Design, synthesis and biological evaluation of dihydrofurocoumarin derivatives as potent neuraminidase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2021.127839

Neuraminidase (NA) is a promising target for development of anti-influenza drugs. In this study a dihydrofurocoumarin derivative ZINC05577497 was discovered as a lead NA inhibitor based on docking-based virtual screening technique. The optimization of lead ZINC05577497 led to the discovery of a series of novel NA inhibitors 5a-5j. Compound 5b has the most potent activity against NA with IC₅₀ = 0.02 μM, which is lower than those of the reference oseltamivir carboxylate (OSC) (IC₅₀ = 0.04 μM) and ZINC05577497 (IC₅₀ = 0.11 μM). Other target compounds also show potential inhibition of NA activity. Molecular docking results indicate that the good potency of 5b may be attributed to the elongation of the dihydrofurocoumarin ring to the 150-cavity. The results of this paper will be useful to discover more potent NA inhibitors.

Full text of DOI:10.1016/j.bmcl.2021.127839

Bioorg. Med. Chem. Lett. 37 (2021) 127839
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and biological evaluation of dihydrofurocoumarin
derivatives as potent neuraminidase inhibitors
Zhi Jian Zhong, Li Ping Cheng*, Wan Pang*, Xue Song Zheng, Shi Kai Fu
School of Chemical and Environmental Engineering, Shanghai Institute of Technology, Shanghai 201418, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Neuraminidase (NA) is a promising target for development of anti-influenza drugs. In this study a dihydrofur-
ocoumarin derivative ZINC05577497 was discovered as a lead NA inhibitor based on docking-based virtual
screening technique. The optimization of lead ZINC05577497 led to the discovery of a series of novel NA in-
hibitors 5a-5j. Compound 5b has the most potent activity against NA with IC50 = 0.02 µM, which is lower than
those of the reference oseltamivir carboxylate (OSC) (IC50 = 0.04 µM) and ZINC05577497 (IC50 = 0.11 µM).
Other target compounds also show potential inhibition of NA activity. Molecular docking results indicate that the
good potency of 5b may be attributed to the elongation of the dihydrofurocoumarin ring to the 150-cavity. The
results of this paper will be useful to discover more potent NA inhibitors.
Neuraminidase inhibitors
Virtual screening
Dihydrofurocoumarin derivatives
Molecule docking
Influenza virus is an RNA virus of the Orthomyxoviridae family, which
can cause acute respiratory infections and lead to the patients to contract
severe pneumonia, as commonly called Influenza.1,2 As its highly con-
tagious, the illness remains a great threat to human health with high
morbidity and mortality rates.3,4 To prevent Influenza effectively, the
world has suffered from serious economic problems.5 Just like most
viruses, during the process of replication and propagation, influenza
virus needs replicate virus offspring. There are two surface membrane
glycoproteins located on the virus surface: hemagglutinin (HA), neur-
aminidase (NA), which are essential for virus infection. Neuraminidase
plays an especially important role in assisting mature influenza viruses
to leave host cells and infect new cells. Due to its highly conserved active
site, neuraminidase (NA) has been the most potential target for the
development of novel anti-influenza drugs.6,7
antiviral therapy.11 Therefore, the development of highly efficient anti-
influenza drugs, which are insusceptible to drug-resistance, is particu-
larly important. At present some small molecules had been discovered
and verified as effective NA inhibitors, such as acylhydrazone de-
rivatives,7 chalcone derivatives,12 thiazole derivatives,13 pyrazole de-
rivatives,14 and some natural products.15 These new NA inhibitors could
give advanced treatments for the drug-resistance strains, and are ur-
gently needed for people to respond to the threat of pandemic influenza.
Recently, two new pockets adjacent to the active site of NA, named
“150-cavity” and “430-cavity”, attract a lot of attention. The crystal
structure of neuraminidase16 (PDB ID: 2HU0) was depicted in Fig. 2, it
can be seen that both the closed 150-cavity and the flat 430-cavity have
large molecular volume and connect directly to the active site, which
makes them promising binding sites for NA inhibitor design. Some
research groups have developed some novel NA inhibitors by targeting
Currently, there are two ways to prevent and treat influenza, one is
vaccine, and the other is antiviral drugs. There is no doubt that vaccine is
the most effective way. However, attempts to control this disease
through immunization have been hampered by the rapid mutation of
virus. Therefore, the development of effective and safe antiviral drugs is
even more important in the event that global pandemics outbreak.8,9
Four NA inhibitors (NAs) had been used as effective treatments for the
influenza infection: oseltamivir (1), zanamivir (2), peramivir (3), and
laninamivir octanoate (4).10 The chemical structures are shown in Fig. 1.
