmid:18094151
View Article
PubMed/NCBI
Google Scholar

[170] View Article

[171] PubMed/NCBI

[172] Google Scholar

[ref48] 48. Ren Z., et al., The newly emerged SARS-like coronavirus HCoV-EMC also has an" Achilles’ heel": current effective inhibitor targeting a 3C-like protease. Protein & cell, 2013. 4(4): p. 248. pmid:23549610
View Article
PubMed/NCBI
Google Scholar

[174] View Article

[175] PubMed/NCBI

[176] Google Scholar

[ref49] 49. Wang F., et al., Structure of main protease from human coronavirus NL63: insights for wide spectrum anti-coronavirus drug design. Scientific reports, 2016. 6(1): p. 22677. pmid:26948040
View Article
PubMed/NCBI
Google Scholar

[178] View Article

[179] PubMed/NCBI

[180] Google Scholar

[ref50] 50. Ma C., et al., Boceprevir, GC-376, and calpain inhibitors II, XII inhibit SARS-CoV-2 viral replication by targeting the viral main protease. Cell research, 2020. 30(8): p. 678–692. pmid:32541865
View Article
PubMed/NCBI
Google Scholar

[182] View Article

[183] PubMed/NCBI

[184] Google Scholar

[ref51] 51. Dai W., et al., Structure-based design of antiviral drug candidates targeting the SARS-CoV-2 main protease. Science, 2020. 368(6497): p. 1331–1335. pmid:32321856
View Article
PubMed/NCBI
Google Scholar

[186] View Article

[187] PubMed/NCBI

[188] Google Scholar

[ref52] 52. Pettersen E.F., et al., UCSF Chimera—a visualization system for exploratory research and analysis. Journal of computational chemistry, 2004. 25(13): p. 1605–1612. pmid:15264254
View Article
PubMed/NCBI
Google Scholar

[190] View Article

[191] PubMed/NCBI

[192] Google Scholar

[ref53] 53. Abraham M.J., et al., GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX, 2015. 1: p. 19–25.
View Article
Google Scholar

[194] View Article

[195] Google Scholar

[ref54] 54. Huang J. and MacKerell A.D. Jr, CHARMM36 all‐atom additive protein force field: Validation based on comparison to NMR data. Journal of computational chemistry, 2013. 34(25): p. 2135–2145. pmid:23832629
View Article
PubMed/NCBI
Google Scholar

[197] View Article

[198] PubMed/NCBI

[199] Google Scholar

[ref55] 55. Vanommeslaeghe K. and MacKerell A.D. Jr, Automation of the CHARMM General Force Field (CGenFF) I: bond perception and atom typing. Journal of chemical information and modeling, 2012. 52(12): p. 3144–3154.
View Article
Google Scholar

[201] View Article

[202] Google Scholar

[ref56] 56. Vanommeslaeghe K., Raman E.P., and MacKerell A.D. Jr, Automation of the CHARMM General Force Field (CGenFF) II: assignment of bonded parameters and partial atomic charges. Journal of chemical information and modeling, 2012. 52(12): p. 3155–3168. pmid:23145473
View Article
PubMed/NCBI
Google Scholar

[204] View Article

[205] PubMed/NCBI

[206] Google Scholar

[ref57] 57. Jorgensen W.L., et al., Comparison of simple potential functions for simulating liquid water. The Journal of chemical physics, 1983. 79(2): p. 926–935.
View Article
Google Scholar

[208] View Article

[209] Google Scholar

[ref58] 58. Haug E., Arora J., and Matsui K., A steepest-descent method for optimization of mechanical systems. Journal of Optimization Theory and Applications, 1976. 19: p. 401–424.
View Article
Google Scholar

[211] View Article

[212] Google Scholar

[ref59] 59. Berendsen H.J., et al., Molecular dynamics with coupling to an external bath. The Journal of chemical physics, 1984. 81(8): p. 3684–3690.
View Article
Google Scholar

[214] View Article

[215] Google Scholar

[ref60] 60. Parrinello M. and Rahman A., Polymorphic transitions in single crystals: A new molecular dynamics method. Journal of Applied physics, 1981. 52(12): p. 7182–7190.
View Article
Google Scholar

[217] View Article

[218] Google Scholar

[ref61] 61. Ryckaert J.-P., Ciccotti G., and Berendsen H.J., Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. Journal of computational physics, 1977. 23(3): p. 327–341.
View Article
Google Scholar

