Journal of Biomolecular Structure and Dynamics

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/tbsd20

Proposing of fungal endophyte secondary

metabolites as a potential inhibitors of 2019-novel
coronavirus main protease using docking and
molecular dynamics

Kosar Sadat Ebrahimi, Mahdieh S. Hosseyni Moghaddam, Mohabbat Ansari,

Amin Nowroozi, Mohsen Shahlaei & Sajad Moradi

To cite this article: Kosar Sadat Ebrahimi, Mahdieh S. Hosseyni Moghaddam, Mohabbat Ansari,
Amin Nowroozi, Mohsen Shahlaei & Sajad Moradi (29 Jan 2024): Proposing of fungal endophyte
secondary metabolites as a potential inhibitors of 2019-novel coronavirus main protease
using docking and molecular dynamics, Journal of Biomolecular Structure and Dynamics, DOI:
10.1080/07391102.2024.2308777

To link to this article: https://doi.org/10.1080/07391102.2024.2308777

View supplementary material

Published online: 29 Jan 2024.

Submit your article to this journal

Article views: 64

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tbsd20
JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
https://doi.org/10.1080/07391102.2024.2308777

Proposing of fungal endophyte secondary metabolites as a potential

inhibitors of 2019-novel coronavirus main protease using docking and
molecular dynamics
Kosar Sadat Ebrahimia, Mahdieh S. Hosseyni Moghaddamb, Mohabbat Ansaria, Amin Nowroozia,
Mohsen Shahlaeic and Sajad Moradia
a
Nano Drug Delivery Research Center, Health Technology Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran;
b
Mycology and Microbiology Center, University of Tartu, Liivi 2, Tartu, Estonia; cMedical Biology Research Center, Health Technology
Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran
Communicated by Ramaswamy H. Sarma

ABSTRACT ARTICLE HISTORY

In this study, the inhibitory potential of 99 fungal derived secondary metabolites was predicted against Received 16 March 2023
SARS-CoV-2 main protease by using of computational approaches. This protein plays an important role Accepted 15 January 2024
in replication and is one of the important targets to inhibit viral reproduction. Among the 99 reported
KEYWORDS
compounds, the 9 of them with the highest binding energy to Mpro obtained from the molecular
Covid-19; entophyte fungi;
docking method were selected for the molecular dynamic simulations. The compounds were then secondary metabolites;
investigated by using the SwissADME serve to evaluate the compounds in terms of pharmacokinetic docking; molecular dynamic
and druglikness properties. The overall results of different analysis show that the compound RKS-1778 simulation
is potentially more effective than others and form strong complexes with viral protease. It also had
better pharmacokinetic properties than other metabolites, so predicted to be a suitable candidate as
anti SARS-CoV-2 bioactive.

Introduction each protomer consists of 306 amino acids which are

encoded from both ORF1a and ORF1ab. The protein com­
Coronaviruses are a large and diverse family of enveloped,
prises of three main domains of I (residues 8–101), II (resi­
positive-stranded RNA viruses (Gallagher & Buchmeier, 2001)
dues 102–184) and III (201–306) and a loop region (residues
that are distributed broadly among mammals and birds and
185–200) which connects the domains II and III. This cysteine
six of their species are known to cause human disease (Su
protease contains a catalytic dyad of Cys145 and His41 in its
et al., 2016). In December 2019, a new type of beta coronavi­ active site which is located in a cleft between the domains I
ruses was identified in Wuhan, Hubei province, China (Huang and II (Zhou et al., 2020). The cleavage site of the substrate
et al., 2020). This Emerging pathogen, with phylogenetic is mainly specified by P4 to P10 positions in which P2, P1
similarity to SARS-CoV (Zhu et al., 2020), has been recently and P10 are more conserved among human coronaviruses.
declared as a global challenge for public health. Due to the Leucine, Glutamine and Serine are the most preferred amino
high infection rate and relatively high mortality of this pan­ acids in these locations respectively and as there is no
demic, finding an efficient therapeutic has become an inter­ known human protease with a similar substrate sequence,
national emergence among researchers, companies and the Mpro is a suitable target to designing antiviral drugs
governments (Wu et al., 2020; Xu et al., 2020; Zhou et al., against the SARS-CoV-2 (Dai et al., 2020).
2020). To the best of our knowledge, there is still no specific One of the urgent therapeutic ways in such emergency
drug for the treatment of coronavirus disease (Chen et al., conditions is the search for pathogen inhibitors among nat­
2020). Accordingly, continued efforts have been made to dis­ ural products. Currently, microorganisms are well explored as
cover, develop, and produce efficient drugs for the treatment one of the most interesting sources of natural products,
of this infection. since they have displayed significant biochemical, biological,
Coronavirus main protease (Mpro) also known as non- and biosynthetic potentials (Demain, 2014). In this case, the
structural protein 5 (nsp5) is one of the critical targets for endophytic fungi have attracted much attention in recent
anti-CoV drug design, because of its vital rule in the matur­ years due to their potential medicinal secondary metabolites
ation of other key functional viral protein (Liu & Wang, (Da Silva, 2017; Negreiros de Carvalho et al., 2016).
2020). The functional form of Mpro is a homodimeric protein; Endophytic fungi are defined as microorganisms that live

