Molecular Diversity (2022) 26:2775–2792

https://doi.org/10.1007/s11030-022-10394-9

ORIGINAL ARTICLE

Systematic virtual screening in search of SARS CoV‑2 inhibitors

against spike glycoprotein: pharmacophore screening, molecular
docking, ADMET analysis and MD simulations
Tejas M. Dhameliya1 · Prinsa R. Nagar1 · Normi D. Gajjar1

Received: 8 July 2021 / Accepted: 22 January 2022 / Published online: 8 February 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022

Abstract
In the absence of efficient anti-viral medications, the coronavirus disease 2019 (COVID-19), stemming from severe acute
respiratory syndrome coronavirus-2 (SARS CoV-2), has spawned a worldwide catastrophe and global emergency. Amidst
several anti-viral targets of COVID-19, spike glycoprotein has been recognized as an essential target for the viral entry into
the host cell. In the search of effective SARS CoV-2 inhibitors acting against spike glycoprotein, the virtual screening of
175,851 ligands from the 2020.1 Asinex BioDesign library has been performed using in silico tools like SiteMap analysis,
pharmacophore-based screening, molecular docking using different levels of precision, such as high throughput virtual
screening, standard precision and extra precision, followed by absorption, distribution, metabolism, excretion and toxicity
analysis, and molecular dynamics (MD) simulation. Following a molecular docking study, seventeen molecules (with a
docking score of less than − 6.0) were identified having the substantial interactions with the catalytic amino acid and nucleic
acid residues of spike glycoprotein at the binding site. In investigations using MD simulations for 10 ns, the hit molecules
(1 and 2) showed adequate compactness and uniqueness, as well as satisfactory stability. These computational research
findings have offered a key starting point in the field of design and development of novel SARS CoV-2 entry inhibitors with
appropriate drug likeliness.

* Tejas M. Dhameliya
[email protected]; [email protected]
1
L. M. College of Pharmacy, Navrangpura, Ahmedabad,
Gujarat 380009, India

13
Vol.:(0123456789)
2776 Molecular Diversity (2022) 26:2775–2792

Graphical abstract

Keywords COVID-19 · SARS CoV-2 · Molecular docking · MD simulations · Spike glycoprotein

Abbreviations HERG Human ether-a-go-go-related gene

ACE2 Angiotensin-converting enzyme 2 HOA Human oral absorption
ADMET Absorption, distribution, metabolism, HTVS High-throughput virtual screening
excretion and toxicity M Membrane protein
ASL Atom specification language N Nucleocapsid
BBB Blood brain barrier nCoV-19 Novel coronaviruses
COVID-19 Coronavirus disease 2019 NSPs Non-structural proteins
E Envelope protein OPLS3e Optimized potentials for liquid simulations
FDA Food and drugs administration pp1a/b Polyproteins
GLIDE Grid-based ligand docking with energetics RdRp RNA-dependent RNA polymerase
HBA Hydrogen bond acceptor S Spike glycoprotein
HBD Hydrogen bond donor

13
Molecular Diversity (2022) 26:2775–2792 2777

SARS CoV-2 Severe acute respiratory syndrome contamination have been regarded as the ways for the illness
coronavirus-2 to spread among human beings [20].
SP Standard precision Despite the devastating effects, no viable medicines to
ssRNA Single-stranded ribonucleic acid tackle COVID-19 have been identified so far. There are cur-
TMPRSS2 Transmembrane protease serine type 2 rently sixty-three diversified vaccines containing inactivated
UTR​ Untranslated region viruses (Covaxin), adeno-based vaccines (Gam-Covid by
XP Extra precision Gameleya Institute), recombinant (Novavax), mRNA (Mod-
erna) and live attenuated vaccines that have been approved
on a global basis by regulatory authorities in order to effec-
tively diminish or uproot the effect of COVID-19. More than
Introduction twenty vaccinations are in phase III clinical investigations
and will be introduced with better efficacy and safety in the
The novel coronavirus disease 2019 (COVID-19), caused future. However, repurposing of existing anti-viral drugs
by deadly, pathogenic and infectious novel coronaviruses and development of vaccines could not act straightforwardly
(nCoV-19), has caused an unprecedented pandemic world- proving themselves an effective therapy to prevent the virus
wide due to increased virulence, morbidity and rapid spread for making its way to animal species through binding with
beyond the previous severe acute respiratory syndrome spike glycoprotein. Additionally, they bring noticeable ill
(SARS) pandemic of 2002–03 [1, 2]. The very first outbreak effects on the imperative physiological processes of human
was reported on December 31, 2019, and suspected to be body upon administration such as revering to the deficit of
assembled in the laboratory of Wuhan, China [3–6]. Citing potential therapy in this line of approach; there is an urgent
the exceedingly concerning situation, COVID-19 has been need for the development of promising candidates [21, 22].
designated as an emergency and a worldwide pandemic by To become an FDA-approved drug, the molecules must
the World Health Organization (WHO) on March 11, 2020 have to pass through the hectic discovery process through
[7]. According to the latest WHO estimates, there were conventional methods at the cost of billions of US dollars
326,279,424 confirmed cases of COVID-19 worldwide as that too might not guarantee a potential therapy. On the
of January 17, 2022, with 5,536,609 fatalities [8]. The severe other hand, drug discovery through the cost effective virtual
effects of quarantine-related counter-measures imposed by screenings could serve the purpose of novel drug discovery
ruling governments, such as lock-down, sanitization of for the eradication of SARS CoV-2 proving itself a need
public premises, social distancing, closure of academies or of an hour [22–25]. In order to get an effective therapeutic
sports clubs, cancellation of public or social events, etc., approach, critical insights into the viral entry and replication
have severely hampered normal life around the world. As cycle into host’s body have been urged to understand the
a result, the socioeconomic consequences of the epidemic pathogenesis of SARS CoV-2 (Fig. 1). The viral replica-
have steadily damaged societal health, education, economy tion process starts with the entry of virion into host cells
and several vital sectors have been greatly challenged or via spike glycoprotein (S), which is followed by the use of
ceased [9, 10]. host cell machinery to assemble and reproduce multiple viral
Coronaviruses (CoVs) are single-strand positive-sense parasites. Spike protein, which is made up of two essential
RNA enclosed viruses that belong to a large family [11]. subunits called S1 and S2, facilitates in virus binding and
Being club-shaped glycoprotein, they are around 60–140 nm fusion to the host cell membrane, which is accomplished
in diameter with the crown-like morphology of the virion as via endocytosis or cleavage at the interface of the two subu-
observed using electron microscopy [12, 13]. The first strain nits by host proteases such as transmembrane protease ser-
of SARS coronavirus species was identified in 2002–3 [14]. ine type 2 (TMPRSS2) and cathepsin. Furthermore, the
As a spillover disease, it has spread throughout the world released 30 kb positive-sense single-stranded ssRNA (5′-3′
through a variety of hosts, including bats, civets, camels, and UTR) into the host cytoplasm uses host cell machinery to
among others [15, 16]. The huge subfamily of CoVs belongs translate into polyproteins, which are then proteolytically
to the Nidovirales family, divided further into four primary cleaved into sixteen distinct non-structural proteins by the
genuses: alpha, beta, gamma and delta-coronaviruses, proteases encoded by the virus (NSPs). These NSPs con-
among which SARS CoV-2 belongs to β-coronaviruses [4, struct a replicase–transcriptase complex, which incorporates
17]. Patients with the condition may experience a variety of RNA-dependent RNA polymerase (RdRp), helicases, endo
clinical manifestations, ranging from minor to severe symp- and exonucleases, and other important enzymes involved in
toms such as headaches, shortness of breath, muscular pains, nucleic acid metabolism. Open reading frames required for
fever and tiredness, to multi-organ failure [18, 19]. Inhala- transcription into structural proteins such as spike (S), mem-
tion of respiratory droplets, usage of personal belongings of brane (M), nucleocapsid (N) and envelope (E) are encoded at
the infected patients, direct or indirect physical exposure and the 3′-end of the genome (E). The proteins generated in the

13
2778 Molecular Diversity (2022) 26:2775–2792

Fig. 1  Life cycle of SARS CoV-2 from entry to replication into host cells

O attachment and entry of coronavirus to human cells [28].