[220] View Article

[221] Google Scholar

[ref62] 62. Humphrey W., Dalke A. and Schulten K., "VMD—Visual Molecular Dynamics", J. Molec. Graphics, 1996. 14: p. 33–38. pmid:8744570
View Article
PubMed/NCBI
Google Scholar

[223] View Article

[224] PubMed/NCBI

[225] Google Scholar

[ref63] 63. Lee B., and Richards F.M. The interpretation of protein structures: estimation of static accessibility. J Mol Biol., 1971. 55(3): p. 379–400. pmid:5551392
View Article
PubMed/NCBI
Google Scholar

[227] View Article

[228] PubMed/NCBI

[229] Google Scholar

In silico studies of anti-oxidative and hot temperament-based phytochemicals as natural inhibitors of SARS-CoV-2 M^pro

In silico studies of anti-oxidative and hot temperament-based phytochemicals as natural inhibitors of SARS-CoV-2 M^pro

Figures

Abstract

Introduction

Results

Bioinformatics analyses of coronavirus M^pro structures with bound ligands/inhibitors

Comparison between experimentally derived structures of five synthetic covalent inhibitors bound to SARS-CoV-2 M^pro and that of molecular docking results

Molecular docking of six antiviral drugs also used for COVID-19

Molecular docking of 38 phytochemicals with SARS-CoV-2 M^pro

Molecular docking using PDB ID 6y7b, containing the deposited monomeric state of SARS-CoV-2 M^pro in the asymmetric unit

Molecular docking using PDB ID 7ali, containing the dimeric and biologically active form of SARS-CoV-2 M^pro

Re-docking of savinin, betulinic acid and curcumin

Covalent versus non-covalent SARS-CoV-2 M^pro inhibitors

Nirmatrelvir as an approved specific inhibitor of SARS-CoV-2 M^pro

ADMET analysis

Validation of the molecular docking results

MD simulation analysis of the complexes

RMSD anlaysis

RMSF analysis

Rg analysis

Number of hydrogen bonds

Binding free energies and residue contributions

Buried surface area analysis

Conclusions

Methods

Bioinformatics analyses

Compound selection

Molecular docking

Re-docking

ADMET calculations

MD simulation

Binding free energies and residue contributions

BSA analysis

Supporting information

S1 File. S1-S3 Figs and S1-S12 Tables are provided in the file entitled “Supporting Information”.

References

Figures

Abstract

Introduction

Results

Bioinformatics analyses of coronavirus Mpro structures with bound ligands/inhibitors

Comparison between experimentally derived structures of five synthetic covalent inhibitors bound to SARS-CoV-2 Mpro and that of molecular docking results

Molecular docking of six antiviral drugs also used for COVID-19

Molecular docking of 38 phytochemicals with SARS-CoV-2 Mpro

Molecular docking using PDB ID 6y7b, containing the deposited monomeric state of SARS-CoV-2 Mpro in the asymmetric unit

Molecular docking using PDB ID 7ali, containing the dimeric and biologically active form of SARS-CoV-2 Mpro

Re-docking of savinin, betulinic acid and curcumin

Covalent versus non-covalent SARS-CoV-2 Mpro inhibitors

Nirmatrelvir as an approved specific inhibitor of SARS-CoV-2 Mpro

ADMET analysis

Validation of the molecular docking results

MD simulation analysis of the complexes

RMSD anlaysis

RMSF analysis

Rg analysis

Number of hydrogen bonds

Binding free energies and residue contributions

Buried surface area analysis

Conclusions

Methods

Bioinformatics analyses

Compound selection

Molecular docking

Re-docking

ADMET calculations

MD simulation

Binding free energies and residue contributions

BSA analysis

Supporting information

S1 File. S1-S3 Figs and S1-S12 Tables are provided in the file entitled “Supporting Information”.

References

Bioinformatics analyses of coronavirus M^pro structures with bound ligands/inhibitors

Comparison between experimentally derived structures of five synthetic covalent inhibitors bound to SARS-CoV-2 M^pro and that of molecular docking results

Molecular docking of 38 phytochemicals with SARS-CoV-2 M^pro

Molecular docking using PDB ID 6y7b, containing the deposited monomeric state of SARS-CoV-2 M^pro in the asymmetric unit

Molecular docking using PDB ID 7ali, containing the dimeric and biologically active form of SARS-CoV-2 M^pro

Covalent versus non-covalent SARS-CoV-2 M^pro inhibitors

Nirmatrelvir as an approved specific inhibitor of SARS-CoV-2 M^pro