However, with the widespread use of these inhibitors, especially the
oseltamivir abuse, virus resistance has become to be a major problem of
these two cavities, and have obtained desired results. For instance, Feng
17
et al.
had designed and synthesized a novel series of C-1 and C-4-
modified zanamivir derivatives as neuraminidase inhibitors based on
the 430-cavity. A series of novel N-substituted oseltamivir derivatives
targeting the 150-cavity were found as potential NA inhibitors by Xie
et al. 18. These findings highlight the importance of 430- and 150-cav-
ities in the development of novel NA inhibitors.
Virtual screening is considered as an effective method to discover
new lead compounds for known target proteins.19 Molecular docking
plays a very important role in new drug discovery.20–22 In this research,
* Corresponding authors.
E-mail addresses: chengliping@sit.edu.cn (L.P. Cheng), pangwan@sit.edu.cn (W. Pang).
https://doi.org/10.1016/j.bmcl.2021.127839
Received 2 November 2020; Received in revised form 15 January 2021; Accepted 30 January 2021
Available online 5 February 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
Fig. 1. Skeleton structures of some representative NA inhibitors.
the discovery of novel dihydrofurocoumarin NA inhibitors was per-
formed by docking-based virtual screening, molecular dynamics (MD)
simulations, chemical synthesis and bioassay. A novel lead compound
was firstly hit by virtual screening and MD simulations, and then the
lead compound was used as a template to design some new NA inhibitors
targeting the 430- and 150-cavities. Finally, the designed 10 new
dihydrofurocoumarin inhibitors were synthesized and tested for bio-
logical activity.
The crystal structure of influenza A virus H5N1 neuraminidase (PDB
ID: 2HU0) was used as a target protein, which was obtained from the
protein database (http://www.rcsb.org/pdb/). The X-ray coordinates of
neuraminidase (PDB ID: 2HU0) had been listed in the Table S1. Some
preparation for the protein structure were necessary, such as removing
water molecules, adding hydrogens and optimizing protein structure,
before the docking analysis was performed. The workflow of virtual
screening is displayed in Fig. 3. Data set 1, consisting of about 450,000
small molecules downloaded from the ZINC 12 database, was selected
for virtual screening in this work. The surflex-dock module of the
SYBYL-X 2.1 (Tripos, Inc., USA) was used for preliminary virtual
screening. During this process, the ring flexibility of molecule also needs
to be considered. The docking scores of all molecules were compared
with the score of the reference molecule that extracted from crystal
protein. As a result, the top 100 compounds with higher total scores
were screened out to make up Data set 2. From the docking results, we
further found that some small molecules can occupy both catalytic site
Fig. 2. The crystal structure of PDB 2HU0.
Fig. 3. The workflow of virtual screening.
Fig. 4. The structures of Data set 4 and the lead compound 5 chemical scaffold.
2

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
Fig. 5. Binding mode of 5 with NA (PDB ID: 2HU0).
and 430-cavity or 150-cavity at the same time. The diversity of the
compound structures were also to be taken into account. Finally, the top
20 compounds were further screened out and to be saved as Data set 3.
The calculation of binding free energy by MD simulation was an
important tool to study the interactions between drug molecules and
targets.23 To discover the novel lead compounds, MD simulations were
carried out for the 20 compounds in the above Data set 3 by Amber 12.0
package.24 The overall convergence of MD simulations and system
equilibration were monitored by using the root-mean-square deviation
(RMSD) of the acceptor backbone atoms. Binding free energies between
the protein and molecules were calculated by using Molecular Me-
chanics/Poisson Surface Area (MM/PBSA) and Molecular Mechanics/
Generalized Born (MM/GBSA) methods.25–27 Compared with these two
methods in predicting binding free energy, MM/PBSA is generally better
than MM/GBSA.25 Furthermore, it is known that there is generally a
good correlation between binding free energy and compound’s activity.
