Synthesis and SARs of dopamine derivatives as potential inhibitors of influenza virus PAN endonuclease

Article abstract of DOI:10.1016/j.ejmech.2020.112048

Currently, influenza PA_N endonuclease has become an attractive target for development of new drugs to treat influenza infections. Herein we report the discovery of new PA_N endonuclease inhibitors derived from a chelating agent dopamine moiety. A series of dopamine amide derivatives and their conformationally constrained 1,2,3,4-tetrahydroisoquinoline-6,7-diol-based analogs were elaborated and assayed against influenza virus A/WSN/33 (H1N1). Most compounds exhibited moderate to excellent antiviral activities, generating a preliminary SARs. Among them, compounds 14 and 19 showed stronger anti-IAV activity compared with the reference Peramivir. Moreover, 14 and 19 demonstrated a concentration-dependent inhibition of PA_N endonuclease based on both FRET assay and SPR assay. Docking studies were also performed to elucidate the binding mode of 14 and 19 with the PA_N protein and to identify amino acids involved in their mechanism of action, which were well consistent with the biological data. This finding was beneficial to laying the foundation for the rational development of more effective PA_N endonuclease inhibitors.

Full text of DOI:10.1016/j.ejmech.2020.112048

Journal Pre-proof
Synthesis and SARs of dopamine derivatives as potential inhibitors of influenza virus
PA endonuclease
N
Yixian Liao, Yilu Ye, Sumei Li, Yilian Zhuang, Liye Chen, Jianxin Chen, Zining Cui,
Lijian Huo, Shuwen Liu, Gaopeng Song
PII:
S0223-5234(20)30015-5
DOI:
https://doi.org/10.1016/j.ejmech.2020.112048
EJMECH 112048
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 25 November 2019
Revised Date: 7 January 2020
Accepted Date: 7 January 2020
Please cite this article as: Y. Liao, Y. Ye, S. Li, Y. Zhuang, L. Chen, J. Chen, Z. Cui, L. Huo, S.
Liu, G. Song, Synthesis and SARs of dopamine derivatives as potential inhibitors of influenza virus
PA endonuclease, European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/
N
j.ejmech.2020.112048.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Masson SAS.

Synthesis and SARs of dopamine Derivatives as Potential Inhibitors
of Influenza Virus PAN Endonuclease
Yixian Liaoa, b*,1, Yilu Ye c ,1, Sumei Lid, Yilian Zhuanga, Liye Chena, Jianxin Chene, Zining Cuib,
Lijian Huoa, Shuwen Liuc*, Gaopeng Songa*
aCollege of Materials and Energy, South China Agricultural University, Guangzhou, 510642,
China
bState Key Laboratory for Conservation and Utilization of Subtropical Agro-bioresources,
Integrative Microbiology Research Centre, Guangdong Province Key Laboratory of Microbial
Signals and Disease Control, South China Agricultural University, Guangzhou 510642, China
cSchool of Pharmaceutical Sciences, Southern Medical University, Guangzhou, 510515, China
dDepartment of human anatomy, School of Medicine, Jinan University, Guangzhou, 510632,
China
eGuangdong Provincial Key Laboratory of Veterinary Pharmaceutics Development and Safety
Evaluation, College of Veterinary Medicine, South China Agricultural University, Guangzhou
510642, China
Abstract: Currently, influenza PAN endonuclease has become an attractive target for development
of new drugs to treat influenza infections. Herein we report the discovery of new PAN
endonuclease inhibitors derived from a chelating agent dopamine moiety. A series of dopamine
amide
derivatives
and
their
conformationally
constrained
1,2,3,4-tetrahydroisoquinoline-6,7-diol-based analogs were elaborated and assayed against
influenza virus A/WSN/33 (H1N1). Most compounds exhibited moderate to excellent antiviral
activities, generating a preliminary SARs. Among them, compounds 14 and 19 showed stronger
anti-IAV activity compared with the reference Peramivir. Moreover, 14 and 19 demonstrated a
concentration-dependent inhibition of PAN endonuclease based on both FRET assay and SPR
assay. Docking studies were also performed to elucidate the binding mode of 14 and 19 with the
PAN protein and to identify amino acids involved in their mechanism of action, which were well
consistent with the biological data. This finding was beneficial to laying the foundation for the
rational development of more effective PAN endonuclease inhibitors.
Keywords: Influenza A virus; PAN endonuclease inhibitors; polyphenols; SARs
1. Introduction
Influenza A viruses (IAVs) are associated with occasional pandemics that usually
* Corresponding author. Tel.: +86- 20- 6164 8538; fax: +86 -20 -6164 8655
E-mail address: liusw@smu.edu.cn (S. Liu).
* Corresponding author. Tel.: + 86-20-85280293; fax: + 86-20-85280292
E-mail address: songgp1021@scau.edu.cn (G. Song).
1These authors contributed equally to this work

cause contagious pandemic respiratory illnesses, high morbidity and mortality rates
[1-2]. Currently, two classes of antiviral drugs have been clinically used that comprise
the M2 protein blockers (amantadine and rimantadine) and neuraminidase (NA)
inhibitors (oseltamivir and zanamivir). However, the growing resistance to above
approved drugs underlines the need for the development of novel therapeutics with
different mechanisms of action on viral life cycles [3-5].
IAVs belong to the Orthomyxoviridae family, and its genome consists of
segmented single stranded RNA (-) that encodes ten different viral protein [6,7].
Thereinto, the three largest RNA segments encode three subunits of viral RdRP:
polymerase acidic protein (PA), polymerase basic protein 1 (PB1) and polymerase
basic protein 2 (PB2) [8]. At the beginning of replication, the PB2 subunit binds the
5’-mRNA cap of a host cell, and in the second step the PA subunit cleaves 10-13
nucleotides downstream to yield 5’-capped RNA fragments. These fragments serve as
primers for viral mRNA elongation catalyzed by the PB1 subunit in the third step [9].
PAN endonuclease is highly conserved across all influenza strains and subtypes
and has no human analogue [10]. Currently, PAN protein has represented an attractive
target for new antiviral drugs. Studies have demonstrated that PAN domain contains
the endonuclease active site where a dinuclear metal active site, employing two Mn2+
or Mg2+ cations, is critical for the endonuclease activity [11]. Thus, PAN endonuclease
inhibitors that coordinate the metal centers of the endonuclease active site, can
effectively inhibit endonuclease activity and prevent the processing of the biological
substrates. To date, several series of direct PAN endonuclease inhibitors have been
reported in recent literature that can be chemically classified as diketo acid derivatives
[12-15], catechol derivatives [16], flutimide derivatives [17,18], hydroxylated
heterocycles (3-hydroxyquinolin-2(1H)-ones, and other scaffolds) [19-23]. Notably,

Xofluza, a typical hydroxylated heterocycles, has been used in clinic in Japan and
approved by FDA in October 2018, which is a highly selective and nontoxic inhibitor
without interfering with function of the host cell [24, 25]. However, widely use of
Xofluza will prompt the influenza A virus PA-I38T variants such as H3N2 and
A/H1N1pdm [26], which confers reduced susceptibility to Xofluza without a loss of
viral fitness. Moreover, the variants also transmitted efficiently between ferrets by
respiratory droplets [26]. Considering this, there is an urgent need for the
development of novel PAN endonuclease inhibitors to prevent and treat the influenza
infection.
Of particular interest, polyphenols have gained a lot of attention due to their
broad efficacy against influenza PAN endonuclease [27, 28]. For example,
polyphenols such as tea catechin (1), marchantin B (2), perrottetin F (3),
phenethylphenylphthalimide analog (4), salicylic acid derivative (5) and gallic acid
derivative (6) (Figure 1) have been found to modulate PA endonuclease function
[27-30]. Structurally, all of the above polyphenols possess a catechol moiety,
indicating its crucial role in the PAN endonuclease inhibitory activity. This finding was
further confirmed based on the co-crystal of the PAN endonuclease with
(-)-epigallocatechin gallate (1) [13], in which the two manganese ions were
cochelated with two of the hydroxyl groups of the gallo group in deep active site cleft.
However, most of polyphenols derivatives produced only modest or weak anti-IAV
activity (EC50 > 15 µM) in the host cells, despite their reasonable PAN endonuclease
inhibition activity.
Insert < Figure 1>
Given the SARs of above polyphenols derivatives, we recognized that there are