CONTACT Mohsen Shahlaei mohsenshahlaei@yahoo.com; mshahlaei@kums.ac.ir Department of Medicinal Chemistry, Faculty of Pharmacy, Kermanshah
University of Medical Sciences, Kermanshah, 67346-67149, Iran; Sajad Moradi sajadmoradi28@gmail.com; sajad.moradi@kums.ac.ir Nano Drug Delivery
Research Center, Health Technology Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Supplemental data for this article can be accessed online at https://doi.org/10.1080/07391102.2024.2308777.
� 2024 Informa UK Limited, trading as Taylor & Francis Group
2 K. S. EBRAHIMI ET AL.

inside healthy plant organs, without causing any apparent Molecular dynamics
symptoms of disease in their hosts (Rodriguez et al., 2009).
In order to evaluate the dynamic of ligand-protein interac­
They are known as an important source of various secondary
tions the molecular dynamic simulation (MDs) approach
metabolites such as alkaloids, terpenoids, steroids, lactones,
were utilized by using a computational package of
phenols, and isocoumarins, with anticancer, antioxidant, cyto­
GROMACS (version 2018). The topology information for pro­
toxic, antiviral, antidiabetic, antileishmaniasis, immunosup­
tein was obtained in the gromos53a6 forcefield (Lindahl
pressive, and antimicrobial activities (Bano et al., 2016; Chen
et al., 2001) and those for ligands prepared from the
et al., 2016; Li et al., 2018; Xie et al., 2016).
PRODRG server (Schu €ttelkopf & Van Aalten, 2004). The
Performing an experimental drug discovery investigation
atomic charges in the obtained ligand topologies were
is too much labor, expensive and time consuming which
manually corrected based on the force field’s data.
make it an insufficient approach for treating of such an
Simulation boxes were filed by the SPC model of water and
urgent pandemic (Fabricant & Farnsworth, 2001). In contrast
the systems were then neutralized by adding the appropriate
the computational methods offer an alternative medium for
amount of counter ion (Mark & Nilsson, 2001). It should be
screening the potential efficacy of hundreds of drug com­
noted that based on the previous study the SPC model of
pounds on specific disease in a time and cost effective man­
water is the most appropriate description for using with
ner (Ebrahimi et al., 2021; Sliwoski et al., 2014). There are
GROMOS force fields in terms of calculate the binding
numerous computational approaches for both pharmacody­
energy of molecular systems (Nguyen et al., 2014). In order
namic and pharmacokinetic explorations of proposed com­ to eliminating atomic clashes in the system, energy mini­
pounds as in silico (Mollazadeh et al., 2021; Shaker et al., mization were done under the steepest descend algorithm
2021). These techniques enable researchers to study the until the maximum force on the system atoms reaches below
atomic details of chemical and biological processes (Gombar the 10 KJ.mol−1 nm−1 (Hirshman & Whitson, 1983). Periodic
et al., 2003; Khaledian et al., 2021). boundary condition was applied in all directions of X, Y and
Because of the high therapeutic potency of endophytic Z. Temperature and pressure were coupled to 310 K and
fungi metabolites, in this research using computational 1 bar in NVT and NPT ensembles respectively. In this regard
methods of docking and molecular dynamics simulations the a Nose-Hoover thermostat (Evans & Holian, 1985; Hoover,
SARS-CoV-2 Mpro inhibitory effects of 99 secondary metabo­ 1985; Nos�e, 1984) and the Parrinello-Rahman barostat
lites extracted from endophytic fungi were investigated. (Parrinello & Rahman, 1980) were used as basic algorithms
Furthermore, the pharmacokinetics properties of the selected for equilibrations. Electrostatic and Van der Waals short-
compounds were predicted in chemoinformatic tools. range interactions were also calculated both in the cut-off
range of 1 nm. The bond constraint for all heavy atoms was
Methods applied by using the LINCS method (Hess et al., 1997).
Finally 200 ns of MD simulations were performed under the
Docking leap frog algorithm (Van Gunsteren & Berendsen, 1988).
The docking studies were performed by autodock4 in two Graphical representations and 2D molecular interactions
were prepared in VMD and LigPlot respectively (Laskowski &
steps; first a blind docking on the whole structure of protein
Swindells, 2011).
and then a targeted one focused on the active site of the
enzyme. The 3D model of protein was obtained from a pro­
tein data bank (www.rcsb.org) (PDB.ID. 6Y84). Molecular Pharmacokinetic
structures for the fungal metabolites (Sharma et al., 2019)
The pharmacokinetic properties of the final compounds were
were prepared in ChemSketch ACD/LAB (www.acdlabs.com)
also evaluated in the SWISSADME server.
and then optimized in Avogadro package using a steep algo­
rithm. For both protein and ligand molecules the polar
hydrogens were added to atoms by the MGLtools package Results and discussions
as well as the gasteiger charges for ligand and Kollman
charges for protein (Weiner et al., 1984). In the case of ligand Docking
preparation it also needs to activate important torsions. In The tendency of fungal compounds for binding to the Mpro
the blind docking step the search space were choose as 126 active site was first evaluated by a blind docking study.
point by the grid spacing of 0.475 A� and for targeted dock­ Among the tested molecules, 23 compounds with the more
ing an 80 point box with 0.375 A� space between each point. binding energies to protein were then docked against the
Energetic maps for all atomic types related to ligands were enzyme active site as targeted manner. From which 9 com­
computed in autogrid4. For all compounds, 250 runs of pound with a binding energy < −9 kcal/mol were selected
molecular dockings were done under the Lamarckian genetic for studding their dynamic of protein binding. The results of
algorithm (Morris et al., 1998). The results were then ana­ targeted docking is shown in Table 1 and as can be seen the
lyzed manually for the best appropriate conformation of the lowest binding energy (−11.177 kcal/mol) is related to RKS-
complexed ligands based on the lower binding energy in the 1778 which is an extracted metabolite from a fungi/plant
cluster with a higher number of runs. symbiosis between Phomopsis theicola and Litsea hypophaea.
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 3