The more negative the ΔGbind of compound is, the more active the
compound is.28,29 The results show that three compounds,
ZINC05577497, ZINC5891426, ZINC1036214 (Fig. 4), with smaller
binding free energies were selected to constitute Data set 4. The binding
free energies of the three compounds are listed in Table S2. The average
RMSD values of each step versus simulation time are shown in Fig. S1.
For comparison, the MD simulation of oseltamivir carboxylate (OSC)
was also carried out. As shown in Table S2, The binding free energies for
ZINC05577497, ZINC5891426, ZINC1036214 are ꢀ 20.94 kcal⋅molꢀ 1
,
ꢀ 15.06 kcal⋅molꢀ 1, and ꢀ 14.67 kcal⋅molꢀ 1 via the prediction by the
MM/GBSA method, respectively. The corresponding values are ꢀ 17.80
kcal⋅molꢀ 1, ꢀ 14.52 kcal⋅molꢀ 1, and ꢀ 12.32 kcal⋅molꢀ 1 via the predic-
tion by the MM/PBSA method, respectively. For OSC, the MM/GBSA
value is ꢀ 13.40 kcal⋅molꢀ 1 and the MM/PBSA value is ꢀ 10.08
kcal⋅molꢀ 1. It is can be seen that compound 5 (ZINC05577497) has the
lowest binding free energy and may form the strongest interaction with
NA crystal. Moreover, the NA inhibitory activity of compound 5 had
been tested to verify the calculation result. The IC50 value against NA is
0.11 µМ, which is comparable to that of the positive control OSC (0.04
µМ). Thus, compound 5 could be selected as the lead compound to
perform further study.
The lead compound 5 was selected as a template to further design
new neuraminidase inhibitors. The bonding model of compound 5 with
NA is shown in Fig. 5. It turned out that the dihydrofurocoumarin ring
Scheme 1. Synthesis of target compounds 5a-5j.
3

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
Fig. 6. Diagram of the relationship between RMSD value and simulation time.
Table 1
The calculated binding free values of all designed compounds.
No.
VDW
EEL
△Ggas
△GGB
△GSA
△Gsolv(GB)
△Gbind(GB)
△GPB
△GSA
△Gsolv(PB)
△Gbind(PB)
OSC
5a
5b
5c
5d
5e
5f
ꢀ 30.36
ꢀ 26.74
ꢀ 32.49
ꢀ 39.55
ꢀ 29.05
ꢀ 30.58
ꢀ 27.83
ꢀ 23.57
ꢀ 24.14
ꢀ 37.67
ꢀ 30.91
ꢀ 24.18
ꢀ 15.21
ꢀ 19.44
ꢀ 9.41
ꢀ 54.54
ꢀ 41.95
ꢀ 51.93
ꢀ 48.96
ꢀ 40.87
ꢀ 52.63
ꢀ 37.07
ꢀ 37.75
ꢀ 36.99
ꢀ 58.40
ꢀ 44.34
45.44
28.04
34.01
25.10
26.65
37.30
19.11
26.30
25.83
38.88
30.04
ꢀ 4.30
ꢀ 3.14
ꢀ 4.23
ꢀ 4.47
ꢀ 3.60
ꢀ 3.28
ꢀ 3.40
ꢀ 2.48
ꢀ 2.93
ꢀ 4.95
ꢀ 3.91
41.14
24.91
29.78
20.63
23.05
34.03
15.71
23.82
22.90
33.93
26.12
ꢀ 13.40
ꢀ 17.04
ꢀ 25.15
ꢀ 28.33
ꢀ 17.82
ꢀ 18.60
ꢀ 21.37
ꢀ 13.93
ꢀ 14.09
ꢀ 24.47
ꢀ 18.22
48.15
28.87
38.08
30.86
27.29
38.97
22.74
25.24
22.92
47.05
33.38
ꢀ 3.69
ꢀ 2.90
ꢀ 3.30
ꢀ 3.83
ꢀ 3.26
ꢀ 3.02
ꢀ 2.85
ꢀ 2.33
ꢀ 2.97
ꢀ 3.88
ꢀ 3.54
44.46
25.97
34.78
27.02
24.04
35.95
19.89
22.92
20.95
43.17
28.84
ꢀ 10.08
ꢀ 15.98
ꢀ 24.19
ꢀ 21.93
ꢀ 16.83
ꢀ 16.68
ꢀ 17.19
ꢀ 14.84
ꢀ 16.04
ꢀ 15.23
ꢀ 15.50
ꢀ 11.82
ꢀ 22.05
ꢀ 9.24
5g
5 h
5i
ꢀ 14.19
ꢀ 12.85
ꢀ 20.73
ꢀ 13.43
5j
Table 2
The IC50 values of all synthesized compounds against neuraminidase from H5N1 subtype.