HN
The multidisciplinary approach performed by Haselhorst
et al. via screening of 57,641 ligands to inhibit human ACE2
O or spike protein led to identification of evans blue (B, Fig. 2)
HO3S
A
N as potent anti-viral agent with ­IC50 of 28.1 µM against vero-
H
NO2
E6 cells incubated with SARS CoV-2 [29]. Three differ-
ent peptide fragments composed of N-terminal amino acid
H2N SO3Na residues of α1 helix of ACE2 peptidase domain have been
HO
reported by de Olivera and co-workers as effective agents
N to inhibit the binding of SARS CoV-2 spike protein with
NaO3S N N SO3Na human ACE2 as identified through molecular modelling
N [30]. Further, Baig et al. identified the 13-amino acid pep-
OH tide (FLDKFNHEAEDLF) as spike protein inhibitor with
B
NaO3S NH2 minimal inhibition (40% at 100 µM concentration) of ACE2-
spike protein–protein interactions through enzyme-linked
Fig. 2  Reported spike-ACE2 inhibitors inhibiting SARS CoV-2 immunosorbent (ELISA)-dependent inhibitory assay [31].
attachments and their entries in host cells The use of computational studies has promoted the
design and discovery of new chemicals scaffolds acting
against the choice of drug targets through molecular model-
endoplasmic reticulum (ER) of the host cell are transformed ling techniques like molecular docking, ADMET analysis
into new viral offspring, which are then released into ER- and molecular dynamics (MD) simulation [32]. Recently,
golgi compartments and infect additional host cells [26, 27] . we have reported the virtual screening of small molecules
In search of SARS CoV-2 inhibitors, Buchwald and co- against RdRp [33] and Mpro [34] using molecular dock-
workers have performed the virtual screening of library ing, ADMET analysis and MD simulations. In continuation
consisting of organic dyes and reported p-nitrobenzamide towards our endeavour for search of SARS CoV-2 inhibi-
derivative (A, Fig. 2) with the I­ C50 of 5.6 µM, as spike- tors, we become interested for the in silico-based virtual
ACE2 protein–protein interaction blockers inhibiting the screening of 175,851 ligands of Asinex BioDesign library

13
Molecular Diversity (2022) 26:2775–2792 2779

considering the massive demand for SARS CoV-2 inhibitors glycoprotein to obtain the steric and electrostatic attributes
having good binding affinity towards the spike glycoprotein. [38]. The model with seven features (RRR​RRR​H) has been
generated comprising of six aromatic rings (R) and a hydro-
phobic feature (H, see supporting information, Fig. S2).
Results and discussion Total 175,851 numbers of compounds have been obtained
from Asinex BioDesign library 2020.1 from online avail-
Pharmacophore hypothesis able sources [39]. With a view of validation of the devel-
oped hypothesis, all the compounds have been processed
SiteMap analysis through preparation of ligands using LigPrep [40] followed
by screening using phase [38]. Total 38,267 molecules from
The protein structure of spike protein (PDB ID: 6VXX) hav- the selected database have matched with the minimum four
ing resolution of 2.8 Å with the symmetrical trimeric struc- features of the developed hypothesis, and they have been
ture possessing two receptor binding domains (S1 and S2 further utilized for the molecular docking.
domains) has been deposited in protein data bank by Veesler
et al. [35]. There has not been any ligand co-crystallized Molecular docking
with the protein of interest. As a result, attempts have been
made to determine the probable binding site of spike gly- Molecular modelling using docking [41–44] has been con-
coprotein using the SiteMap tool of Schrödinger [36]. Five sidered as an important computational tool to predict the
possible active sites have been predicted, and their druga- binding interactions of the ligands with the active site of pro-
bility scores (Dscore) and SiteScores have been calculated tein in search of their mode of action against SARS CoV-2
by consideration of the pocket volume, enclosure, and the [45–52]. In this context, we have performed the molecular
degree of hydrophobicity (Table 1) [37]. Next, the site hav- docking in the search of potent SARS CoV-2 inhibitors using
ing the highest Dscore (1.103642) has been utilized for the the obtained hypothesis model. The receptor grid box with
pharmacophore hypothesis development which includes 10 Å has been generated to perform the molecular docking
residues Lys41, Ile197, Asp198, Tyr200, Lys202, Asp228, on the selected active site from the SiteMap analysis using
Arg355, Cys379, Tyr380, Gly381, Val382, Ser383, Tyr396, the Grid-based Ligand Docking with Energetics (GLIDE)
Ala411, Pro412, Gly413, Gln414, Thr415, Ala419, Asp420, module of Schrödinger [53]. To get the detailed knowledge
Lys424, Pro426, Asp427, Asp428, Phe429, Thr430, Phe464, about the binding strength and types of interactions of the
Ser514, Phe515, Glu516, Leu517, Leu518, His519, Glu748, ligand with receptor, we have performed sequential docking
Asn751, Leu752, Gln755, Tyr756, Gln762, Leu763, Arg765, at three different precision levels such as high throughput
Ala766, Gly769, Ile770, Glu773, Val951, Gln954, Gln957, virtual screening (HTVS), standard precision (SP) and extra
Ala958, Thr961, Leu962, Gln965, Phe970, Gly971, Ile973, precision (XP). First of all, the HTVS has been performed
Ser974, Asp979, Leu981, Arg983, Leu984, Asp985, Pro986, using 38,267 ligands, qualified in phase screening, from
Pro987, Glu988, Ala989, Glu990, Val991, Gln992, Asp994, which 1094 molecules have been found with the docking
Arg995, Thr998, Gly999, Leu1001, Gln1002, Ser1003, score of ≤ − 4.6. Next, they have been further processed for
Gln1005, Thr1006, Tyr1007, Val1008, Thr1009, Gln1010, docking with SP mode and 175 molecules have been identi-
Gln1011, Leu1012, Ile1013, Arg1014, Ala1015, Ala1016, fied with the docking score ≤ − 5.9 in SP docking which have
Glu1017, Ile1018 and Arg1019 (Table 1, Entry 1). been passed through the XP docking with the generation
of 10 poses for each ligand molecule. From that, seventeen
Pharmacophore development and ligand screening compounds with highest docking scores (≤ − 6.0) have been
identified and summarized in Table 2, and 2-D interactions
Next, using the phase module of Schrödinger, an energy- are presented in Figs. 3, 4 and 5.
based pharmacophore (e-pharmacophore) hypothesis model Further, we have studied the detailed 3D interactions of
has been generated around the predicted binding site of spike the docked ligands having the docking score ≤ − 6.0 using
the PyMol 2.4.0 [54]. Compound 1 has shown the high-
est docking score of − 7.34 by formation of the hydrogen
Table 1  Predicted five sites Entry SiteScore Dscore bond (HB) interactions with Phe970 and Asp994 along
from the sitemap analysis for
the selected spike protein (PDB with the two salt bridges with Asp994 as presented in
1 1.069965 1.103642
ID: 6VXX) Fig. 6a. Hydroxy group and nitrogen atoms were involved
2 1.054111 1.057707
in the bond formation. The nitrogen atom present in the
3 1.035711 1.025884
five membered ring of compound 2 has interacted with
4 1.128571 1.011641
Arg995 through π-cation interaction and ionic inter-
5 1.027926 0.997882
actions (salt bridge) with Asp994 (Fig. 6b). Another