Table 1. Results of targeted docking studies on the active site of coronavirus main protease.
Compound Lowest binding energy (kcal/mol) Number of runs in cluster
3,7-dihydroxy-9-methoxy- 2-methyl-6Hbenzo [c] chromen-6-one −7.31 127
4-des-hydroxyl altersolanol A −6.49 117
Altertoxin I −7.32 183
Altertoxin II −7.47 140
Altertoxin III −7.47 122
Altertoxin V −7.49 166
Camptothecin −7.96 193
Colletotrichones A −6.91 128
GKK1032B 29.15 173
Outovirin C −7.27 109
Phomocytochalasin −7.37 104
Purpureone 29.31 127
Pyrrocidine A 210.08 194
RKS-1778 211.17 121
Thielavins A 210.18 122
Thielavins J 210.61 159
Beauvericin 29.25 189
Thielavins K −8.10 163
Trichothecinol-A −7.91 237
11-dehydroxy epoxyphomalin A −7.91 144
Viriditoxin 210.91 218
a-viridin −8.12 169
b-Viridin 29.23 121

Table 2. The most potent metabolites extracted from molecular docking studies.
Compound Endophytic fungi Host plant
GKK1032B Penicillium citrinum Garcinia mangostana
b-Viridin Trichoderma sp. Centaurea stoebe
Purpureone Purpureocillium lilacinum Rauvolfia macrophylla
RKS-1778 Phomopsis theicola Litsea hypophaea
Thielavins A Unidentified Hintonia latiflora
Thielavins J Unidentified Hintonia latiflora
Beauvericin Epicoccum nigrum Entada abyssinica
Pyrrocidine A Neonectria ramulariae Cylindrocarpon sp.
Viriditoxin Paecilomyces variotii Laguncularia racemosa

The information about all selected compound is reported in main position of the ligand did not change significantly. Also
Table 2. the number of VdW interactions after the simulations is
Further details about molecular interactions between pro­ increased which is a sign for a more potent interaction after
tease and selected ligands in both docking and after 200 ns the MDs. Thielavins A also formed five hydrogen bonds with
of molecular simulations are represented in Figure 1. The residues phe140, His164, Glu166, His41, and Thr26 from
compound GKK1032B found to interact with His41, and which Glu166 and His41 play an important role in the dimer­
Cys145 and Glu166 residues which mark the cleavage site of ization and catalytic activity of Mpro respectively. The only
Mpro. The interaction pattern is then undergoing some Thr26 H-bonding is maintained after the molecular dynamics
changes between protein residues however the basic loca­ for this ligand. Based on this result it can be concluded the
tion is maintained. This is confirms by the presence of the whole interaction energy of the ligand is decreased during
interacted residues of ducking results after the MD simula­ the simulation and also the molecular position of the ligand
tion but changing the H-bond forming residues. His163, did not changed. Thielavins J showed 3 hydrogen bonds
Gly143, and Glu166 residues are involved on the hydrogen with amino acids Asn142, and two others with Thr26, and
bonding of protein with B-viridin. After the MD simulation Thr24. Likewise the others the changes in the H-bonding pat­
there are only a few VdW interactions which make it the tern along with consistency in VdW residue patterns confirms
weakest interacted compound and this is in line with inter­ some relocation of ligand on the protein active site. Asn142
action energy results represented in Table 3. Purpureone and Gln189 formed two hydrogen bonds with Beauvericin
demonstrated three hydrogen bonds with Glu166 and four before and after the MD simulation respectively along with
others with Cys145 as well Thr24, Thr25 and Thr26. Despite other VdW interactions confirms the stability of ligand on
changing in the h-bonding pattern the compound in steel the active site of the protein. In the case of Pyrrocidine A
tightly attached to the active site of protein as it forms a the only hydrogen bonding is occurred with Glu166 which
hydrogen bond with His41 during the molecular dynamics. then shifted to two H-bondings with Gls189 and Ser46.
The compound RKS-1778 formed two hydrogen bonding Finally in the case of Viriditoxin there are five potent hydro­
with the residues His164 and Gly143 which are located gen bondings with the residues Thr26, His163, Gln189,
around the active site of the enzyme. During the simulation Asn142 and Phe149. From which the two H-bond forming
the h-bondings have changed to Cys44 and His164 but the residues maintained stable during the simulations including
4 K. S. EBRAHIMI ET AL.