Compound
IC50
(μ
M)
Compound
IC50 (μM)
5a
0.15 ± 0.02
5f
0.04 ± 0.02
5b
5c
5d
0.02 ± 0.01
0.03 ± 0.01
30.01 ± 5.27
5g
5h
5i
99.84 ± 12.72
39.46 ± 3.26
36.25 ± 7.88
5e
32.43 ± 7.65
0.04 ± 0.01
5j
45.76 ± 4.15
0.10 ± 0.01
OSC
–
OS
–
4

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
Fig. 7. The inhibitory profiles of the NA inhibitors 5b and 5c.
could bind well with the 430-cavity, but there is only one hydrogen bond
formed between one methoxy of phenyl and the residue (Arg152)
around the active site. So, the introduction of some hydrogen bond re-
ceptors (such as oxygen atom, nitro, and hydroxyl) into the benzene
ring, which may increase the interactions between the substitutes and
the active site, and thus increase the NA inhibitory activity. Accordingly,
a total of 10 new compounds of 5a-5j were designed (see Scheme 1).
Subsequently, MD simulations were carried out for all designed com-
pounds to predict their theoretical activity. The RMSD value (Å) with
respect to simulation time (ns) is shown in Fig. 6. It can be seen that all
complexes are mainly stable after 7.5 ns simulation time, and it can be
inferred that the binding pocket and the conformation of the ligand is
stable. All calculated binding free energy values were listed in Table 1. It
can be seen that all binding free energies of 5a–5j are more negative
than that of OSC based on both MM/GBSA and MM/PBSA methods,
indicating their good activity. Therefore, from the binding free energy
viewpoint, all designed compounds 5a–5j have potential values to
further synthesize and study. In addition, as the former mentioned, MM/
PBSA is generally considered superior to MM/GBSA in predicting
binding free energy. Among the ten newly designed compounds, 5b has
the most negative binding free energy via the prediction by MM/PBSA,
implying that it perhaps possesses the best inhibitory activity among
Fig. 8. Binding modes of 5a, 5b, 5c and 5f with NA (2HU0).
5

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
Fig. 8. (continued).
these ten compounds.
(30.01 µM and 32.43 µM). 5 g has the poorest activity (99.84 µМ) and
the highest ΔGbind value (ꢀ 14.84 kcal⋅molꢀ 1). Hence one can see that
the theoretical results are consistent well with the experimental results.
The inhibitory profiles of the most potent NA inhibitors 5b and 5c are
shown in Fig. 7.
The designed compounds (5a-5j) were synthesized by feasible syn-
thetic strategies as depicted in Scheme 1.30–33 The target compounds
were obtained using resorcinol and ethyl acetoacetate as the starting
materials in four steps including Pechmann reaction, acylation, unusual
Fries rearrangement reaction, and condensation reaction. The structures
of all the synthesized compounds were confirmed by 1H NMR, 13C NMR
and HRMS, and the purity was detected by high performance liquid
chromatography.