several key pharmacophoric regions where appropriate substituents are essential
(Figure 2). a) 3,4-dihydroxyphenethyl moiety is indispensable, whereas an additional
aromatic moiety is important at the para-position of 3,4-dihydroxyphenethyl fragment
through appropriate linkers. b) In general, the length of linkers was less than 4 C. c)
both different substituents such as fluorine, chlorine and the aromatic moiety could
bind to the active pocket via hydrophobic interaction. Encouragingly, the
dihydroxyindole derivative 7 as an effective PAN endonuclease inhibitor displayed
remarkable anti-IAV activity in MDCK cells (EC90 = 3.2 µM), of which both
hydroxyl groups on the dihydroxyindole moiety chelated two metal ions and the
catechol functional group was engaged in the Val122-Arg124-Tyr130 cavity that is of
crucial importance [31]. Careful analysis of docking studies on the endonuclease with
7 and its analogues, we inferred that not only the active site of PA protein can
accommodate aromatic molecules connected by longer linkage (more than 3 C), but
also there is a preference for the aromatic molecules with hydroxyl groups, which
may also make very favorable interactions with amino acids in hydrophobic pockets.
Insert < Figure 2>
More recently, Ferro et al carried out a three-dimensional pharmacophore model
and found the dopamine moiety (3,4-dihydroxyphenethylamine) seemed a great
fragment capable of chelating the two metal ions [32]. Guided by both the above
SARs outlined in Figure 2 and prior docking studies, we started our structure
modification study of dopamine moiety to design the title series A (8-17). Initial
efforts focused on exploring aryl amides 8-13 derived from the coupling of dopamine
and several hydroxybenzoic acids. With the aim to investigate the influence of

linkage-length on antiviral activity, introduction of methylene or ethylene spacers
between the amide and aryl ring gave rise to derivative 14 or 15, respectively.
Moreover, compounds 16 and 17, bioisosteric surrogates of 14, were designed to
study the effects of type of linkages on antiviral activity. Considering the
conformational restriction has been frequently introduced to some structurally flexible
ligands for optimizing the biological activity and specificity [33, 34], we prepared the
title series
B
(18-20) by replacing the flexible dopamine with
1,2,3,4-tetrahydroisoquinoline-6,7-diol fragment that bears more conformational
rigidity. Herein, we report an extensive SARs study leading to the identification of the
lead compound 14 with potential antiviral activity in vitro (EC50 = 2.5 µM for WSN33)
and excellent PAN endonuclease inhibition (IC50 = 0.3 µM).
Insert < Figure 3>
2. Results and discussion
2.1. Chemistry
Synthesis
of
dopamine
derivative
derivatives
8-15
and
1,2,3,4-tetrahydroisoquinoline-6,7-diol 18-20 from commercially available
3,4-dihydroxybenzaldehyde was outlined in Scheme 1. The phenolic hydroxyl groups
of 21 were selectively protected as a benzyl ether by treatment with benzyl bromide in
the presence of potassium carbonate to yield 22. The intermediate 23 was derived
from 22 and then reduced to benzyl-protected dopamine 24 using lithium aluminum
hydride. Condensation of various carboxylic acid followed by selective
hydrogenolysis of benzyl ether over palladium/carbon furnished 8-15 in good yields.
1,2,3,4-Tetrahydroisoquinoline ring 33 was accomplished via the Pictet-Spengler

reactions. Amide derivatives 18-20 were generated by the coupling of various acids to
33, followed by hydrogenolysis of benzyl ether.
Insert <Scheme 1>
Derivative 16 was done by the similar route as that for compound 14 from the
known compound 37, as shown in Scheme 2. Consider palladium carbon was
sensitive
to
sulfur
element,
the
commercially
available
2-(3,4-dimethoxyphenyl)ethan-1-amine 38 was treated with the known
2-(3,4-dimethoxyphenyl)acetic acid in the presence of EDC•HCl and DMAP to give
40, which was subject to Lawesson’s reagent to provide thioamide intermediate 41.
Hydrolysis of methoxy groups in the presence of BBr3 efficiently afforded 17.
Insert <Scheme 2>
2.2. Biological assays
2.2.1. Anti-influenza virus activity of polyphenol derivatives 8-20
The cytopathic effect (CPE) assay and the antiviral effect quantitatively evaluated
with the cell-based MTT assay were utilized in parallel to demonstrate their antiviral
abilitiws of these polyphenols derivatives 8-20 [35]. And the cytotoxicity of
polyphenols
derivatives
was
assessed
by
the
3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay [36]. As
shown in Figure 4, compound 14 dramatically reduced the CPE in 20 µM, indicating
that these polyphenols derivatives were able to protect MDCK cells from influenza
virus-induced CPE. Among them, 14 displayed the strongest antiviral activity (see

Table 1) and had an increased 2-fold or 12-fold cellular activity than the control
Peramivir or EGCG, respectively.
Insert <Figure 4>
Firstly, we analysed the effect of binding hydroxyl group in different positions of
the second phenyl ring, ortho- (8, EC50 > 20.0 µM), meta- (9, EC50 = 13.1 µM) and
para- (10, EC50 = 12.7 µM), suggesting the importance of this structural modification
for antiviral activity. Specifically, moving the para-OH substituent to the meta- or
ortho -position on the second phenyl ring resulted in a slightly decreased or total loss
of activity, revealing that the hydroxyl substitution at the para-position of the phenyl
ring was indispensable and the most optimal substituent for this position. There was a
similar trend in 2,3-dihydroxyl substitution (11) as that of ortho-OH analogue 8,
probably due to the presence of intramolecular hydrogen bond formed by ortho-OH
and carbonyl group. Of particular interest, 3,4-dihydroxyl substitution compound (12)
showed enhanced inhibition over monosubstituted 9 or 10. With the identification of
potent analog 12 with 3,4-dihydroxyl substitution, we attempted to attach a third
hydroxyl group in the meta-position of the second phenyl ring. However,
3,4,5-trihydroxyl derivative 13 exhibited a significant loss of activity relative to the
disubstituted compound 12. As observed, both the position and number of the
substituents in the second phenyl ring were responsible for the potentiation of
antiviral activity. Having optimized the different positions of the phenyl ring, we
turned our attention to both the length and type of linkers. Comparing the anti-IAV
activity of 14 or 15 with 12, it was found that the length of the carbon chain (n) from
0 to 2 resulted in a substantial increase in potency though there was no correlation

between the length of the carbon chain and antiviral activity. The activity in the order
of compounds 14 > 15 > 12 revealed that 3,4-dihydroxyphenylmethylene (n =1) was
the most potent elaboration at this position, likely owing to both favorable
hydrophobic interactions with pocket 3 or 4 (Fig. 7) and possible additional hydrogen
bond interactions with key hydrophilic residues on the external surface. Most
importantly, 14 not only exhibited the best anti-IAV activity among this set of
compounds, but also the remarkablly improved selective index (CC50/EC50)
comparable to 12. We next explored the linkage moiety by replacement of amide with
ester or thioamide and found such a small change substantially had an important
effect on anti-IAV activity (14 vs 16, 14 vs 17). The CPE assay demonstrated that 17,
an analogue of 14, in which an amide bond was replaced by the ester link had a
slightly reduced potency compared to 14, indicating the compatibility of amide and
ester linkers for anti-A/WSN/33 virus. However, replacement of the amide group in
14 by the thioamide group, yielding 16, led to the total loss of activity, revealing that
the crucial nature of the amide linkage for anti-IAV activity. This result indicated that
both the style and length of the linkers were a critical factor for antiviral activity.
Insert <Table 1>
With 12, 14, 15 as our most potent analogs in the series A, we made other push to
enhance inhibition by restricting conformational freedom with a planarizing fusion of
the
catechol
fragment
and
primary
amine
to
furnish
the
1,2,3,4-tetrahydroisoquinoline derivatives 18-20. Surprisingly, compound 18
displayed a significant 3-fold increase in potency compared with the open-chain
congener 12. Of the other two analogues 19-20, 19 performed slightly less

comparable with 14, whereas 20 showed approximately 4-fold decreased activity than
15. However, restricting conformational modification was detrimental for selective
index (Table 2). Base on this result, we envisioned that restriction of rotation across
the amide linkage might be crucial for antiviral activity and
1,2,3,4-tetrahydroisoquinoline derivatives maybe bind different regions and key
amino acids in PA endonuclease active site from open-chain analogues in the series A.
Insert <Table 2>
2.2.2. The title compounds targeted PAN endonuclease
In the next stage, compounds 14-15 and 17-19 were subjected to PAN endonuclease
enzymatic assay using fluorescence resonance energy transfer (FRET) inhibition
assay [37, 38]. As shown in Table 3, the most potent antiviral compounds 14 and 19
within this series, exhibited stronger inhibition against the endonuclease activity than
other three analogs, yielding EC50 values of 312.36 nM and 489.39 nM, respectively,
which were superior to EGCG (EC50 = 6454.75 nM) but inferior to Baloxavir acid
(BXA) (EC50 = 1.37 nM). As depicted in Figure 5A and 5B, a
concentration-dependent inhibitory effect of compound 14 or 19 was observed on the
PAN endonuclease, which was similar to that of the positive control EGCG (Figure
S1A) or BXA (Figure S1B). The good correlation between the results from the
enzymatic assay and cell-based methods supported our hypothesis that the anti-IAV
activity of compound 14 or 19 in cell culture was related to inhibition of PA-Nter.
Insert <Table 3>
Insert <Figure 5>