Figure 1. The 2D results of molecular interactions between 9 final selected compounds and the main protease of SARS-CoV-2, section A represents the docking
results and B after 200 ns of MD simulations.

the His 163 and Thr 26. This further indicates a strung inter­ mean squared deviation of the system atoms during the
action with the protease active site both before and after simulation time. The RMSD for all complexes were calculated
MD simulations. and their results are shown in Figure 2. In this figure the
RMSD curves for the main chain of protein in complex with
different metabolites are compared with that for free protein.
Molecular dynamic simulations
As can be seen, in the case of free protein after a relaxation
The dynamics of interactions between fungal metabolites jump in RMSD value for about 2.5 Ao, there are some fluctua­
with the highest score in docking studies and covid-19 main tions in the curve until the time of 80 ns. Thereafter the
protease were further investigated using a molecular curve is clearly smooth by the rest of the simulation which
dynamic approach. In this regard 200 ns of MD simulations indicating the system reached to its equilibrium state and
were done in an aqueous medium at 310 K and 1 bar. Based the applied simulation time is enough for corona virus main
on the binding free energies and number of conformations protease. In the case of protein in interaction with metabo­
in cluster, 9 most potent compounds were selected for fur­ lites, there are some differences in the mean value and
ther analysis in dynamic state (Table 2). severity of RMSD fluctuations as a result of binding the com­
pounds to protein. A higher amount of RMSD imply a higher
relocation of protein atoms and more fluctuations indicate
Root mean squared deviation (RMSD)
the most sever changes in atom position during the simula­
At the first it is necessary to ensuring of equilibration in the tion. The highest value of deviations is related to the protein
system which is a good criterion to validate that the simula­ in complex with RKS-1778, Pyrrocidine A and Viriditoxin
tion time is sufficient. This is addressed by analyzing the root while the most sever fluctuations are observed in the
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 5

Figure 1. Continued.

systems contain Thielavins A, Beauvericin and Viriditoxin. important is the dimerization of enzyme. Another location in
These results indicate some changes in protein structure as a which the ligand binding is resulted in a decrement in RMSF
result of binding the fungal metabolites to the protein which is position 72–91 that is a beta sheet just behind the active
can lead to malfunction in its enzymatic activity. site if the enzyme. Binding the metabolites to the active site
of the protease has also changed the fluctuation of the
important residue of Cys145 and its milieu amino acids. All
Root mean squared fluctuation (RMSF)
ligands besides the Purpureone and Viriditoxin have
The relative displacement of the residues to the mean of increased the residual fluctuations in this site.
their displacement in the entire time of simulation can be
evaluated by root mean squared fluctuation analysis. In other
Radius of gyration
word the RMSF specifies that which amino acid in the pro­
tein has experienced more or less mobility during the simu­ Attaching of the ligands can cause swelling or compression
lation. The results of the RMSF analysis are shown in Figure 3 in the protein structure and this can be evaluated by meas­
and from these results it can be seen that binding the uring its radius of gyration. The results of Rg for ligand-pro­
metabolites to protein has decreased the residue fluctuation tein complex in comparison with free protein are
in most parts of the protein. The fluctuations of the region represented in Figure 4. In the case of free protein, the value
between residue 240–265 in all complexes besides of Rg is decreased during the simulation by about 10 Ao. The
Purpureone and Viriditoxin are decreased. The sequence binding of different metabolites to the protein have changed
belongs to an alpha helix in protein domain II which is the Rg pattern of protein in related to free protein. Despite
6 K. S. EBRAHIMI ET AL.

Figure 1. Continued.

of decrement in Rg value at the early nanoseconds of the swelling-compression in its structure during the simulation.
simulations in most cases at the end of time all metabolite Such events can lead to instability in protein structure or dis­
binding systems have equal Rg values with free protein. The rupt the attachments of the enzyme to its substrate.
exceptions for this are the proteins in complex with
Purpureone and RKS-1778. This means that the binding of
Principal component analysis (PCA)
metabolites prevents the compression in the protein struc­
ture in these systems. In the case of RKS-1778, Beauvericin, The principal component analysis is extensively used as a
Pyrrocidine A and Viriditoxin, the molecules caused the most useful tool in order to analyze the main motional pattern of
severe fluctuations in the Rg pattern that is implying on the proteins. This is because the main motional component
the fact that protein undergoes alternating courses of is specific for each protein and has a critical rule in their
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 7

Figure 1. Continued.

functions. The results of Eigenvalues related to the first 6 first two PCs have more eigenvalues and can be used for fur­
eigenvectors are represented in the Figure S1. As can be ther analysis on the other properties of protein movement
seen the eigenvalues are significantly decreased when the components. The results of two dimensional PCA analyses for
protein is in interaction with ligands. This indicates a per­ corona virus main protease if free and complex with different
turbation in the pattern of regular protein movement that secondary metabolites is represented in Figure 5. As can be
can lead to disruption in substrate binding or catalytic activ­ seen from the figure, the protein has well fitted along with
ity of the enzyme. From the figure it is also confirms that the the first PC and in two distinct movement clusters. Binding
8 K. S. EBRAHIMI ET AL.