To further understand the discrepancy in activity of the target
compounds, molecular docking analysis was performed by the surflex-
dock module of SYBYL-X 2.1. Fig. 8 describes the binding modes of
compounds 5a, 5b, 5c, and 5f with NA. The binding modes of the other
synthesized compounds and NA are shown in Fig. S2. As shown in Fig. 8,
the phenyl moiety of 5a and 5c can both well occupied the active site of
neuraminidase. Furthermore, their dihydrofurocoumarin rings could be
elongated into the 430-cavity. 430-cavity is mainly composed of Trp403,
Lys432, Ile427, and Thr439, etc., which could be regarded as a hydro-
phobic pocket. From 2D figures one can see that the dihydrofur-
ocoumarin ring rings of 5a and 5c were projected toward the 430-cavity
to form hydrogen bonds and hydrophobic interactions. For example, for
5a, 3-methoxy of benzene ring may form one hydrogen bond with the
residue Arg156; the 4-hydroxyl group may form three hydrogen bonds
with residues Arg156 and Glu119. The oxygen atom of the tetrahydro-
furan ring forms two hydrogen bonds with Arg118. The carbon-
yl oxygen atom linked to the tetrahydrofuran ring forms one hydrogen
bond with Tyr347. A few groups, such as the heterocyclic 2H-chromene
moiety could produce hydrophobic interactions and van der Waals
forces with the residues Arg371 and Pro431 of the 430-cavity. For 5c,
the two carbonyl oxygen atoms of the dihydrofurocoumarin ring may
form six hydrogen bonds with three positively charged arginine residues
(Arg118, Arg292, Arg371) and Tyr406 of 430-cavity. The oxygen of 3,6-
dihydro-2H-pyran ring can also form two hydrogen bonds with the
residue Arg371 and Arg118, respectively. The dihydrofurocoumarin
ring can form extensive hydrophobic interactions with some key
The NA inhibitory activities of the synthesized compounds were
evaluated. Preliminary screening results show that most compounds
have excellent activity. The IC50 values are listed in Table 2. As shown in
Table 2, among all the synthesized compounds, 5b and 5c have signif-
icantly activity against NA with IC50 value of 0.02 µМ and 0.03 µМ,
respectively. The inhibitory activity of 5f is comparable to the positive
control OSC and its IC50 value is 0.04 µM. The activities of 5d (IC50
=
30.01 µM), 5 h (IC50 = 39.46 µM), 5i (IC50 = 36.25 µM) and 5j (45.76
µM) are comparable. Particularly, the IC50 value of 5b is about 1.5-fold,
2.0-fold against NA compared with 5c, and 5f. It is worth noting that the
difference between 5b, 5c and the other synthesized compounds is the
coexistence of strong electron-withdrawing nitro and hydroxyl groups at
the meta- or para- position, which may increase their activity. Compound
5 g has the poorest activity with IC50 value of 99.84 µМ. The only dif-
ference between 5 g and other compounds is the existence of a fluorine
group at the ortho-position, which decreases its activity. This trend could
be rationalized with the strong electron-withdrawing effect of fluorine.
Combined with the MD simulation results, as seen from Table 1 and
Table 2, 5b and 5c have the lowest ΔGbind values among all synthesized
compounds and they show the most potency against NA. The ΔGbind
values of 5d (ꢀ 16.83 kcal⋅molꢀ 1) and 5e (ꢀ 16.68 kcal⋅molꢀ 1) predicted
by MM/PBSA are very similar, and their IC50 values are also very close
6

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
residues Ile427, Pro431, and Arg430 of the 430-cavity. The nitro per-
formed hydrophobic interaction and van der Waals forces with some key
residues, namely, Ser246, Ile222 and Arg224 of the active site. The
hydroxyl oxygen atom linked to the benzene ring form one hydrogen
bond with the carboxylate of Glu276. Compared with 5a and 5c, the
total number of hydrogen bonds formed by 5c (9) is slightly larger than
that formed by 5a (7), the hydrophobic interactions between 5a and
430-cavity seems weaker since fewer residues were involved. It is
revealed that the interactions between designed compounds and key
amino acids around the 430-cavity play an important role in improve-
ment of NA inhibitory activity. The enzyme inhibition test results are in
good agreement with the docking results of 5a and 5c. The phenyl
moiety of 5b and 5f can both well implant into the active site of neur-
aminidase. Different from 5a and 5c, it is interesting that the dihy-
drofurocoumarin rings of 5b and 5f fit into not 430-cavity but 150-
cavity. Compared with 5b and 5f, the strong electron-withdrawing
nitro of the benzene ring of 5b may form ten hydrogen-bonds with
residues Tyr406, Tyr347 and three positively charged arginine residues
(Arg118, Arg292, Arg371), which are essential for the NA inhibitory
activity.34 The hydroxyl oxygen atom linked to the benzene ring form
two hydrogen bonds with Arg292 and Tyr406, respectively. The oxygen
atom of the tetrahydrofuran ring could develop one hydrogen-bond with
Arg118. In addition, the aromatic or heterocyclic group of dihydrofur-
ocoumarin ring in 5b fit into the 150-cavity and interact with residues
Gly147, Val116, Thr439, and Ile437, which form a big hydrophobic
pocket. The oxygen of 3,6-dihydro-2H-pyran ring can also form one
hydrogen bond with the residue Gln136. Although the dihydrofur-
ocoumarin ring forms only two hydrogen bonds with the residues
Arg118 and Gln136, the strong hydrophobic interaction between the
ring and the hydrophobic pocket of 150-cavity can compensate for this
loss, resulting in its good inhibition potency. 5f can also form five
hydrogen bonds with the key residues of 150-cavity and actives site of
NA. However, there is only one of the three essential arginine residues
(Arg118, Arg292, Arg371) involved to form hydrogen bond. On the
other hand, the hydrophobic interaction between the dihydrofur-
ocoumarin ring and 150-cavity seems weaker because the number of
involved residues (Gly147, Gln136, and Thr439) is less than that
interacted with 5b. The docking results can easily explain why 5b have
better potency compared with 5f. At the same time, the results show the
importance of 150- and 430-cavities for this series of NA inhibitors.