Considering that PAN protein is a valuable target to develop antiviral agents, the
possible interaction between 14 or 19 and influenza PAN protein was further evaluated
using SPR spectroscopy. It was found that compounds 14 and 19 immobilized on the
Photo-cross-linker Sensor CHIP could bind tightly to PAN endonuclease at
concentrations of 10-160 nM, which was observed in a dose-dependent manner. As
shown in Figure 6, 14 and 19 exhibited potential binding affinity to PAN protein with
KD values of 26.33 nM and 94.19 nM, respectively, which was substantially higher
relative to EGCG (Figure S2A, KD = 956.32 nM) but lower than BXA (KD = 207.51
pM, Figure S2B). In the light of these findings, it was reasonable to assume that 14
and 19 had the ability to exert anti-influenza activity by targeting PAN endonuclease.
Insert <Figure 6>
2.2.3. Molecular docking analysis
In view of PAN endonuclease as the potential target, we designed molecular
docking calculations of compounds 14 and 19 with influenza PAN endonuclease to
further understand the molecular basis of the inhibitory properties of polyphenols
derivatives through Surflex-Dock program in SYBYL 7.3 [39]. The optimum docked
conformations of PA-compounds were determined based on the Total_Score and
CSCORE. Hence, the selected compound 14 or 19 was docked into the published
structure of H1N1 PAN endonuclease bound with BXA (PDB: 6FS6) [40].
Our
result
indicated
that
either
the
dopamine
moiety
or
1,2,3,4-tetrahydroisoquinoline-6,7-diol fragment was able to direct toward to PAN
endonuclease active site for metal chelation and form hydrophobic contact with the

key residue His41. Specifically, it was found that the hydroxyl in 4-position of
dopamine moiety could not only chelate with two catalytic metal ions but also form a
hydrogen bond with the key residues Asp108 while the 3-hydroxyl substituent of
dopamine could chelate with Mn1. As we have hypothesized above, the
3,4-dihydroxyphenylacetamide moiety of 14 was able to fit well into pocket 3, with
the aryl ring making hydrophobic interactions with Tyr24 and the hydroxyl in
para-position making a favorable hydrogen bond to Glu26. In contrast, the
1,2,3,4-tetrahydroisoquinoline-6,7-diol fragment was found to be too rigid to allow
for chelating simultaneously with two catalytic metal ions but allow for orientating
toward the pocket 4 and forming double hydrogen bonds with Ser194. Moreover, the
aryl ring in 19 was able to form multiple hydrophobic contacts with Lys34 and Ala134,
indicating that the larger size and greater number of available hydrophobic contacts
and hydrogen interactions may account for the retained similar inhibitory activity as
14, although with slightly less intensity of metal-ligand interactions than 14. Of note,
neither 14 nor 19 could form contact with Ile38. Therefore, we hypothesized that 14
or 19 might not change their binding affinity with PAN endonuclease with I38T
mutation.
Insert <Figure 7>
3. Conclusions
With the aim of developing a new class of PAN endonuclease inhibitors, we
designed
and
synthesized
liner
dopamine
amide
derivatives
and
1,2,3,4-tetrahydroisoquinoline-6,7-diol-based conformationally constrained analogs,
respectively, of which anti-IAV activities were evaluated in MDCK cells. The
bioactivity results allowed for the generation of a preliminary SARs. The presence of

catechol moiety chelated to manganese ions was crucial for activity retention and both
the position and number of the hydroxyl substituents in the second phenyl ring were
responsible for the potentiation of anti-IAV activity, with the simultaneous
replacement in the position meta- and para-positions most favorable. Both the style
and length of the amide linkages played a key influence on antiviral activity.
Conversion of the dopamine to 1,2,3,4-tetrahydroisoquinoline-6,7-diol ring led to an
increase in cytotoxicity while keeping antiviral potency. Among all compounds tested,
14 and 19 exhibited comparable anti-IAV activity in relative to the reference
Peramivir with a high SI. Based on FRET assay and SPR assay, the most potential 14
and 19 were found to show good antiviral ability by binding PAN endonuclease,
consistent well with the analysis result of docking study. To summarize, 14 and its
active derivatives are promising antiviral agents that should be further investigated to
develop new PAN endonuclease inhibitors with preclinical relevance.
4. Experimental section
4.1. Chemistry: General Methods.
All reactions, unless otherwise stated, were done under nitrogen atmosphere. Reaction
monitoring and follow-up were done using precoated E. Merck silica gel 60 F254
plates, visualizing with ultraviolet light. Flash column chromatography was done on
silica gel (200-300 mesh, Qingdao, China) using petroleum ether, ethyl acetate,
dichloromethane and methanol. The 1H (600 MHz) and 13C (150 MHz) NMR spectra
were done in CDCl3 and DMSO-d6 and recorded on a JEOL JNM-ECP 600
spectrometer with tetramethylsilane as an internal standard, and chemical shifts were
recorded in ppm values. Mass spectra were recorded on a Q-TOF Global mass
spectrometer.

4.2. 3,4-Bis(benzyloxy)benzaldehyde (22)
To a solution of 3,4-dihydroxybenzaldehyde (5.0 g, 36.20 mmol), potassium
carbonate (20.1 g, 144.80 mmol) in DMF (100 mL) was added benzyl bromide (17.19
mL, 144.80 mmol) at room temperature. The mixture was heated to 65 ºC for 10 h.
After cooling, the solvent was evaporated under vacuum. The residue was partitioned
between EtOAc and water. The combined organic layers were washed with brine,
dried over Na2SO4, filtered and concentrated under reduced pressure. The crude
product was purified by silica gel chromatography [EtOAc : petroleum ether : CH2Cl2,
1
15 : 1 : 1] to afford 22 as a white solid (11.10 g, 96%). H NMR (600 MHz,
Chloroform-d) δ 9.83 (s, 1H, Ar-CHO), 7.51 (d, 1H, J = 1.9 Hz, Ar-H), 7.50-7.45 (m,
4H, Ar-H), 7.43 (dd, 1H, J = 8.2, 1.9 Hz, Ar-H), 7.42-7.38 (m, 4H, Ar-H), 7.37-7.32
(m, 2H, Ar-H), 7.04 (d, 1H, J = 8.2 Hz, Ar-H), 5.28, 5.23 (each s, each 2H, Ar-CH2-O);
ESIMS: calcd for [M+H]+ m/z 319.1; found, 319.1.
4.3. (E)-(((4-(2-Nitrovinyl)-1,2-phenylene)bis(oxy))bis(methylene))dibenzene (23)
A mixture of 22 (11.52 g, 36.18 mmol) and nitromethane (5.81 mL, 108.54 mmol) in
glacial acetic acid (35 mL) was stirred at r. t. Then ammonium acetate (2.79 g, 36.18
mmol) was added and the mixture was stirred at 90 ºC for 5 h. After removing the
solvents under vacuum, the residue was poured into water and extracted with EtOAc.
The combined organic layers were washed with brine, dried over Na2SO4, filtered and
evaporated. The crude product was subjected to silica gel chromatography and eluted
with EtOAc : petroleum ether : CH2Cl2 [15 : 1 : 1] to provide 23 as a yellow solid
1
(11.79 g, 90%). H NMR (600 MHz, Chloroform-d) δ 7.91 (d, 1H, J = 13.5 Hz,
CH=C), 7.49-7.46 (m, 3H, Ar-H), 7.45 (d, 1H, J = 13.5 Hz C=CH), 7.44-7.37 (m, 5H,
Ar-H), 7.37-7.34 (m, 2H, Ar-H), 7.13 (dd, 1H, J = 8.3, 2.1 Hz, Ar-H), 7.09 (d, 1H, J =