ligands to the enzyme in some cases have changed the PCA dependent changes in the SASA value of SARS-CoV-2 main
pattern of protein in different manners. In the case of B-viri­ protease in complex with different ligands is represented in
din, Thielavins J and Beauvericin the results showed that the Figure 6. As this value is commonly related to swilling and
dispersion pattern of protein has changed in shape and also compression of the protein structure, its changes are mainly
the value of expansion along the first two PCs have similar to that of the Rg diagram. Also in this study it can
increased. In contrast the PC patterns of protein in inter­ be seen that the diagram of Rg and SASA analyzes are in
action with RKS-1778, Thielavins A, Pyrrocidine A and line and in both analysis. As can be seen the SASA fluctua­
Viriditoxin showed significant discretion in 2D dispersion tions in the presence of ligands RKS-1778, Beauvericin,
along the first two PCs. These changes in movement compo­ Pyrrocidine A and Viriditoxin are higher than other
nents can significantly affect the protein function and also a complexes.
tendency to its substrate which together potentially lead to
prevent its correct protease activity.
Analysis the secondary structure of protein
Solvent accessible surface area (SASA) The functional structures of proteins are consisting of several
The value of the protein surface to be accessible for solvent secondary structures including Alpha Helixes, Beta Sheets,
molecules can be measured using SASA analysis. Time bends, turns and coils. Any removal or displacement in the
amount and location of these structures can seriously impair
Table 3. The mean interaction energies of drug binding to protein obtained protein function. The effects of selected fungal metabolites
from the MMPBSA method.
on the secondary structure of the coronavirus main protease
VdW (KJ/mol) Elec (KJ/mol) Total
were evaluated by using of DSSP analysis (supporting data:
GKK1032B −171.7 −42.5 −214.2
b-viridin −150.8 −31.6 −181.6 Figures S2–S11). From the results it can be seen that an
Purpureone −167.0 −111.7 280.7
alpha helix (Blue colored in figures) in the position 245–255
RKS-1778 −252.2 −53.3 2305.5
Thielavins A −195.6 −73.6 −269.2 is completely degraded after the protein is bind to RKS-1778.
Thielavins J −217.9 −62.1 −280.0
Beauvericin −272.1 −39.8 2311.9
The compound Viriditoxin has decreased two beta sheet
Pyrrocidine A −169.7 −60.3 −230.0 structures (Red colored in figures) in the position of residues
Viriditoxin −245.7 −78.5 2324.2
110–130.

Figure 2. Changes in Root Mean Squared Deviation of main chain atoms of protein in the presence of (a) Gkk1032B, (b) B-viridin, (c) Purpureone, (d) RKS-1778, (e)
Thielavins A, (f) Thielavins J, (g) Beauvericin, (h) Pyrrocidine A and (i) Viriditoxin.
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 9

Figure 3. Comparison of residue fluctuation in different systems of (a) Gkk1032B, (b) B-viridin (c) Purpureone, (d) RKS-1778, (e) Thielavins A, (f) Thielavins J, (g)
Beauvericin, (h) Pyrrocidine A and (i) Viriditoxin.

Figure 4. Comparison of changes in Rg value of protein in interaction with different ligands (a) Gkk1032B, (b) B-viridin (c) Purpureone, (d) RKS-1778, (e) Thielavins
A, (f) Thielavins J, (g) Beauvericin, (h) Pyrrocidine A and (i) Viriditoxin.
10 K. S. EBRAHIMI ET AL.

Figure 5. Principle component analysis of the different protein ligand complex in comparison with ligand free protein, (a) Gkk1032B, (b) B-viridin, (c) Purpureone,
(d) RKS-1778, (e) Thielavins A, (f) Thielavins J, (g) Beauvericin, (h) Pyrrocidine A and (i) Viriditoxin.

Interaction energies absorption but can potentially inhibit the activity of CYP2C9
which is a metabolic enzyme involved in the metabolism of
The mean dynamic interaction energies between secondary
several drugs such as NSAIDs.
metabolites and viral protease were computed using the
Finally the druglikness of 9 selected fungal metabolites
MMPBSA method and the results are reported in Table 3.
were predicted based on different methods and the results
From these results it is clear that three of the compounds
are reported in Table 5. Based on the data in this table it
including RKS-1778, Beauvericin and Viriditoxin have more can be seen that B-viridin, RKS-1778 and PyrrocidineA have
potent interaction with the protein than others. It should be passed the much of the factors defined in different methods
noted that the other metabolites also have made a strong and it is predicted to show better bioavailability properties
interaction with protein. The difference between the than others.
obtained binding energies from docking analysis and MD
results indicated that there is some relocation in ligand pos­
ition on the protein during the simulation. Conclusion
To propose natural based drug candidates for COVID-19 dis­
ease, the inhibitory potential of 99 secondary metabolites
Pharmacokinetic properties
extracted from endophytic fungi were evaluated by using
The 9 selected secondary metabolites were then evaluated computational methods against Mpro of SARS-CoV-2. Eleven
to predict their pharmacokinetic properties and the results compounds were identified and selected through molecular
are reported in Tables 4 and 5. As in Table 4 better pharma­ docking with the highest binding energy to the active site of
cokinetic is predicted for compounds B-viridin and the enzyme. The molecular dynamics simulation study
Beauvericin which have high GI absorption and are not sub­ showed that the selected compounds change the dynamic
strate for any CYP metabolite enzyme. Also the compounds behavior of protein in terms of RMSD, RMSF, Rg, SASA and
Gkk1032B, RKS-1778 and Pyrrocidine A showed high GI PCA. In this case the ligands RKS-1778, Beauvericin,
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 11

Figure 6. Comparison of changes in SASA value of protein in interaction with different ligands (a) Gkk1032B, (b) B-viridin (c) Purpureone, (d) RKS-1778, (e)
Thielavins A, (f) Thielavins J, (g) Beauvericin, (h) Pyrrocidine A and (i) Viriditoxin.