In this study, a novel lead NA inhibitor ZINC05577497 (IC50 = 0.11
µM) was discovered by docking-based virtual screening, MD simulation,
and biological assay. Structural optimization of ZINC05577497 leads to
discovery of a series of 10 novel NA inhibitors (5a–5j). Most inhibitors
have excellent inhibitory activity in the range of potency of FDA-
approved oseltamivir carboxylate. Particularly, compound 5b exerts
the most potency with IC50 value of 0.02 µM, which is better than the
reference oseltamivir carboxylate (OSC) (IC50 = 0.04 µM). Molecular
docking results show that the good inhibition activity of 5b may be
attributed to the elongation of dihydrofurocoumarin ring to the 150-cav-
ity. The 430- and 150-cavities play important roles in determining the
activities against NA for this series of dihydrofurocoumarin derivates.
The obtained results here hope to be helpful for the development of new
neuraminidase inhibitors.
Innovation Found (No. XTCX2016-14), Shanghai Municipal Education
Commission (Plateau Discipline Construction Program).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.127839.
References
1 Forrest HL, Webster RG. Perspectives on influenza evolution and the role of research.
Anim Health Res Rev. 2010;11:3–18.
2 Jusoh N, Zainal H, Abdul Hamid AA, Bunnori NM, Abd Halim KB, Abd Hamid S. In
silico study of carvone derivatives as potential neuraminidase inhibitors. J Mol
Model. 2018;24:1–13.
3 Zhang WQ, Webster RG. Can we beat influenza? Science. 2017;357, 111-111.
4 Ye DJ, Shin WJ, Li N, et al. Synthesis of C-4-modified zanamivir analogs as
neuraminidase inhibitors and their anti-AVI activities. Eur J Med Chem. 2012;54:
764–770.
5 Jia RF, Zhang J, Ai W, et al. Design, synthesis and biological evaluation of “Multi-
site”-binding influenza virus neuraminidase inhibitors. Eur J Med Chem. 2019;178:
64–80.
6 Hsieh HP, Hsu J-T-A. Strategies of development of antiviral agents directed against
influenza virus replication. Curr Pharm Des. 2007;13:3531–3542.
7 Xiao MW, Ye J, Lian WW, et al. Microwave-assisted synthesis, characterization and
bioassay of acylhydrazone derivatives as influenza neuraminidase inhibitors. Med
Chem Res. 2017;26:3216–3227.
8 Noh HJ, Chowdhury MYE, Cho S, et al. Programming of influenza vaccine broadness
and persistence by mucoadhesive polymer-based adjuvant systems. J Immunol. 2015;
195:2472–2482.
9 Wu XA, Wu XL, Sun QZ, et al. Progress of small molecular inhibitors in the
development of anti-influenza virus agents. Theranostics. 2017;7:826–845.
10 Shie JJ, Fang JM. Development of effective anti-influenza drugs: congeners and
conjugates-a review. J Biomed Sci. 2019;26:1–20.
11 Gong JZ, Xu WF, Zhang J. Structure and functions of influenza virus neuraminidase.
Curr Med Chem. 2007;14:113–122.
12 Yaeghoobi M, Frimayanti N, Chee CF, et al. QSAR, in silico docking and in vitro
evaluation of chalcone derivatives as potential inhibitors for H1N1 virus
neuraminidase. Med Chem Res. 2016;25:2133–2142.
13 Liu Y, Zhang L, Gong JZ, et al. Design, synthesis, and biological activity of thiazole
derivatives as novel influenza neuraminidase inhibitors. J Enzym Inhib Med Ch. 2011;
26:506–513.