2.1 Hz, Ar-H), 6.98 (d, 1H, J = 8.3 Hz, Ar-H), 5.25, 5.21 (each s, each 2H, Ar-O-CH2);
ESIMS: calcd for [M+Na]+ m/z 384.1; found, 384.2.
4.4. 2-(3,4-Bis(benzyloxy)phenyl)ethan-1-amine (24)
To a stirring solution of 23 (6.15 g, 17.02 mmol) in anhydrous THF (100 mL) was
added a portion of lithium aluminum hydride (4.0 g, 105.52 mmol) at 0 ºC under N2
atmosphere and the resulting mixture was warmed gradually to 65 ºC. The mixture
was stirred overnight at the same temperature until TLC indicated the reaction was
complete. After cooling to 0 ºC, the mixture was carefully treated in succession with
water (4 mL), 15% NaOH (4 mL) and water again (4 mL). The mixture was filtered
through Celite, and the filter cake was washed with tetrahydrofuran. After the filtrate
was concentrated under reduced pressure, the residue was diluted with EtOAc,
washed with brine, and then dried over Na2SO4. The solvent was distilled off, and the
resulting oil was purified by flash chromatography [EtOAc : CH2Cl2, 1 : 1] to provide
24 as a red oil (3.45 g, 61%). 1H NMR (600 MHz, Chloroform-d) δ 7.49-7.45 (m, 4H,
Ar-H), 7.38 (dd, 4H, J = 8.4, 6.7 Hz, Ar-H), 7.34-7.30 (m, 2H, Ar-H), 6.90 (d, 1H, J =
8.1 Hz, Ar-H), 6.81 (d, 1H, J = 2.0 Hz, Ar-H), 6.73 (dd, 1H, J = 8.1, 2.0 Hz, Ar-H),
5.18, 5.16 (each s, each 2H, Ar-O-CH2), 2.90 (t, 2H, J = 6.8 Hz, N-CH2), 2.65 (t, 2H,
J = 6.8 Hz, Ar-CH2); ESIMS: calcd for [M+Na]+ m/z 356.2; found, 356.2.
4.5. 6,7-Bis(benzyloxy)-1,2,3,4-tetrahydroisoquinoline (33)
To a solution of 24 (2.08 g, 6.24 mmol) in anhydrous MeOH (30 mL) was added 5%
paraformaldehyde (237 mg, 7.5 mmol) at room temperature under N2 atmosphere.
After being stirred at 80 ºC for 2 h, trifluoroacetic acid (1 mL, 12.5 mmol) was added
to the reaction mixture and kept at the same temperature for 5 h. Then the reaction
was neutralized with saturated NaHCO3 solution at 0 °C and then extracted with
CH2Cl2 three times. The combined organic extracts were washed with water, dried

over Na2SO4, concentrated in vacuo. A solution of the above crude product in
anhydrous MeOH (5 mL) at 0 °C under N2 atmosphere was treated dropwise with a
suspension of sodium borohydride (1.18 g, 31.20 mmol) in anhydrous MeOH (20 mL).
The mixture was stirred at r.t. for 5 h and concentrated under reduced pressure. The
residue was poured into water and extracted with EtOAc. The combined organic
layers were washed with brine, dried over anhydrous Na2SO4, filtered, and condensed.
The resulting residue was purified via flash chromatography [CH2Cl2 : MeOH, 60 : 1]
to obtain 33 as a light yellow solid (600 mg, 33%). 1H NMR (600 MHz, Chloroform-d)
δ 7.48-7.43 (m, 4H, Ar-H), 7.39-7.34 (m, 4H, Ar-H), 7.34-7.29 (m, 2H, Ar-H), 6.69 (s,
1H, Ar-H), 6.61 (s, 1H, Ar-H), 5.13, 5.12 (each s, each 2H, Ar-O-CH2), 3.92 (s, 2H,
Ar-CH2-N), 3.12 (t, J = 6.0 Hz, 2H, N-CH2), 2.71 (t, 2H, J = 6.0 Hz, Ar-CH2); ESIMS:
calcd for [M+Na]+ m/z 368.2; found, 368.2.
4.6. 2-(3,4-Bis(benzyloxy)phenyl)ethan-1-ol (37)
Compound 37 was prepared by the procedure used for compound 24 from the known
1
compound methyl 2-(3,4-bis(benzyloxy)phenyl)acetate in 85% yield. H NMR (600
MHz, Chloroform-d) δ 7.50-7.44 (m, 4H, Ar-H), 7.42-7.35 (m, 4H, Ar-H), 7.35-7.31
(m, 2H, Ar-H), 6.91 (d, 1H, J = 8.1 Hz, Ar-H), 6.85 (d, 1H, J = 2.0 Hz), 6.76 (dd, 1H,
J = 8.1, 2.0 Hz, Ar-H), 5.18, 5.16 (each s, each 2H, Ar-O-CH2), 3.79 (t, 2H, J = 6.5 Hz,
O-CH2), 2.77 (t, 2H, J = 6.5 Hz, Ar-CH2); ESIMS: calcd for [M+Na]+ m/z 357.1;
found, 357.1.
4.7. General procedure for the preparation of compounds (25-32), (34-36), (39-40)
To a solution of an appropriate acid (0.12 mmol) and related amine or alcohol (0.10
mmol) in an anhydrous CH2Cl2 (50 mL) was added EDC·HCl (0.13 mmol) and
DMAP (0.01 mmol) at 0 ºC under N2 atmosphere. Then the mixture was stirred at
room temperature for 12 h. The mixture was diluted with CH2Cl2, washed with 1 M

HCl, saturated NaHCO3 and brine, dried over Na2SO4. After the solvent was distilled
off under reduced pressure, the resulting residues were purified by flash
chromatography to furnish compounds 25-32, 34-36, 39-40, respectively.
4.7.1. 2-(Benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (25)
1
25 was synthesized as a colourless oil in 85% yield. H NMR (600 MHz,
Chloroform-d) δ 8.28 (dd, 1H, J = 7.8, 1.9 Hz, Ar-H), 8.01 (t, 1H, J = 5.5 Hz, NH),
7.45-7.42 (m, 4H), 7.40-7.34 (m, 8H), 7.33-7.29 (m, 4H), 7.12 (td, 1H, J = 7.5, 1.0 Hz,
Ar-H), 7.00 (dd, 1H, J = 8.4, 1.0 Hz, Ar-H), 6.82 (d, 1H, J = 8.1 Hz, Ar-H), 6.79 (d,
1H, J = 2.1 Hz, Ar-H), 6.64 (dd, 1H, J = 8.1, 2.1 Hz), 5.09, 5.09, 5.08 (each s, each
2H, Ar-O-CH2), 3.61 (td, 2H, J = 7.1, 5.5 Hz, N-CH2), 2.69 (t, 2H, J = 7.1 Hz,
Ar-CH2); HRMS (ESI) calculated for C36H33NO4Na (M+Na)+ 566.2307, found
566.2311.
4.7.2. 3-(Benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (26)
26 was synthesized as a white solid in 89% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.47-7.42 (m, 5H, Ar-H), 7.42-7.37 (m, 4H, Ar-H), 7.37-7.29 (m, 7H, Ar-H), 7.19
(dt, 1H, J = 7.7, 1.2 Hz, Ar-H), 7.13-7.08 (m, 2H), 6.91 (d, 1H, J = 8.1 Hz, Ar-H),
6.85 (d, 1H, J = 2.0 Hz, Ar-H), 6.76 (dd, 1H, J = 8.1, 2.0 Hz, Ar-H), 6.09 (t, 1H, J =
5.9 Hz, NH), 5.15, 5.13, 5.10 (each s, each 2H, Ar-O-CH2), 3.66 (q, 2H, J = 6.6 Hz,
N-CH2), 2.84 (t, 2H, J = 6.8 Hz, Ar-CH2) ; HRMS (ESI) calculated for C36H33NO4Na
(M+Na)+ 566.2307, found 566.2312.
4.7.3. 4-(Benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (27)
27 was synthesized as a white solid in 85% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.65 (d, 2H, J = 8.8 Hz, Ar-H), 7.47-7.28 (m, 15H, Ar-H), 6.98 (d, 1H, J = 8.8 Hz,
Ar-H), 6.91 (d, 1H, J = 8.1 Hz, Ar-H), 6.85 (d, 1H, J = 2.0 Hz, Ar-H), 6.75 (dd, 1H, J
= 8.1, 2.0 Hz, Ar-H), 6.03 (t, 1H, J = 5.8 Hz, NH), 5.16, 5.12, 5.11 (each s, each 2H,