Table 4. Pharmacokinetic prediction of 9 selected fungal metabolites.

No. compound GI absorption CYP1A2 inhibitor CYP2C19 inhibitor CYP2C9 inhibitor CYP2D6 inhibitor CYP3A4 inhibitor
1 Gkk1032B High No No Yes No No
2 B-viridin High No No No No No
3 Purpureone Low No No No No No
4 RKS-1778 High No No Yes No No
5 Thielavins A Low No No No No No
6 Thielavins J Low No No No No No
7 Beauvericin High No No No No No
8 Pyrrocidine A High No No Yes No No
9 Viriditoxin Low No No No No No

Table 5. Drug likeness property of 9 selected fungal metabolites. less counter indication with other drugs. Therefore as a con­
No. compound Lipinski Ghose Veber Egan Muegge Bioavailability score clusion of all results obtained in this study it can be suggested
1 Gkk1032B No No Yes Yes No 0.17 that the compounds RKS-1778 can be a good candidate as an
2 B-viridin Yes Yes Yes Yes Yes 0.55
3 Purpureone No No Yes No No 0.17
Mpro inhibitor, and this requires further experiments to inves­
4 RKS-1778 Yes No Yes Yes No 0.55 tigate its efficiency in wet laboratory conditions.
5 Thielavins A Yes No No No No 0.11
6 Thielavins J Yes No No No No 0.11
7 Beauvericin No No Yes Yes No 0.17 Disclosure statement
8 PyrrocidineA Yes No Yes Yes No 0.55
9 Viriditoxin No No Yes No No 0.17 No potential conflict of interest was reported by the author(s).

Pyrrocidine A and Viriditoxin showed more potential to des­ Funding

troy the structure and dynamic of protein. The results of the The authors gratefully acknowledge Kermanshah University of Medical
mmpbsa method showed that the compounds RKS-1778, Sciences for financial support. Grant number [4020578].
Beauvericin and Viriditoxin have more stable complex with
Mpro of SARS-CoV-2 than others. In addition, investigating the
pharmacokinetic and druglikness properties predicted that References
overall the compounds B-viridin, RKS-1778 and Pyrrocidine A Bano, N., Rizvi, I. F., Sharma, N., Siddiqui, M. H., Kalim, M., Khan, A., &
had the best results in the case of better bioavailability and Akhtar, S. (2016). Production of bioactive secondary metabolites from
12 K. S. EBRAHIMI ET AL.