14 Meng FJ, Sun T, Dong WZ, Li MH, Tuo ZZ. Discovery of novel pyrazole derivatives as
potent neuraminidase inhibitors against influenza H1N1 virus. Arch Pharm. 2016;
349:168–174.
15 Grienke U, Schmidtke M, Kirchmair J, et al. Antiviral potential and molecular insight
into neuraminidase inhibiting diarylheptanoids from Alpinia Katsumadai. J Med
Chem. 2010;53:778–786.
16 Russell RJ, Haire LF, Stevens DJ, et al. The structure of H5N1 avian influenza
neuraminidase suggests new opportunities for drug design. Nature. 2006;443:45–49.
17 Feng EG, Shin WJ, Zhu XL, et al. Structure-Based design and synthesis of C-1 and C-4-
Modified analogs of zanamivir as neuraminidase inhibitors. J Med Chem. 2013;56:
671–684.
18 Xie YC, Xu DQ, Huang B, et al. Discovery of N-Substituted oseltamivir derivatives as
potent and selective inhibitors of H5N1 influenza neuraminidase. J Med Chem. 2014;
57:8445–8458.
19 Lill M. Virtual screening in drug design. Methods Mol Biol. 2013;993:1–12.
20 Jain AN. Scoring functions for protein-ligand docking. Curr Protein Pept Sc. 2006;7:
407–420.
21 Bursulaya BD, Totrov M, Abagyan R, Brooks CL. Comparative study of several
algorithms for flexible ligand docking. J Comput Aid Mol Des. 2003;17:755–763.
22 Onodera K, Satou K, Hirota H. Evaluations of molecular docking programs for virtual
screening. J Chem Inf Model. 2017;47:1609–1618.
23 Liu XW, Shi DF, Zhou SY, Liu HL, Liu HX, Yao XJ. Molecular dynamics simulations
and novel drug discovery. Expert Opin Drug Dis. 2018;13:23–37.
24 Vay JL, Fawley W, AMBER User’s Manual, 2000.
25 Hou TJ, Wang JM, Li YY, Wang W. Assessing the performance of the MM/PBSA and
MM/GBSA methods. 1. The accuracy of binding free energy calculations based on
molecular dynamics simulations. J Chem Inf Model. 2011;51:69–82.
26 Liu M, Yuan MG, Luo MX, Bu XZ, Luo HB, Hu XP. Binding of curcumin with
glyoxalase I: molecular docking, molecular dynamics simulations, and kinetics
analysis. Biophys Chem. 2010;147:28–34.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
27 Massova I, Kollman PA. Combined molecular mechanical and continuum solvent
approach (MM-PBSA/GBSA) to predict ligand binding. Perspect Drug Discov. 2000;18:
113–135.
28 Kollaman PA, Massova I, Reyes C, et al. Calculating structures and free energies of
complex molecules: combing molecular mechanics and continuum models. Acc Chem
Res. 2000;33:889–897.
Acknowledgments
29 Homeyer N, Gohlke H. Free energy calculations by the molecular mechanics Poisson-
Boltzmann Surface Area Method. Mol Inform. 2012;31:114–122.
30 Jiang N, Huang QC, Liu J, et al. Design, synthesis and biological evaluation of new
coumarin-dithiocarbamate hybrids as multifunctional agents for the treatment of
Alzheimer’s disease. Eur J Med Chem. 2018;146:287–298.
Thanks for Natural Science Foundation of Shanghai (No.
15ZR1440400), Middle and Youth teachers Scientific and Technological
Talents Developing Fund (No. 2Q2018-20) and Collaborative
7

Z.J. Zhong et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127839
31 Liu RJ, Tang X, Wang Y, et al. A fluorescent chemosensor for relay recognition of Fe3
+ and PO34- in aqueous solution and its applications. Tetrahedron. 2017;73:
5229–5238.
33 Haudecoeur R, Ahmed-Belkacem A, Yi W, et al. Discovery of naturally occurring
aurones that are potent allosteric inhibitors of hepatitis C virus RNA-Dependent RNA
polymerase. J Med Chem. 2011;54:5395–5402.
32 Traven VF. Dihydrofurocoumarinones-new useful intermediates for substituted and
condensed furocoumarins. Arkivoc. 2000;1:523–562.
34 Varghese JN, Epa VC, Colman PM. Three-dimensional structure of the complex of 4-
guanidino-Neu5Ac2en and influenza virus neuraminidase. Protein Sci. 1995;4:
1081–1087.
8