Ar-O-CH2), 3.65 (q, 2H, J = 6.7 Hz, N-CH2), 2.83 (t, 2H, J = 6.8 Hz, Ar-CH2);
HRMS (ESI) calculated for C36H33NO4Na (M+Na)+ 566.2307, found 566.2313.
4.7.4. 2,3-Bis(benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (28)
1
28 was synthesized as a yellow oil in 83% yield. H NMR (600 MHz, Chloroform-d)
δ 8.04 (t, 1H, J = 5.7 Hz, NH), 7.80 (dd, 1H, J = 7.2, 2.3 Hz, Ar-H), 7.52-7.28 (m,
18H, Ar-H), 7.25-7.21 (m, 2H, Ar-H), 7.20-7.15 (m, 2H, Ar-H), 6.85 (d, 1H, J = 8.1
Hz, Ar-H), 6.79 (d, 1H, J = 1.9 Hz, Ar-H), 6.67 (dd, 1H, J = 8.1, 1.9 Hz, Ar-H), 5.16,
5.12, 5.11, 5.02 (each s, each 2H, Ar-O-CH2), 3.51 (td, 2H, J = 7.2, 5.7 Hz, N-CH2),
2.64 (t, 2H, J = 7.3 Hz, Ar-CH2); HRMS (ESI) calculated for C43H39NO5Na (M+Na)+
672.2726, found 672.2732.
4.7.5. 3,4-Bis(benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (29)
29 was synthesized as a white solid in 90% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.49-7.41 (m, 9H, Ar-H), 7.39-7.28 (m, 12H, Ar-H), 7.10 (dd, 1H, J = 8.4, 2.1 Hz,
Ar-H), 6.91 (d, 1H, J = 8.1 Hz, Ar-H), 6.88 (d, 1H, J = 8.4 Hz, Ar-H), 6.84 (d, 1H, J =
2.0 Hz, Ar-H), 6.74 (dd, 1H, J = 8.1, 2.0 Hz, Ar-H), 5.96 (t, 1H, J = 5.9 Hz, NH), 5.20,
5.19, 5.15, 5.12 (each s, each 2H, Ar-O-CH2), 3.62 (q, 2H, J = 6.6 Hz, N-CH2), 2.82 (t,
2H, J = 6.8 Hz, Ar-CH2); HRMS (ESI) calculated for C43H39NO5Na (M+Na)+
672.2726, found 672.2734.
4.7.6. 3,4,5-Tris(benzyloxy)-N-(3,4-bis(benzyloxy)phenethyl)benzamide (30)
30 was synthesized as a white solid in 84% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.47-7.20 (m, 25H, Ar-H), 6.97 (s, 2H, Ar-H), 6.88 (d, 1H, J = 7.9 Hz, Ar-H), 6.83
(s, 1H, Ar-H), 6.72 (d, 1H, J = 7.9 Hz, Ar-H), 5.94 (d, 1H, J = 6.1 Hz, NH), 5.11 (s,
4H, 2 × Ar-O-CH2), 5.07 (s, 6H, 3 × Ar-O-CH2), 3.60 (d, 2H, J = 6.7 Hz, N-CH2),
2.80 (t, 2H, J = 7.1 Hz, Ar-CH2); HRMS (ESI) calculated for C50H45NO6Na (M+Na)+
778.3145, found 778.3150.

4.7.7. N-(3,4-Bis(benzyloxy)phenethyl)-2-(3,4-bis(benzyloxy)phenyl)acetamide (31)
31 was synthesized as a white solid in 99% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.47-7.28 (m, 20H, Ar-H), 6.86 (d, 1H, J = 8.1 Hz, Ar-H), 6.80 (d, 1H, J = 8.2 Hz,
Ar-H), 6.78 (d, 1H, J = 2.1 Hz, Ar-H), 6.72 (d, 1H, J = 2.1 Hz, Ar-H), 6.65 (dd, 1H, J
= 8.2, 2.1 Hz, Ar-H), 6.50 (dd, 1H, J = 8.1, 2.1 Hz, Ar-H), 5.32 (t, 1H, J = 5.8 Hz,
NH), 5.14, 5.12, 5.11, 5.10 (each s, each 2H, Ar-O-CH2), 3.41 (s, 2H, Ar-CH2-CO),
3.37 (q, 2H, J = 6.8 Hz, NH-CH2), 2.60 (t, 2H, J = 6.9 Hz, Ar-CH2); HRMS (ESI)
calculated for C44H41NO5Na (M+Na)+ 686.2882, found 686.2892.
4.7.8. (E)-N-(3,4-Bis(benzyloxy)phenethyl)-3-(3,4-bis(benzyloxy)phenyl)acrylamide
(32)
32 was synthesized as a yellow solid in 87% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.50 (d, 1H, J = 15.5 Hz, Ar-CH=C), 7.48-7.40 (m, 8H, Ar-H), 7.40-7.31 (m, 12H,
Ar-H), 7.09 (d, 1H, J = 1.8 Hz, Ar-H), 7.04 (dd, 1H, J = 8.3, 1.8 Hz, Ar-H), 6.91 (d, J
= 8.4 Hz, 1H), 6.90 (d, J = 8.2 Hz, 1H), 6.83 (d, 1H, J = 1.9 Hz, Ar-H), 6.74 (dd, 1H,
J = 8.2, 1.9 Hz, Ar-H), 6.11 (d, 1H, J = 15.5 Hz, CO-CH=C), 5.54 (t, 1H, J = 5.8 Hz,
NH), 5.20, 5.16, 5.16, 5.15 (each s, each 2H, Ar-O-CH2), 3.58 (q, 2H, J = 6.6 Hz,
N-CH2), 2.78 (t, 2H, J = 6.8 Hz, Ar-CH2); HRMS (ESI) calculated for C45H41NO5Na
(M+Na)+ 698.2882, found 698.2886.
4.7.9. (6,7-Bis(benzyloxy)-3,4-dihydroisoquinolin-2(1H)-yl)(3,4-bis(benzyloxy)phenyl)
methanone (34)
34 was synthesized as a white solid in 82% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.48-7.42 (m, 8H, Ar-H), 7.42-.34 (m, 12H, Ar-H), 7.05 (d, 1H, J = 1.9 Hz, Ar-H),
7.02 (dd, 1H, J = 8.2, 1.9 Hz, Ar-H), 6.96 (d, 1H, J = 8.2 Hz, Ar-H), 6.71 (s, 2H,
Ar-H), 5.22, 5.20 (each s, each 2H, Ar-O-CH2), 5.16 (s, 4H, 2 × Ar-O-CH2), 4.85-4.51