13
2780 Molecular Diversity (2022) 26:2775–2792

Table 2  Molecular docking Compd. Docking G

result of identified hits with the Dataset ID 2D-Structures
No. Scores score
docking score ≤ 6 in XP
LAS
1 -7.34 -7.43
52153344

LAS
2 -7.25 -7.34
52165865

LAS
3 -7.09 -7.09
33586443

BDE
4 -6.78 -6.86
33953587

BDD
5 -6.77 -6.88
27849696

BDE
6 -6.63 -6.64
19706336

LAS
7 -6.62 -6.93
52167276

BDE
8 -6.54 -6.55
19706291

BDF
9 -6.53 -6.54
26908734

LAS
10 -6.50 -6.63
52164699

LAS
11 -6.48 -7.52
52126337

LAS
12 -6.33 -6.35
52160955

LAS
13 -6.23 -6.37
52154116

13
Molecular Diversity (2022) 26:2775–2792 2781

Table 2  (continued) Compd. Docking G

Dataset ID 2D-Structures
No. Scores score

LAS
14 -6.06 -6.08
34121280

LAS
15 -6.06 -6.20
52165156

BDD
16 -6.04 -6.04
22945640

LAS
17 -6.03 -6.13
52160889

Fig. 3  2D interactions of compounds having docking score of ≤ − 6.0 using XP module against spike glycoprotein. HB has been represented as
purple-coloured arrows and π–π bond as green lines; solvent exposures have been presented through grey spot and salt bridge as red-blue lines

compound 3 has also shown the similar interactions with site of spike glycoprotein. Benzene rings of compound 4
Asp994 residue via formation of salt bridge with nitrogen (Fig. 6d) and nitrogen atom present in the benzene ring of
atom (Fig. 6c) being tight binding in the predicted active 5 (Fig. 6e) have been found to form double and single and

13
2782 Molecular Diversity (2022) 26:2775–2792

Fig. 4  2D interactions of compounds having docking score of ≤ − 6.0 using XP module against spike glycoprotein. HB has been represented as
purple-coloured arrows and π–π bond as green lines; solvent exposures have been presented through grey spot and salt bridge as red-blue lines

Fig. 5  2D interactions of compounds having docking score of ≤ − 6.0 using XP module against spike glycoprotein. HB has been represented as
purple-coloured arrows and π-π bond as green lines; solvent exposures have been presented through grey spot and salt bridge as red-blue lines

13
Molecular Diversity (2022) 26:2775–2792 2783

Fig. 6  3D interactions of identified hits including 1 (a), 2 (b), 3 (c), 4 (d), 5 (e) and 6 (f) through XP docking. The poses for the represented
ligands (yellow coloured ball and stick models) and protein (coloured cartoons) have been generated and visualized using PyMol 2.4.1 [54]

π-π interactions, respectively, with Tyr756. Additionally, Compound 13 has been found to form the π-cation inter-
compound 6 has been fitted in the active site with maxi- action with Arg995 (Fig. 8a) along with formation of salt
mum exposures to solvent molecules specifically with the bridge and HB formation with Asp994 resulting in docking
fluorine atom (Fig. 6f). score of − 6.23 (Entry 13, Table 2). Nitrogen of the amine
Hydroxy and amine group of 1H-imidazole derivative (7) group of non-heterocyclic compound 14 has formed HB
has been involved in π-cation interaction with Arg995 along with Thr998 residue and π–π bond interaction with Tyr756
with the formation of salt bridge (Asp994) and HB (Asp994 residue (Fig. 8b), whereas nitrogen atom and hydroxy group
and Phe970, Fig. 7a). Biologically divergent 1,2,4-oxadia- present in compound 15 have been found to forms HB with
zole derivative (8) [55] has been tightly accommodated with Phe970 and Arg995 residues (Fig. 8c). Synthetically diver-
several solvent molecules and surrounded by amino acid res- gent indole [56] compound 16 has been found in the active
idues in the active site of spike glycoprotein (Fig. 7b). The site with mainly solvent exposure (Fig. 8d) with the docking
nitrogen atoms of hybrid heterocyclic scaffold 9 have been score of − 6.04 (Entry 16, Table 2). Nitrogen of the amine
found to form HB with Thr998 (Fig. 7c) with docking score group present in the 2-ethyl imidazole derivative 17 has
of − 6.53 (Entry 9, Table 2). Compound 10 has been found been observed with the formation of HB with Thr998 and
with formation of HB with Thr998 and Thr970, salt bridge π-π bonds with the Tyr756 residues of both chains A and B
with Asp994, π-cat with Arg995 and π–π bond with Tyr756 (Fig. 8e).
(Fig. 7d) showing the string interactions in the active site of
protein. Hydroxy and amine groups were found to involve ADMET analysis
in the interaction. Pyrimidine derivative compound 11 has
been found to fit into the active site by formation of HB with Further, we have performed the analysis of absorption, dis-
Thr998 along with two π-π bonds with Tyr756 (Fig. 7e). tribution, metabolism, excretion and toxicity (ADMET)
The HB formation has been observed between Gln1002 and parameters for the identified hits through XP docking using
ethereal oxygen (HB acceptor) attached to 4-chlorophenyl QikProp module of Schrödinger [57]. Various physicochem-
ring of compound 12 (Fig. 7f) having docking score of -6.33 ical properties like molecular weight (MW), oil/water parti-
(Entry 12, Table 2). tion coefficient (LogP), water solubility (LogS), I­ C50 value