endophytic fungi. International Research Journal of Engineering and Molecular Modeling, 7(8), 306–317. https://doi.org/10.1007/
Technology, 3(6), 1859–1866. s008940100045
Chen, L., Zhang, Q.-Y., Jia, M., Ming, Q.-L., Yue, W., Rahman, K., Qin, L. P., Liu, X., & Wang, X.-J. (2020). Potential inhibitors against 2019-nCoV cor­
& Han, T. (2016). Endophytic fungi with antitumor activities: Their onavirus M protease from clinically approved medicines. Journal of
occurrence and anticancer compounds. Critical Reviews in Genetics and Genomics, 47(2), 119–121. https://doi.org/10.1016/j.jgg.
Microbiology, 42(3), 454–473. 2020.02.001
Chen, T., Wu, D., Chen, H., Yan, W., Yang, D., Chen, G., Ma, K., Xu, D., Yu, Mark, P., & Nilsson, L. (2001). Structure and dynamics of the TIP3P, SPC,
H., Wang, H., Wang, T., Guo, W., Chen, J., Ding, C., Zhang, X., Huang, and SPC/E water models at 298 K. The Journal of Physical Chemistry A,
J., Han, M., Li, S., Luo, X., Zhao, J., & Ning, Q. (2020). Clinical character­ 105(43), 9954–9960. https://doi.org/10.1021/jp003020w
istics of 113 deceased patients with coronavirus disease 2019: Mollazadeh, S., Sahebkar, A., Shahlaei, M., & Moradi, S. (2021). Nano drug
Retrospective study. BMJ, 368, m1091. https://doi.org/10.1136/bmj. delivery systems: Molecular dynamic simulation. Journal of Molecular
m1091 Liquids, 332, 115823. https://doi.org/10.1016/j.molliq.2021.115823
Da Silva, R. R. (2017). Bacterial and fungal proteolytic enzymes: Morris, G. M., Goodsell, D. S., Halliday, R. S., Huey, R., Hart, W. E., Belew,
Production, catalysis and potential applications. Applied Biochemistry R. K., & Olson, A. J. (1998). Automated docking using a Lamarckian
and Biotechnology, 183(1), 1–19. https://doi.org/10.1007/s12010-017- genetic algorithm and an empirical binding free energy function.
2427-2 Journal of Computational Chemistry, 19(14), 1639–1662. https://doi.
Dai, W., Zhang, B., Jiang, X.-M., Su, H., Li, J., Zhao, Y., Xie, X., Jin, Z., Peng, org/10.1002/(SICI)1096-987X(19981115)19:14<1639::AID-JCC10>3.0.
J., Liu, F., Li, C., Li, Y., Bai, F., Wang, H., Cheng, X., Cen, X., Hu, S., Yang, CO;2-B
X., Wang, J., … Liu, H. (2020). Structure-based design of antiviral Negreiros de Carvalho, P. L., de Oliveira Silva, E., Aparecida Chagas-Paula,
drug candidates targeting the SARS-CoV-2 main protease. Science, D., Honorata Hortolan Luiz, J., & Ikegaki, M. (2016). Importance and
368(6497), 1331–1335. https://doi.org/10.1126/science.abb4489 implications of the production of phenolic secondary metabolites
Demain, A. L. (2014). Importance of microbial natural products and the by endophytic fungi: A mini-review. Mini Reviews in Medicinal
need to revitalize their discovery. Journal of Industrial Microbiology & Chemistry, 16(4), 259–271. https://doi.org/10.2174/138955751566615
Biotechnology, 41(2), 185–201. https://doi.org/10.1007/s10295-013- 1016123923
1325-z Nguyen, T. T., Viet, M. H., & Li, M. S. (2014). Effects of water models on
Ebrahimi, K. S., Ansari, M., Hosseyni Moghaddam, M. S., Ebrahimi, Z., binding affinity: Evidence from all-atom simulation of binding of tami­
Salehi, Z., Shahlaei, M., & Moradi, S. (2021). In silico investigation on flu to A/H5N1 neuraminidase. The Scientific World Journal, 2014, 1–14.
the inhibitory effect of fungal secondary metabolites on RNA depend­ https://doi.org/10.1155/2014/536084
ent RNA polymerase of SARS-CoV-II: A docking and molecular Nos�e, S. (1984). A unified formulation of the constant temperature
dynamic simulation study. Computers in Biology and Medicine, 135, molecular dynamics methods. The Journal of Chemical Physics, 81(1),
104613. https://doi.org/10.1016/j.compbiomed.2021.104613 511–519. https://doi.org/10.1063/1.447334
Evans, D. J., & Holian, B. L. (1985). The nose–hoover thermostat. The Parrinello, M., & Rahman, A. (1980). Crystal structure and pair potentials:
Journal of Chemical Physics, 83(8), 4069–4074. https://doi.org/10.1063/ A molecular-dynamics study. Physical Review Letters, 45(14), 1196–
1.449071 1199. https://doi.org/10.1103/PhysRevLett.45.1196
Fabricant, D. S., & Farnsworth, N. R. (2001). The value of plants used in Rodriguez, R., White, J., Arnold, A. E., & Redman, A R. A (2009). Fungal
traditional medicine for drug discovery. Environmental Health endophytes: Diversity and functional roles. The New Phytologist,
Perspectives, 109(suppl 1), 69–75. https://doi.org/10.1289/ehp.01109s169 182(2), 314–330. https://doi.org/10.1111/j.1469-8137.2009.02773.x
Gallagher, T. M., & Buchmeier, M. J. (2001). Coronavirus spike proteins in Sch€uttelkopf, A. W., & Van Aalten, D. M. (2004). PRODRG: A tool for high-
viral entry and pathogenesis. Virology, 279(2), 371–374. https://doi. throughput crystallography of protein–ligand complexes. Acta
org/10.1006/viro.2000.0757 Crystallographica, 60(Pt 8), 1355–1363. https://doi.org/10.1107/
Gombar, V. K., Silver, I. S., & Zhao, Z. (2003). Role of ADME characteristics S0907444904011679
in drug discovery and their in silico evaluation: In silico screening of Shaker, B., Ahmad, S., Lee, J., Jung, C., & Na, D. (2021). In silico methods
chemicals for their metabolic stability. Current Topics in Medicinal and tools for drug discovery. Computers in Biology and Medicine, 137,
Chemistry, 3(11), 1205–1225. https://doi.org/10.2174/1568026033452014 104851. https://doi.org/10.1016/j.compbiomed.2021.104851
Hess, B., Bekker, H., Berendsen, H. J., & Fraaije, J. G. (1997). LINCS: A lin­ Sharma, N., Sharma, V., Abrol, V., Panghal, A., & Jaglan, S. (2019). An
ear constraint solver for molecular simulations. Journal of update on bioactive natural products from endophytic fungi of medi­
Computational Chemistry, 18(12), 1463–1472. https://doi.org/10.1002/ cinal plants. In Pharmaceuticals from microbes: Impact on drug discov­
(SICI)1096-987X(199709)18:12<1463::AID-JCC4>3.3.CO;2-L ery (pp. 121–153). Springer. https://doi.org/10.1007/978-3-030-04675-
Hirshman, S. P., & Whitson, J. (1983). Steepest-descent moment method 0_6
for three-dimensional magnetohydrodynamic equilibria. The Physics of Sliwoski, G., Kothiwale, S., Meiler, J., & Lowe, E. W. (2014). Computational
Fluids, 26(12), 3553–3568. https://doi.org/10.1063/1.864116 methods in drug discovery. Pharmacological Reviews, 66(1), 334–395.
Hoover, W. G. (1985). Canonical dynamics: Equilibrium phase-space distri­ https://doi.org/10.1124/pr.112.007336
butions. Physical Review. A, General Physics, 31(3), 1695–1697. https:// Su, S., Wong, G., Shi, W., Liu, J., Lai, A. C. K., Zhou, J., Liu, W., Bi, Y., &
doi.org/10.1103/physreva.31.1695 Gao, G. F. (2016). Epidemiology, genetic recombination, and patho­
Huang, X., Wei, F., Hu, L., Wen, L., & Chen, K. (2020). Epidemiology and genesis of coronaviruses. Trends in Microbiology, 24(6), 490–502.
clinical characteristics of COVID-19. Archives of Iranian Medicine, 23(4), https://doi.org/10.1016/j.tim.2016.03.003
268–271. https://doi.org/10.34172/aim.2020.09 Van Gunsteren, W. F., & Berendsen, H. J. (1988). A leap-frog algorithm for
Khaledian, S., Kahrizi, D., Moradi, S., & Martinez, F. (2021). An experimen­ stochastic dynamics. Molecular Simulation, 1(3), 173–185. https://doi.
tal and computational study to evaluation of chitosan/gum traga­ org/10.1080/08927028808080941
canth coated-natural lipid-based nanocarriers for sunitinib delivery. Weiner, S. J., Kollman, P. A., Case, D. A., Singh, U. C., Ghio, C., Alagona,
Journal of Molecular Liquids, 334, 116075. https://doi.org/10.1016/j. G., Profeta, S., & Weiner, P. (1984). A new force field for molecular
molliq.2021.116075 mechanical simulation of nucleic acids and proteins. Journal of the
Laskowski, R. A., & Swindells, M. B. (2011). LigPlotþ: Multiple ligand–pro­ American Chemical Society, 106(3), 765–784. https://doi.org/10.1021/
tein interaction diagrams for drug discovery. ACS Publications. ja00315a051
Li, S.-J., Zhang, X., Wang, X.-H., & Zhao, C.-Q. (2018). Novel natural com­ Wu, F., Zhao, S., Yu, B., Chen, Y.-M., Wang, W., Song, Z.-G., Hu, Y., Tao, Z.-
pounds from endophytic fungi with anticancer activity. European W., Tian, J.-H., Pei, Y.-Y., Yuan, M.-L., Zhang, Y.-L., Dai, F.-H., Liu, Y.,
Journal of Medicinal Chemistry, 156, 316–343. https://doi.org/10.1016/j. Wang, Q.-M., Zheng, J.-J., Xu, L., Holmes, E. C., & Zhang, Y.-Z. (2020).
ejmech.2018.07.015 A new coronavirus associated with human respiratory disease in
Lindahl, E., Hess, B., & Van Der Spoel, D. (2001). GROMACS 3.0: A pack­ China. Nature, 579(7798), 265–269. https://doi.org/10.1038/s41586-
age for molecular simulation and trajectory analysis. Journal of 020-2008-3
 JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS 13