(m, 2H, Ar-CH2-N), 3.67-3.40 (m, 2H, N-CH2), 2.68-2.46 (m, 2H, Ar-CH2); HRMS
(ESI) calculated for C44H39NO5Na (M+Na)+ 684.2726, found 684.2734.
4.7.10. 1-(6,7-Bis(benzyloxy)-3,4-dihydroisoquinolin-2(1H)-yl)-2-(3,4-bis(benzyloxy)-
phenyl)ethan-1-one (35)
35 was synthesized as a white solid in 90% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.48-7.42 (m, 12H, Ar-H), 7.40-7.36 (m, 8H, Ar-H), 6.89 (s, 1H, Ar-H), 6.88-6.86
(m, 2H, Ar-H), 6.71 (s, 1H, Ar-H), 6.64 (s, 1H, Ar-H), 5.16 (s, 2H, Ar-O-CH2), 5.15 (s,
4H, 2 × Ar-O-CH2), 5.14 (s, 2H, Ar-O-CH2), 4.58 (s, 2H, Ar-CH2-N), 3.70 (s, 2H,
Ar-CH2-CO), 3.49 (t, 2H, J = 5.8 Hz, N-CH2), 2.44 (t, 2H, J = 5.8 Hz, Ar-CH2);
HRMS (ESI) calculated for C45H41NO5Na (M+Na)+ 698.2882, found 698.2888.
4.7.11. (E)-1-(6,7-Bis(benzyloxy)-3,4-dihydroisoquinolin-2(1H)-yl)-3-(3,4-bis-
(benzyloxy)phenyl)prop-2-en-1-one (36)
36 was synthesized as a white solid in 94% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.60 (d, 1H, J = 15.3 Hz, Ar-CH=), 7.51-7.43 (m, 8H, Ar-H), 7.42-7.36 (m, 8H,
Ar-H), 7.36-7.30 (m, 4H, Ar-H), 7.14 (s, 1H, Ar-H), 7.11 (d, 1H, J = 8.3 Hz, Ar-H),
6.94 (d, 1H, J = 8.3 H, Ar-Hz), 6.76 (s, 1H, Ar-H), 6.74 (s, 1H, Ar-H), 6.71 (d, 1H, J =
15.3 Hz, CO-CH=), 5.21 (s, 4H, 2 × Ar-O-CH2), 5.16 (s, 4H, 2 × Ar-O-CH2),
4.74-4.64 (m, 2H, Ar-CH2-N), 3.95-3.74 (m, 2H, N-CH2), 2.86-2.74 (m, 2H, Ar-CH2);
HRMS (ESI) calculated for C46H41NO5Na (M+Na)+ 710.8258, found 710.8265.
4.7.12. 3,4-Bis(benzyloxy)phenethyl 2-(3,4-bis(benzyloxy)phenyl)acetate (39)
39 was synthesized as a white solid in 92% yield. 1H NMR (600 MHz, Chloroform-d)
δ 7.49-7.42 (m, 8H, Ar-H), 7.42-7.32 (m, 8H, Ar-H), 7.35-7.29 (m, 4H, Ar-H), 6.90 (d,
1H, J = 2.1 Hz, Ar-H), 6.88 (d, 1H, J = 8.2 Hz, Ar-H), 6.86 (d, 1H, J = 8.2 Hz, Ar-H),
6.81 (d, 1H, J = 2.1 Hz, Ar-H), 6.76 (dd, 1H, J = 8.2, 2.1 Hz, Ar-H), 6.67 (dd, 1H, J =
8.2, 2.1 Hz, Ar-H), 5.14 (s, 4H, 2 × Ar-O-CH2), 5.14, 5.13 (each s, each 2H,

Ar-O-CH2), 4.24 (t, 2H, J = 7.0 Hz, N-CH2), 3.49 (s, 2H, Ar-CH2-CO), 2.81 (t, 2H, J
= 7.0 Hz, Ar-CH2); HRMS (ESI) calculated for C44H40O6Na (M+Na)+ 687.2723,
found 687.2730.
4.7.13. N-(3,4-Dimethoxyphenethyl)-2-(3,4-dimethoxyphenyl)acetamide (40)
40 was synthesized as a white solid in 95% yield. 1H NMR (600 MHz, Chloroform-d)
δ 6.79 (dd, J = 7.9, 1.6 Hz, 1H), 6.73-6.69 (m, 2H), 6.68 (s, 1H), 6.62 (d, J = 1.7 Hz,
1H), 6.53 (dd, J = 8.2, 1.7 Hz, 1H), 3.87, 3.84 (each s, each 3H, Ar-O-CH3), 3.81 (s,
6H, 2 × OCH3), 3.46 (s, 2H, Ar-CH2-CO), 3.43 (q, 2H, J = 6.5 Hz, N-CH2), 2.67 (t,
2H, J = 7.0 Hz, Ar-CH2); HRMS (ESI) calculated for C20H25NO5Na (M+Na)+
382.1630, found 382.1638.
4.8. N-(3,4-Dimethoxyphenethyl)-2-(3,4-dimethoxyphenyl)ethanethioamide (41)
To a solution of 40 (200 mg, 0.56 mmol) in 10 mL THF was added Lawesson’s
reagent (151 mg, 0.61 mmol) at r. t. and the mixture was stirred at reflux for 6 h. After
the solvent was evaporated, the residue was purified by flash chromatography
[petroleum ether : EtOAc : CH2Cl2, 1:1:1] to afford 40 as a yellow oil (208 mg, 99%).
1H NMR (600 MHz, Chloroform-d) δ 7.05 (s, 1H, NH), 6.79 (d, 1H, J = 8.1 Hz,
Ar-H), 6.66 (d, 1H, J = 8.2 Hz, Ar-H), 6.63 (dd, 1H, J = 8.2, 1.9 Hz, Ar-H), 6.60 (d,
1H, J = 1.9 Hz, Ar-H), 6.59 (d, 1H, J = 1.9 Hz, Ar-H), 6.41 (dd, 1H, J = 8.1, 1.9 Hz,
Ar-H), 4.02 (s, 2H, Ar-CH2-CS), 3.90, 3.87 (each s, each 3H, each OCH3), 3.85 (t, 2H,
J = 6.5 Hz, N-CH2), 3.83, 3.78 (each s, each 3H, each OCH3), 2.78 (t, J = 6.7 Hz, 2H,
Ar-CH2); HRMS (ESI) calculated for C20H25NSO4Na (M+Na)+ 398.1402, found
398.1410.
4.9. General procedure for the preparation of compounds (8-16), (18-20)

To a solution of compounds 25-32 or 34-36 in MeOH was added a catalytic amount of
10% Pd/C, and the mixture was stirred at room temperature for 12 h under H2
atmosphere. The reaction mixture was diluted with ether, filtered through Celite. The
solvent was distilled off, and the residue was purified by flash chromatography to give
compounds 8-16, 18-20 respectively.
4.9.1. N-(3,4-Dihydroxyphenethyl)-2-hydroxybenzamide (8)
1
Compound 8 was prepared as a yellow solid in 86% yield. H NMR (600 MHz,
DMSO-d6) δ 12.61 (brs, 1H, Ar-OH), 8.85 (t, 1H, J = 5.6 Hz, CO-NH), 8.77 (brs, 1H,
Ar-OH), 8.67 (brs, 1H, Ar-OH), 7.82 (dd, 1H, J = 8.0, 1.6 Hz, Ar-H), 7.38 (ddd, 1H, J
= 8.5, 7.1, 1.6 Hz, Ar-H), 6.91-6.85 (m, 2H, Ar-H), 6.64 (d, 1H, J = 8.0 Hz, Ar-H),
6.63 (d, 1H, J = 2.1 Hz, Ar-H), 6.48 (dd, 1H, J = 8.0, 2.1 Hz, Ar-H), 3.46-3.41 (m, 2H,
N-CH2), 2.67 (t, 2H, J = 7.5 Hz, Ar-CH2); 13C NMR (150 MHz, DMSO-d6) δ 169.23,
160.49, 145.57, 144.06, 134.04, 130.43, 128.11, 119.70, 118.98, 117.81, 116.45,
116.00, 115.73, 41.39, 34.79; HRMS (ESI) calculated for C15H14NO4 (M-H)-
272.0923, found 272.0945.
4.9.2. N-(3,4-Dihydroxyphenethyl)-3-hydroxybenzamide (9)
1
Compound 9 was prepared as a yellow solid in 91% yield. H NMR (600 MHz,
Chloroform-d) δ 9.61 (s, 1H, Ar-OH), 8.75 (s, 1H, Ar-OH), 8.63 (s, 1H, Ar-OH), 8.38
(t, 1H, J = 5.6 Hz, NH), 7.23 (s, 1H, Ar-H), 7.23-7.21 (m, 2H, Ar-H), 6.92-6.86 (m,
1H, Ar-H), 6.64 (d, 1H, J = 8.0 Hz, Ar-H), 6.62 (d, 1H, J = 2.1 Hz, Ar-H), 6.47 (dd,
1H, J = 8.0, 2.1 Hz, Ar-H), 3.41-3.33 (m, 2H, N-CH2), 2.67-2.59 (m, 2H, Ar-CH2);
13C NMR (150 MHz, DMSO-d6) δ 166.58, 157.72, 145.51, 143.95, 136.67, 130.79,
129.65, 119.69, 118.35, 118.04, 116.45, 115.95, 114.65, 41.71, 35.08; HRMS (ESI)
calculated for C15H14NO4 (M-H)- 272.0923, found 272.0945.
4.9.3. N-(3,4-Dihydroxyphenethyl)-4-hydroxybenzamide (10)