13
2784 Molecular Diversity (2022) 26:2775–2792

Fig. 7  3D interactions of identified hits including 7 (a), 8 (b), 9 (c), 10 (d), 11 (e) and 12 (f) through XP docking. The poses for the represented
ligands (yellow coloured ball and stick models) and protein (coloured cartoons) have been generated and visualized using PyMol 2.4.1 [54]

for blockage of (human ether-a-go-go-related gene (HERG) The Brain Or IntestinaL EstimateD permeation method
­K+ channels, gastrointestinal (GI) cell for cell permeability (BOILED-Egg) gives the estimation of accessibility of
(QPP caco2), brain/blood partition coefficient (Log BB), compounds to gastrointestinal (GI) tract and blood–brain
number of metabolic reactions (Metab), binding affinity barrier [60]. So, we also evaluated the accessibility of
to human serum albumin (QPLog Khsa) and human oral the hits using SwissADME [61]. The boiled egg model
absorption in percentage (% HOA) have been studied for the revealed all the hit molecules possessed satisfactory GI
selected compounds and are presented in Table 3. Mostly absorption along with inhibition of the P-glycoprotein, a
all the hits have satisfied the ADMET criteria, and all the protein responsible for efflux of drugs from cells (Fig. 9).
hit molecules may have good oral absorption. This provides All the compounds (except 3 and 16) may have sufficient
better scope for the drug molecule development in the future. permeability across the blood–brain permeability (BBB)
The Lipinski’s rule of five (Ro5) has been recognized indicating their usefulness to treat the infectious disorders
as the suitable in silico tool for the assessment of essential related to central nervous system.
physico-chemical properties of the hits or leads identified The presently identified compounds (1–17) gain the
during pre-clinical settings [59]. To ensure the drug likeli- advantage of sufficient metabolic stability and free from
ness of the hit molecules, the physico-chemical properties nitro-aryl derived toxicity over the literature reported com-
for Ro5 have been also evaluated using the QikProp module pounds (A and B, Fig. 2), whereas these organic (azo)dyes
of Schrödinger [57]. The parameters described under Lipin- (A and B) suffer from the limitations of intense colour and
ski’s rule of five such as molecular weight (MW), number metabolic instabilities limiting their therapeutic potential
of hydrogen bond donor (HBD) and acceptor (HBA), oil to against pathogenic diseases [62]. Further, compound A
water partition coefficient (LogP) and number of rotatable possesses nitro-aryl features imparting the potential risk of
bonds (RB) have been studied (Table 4). The analysis of mutagenicity due to interaction of nitro group with DNA
these properties revealed that none of the hits has violated [63]. Peptides identified by de Oliveira et al. [30] and Baig
more than one rule as per the criteria of the Lipinski’s Ro5, et al. [31] may not have sufficient oral bioavailability as
highlighting the significance of these hits in the search of compared to identified small molecules (1–17) due to
SARS CoV-2 inhibitors.

13
Molecular Diversity (2022) 26:2775–2792 2785

Fig. 8  3D interactions of identified hits including 13 (a), 14 (b), 15 (c), 16 (d) and 17 (e), through XP docking. The poses for the represented
ligands (yellow coloured ball and stick models) and protein (coloured cartoons) have been generated and visualized using PyMol 2.4.1 [54]

Table 3  ADMET parameters of the identified hits

Compd. No MWa LogPb Log ­Sc QPlog ­HERGd QPP ­caco2e Log ­BBf Metabg QPLog ­Khsah % ­HOAi

1 453.56 5.18 − 3.90 − 5.69 591.32 − 0.16 6 0.83 93.95

2 483.58 5.96 − 6.25 − 7.59 760.69 − 0.26 7 1.03 100
3 362.49 4.35 − 5.49 − 7.58 401.76 − 0.13 4 0.76 100
4 449.03 5.06 − 3.73 − 8.30 287.78 0.54 6 0.94 87.64
5 397.44 5.34 − 5.57 − 7.05 1267.41 0.70 6 0.85 100
6 391.37 4.57 − 5.60 − 7.03 746.09 0.66 4 0.60 100
7 453.56 4.77 − 3.58 − 6.41 685.05 − 0.18 5 0.61 100
8 373.38 3.88 − 4.46 − 6.58 524.52 0.43 4 0.41 100
9 414.51 3.61 − 4.30 − 5.93 194.37 − 0.31 3 0.47 89.06
10 451.56 5.43 − 4.82 − 6.42 682.53 − 0.28 7 0.97 96.48
11 407.51 4.59 − 4.54 − 7.04 716.06 − 0.09 6 0.68 100
12 484.04 6.13 − 6.27 − 7.25 918.95 − 0.07 7 1.14 100
13 470.01 5.63 − 4.56 − 6.12 1244.17 0.23 5 0.80 100
14 407.48 5.41 − 5.51 − 7.90 959.10 0.14 3 0.91 100
15 469.56 5.21 − 4.48 − 6.20 713.70 − 0.16 7 0.82 95.57
16 416.49 6.73 − 8.63 − 7.49 3518.97 − 0.27 6 1.41 100
17 467.58 5.18 − 5.38 − 6.81 872.36 − 0.13 6 1.06 100
Standard value < 500 2.0–6.5 >−4 >−5 < 25 (poor) – 1–8 – 1.5 to 1.5 0–100%
and > 500 (excel-
lent)
a
MW: Molecular weight; (Da); bOil/water partition coefficient; cSolubility; dIC50 value for blockage of HERG K
­ + channels; eGut–blood barrier
f g h
in nm/s permeability of the cell model; Brain/blood partition coefficient; Number of metabolic reactions; Binding to human serum albumin;
i
Human oral absorption. The reference values for ADMET have been taken from the Ref [58]

13
2786 Molecular Diversity (2022) 26:2775–2792

Table 4  Analysis of physico- Compd. No MWa HBAb HBDc LogPd RBe Violationf
chemical properties using
Lipinski’s rule of Five (Ro5) 1 453.56 5.75 1 5.18 11 1
2 483.58 6.5 1 5.96 12 1
3 362.49 4.5 2 4.35 5 0
4 449.03 5.2 2 5.06 9 1
5 397.44 4.5 0 5.34 4 1
6 391.37 5 0 4.57 2 0
7 453.56 6.25 1 4.77 11 0
8 373.38 5 0 3.88 2 0
9 414.51 6 2 3.61 5 0
10 451.56 6.5 1 5.43 11 1
11 407.51 6.5 1 4.59 8 0
12 484.04 5.75 1 6.13 11 1
13 470.01 5.75 0 5.63 10 1
14 407.48 4.45 2 5.41 8 1
15 469.56 6.5 1 5.21 11 1
16 416.49 3.25 2 6.73 6 1
17 484.04 5.75 1 6.13 11 1
Standard value < 500 < 10 <5 <5 <5 –
a
Molecular weight (Da), bNumber of hydrogen bond acceptors, cNumber of hydrogen bond donors, dParti-
tion coefficient in oil to water, eNumber of rotational bonds, fNumber of violated parameters from Ro5

Fig. 9  The BOILED-egg model

of identified hit molecules
(1–17) obtained through Swis-
sADME. BBB and GI perme-
ability have been indicated
through yellow and colourless
regions, respectively. The blue
circles denote inhibition of
P-glycoprotein by the corre-
sponding hits

degradation of these peptides under the acidic environ- MD simulation

ment of stomach and peptidase enzymes [64].
Molecular dynamics simulations have been recognized as
significant tool to claim the stability of the ligand and
to reveal macromolecular structure to function relation-
ships in the field of drug discovery [65–68]. Henceforth,

13
Molecular Diversity (2022) 26:2775–2792 2787

we performed the molecular dynamics (MD) simula- as evident from Fig. 10a, b. The radius of gyration (RoG)
tions at various time points up to 10 ns using GROMACS for the same complex ranging from 7.13 to 8.90 nm with
2020.1 to assess the stability of the hits into the complexes an average of 8.311 nm (Fig. 10c) indicated the compact-
obtained from ADMET and Lipinski rules analysis [69, ness of the complex. The surface area accessed by the
70]. The graphical representation of plots of statistical solvent molecules was found within the range from 1550
parameters obtained by the MD simulation of compound to 1590 n­ m2 with an average of 1586.597 n­ m2 (Fig. 10d).
1 is presented in Fig. 10. The ligand–receptor complex The plot of number of hydrogen bonds (HB) vs simula-
of 1 with spike was found with root mean square devia- tion time revealed maximum five HBs between the ligand
tion (RMSD) value ranging from 5.81 to 10.14 nm with and receptor within the time period of 10 ns (Fig. 10e).
an average of 8.889 nm (Fig. 10a) for the ligand and The electrostatic (coulombic short-range, Coul-SR) and
7.06–10.14 nm with an average of 9.710 nm (Fig. 10b) for van der Waals/hydrophobic interactions (LJ-SR) for the
the protein. Although these values were little high, the sta- complex were found − 230.361 ± 7.5 and − 146.553 ± 5 kJ/
bility of the ligand into the active site was found consistent mol, respectively, which indicated the key role of the