Xie, X.-G., Huang, C.-Y., Fu, W.-Q., & Dai, C.-C. (2016). Potential of endo­ Zhou, P., Yang, X.-L., Wang, X.-G., Hu, B., Zhang, L., Zhang, W., Si, H.-R.,
phytic fungus Phomopsis liquidambari for transformation and degrad­ Zhu, Y., Li, B., Huang, C.-L., Chen, H.-D., Chen, J., Luo, Y., Guo, H.,
ation of recalcitrant pollutant sinapic acid. Fungal Biology, 120(3), Jiang, R.-D., Liu, M.-Q., Chen, Y., Shen, X.-R., Wang, X., … Shi, Z.-L.
402–413. https://doi.org/10.1016/j.funbio.2015.11.010 (2020). A pneumonia outbreak associated with a new coronavirus of
Xu, B., Gutierrez, B., Mekaru, S., Sewalk, K., Goodwin, L., Loskill, A., probable bat origin. Nature, 579(7798), 270–273. https://doi.org/10.
Cohn, E. L., Hswen, Y., Hill, S. C., Cobo, M. M., Zarebski, A. E., Li, 1038/s41586-020-2012-7
S., Wu, C.-H., Hulland, E., Morgan, J. D., Wang, L., O’Brien, K., Zhu, N., Zhang, D., Wang, W., Li, X., Yang, B., Song, J., Zhao, X., Huang,
Scarpino, S. V., Brownstein, J. S., … Kraemer, M. U. G. (2020). B., Shi, W., Lu, R., Niu, P., Zhan, F., Ma, X., Wang, D., Xu, W., Wu, G.,
Epidemiological data from the COVID-19 outbreak, real-time case Gao, G. F., & Tan, W. (2020). A novel coronavirus from patients with
information. Scientific Data, 7(1), 106. https://doi.org/10.1038/ pneumonia in China, 2019. New England Journal of Medicine, 382(8),
s41597-020-0448-0 727–733. https://doi.org/10.1056/NEJMoa2001017