1
Compound 10 was prepared as a white solid in 95% yield. H NMR (600 MHz,
DMSO-d6) δ 9.92 (s, 1H, Ar-OH), 8.75 (s, 1H, Ar-OH), 8.63 (s, 1H, Ar-OH), 8.25 (t,
1H, J = 5.6 Hz, CO-NH), 7.69 (d, 2H, J = 8.6 Hz, Ar-H), 6.78 (d, 2H, J = 8.6 Hz,
Ar-H), 6.63 (d, 1H, J = 8.0 Hz, Ar-H), 6.61 (d, 1H, J = 2.1 Hz, Ar-H), 6.46 (dd, 1H, J
13
= 8.0, 2.1 Hz, Ar-H), 3.39-3.31 (m, 2H, N-CH2), 2.65-2.59 (m, Ar-CH2); C NMR
(150 MHz, DMSO-d6) δ 166.24, 160.40, 145.51, 143.94, 130.87, 129.44(two), 125.88,
119.67, 116.45, 115.95, 115.18 (two), 41.67, 35.28; HRMS (ESI) calculated for
C15H14NO4 (M-H)- 272.0923, found 272.0945.
4.9.4. N-(3,4-Dihydroxyphenethyl)-2,3-dihydroxybenzamide (11)
1
Compound 11 was prepared as a purple solid in 86% yield. H NMR (600 MHz,
DMSO-d6) δ 12.79 (s, 1H, Ar-OH), 9.12 (s, 1H, Ar-OH), 8.83 (t, 1H, J = 5.7 Hz), 8.76
(s, 1H, Ar-OH), 8.66 (s, 1H, Ar-OH), 8.83 (t, 1H, J = 5.7 Hz, CO-NH), 7.25 (dd, 1H, J
= 8.1, 1.4 Hz, Ar-H), 6.90 (dd, 1H, J = 7.8, 1.4 Hz, Ar-H), 6.67 (t, 1H, J = 7.9 Hz,
Ar-H), 6.64 (d, 1H, J = 8.0 Hz, Ar-H), 6.62 (d, 1H, J = 2.0 Hz, Ar-H), 6.47 (dd, 1H, J
= 8.0, 2.0 Hz, Ar-H), 3.41 (q, 2H, J = 6.6 Hz, N-CH2), 2.67 (t, 2H, J = 7.6 Hz,
13
Ar-CH2); C NMR (150 MHz, DMSO-d6) δ 170.08, 150.11, 146.65, 145.56, 144.06,
130.44, 119.70, 119.21, 118.34, 117.58, 116.46, 116.00, 115.48, 41.45, 34.79; HRMS
(ESI) calculated for C15H14NO5 (M-H)- 288.0827, found 288.0875.
4.9.5. N-(3,4-Dihydroxyphenethyl)-3,4-dihydroxybenzamide (12)
1
Compound 12 was prepared as a white solid in 93% yield. H NMR (600 MHz,
DMSO-d6) δ 8.84 (brs, 4H), 8.16 (t, 1H, J = 5.6 Hz, CO-NH), 7.26 (d, 1H, J = 2.0 Hz,
Ar-H), 7.16 (dd, 1H, J = 8.2, 2.0 Hz, Ar-H), 6.74 (d, 1H, J = 8.2 Hz, Ar-H), 6.63 (d,
1H, J = 7.9 Hz, Ar-H), 6.61 (d, 1H, J = 2.0 Hz, Ar-H), 6.45 (dd, 1H, J = 7.0, 2.0 Hz,
13
Ar-H), 3.33 (q, 2H, J = 6.9 Hz, N-CH2), 2.61 (t, 2H, J = 7.6 Hz, Ar-CH2); C NMR
(150 MHz, DMSO-d6) δ 166.43, 148.64, 145.50, 145.23, 143.92, 130.89, 126.43,

119.66, 119.27, 116.44, 115.95, 115.51, 115.25, 41.66, 35.25; HRMS (ESI) calculated
for C15H14NO5 (M-H)- 288.0827, found 288.0875.
4.9.6. N-(3,4-Dihydroxyphenethyl)-3,4,5-trihydroxybenzamide (13)
1
Compound 13 was prepared as a brown solid in 90% yield. H NMR (600 MHz,
DMSO-d6) δ 8.75 (s, 5H, Ar-OH), 8.08 (t, 1H, J = 5.6 Hz, CO-NH), 6.80 (s, 2H,
Ar-H), 6.63 (d, 1H, J = 8.0 Hz, Ar-H), 6.60 (d, 1H, J = 2.0 Hz, Ar-H), 6.45 (dd, 1H, J
= 8.0, 2.0 Hz, Ar-H), 3.33-3.28 (m, 2H, N-CH2), 2.60 (t, 2H, J = 7.7 Hz, Ar-CH2); 13C
NMR (150 MHz, DMSO-d6) δ 166.69, 145.84 (two), 145.51, 143.92, 136.57, 130.92,
125.50, 119.68, 116.43, 115.95, 107.12 (two), 41.68, 35.23; HRMS (ESI) calculated
for C15H14NO6 (M-H)- 362.0876, found 362.0887.
4.9.7. N-(3,4-Dihydroxyphenethyl)-2-(3,4-dihydroxyphenyl)acetamide (14)
1
Compound 14 was prepared as a white solid in 96% yield. H NMR (600 MHz,
DMSO-d6) δ 8.76 (brs, 4H), 7.90 (t, 1H, J = 5.5 Hz, CO-NH), 6.66 (s, 1H, Ar-H), 6.63
(d, 1H, J = 7.9 Hz, Ar-H), 6.62 (d, 1H, J = 7.9 Hz, Ar-H), 6.57 (s, 1H, Ar-H), 6.46 (d,
1H, J = 8.0 Hz), 6.40 (d, 1H, J = 8.0 Hz, Ar-H), 3.18 (s, 2H, Ar-CH2-CO), 3.16-3.12
(m, 2H, N-CH2), 2.51-2.49 (m, 2H, Ar-CH2); 13C NMR (150 MHz, DMSO-d6) δ
170.97, 145.52, 145.36, 144.23, 143.98, 130.66, 127.60, 120.20, 119.68, 116.93,
116.41, 115.97, 115.79, 42.34, 41.20, 35.18; HRMS (ESI) calculated for C16H16NO5
(M-H)- 302.1028, found 302.1049.
4.9.8. N-(3,4-Dihydroxyphenethyl)-3-(3,4-dihydroxyphenyl)propanamide (15)
1
Compound 15 was prepared as a white solid in 89% yield. H NMR (600 MHz,
DMSO-d6) δ 8.67 (s, 4H, Ar-OH), 7.82 (t, 1H, J = 5.7 Hz, CO-NH), 6.62 (d, 1H, J =
8.0 Hz, Ar-H), 6.61 (d, 1H, J = 8.1 Hz, Ar-H), 6.57 (s, 1H, Ar-H), 6.56 (s, 1H, Ar-H),
6.43-6.39 (m, 2H, Ar-H), 3.18-3.12 (m, 2H, N-CH2), 2.60 (dd, 2H, J = 9.0, 6.7 Hz,
13
Ar-CH2), 2.50-2.46 (m, 2H, Ar-CH2), 2.25 (dd, 2H, J = 8.8, 7.0 Hz, CO-CH2); C

NMR (150 MHz, DMSO-d6) δ 171.79, 145.48, 145.41, 143.93, 143.72, 132.68,
130.75, 119.66, 119.16, 116.40, 116.11, 115.93, 115.87, 41.03, 38.08, 35.20, 31.09;
HRMS (ESI) calculated for C17H18NO5 (M-H)- 316.1185, found 316.1200.
4.9.9. 3,4-Dihydroxyphenethyl 2-(3,4-dihydroxyphenyl)acetate (16)
1
Compound 16 was prepared as a yellow solid in 88% yield. H NMR (600 MHz,
DMSO-d6) δ 8.78 (s, 4H, Ar-OH), 6.65-6.62 (m, 3H, Ar-H), 6.61 (d, 1H, J = 2.1 Hz,
Ar-H), 6.46 (dd, 1H, J = 8.1, 2.1 Hz, Ar-H), 6.44 (dd, 1H, J = 8.0, 2.1 Hz, Ar-H), 4.11
(t, 2H, J = 7.1 Hz, N-CH2), 3.42 (s, 2H, Ar-CH2-CO), 2.67 (t, 2H, J = 7.1 Hz, Ar-CH2);
13C NMR (150 MHz, DMSO-d6) δ 171.92, 145.54, 145.50, 144.63, 144.22, 128.91,
125.40, 120.47, 119.99, 117.11, 116.65, 115.99, 115.89, 65.48, 40.29, 34.25; HRMS
(ESI) calculated for C16H15O6 (M-H)- 303.0869, found 303.0909.
4.9.10.
(6,7-Dihydroxy-3,4-dihydroisoquinolin-2(1H)-yl)(3,4-dihydroxyphenyl)
methanone (18)
1
Compound 18 was prepared as a white solid in 91% yield. H NMR (600 MHz,
DMSO-d6) δ 9.28 (brs, 2H, Ar-OH), 8.84 (brs, 2H, Ar-OH), 6.89-6.70 (m, 3H, Ar-H),
6.52 (s, 2H, Ar-H), 4.46 (s, 2H, Ar-CH2-N), 3.65-3.53 (m, 2H, N-CH2), 2.65 (t, 2H, J
= 5.7 Hz, Ar-CH2); 13C NMR (150 MHz, DMSO-d6) δ 170.12, 147.37, 145.37, 144.41,
144.30, 127.42, 125.01, 124.03, 119.23, 115.81, 115.60, 115.35, 113.46, 51.00, 49.06,
28.41; HRMS (ESI) calculated for C16H14NO5 (M-H)- 300.0872, found 300.0880.
4.9.11. 1-(6,7-Dihydroxy-3,4-dihydroisoquinolin-2(1H)-yl)-2-(3,4-dihydroxyphenyl)-
ethan-1-one (19)
1
Compound 19 was prepared as a white solid in 96% yield. H NMR (600 MHz,
DMSO-d6) δ 8.86 (brs, 4H, Ar-OH), 6.65-6.61 (m, 2H, Ar-H), 6.51-6.48 (m, 2H,
Ar-H), 6.46 (d, 1H, J = 8.1 Hz, Ar-H), 4.40 (s, 2H, Ar-CH2-N), 3.58 (q, 2H, J = 5.6
13
Hz, N-CH2), 3.56 (s, 2H, Ar-CH2-CO), 2.45 (t, 2H, J = 5.7 Hz, Ar-CH2); C NMR