Fig. 10  The schematic plots of

(6a) RMSD-L, (6b) RMSD-P,
(6c) RoG, (6d) SASA and (6e)
HB for the complex of com-
pound 1 with spike glycoprotein

13
2788 Molecular Diversity (2022) 26:2775–2792

electrostatic interactions to stabilize the complex of 1 with was found within the range from 1030 to 1070 n­ m2 with
spike glycoprotein. an average of 1060.85 n­ m2 (Fig. 11d). The plot of number
The graphical representation of plots of statistical param- of hydrogen bonds (HB) vs simulation time revealed maxi-
eters obtained by the MD simulation of compound 2 is rep- mum four HBs between the ligand and receptor within the
resented in Fig. 11. The ligand–receptor complex of 2 with time period of 10 ns (Fig. 11e). The electrostatic (coulombic
the spike glycoprotein was found with RMSD value ranging short-range, Coul-SR) and van der Waals/hydrophobic inter-
from 7.08 to 11.32 nm with an average of 9.05 nm (Fig. 11a) actions (LJ-SR) for the complex were found − 87.9052 ± 18
for the ligand and 7.39–7.80 nm with an average of 7.70 nm and − 72.2626 ± 13 kJ/mol, respectively.
(Fig. 11b) for the protein. Although these values were high,
the stability of the ligand into the active site was found con-
sistent during 0–8 ns, and little instability was observed after Conclusions
8 ns in the RMSD. The radius of gyration (RoG) for the
same complex ranging from 8.4 to 8.8 nm with an average In the search of effective SARS CoV-2 inhibitors inhibiting
of 8.586 nm (Fig. 11c) indicated the compactness of the the viral entry of coronavirus into host cells, the systematic
complex. The surface area accessed by the solvent molecules virtual screening of ligands from Asinex BioDesign library

Fig. 11  The schematic plots of

(7a) RMSD-L, (7b) RMSD-P,
(7c) RoG, (7d) SASA and (7e)
HB for the complex of com-
pound 2 with spike glycoprotein

13
Molecular Diversity (2022) 26:2775–2792 2789

2020.1 has been performed against spike glycoprotein, a has been performed to identify specific receptor binding site
promising target of SARS CoV-2. The predicted binding [36]. The possible binding site regions with minimum fif-
site of spike glycoprotein has been used for pharmacophore- teen site points have been generated under more restrictive
based screening, wherein the selected ligands with essential hydrophobic environment using standard grid to visualize
pharmacophoric features have been subjected for three dif- and evaluate the top-most binding sites. For further molecu-
ferent levels of precise molecular docking to shortlist the lar modelling, binding site with the highest drugability score
identified hits with the desired docking score. Total seven- (DScore) has been selected as an active site from the nomi-
teen hits having the promising docking score have qualified nated top five binding sites.
the ADMET parameters and Lipinski’s rule of five proper-
ties to claim their oral bioavailability. The dynamics simu- Pharmacophore development and screening
lation studies have demonstrated the stability of identified
hits (1 and 2) into the predicted catalytic site of protein with The selected protein of interest (PDB ID: 6VXX) does
sufficient compactness, uniqueness and stability. In sum- not contain any co-crystallized ligand [35]. The residues
mary, the present studies accomplishes that the identified have been specified through the atom specification lan-
hits may interfere with the key residues of spike glycoprotein guage (ASL) for the selected best active site with highest
effectively as SARS CoV-2 inhibitors with sufficient binding D score followed by the generation of receptor cavity-based
affinity, good stability into the active site with the acceptable E-pharmacophore hypothesis using the phase module of
drug likeliness. These studies have provided SARS CoV-2 Schrödinger [38]. The model has been generated for the
inhibitors having the potential of inhibiting the viral entry selected active site along with the generation of seven fea-
into the host cell by acting on the viral spike protein against tures (RRR​RRR​H), comprising of aromatic ring (R) and
COVID-19. hydrophobic feature (H). The molecules prepared after the
LigPrep have been subjected to phase ligand screening to
Computational details fit with minimum four of the identified features of the gen-
erated pharmacophore hypothesis model. Maximum fifty
Preparation of ligands conformers for each ligand have been generated and sub-
jected to energy minimization. The phase screen score for
Asinex BioDesign library 2020.1, comprising of 175,851 each conformer has been calculated based on preset acceptor
molecules with pharmacologically important structural and donor as negative and positive equivalent, respectively.
features in the chemical scaffolds with synthetic feasibility, Total 38,267 molecules have qualified with the set features
has been accessed through open source databases [39]. The of hypothesis from the selected drug library.
ligands were prepared for molecular modelling using Lig-
Prep, where the possible states of ionization were generated Molecular docking
using Epik at physiological pH (7.0 ± 2.0) with consideration
of metal binding states, removal of salt forms and generation Receptor grid generation
of tautomeric isomers or stereoisomers using the OPLS3e
force field [40]. The grid box of size 10 Å has been generated using GLIDE
module of Schrödinger on the identified binding site from
Preparation of proteins site map analysis with the highest DScore [53]. The van der
Waals radius scaling factor and the partial charge cut-off
The 3D structure of SARS CoV-2 spike glycoprotein (PDB have been set 1 and 0.25, respectively, to soften the potential
ID: 6VXX) having resolution of 2.8 Å has been accessed for nonpolar parts of the protein.
from RCSB protein data bank [71]. The protein has been
modified using protein preparation wizard (PrepWizard) of Ligand docking
Schrödinger suite (Maestro 12.5) [72]. The hydrogen con-
sistency, steric relations, bond orders and total charges have Next, molecular docking has been performed using GLIDE
been generated followed by optimization and energy mini- module of Schrödinger. The 38,267 ligands which have
mization of protein using the force field (OPLS3e) to assure qualified through the phase screening, have been docked on
the structural accuracy of final protein. selected active sites of the protein using high throughput
virtual screening (HTVS) through flexible ligand sampling
Site map analysis and one pose per each ligand has been generated. A total
of 1094 molecules with docking score of ≤ − 4.6 obtained
The selected protein does not contain any co-crystallized in HTVS docking have been further docked on the same
ligand or inhibitor. In a view of this, the site map analysis active site using standard precision (SP) to generate one pose

13
2790 Molecular Diversity (2022) 26:2775–2792

for each ligand. Again 175 top compounds having docking New Delhi, Council of Scientific and Industrial Research (CSIR), New
score ≤ − 5.9 have been further docked with extra precision Delhi, and Government of India.
(XP) to generate 10 poses per ligand. ADMET analysis with
Author contributions All the authors contributed equally for this work.
Lipinski parameters has been performed for 17 hits having TMD performed docking, data generation, conceptualization, supervi-
docking score of ≤ − 6.0 in XP docking. The interactions of sion, writing—original draft and revised manuscript preparation, and
the ligand with the amino acids of the active site have been reviewing and editing; PRN contributed to methodology, data genera-
visualized using PyMol 2.4.1 [54]. tion, docking, writing—original draft and revised manuscript prepara-
tion; and NDG contributed to methodology, data generation, docking,
ADMET assay, MD simulations, writing—original draft and revised
manuscript preparation.
ADMET analysis and Lipinski rule of five

The absorption, distribution, metabolism, excretion and

toxicity (ADMET) analysis and assessment of Lipinski
parameters have been carried out using QikProp module of
Schrödinger for the obtained 16 hits which have achieved References
the docking score less than − 6.33 in XP docking [57]. The
1. Hu B, Guo H, Zhou P, Shi ZL (2020) Characteristics of SARS-
various physical, chemical and functional group properties
CoV-2 and COVID-19. Nat Rev Microbiol 19:141–154. https://​
such as molecular weight, oil/water partition coefficient, doi.​org/​10.​1038/​s41579-​020-​00459-7
solubility, ­IC50 value for blockage of HERG ­K+ channels, 2. Cui J, Li F, Shi ZL (2019) Origin and evolution of pathogenic
gut–blood barrier in nm/s permeability of the cell model, coronaviruses. Nat Rev Microbiol 17:181–192. https://d​ oi.o​ rg/1​ 0.​
1038/​s41579-​018-​0118-9
brain/blood partition coefficient, number of metabolic reac-
3. Ashour HM, Elkhatib WF, Rahman MM, Elshabrawy HA (2020)
tions, binding to human serum albumin and human oral Insights into the recent 2019 novel coronavirus (SARS-CoV-2)
absorption have been studied from the manual analysis of in light of past human coronavirus outbreaks. Pathogens 9:e186.
the result of ADMET analysis. The drug likeliness features https://​doi.​org/​10.​3390/​patho​gens9​030186
4. Pillaiyar T, Wendt LL, Manickam M, Easwaran M (2021) The
like molecular weight, hydrogen bond donors, hydrogen
recent outbreaks of human coronaviruses: a medicinal chemistry
bond acceptors, partition coefficient (oil/water) and number perspective. Med Res Rev 41:72–135. https://​doi.​org/​10.​1002/​
of rotatable bonds have been studied through Lipinski rule med.​21724
of five. 5. Wu JT, Leung K, Leung GM (2020) Nowcasting and forecasting
the potential domestic and international spread of the 2019-nCoV
outbreak originating in Wuhan, China: a modeling study. Obstet
Gynecol Surv 75:399–400. https://​doi.​org/​10.​1097/​01.​ogx.​00006​
Molecular dynamics (MD) simulation 88032.​41075.​a8
6. Hui DS, Azhar EI, Madani TA et al (2020) The continuing 2019-
Hit molecule obtained from the manual analysis of ADMET nCoV epidemic threat of novel coronaviruses to global health:
and Lipinski parameters have been incorporated to perform the latest 2019 novel coronavirus outbreak in Wuhan, China. Int J
Infect Dis 91:264–266. https://​doi.​org/​10.​1016/j.​ijid.​2020.​01.​009
the MD simulation using GROningen MAchine for Chemi- 7. World Health Organization. (2020) Coronavirus disease 2019
cal Simulations (GROMACS) 2020.1 [69, 70] software. (COVID-19). Situation report: 51. https://​www.​who.​int/​docs/​
CHARMM36 (Chemistry at Harvard Macromolecular defau​lt-​source/​coron​aviru​se/​situa​tion-​repor​ts/​20200​311-​sitrep-​
Mechanics) was used as an all atom force field [73] and 51-​covid-​19.​pdf?​sfvrsn=​1ba62​e57_​10. Accessed 2 Feb 2021
8. WHO Coronavirus Disease (COVID-19) Dashboard. https://c​ ovid​
CHARMM General Force Field (CGenFF) server [74, 75] 19.​who.​int/. Accessed 18 Jan 2022
was used to generate the topology of the ligands. After sol- 9. Nicola M, Alsafi Z, Sohrabi C et al (2020) The socio-economic
vation (TIP3P water model) and neutralization (sodium implications of the coronavirus pandemic (COVID-19): a review.
chloride), the complex was subjected to energy minimiza- Int J Surg 78:185–193. https://​doi.​org/​10.​1016/j.​ijsu.​2020.​04.​018
10. Bashir MF, Ma B, Shahzad L (2020) A brief review of socio-eco-
tion followed by equilibration through the canonical isovol- nomic and environmental impact of COVID-19. Air Qual Atmos
umetric-isothermal (NVT) and isobaric-isothermic (NPT) Heal 13:1403–1409. https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 11869-0​ 20-0​ 0894-8
canonicals for 100 ps. The energy minimized complexes 11. Mousavizadeh L, Ghasemi S (2020) Genotype and phenotype
have been subjected for final MD run of 10 ns for success- of COVID-19: their roles in pathogenesis. J Microbiol Immunol
Infect 54:159–163. https://​doi.​org/​10.​1016/j.​jmii.​2020.​03.​022
ful completion. 12. Ahidjo BA, Loe MWC, Ng YL et al (2020) Current perspective
of antiviral strategies against COVID-19. ACS Infect Dis 6:1624–
Supplementary Information The online version contains supplemen- 1634. https://​doi.​org/​10.​1021/​acsin​fecdis.​0c002​36
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 11030-0​ 22-1​ 0394-9. 13. Tyrrell DAJ, Myint SH (1996) Coronaviruses. In: Baron S (ed)
Medical Microbiology. University of Texas Medical Branch at
Acknowledgements The authors would like to acknowledge the sup- Galveston, Galveston
port provided by Schrödinger Inc. for performing molecular modelling 14. Zhu N, Zhang D, Wang W et al (2020) A novel coronavirus from
under the banner of Drug Discovery Hackathon 2020 (DDH-2020), a patients with pneumonia in China, 2019. N Engl J Med 382:727–
joint initiative of All India Council of Technical Education (AICTE), 733. https://​doi.​org/​10.​1056/​nejmo​a2001​017

13
Molecular Diversity (2022) 26:2775–2792 2791

15. Freitas BT, Durie IA, Murray J et al (2020) Characterization and 32. Bharatam PV (2021) Computer-aided drug design. In: Poduri R
noncovalent inhibition of the deubiquitinase and deISGylase (ed) Drug discovery and development. Springer, Singapore, pp
activity of SARS-CoV-2 papain-like protease. ACS Infect Dis 137–210
6:2099–2109. https://​doi.​org/​10.​1021/​acsin​fecdis.​0c001​68 33. Nagar PR, Gajjar ND, Dhameliya TM (2021) In search of SARS
16. Lu R, Zhao X, Li J et al (2020) Genomic characterisation and CoV-2 replication inhibitors: virtual screening, molecular dynam-
epidemiology of 2019 novel coronavirus: implications for virus ics simulations and ADMET analysis. J Mol Struct 1246:131190.
origins and receptor binding. Lancet 395:565–574. https://d​ oi.o​ rg/​ https://​doi.​org/​10.​1016/j.​molst​ruc.​2021.​131190
10.​1016/​S0140-​6736(20)​30251-8 34. Gajjar ND, Dhameliya TM, Shah GB (2021) In search of RdRp
17. Woo PCY, Lau SKP, Huang Y, Yuen KY (2009) Coronavirus and Mpro inhibitors against SARS CoV-2: molecular docking,
diversity, phylogeny and interspecies jumping. Exp Biol Med molecular dynamic simulations and ADMET analysis. J Mol
234:1117–1127. https://​doi.​org/​10.​3181/​0903-​MR-​94 Struct 1239:130488. https://​doi.​org/​10.​1016/j.​molst​r uc.​2021.​
18. Guan W, Ni Z, Hu Y et al (2020) Clinical characteristics of coro- 130488
navirus disease 2019 in China. N Engl J Med 382:1708–1720. 35. Walls AC, Park YJ, Tortorici MA et al (2020) Structure, func-
https://​doi.​org/​10.​1056/​nejmo​a2002​032 tion, and antigenicity of the SARS-CoV-2 spike glycoprotein. Cell
19. Rothan HA, Byrareddy SN (2020) The epidemiology and patho- 181:281-292.e6. https://​doi.​org/​10.​1016/j.​cell.​2020.​02.​058
genesis of coronavirus disease (COVID-19) outbreak. J Autoim- 36. Schrödinger Release 2020–3: SiteMap, Schrödinger, LLC, New
mun 109:102433. https://​doi.​org/​10.​1016/j.​jaut.​2020.​102433 York, NY, 2020
20. Singhal T (2020) A review of coronavirus disease-2019 (COVID- 37. Vidler LR, Brown N, Knapp S, Hoelder S (2012) Druggability
19). Indian J Pediatr 87:281–286. https:// ​ d oi. ​ o rg/ ​ 1 0. ​ 1 007/​ analysis and structural classification of bromodomain acetyl-
s12098-​020-​03263-6 lysine binding sites. J Med Chem 55:7346–7359. https://​doi.​org/​
21. Ghosh AK, Brindisi M, Shahabi D et al (2020) Drug development 10.​1021/​jm300​346w
and medicinal chemistry efforts toward SARS-coronavirus and 38. Schrödinger Release 2020–3: Phase, Schrödinger, LLC, New
COVID-19 therapeutics. ChemMedChem 15:907–932. https://d​ oi.​ York, NY, 2020
org/​10.​1002/​cmdc.​20200​0223 39. Asinex. http://​www.​asinex.​com/. Accessed 17 Aug 2019
22. Mohamed K, Yazdanpanah N, Saghazadeh A, Rezaei N (2021) 40. Schrödinger Release 2020–3: LigPrep, Schrödinger, LLC, New
Computational drug discovery and repurposing for the treatment York, NY, 2020
of COVID-19: a systematic review. Bioorg Chem 106:104490. 41. Dhameliya TM, Tiwari R, Banerjee A et al (2018) Benzo[d]thia-
https://​doi.​org/​10.​1016/j.​bioorg.​2020.​104490 zole-2-carbanilides as new anti-TB chemotypes: design, synthesis,
23. Wang X, Guan Y (2021) COVID-19 drug repurposing: A review biological evaluation, and structure-activity relationship. Eur J
of computational screening methods, clinical trials, and protein Med Chem 155:364–380. https://​doi.​org/​10.​1016/j.​ejmech.​2018.​
interaction assays. Med Res Rev 41:5–28. https://d​ oi.o​ rg/1​ 0.1​ 002/​ 05.​049
med.​21728 42. Jadhavar PS, Dhameliya TM, Vaja MD et al (2016) Synthesis,
24. Chakravarti R, Singh R, Ghosh A et al (2021) A review on poten- biological evaluation and structure-activity relationship of 2-sty-
tial of natural products in the management of COVID-19. RSC rylquinazolones as anti-tubercular agents. Bioorg Med Chem Lett
Adv 11:16711–16735. https://​doi.​org/​10.​1039/​d1ra0​0644d 26:2663–2669. https://​doi.​org/​10.​1016/j.​bmcl.​2016.​04.​012
25. Shagufta AI (2021) The race to treat COVID-19: Potential thera- 43. Shah P, Dhameliya TM, Bansal R et al (2014) N-Arylalkylbenzo[d]
peutic agents for the prevention and treatment of SARS-CoV-2. thiazole-2-carboxamides as anti-mycobacterial agents: design,
Eur J Med Chem 213:113157. https://​doi.​org/​10.​1016/j.​ejmech.​ new methods of synthesis and biological evaluation. Med Chem
2021.​113157 Commun 5:1489–1495. https://​doi.​org/​10.​1039/​C4MD0​0224E
26. Tang T, Bidon M, Jaimes JA et al (2020) Coronavirus membrane 44. Bhakhar KA, Gajjar ND, Bodiwala KB et al (2021) Identification
fusion mechanism offers a potential target for antiviral develop- of anti-mycobacterial agents against mmpL3: virtual screening,
ment. Antiviral Res 178:104792. https://​doi.​org/​10.​1016/j.​antiv​ ADMET analysis and MD simulations. J Mol Struct 1244:130941.
iral.​2020.​104792 https://​doi.​org/​10.​1016/j.​molst​ruc.​2021.​130941
27. Varghese PM, Tsolaki AG, Yasmin H et al (2020) Host-pathogen 45. Alexpandi R, De Mesquita JF, Pandian SK, Ravi AV (2020) Qui-
interaction in COVID-19: pathogenesis, potential therapeutics and nolines-based SARS-CoV-2 3CLpro and RdRp inhibitors and
vaccination strategies. Immunobiology 225:152008. https://​doi.​ Spike-RBD-ACE2 inhibitor for drug-repurposing against COVID-
org/​10.​1016/j.​imbio.​2020.​152008 19: an in silico analysis. Front Microbiol. https://​doi.​org/​10.​3389/​
28. Bojadzic D, Alcazar O, Chen J et al (2021) Small-molecule inhibi- fmicb.​2020.​01796
tors of the coronavirus spike: ACE2 protein-protein interaction 46. Ferraz WR, Gomes RA, Novaes ALS, Goulart Trossini GH (2020)
as blockers of viral attachment and entry for SARS-CoV-2. ACS Ligand and structure-based virtual screening applied to the SARS-
Infect Dis 7:1519–1534. https://d​ oi.o​ rg/1​ 0.1​ 021/a​ csinf​ ecdis.1​ c000​ CoV-2 main protease: an in silico repurposing study. Future Med
70 Chem 12:1815–1828. https://​doi.​org/​10.​4155/​fmc-​2020-​0165
29. Day CJ, Bailly B, Guillon P et al (2021) Multidisciplinary 47. Shahinshavali S, Hossain KA, Kumar AVDN et al (2020) Ultra-
approaches identify compounds that bind to human ACE2 or sound assisted synthesis of 3-alkynyl substituted 2-chloroquinoxa-
SARS-CoV-2 spike protein as candidates to block SARS-CoV-2– line derivatives: their in silico assessment as potential ligands for
ACE2 receptor interactions. MBio 12:e03681-e3720. https://​doi.​ N-protein of SARS-CoV-2. Tetrahedron Lett 61:152336. https://​
org/​10.​1128/​mBio.​03681-​20 doi.​org/​10.​1016/j.​tetlet.​2020.​152336
30. Freitas FC, Ferreira PHB, Favaro DC, De ORJ (2021) Shedding 48. Olubiy OO, Olagunju M, Keutmann M et al (2020) High through-
light on the inhibitory mechanisms of SARS-CoV-1/CoV-2 spike put virtual screening to discover inhibitors of the main protease
proteins by ACE2-designed peptides. J Chem Inf Model 61:1226– of the coronavirus SARS-CoV-2. Molecules 25:3193. https://​doi.​
1243. https://​doi.​org/​10.​1021/​acs.​jcim.​0c013​20 org/​10.​3390/​molec​ules2​51431​93
31. Rajpoot S, Ohishi T, Kumar A et al (2021) A novel therapeutic 49. Ngo ST, Quynh Anh Pham N, Le Thi L et al (2020) Computa-
peptide blocks SARS-CoV-2 spike protein binding with host cell tional determination of potential inhibitors of SARS-CoV-2 main
ACE2 receptor. Drugs R D 21:273–283. https://​doi.​org/​10.​1007/​ protease. J Chem Inf Model 60:5771–5780. https://​doi.​org/​10.​
s40268-​021-​00357-0 1021/​acs.​jcim.​0c004​91

13
2792 Molecular Diversity (2022) 26:2775–2792

50. Hagar M, Ahmed HA, Aljohani G, Alhaddad OA (2020) Inves- 63. Purohit V, Basu AK (2000) Mutagenicity of nitroaromatic com-
tigation of some antiviral N-heterocycles as COVID 19 drug: pounds. Chem Res Toxicol 13:673–692. https://​doi.​org/​10.​1021/​
molecular docking and DFT calculations. Int J Mol Sci 21:3922. tx000​002x
https://​doi.​org/​10.​3390/​ijms2​11139​22 64. Sewald N, Jakubke H-D (2009) Peptides: chemistry and biology,
51. Gentile D, Patamia V, Scala A et al (2020) Putative inhibitors 2nd edn. Wiley-VCH, New York
of SARS-COV-2 main protease from a library of marine natural 65. Naqvi AAT, Mohammad T, Hasan GM, Hassan MI (2019)
products: a virtual screening and molecular modeling study. Mar Advancements in docking and molecular dynamics simulations
Drugs 18:225. https://​doi.​org/​10.​3390/​md180​40225 towards ligand-receptor interactions and structure-function rela-
52. Chemboli R, Kapavarapu R, Deepti K et al (2021) Pyrrolo[2,3-b] tionships. Curr Top Med Chem 18:1755–1768. https://​doi.​org/​10.​
quinoxalines in attenuating cytokine storm in COVID-19: their 2174/​15680​26618​66618​10251​14157
sonochemical synthesis and in silico/in vitro assessment. J Mol 66. Salo-Ahen OMH, Alanko I, Bhadane R et al (2021) Molecular
Struct 1230:129868. https://​doi.​org/​10.​1016/j.​molst​r uc.​2020.​ dynamics simulations in drug discovery and pharmaceutical
129868 development. Processes 9:71. https://​doi.​org/​10.​3390/​pr901​0071
53. Schrödinger Release 2020-3: Glide, Schrödinger, LLC, New York, 67. Bera I, Payghan PV (2019) Use of molecular dynamics simula-
NY, 2020 tions in structure-based drug discovery. Curr Pharm Des 25:3339–
54. DeLano WL (2002). The PyMOL molecular genetics graphics 3349. https://​doi.​org/​10.​1021/​acs.​jpcb.​1c045​56
system, DeLano Scientific LLC, San Carlos 68. Padhi AK, Rath SL, Tripathi T (2021) Accelerating COVID-19
55. Dhameliya TM, Chudasma SJ, Patel TM, Dave BP (2022) A research using molecular dynamics simulation. J Phys Chem B
review on synthetic account of 1,2,4-oxadiazoles as anti-infective 125:9078–9091. https://​doi.​org/​10.​1021/​acs.​jpcb.​1c045​56
agents. Mol Divers. https://​doi.​org/​10.​1007/​s11030-​021-​10375-4 69. Abraham MJ, Berk Hess, Lindahl E, Spoel D van der (2020)
56. Bhakhar KA, Sureja DK, Dhameliya TM (2022) Synthetic account GROMACS 2020.1 (Manual Version 2020.1) Zenodo. https://​
of indoles in search of potential anti-mycobacterial agents: a doi.​org/​10.​5281/​zenodo.​40549​96. Accessed 10 Sep 2020
review and future insights. J Mol Struct 1248:131522. https://​ 70. Abraham MJ, Murtola T, Schulz R et al (2015) GROMACS: High
doi.​org/​10.​1016/j.​molst​ruc.​2021.​131522 performance molecular simulations through multi-level parallel-
57. Schrödinger Release 2020-3: QikProp, Schrödinger, LLC, New ism from laptops to supercomputers. SoftwareX 1–2:19–25
York, NY, 2020 71. Protein Data Bank. https://w ​ ww.r​ csb.o​ rg/. Accessed 30 Aug 2020
58. Modi P, Patel S, Chhabria M (2019) Structure-based design, syn- 72. Protein Preparation Wizard; Epik, Schrödinger, LLC, New York,
thesis and biological evaluation of a newer series of pyrazolo[1,5- NY, (2020) Impact, Schrödinger, LLC, New York, NY, 2020;
a]pyrimidine analogues as potential anti-tubercular agents. Bioorg Prime, Schrödinger, LLC, New York, NY, 2020
Chem 87:240–251. https://​doi.​org/​10.​1016/j.​bioorg.​2019.​02.​044 73. Huang J, Rauscher S, Nawrocki G et al (2016) CHARMM36m:
59. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (2001) Experi- an improved force field for folded and intrinsically disordered
mental and computational approaches to estimate solubility and proteins. Nat Methods 14:71–73. https://​doi.​org/​10.​1038/​nmeth.​
permeability in drug discovery and devlopment settings. Adv 4067
Drug Deliv Rev 46:3–26. https://d​ oi.o​ rg/1​ 0.1​ 016/s​ 0169-4​ 09x(00)​ 74. Vanommeslaeghe K, Hatcher E, Acharya C et al (2010)
00129-0 CHARMM general force field: a force field for drug-like mol-
60. Daina A, Zoete V (2016) A BOILED-Egg to predict gastroin- ecules compatible with the CHARMM all-atom additive biologi-
testinal absorption and brain penetration of small molecules. cal force fields. J Comput Chem 31:671–690. https://​doi.​org/​10.​
ChemMedChem 11:1117–1121. https://​doi.​org/​10.​1002/​cmdc.​ 1002/​jcc.​21367
20160​0182 75. Yu W, He X, Vanommeslaeghe K, MacKerell AD (2012) Exten-
61. Daina A, Michielin O, Zoete V (2017) SwissADME: a free web sion of the CHARMM general force field to sulfonyl-containing
tool to evaluate pharmacokinetics, drug-likeness and medicinal compounds and its utility in biomolecular simulations. J Comput
chemistry friendliness of small molecules. Sci Reports 7:42717. Chem 33:2451–2468. https://​doi.​org/​10.​1002/​jcc.​23067
https://​doi.​org/​10.​1038/​srep4​2717
62. Levine WG (1991) Metabolism of aZO dyes: implication for Publisher's Note Springer Nature remains neutral with regard to
detoxication and activation. Drug Metab Rev 23:253–309. https://​ jurisdictional claims in published maps and institutional affiliations.
doi.​org/​10.​3109/​03602​53910​90297​61

13