(150 MHz, DMSO-d6) δ 170.01, 145.58, 144.25, 126.85, 125.38, 125.00, 124.16,
123.90, 119.93, 116.46, 115.98, 115.61, 113.60, 46.94, 43.87 , 43.71, 28.44; HRMS
(ESI) calculated for C17H16NO5 (M-H)- 314.1028, found 314.1036.
4.9.12. 1-(6,7-Dihydroxy-3,4-dihydroisoquinolin-2(1H)-yl)-3-(3,4-dihydroxyphenyl)
propan-1-one (20)
1
Compound 20 was prepared as a white solid in 97% yield. H NMR (600 MHz,
DMSO-d6) δ 8.74 (brs, 4H, Ar-OH), 6.62-6.60 (m, 2H, Ar-H), 6.51 (s, 1H, Ar-H), 6.50
(s, 1H, Ar-H), 6.47 (s, 1H, Ar-H), 4.39 (s, 1H, Ar-CH2-N), 3.59 (t, 2H, J = 5.9 Hz,
N-CH2), 3.54 (t, 2H, J = 5.9 Hz, Ar-CH2), 2.59-2.54 (m, 4H, Ar-CH2CH2); 13C NMR
(150 MHz, DMSO-d6) δ 170.83, 145.42, 144.25, 144.18, 143.74, 132.70, 125.20,
123.96, 119.32, 116.29, 115.86, 115.60, 113.64, 46.40, 43.63, 35.57, 30.64, 28.55;
HRMS (ESI) calculated for C18H18NO5 (M-H)- 328.1185, found 318.1193.
4.10. N-(3,4-Dihydroxyphenethyl)-2-(3,4-dihydroxyphenyl)ethanethioamide (17)
A solution of 41 (229 mg, 0.61 mmol) in anhydrous dichloromethane (5 mL) was
cooled to -78 ºC and 1 M boron tribromide solution in dichloromethane (4.88 mL,
4.88 mmol) was added dropwise under N2 atmosphere. The reaction was gradually
warmed to -40 ºC and stirred for 4 h, and then carefully quenched with cold methanol.
After removing the solvents in vacuum, the crude residue was purified by flash
chromatography [CH2Cl2 : MeOH, 8 :1 ] to get 17 as a yellow oil (171 mg, 88%). 1H
NMR (600 MHz, DMSO-d6) δ 10.07 (t, 1H, J = 5.4 Hz, NH), 8.22 (brs, 1H, Ar-OH),
6.74 (d, 1H, J = 2.0 Hz, Ar-H), 6.68-6.59 (m, 3H, Ar-H), 6.55 (dd, 1H, J = 8.0, 2.0 Hz,
Ar-H), 6.44 (dd, 1H, J = 8.0, 2.1 Hz, Ar-H), 3.70 (s, 2H, Ar-CH2-CS), 3.61 (q, 2H, J =
6.2 Hz, N-CH2), 2.68 (t, 2H, J = 7.7 Hz, Ar-CH2); 13C NMR (150 MHz, DMSO-d6) δ
202.00, 145.59, 145.30, 144.48, 144.14, 130.11, 128.65, 120.09, 119.71, 116.76,

116.37, 116.00, 115.74, 51.41, 47.63, 32.94; HRMS (ESI) calculated for C16H16NO4S
(M-H)- 318.0800, found 318.0802.
4.11. Cells, viruses and agents
Madin-Darby canine kidney (MDCK) cells were cultured in DMEM (Gibco BRL,
USA) supplemented with 10% fetal bovine serum (FBS) (Excell Bio, China), and 1%
penicillin/streptomycin (Gibco BRL, USA) at 37ꢀ°C with 5% CO2. The type of
influenza A virus applied in this work was A/WSN/33 (H1N1). After infected by the
virus, the cells were maintained in FBS-free medium containing 0.6 µg/mL
TPCK-Treated Trypsin (SIGMA, USA). Peramivir Trihydrate (Meilun Biological,
China) dissolved in DNase/RNase-free ddH2O to 5 mM as a stock solution was
applied as a positive control in the antiviral assay.
4.12. Antiviral assay and microscopy
The inhibitory activity of title compounds against influenza A virus was reviewed
by the Thiazolyl Blue Tetrazolium Bromide (MTT) (SIGMA, USA) assay [35].
MDCK cells were seeded into a 96-well tissue culture plate overnight, and then
infected with a 100 tissue culture infective doses (TCID50) (MOI=3.5×10-3) for 1 h.
The test compounds were added to the infected cells with increasing amounts and
then cultivated about 48 h. MTT solution (5ꢀmg/ꢀmL) in PBS was added to each well.
Prior to detection, DMSO was added to dissolve the formazan crystals. And the
absorbance was recorded by a microplate reader (GENios Pro, TECAN, Bedford, MA,
USA) at 570 nm. In addition, the cytopathic effect (CPE) that induced by the virus
was observed by microscopy.
4.13. Cytotoxicity assay
MDCK cells were grown in a 96-well tissue culture plate overnight. Diluted test
compounds with increasing amounts or blank control solutions (0.1% DMSO) were

supplemented in cells. After 48 h, the cytotoxicity of the test compounds was
submitted to the MTT assay as previously described [36]. The concentration of the
test compounds that resulted in the death of 50% cells was recognized as the 50%
cytotoxic concentration (CC50).
4.14. Endonuclease activity assay
The PAN endonuclease (MyBiosource MBS7111040, San Diego, USA) assay was
performed as described previously [37, 38]. The influenza A PAN domain has been
shown to cleave ssRNA as well as ssDNA. A TaqMan-like oligonucleotide was used
containing a 6-carboxy-fluorescein (FAM) fluorophore at the 5’-end followed by 19
nucleotides and a minor groove binding non-fluorescent quencher (MGBNFQ,
Applied Biosystems) at the 3’-end. When excited by light at a wavelength of 492 nm,
MGBNFQ quenches the fluorescence of FAM via fluorescence resonance energy
transfer. If the endonuclease cleaves the oligonucleotide, the quencher is no longer
coupled to the fluorophore, and therefore, FAM fluoresces. The assay can be used to
evaluate the inhibitory characteristics of compounds that are found to bind PAN. The
assay uses the probe 6FAM-TGGCAATATCAGCTCCACA-MGBNFQ. EGCG and
BXA were included as a reference compound, respectively.
4.15. Surface Plasmon Resonance (SPR) Measurement
In order to study the interaction between candidate compounds and H1N1 virus
PAN endonuclease (MyBiosource MBS7111040, San Diego, USA), we performed an
affinity measurement by means of Surface Plasmon Resonance (SPR) technology. The
SPR validation experiment was perform with the Screen LB 991 Label-free
Microarray System (BERTHOLD TECHNOLOGIES, Germany). The chemical
modified label-free photo-cross-linker sensor chips were provided by Betterways Inc.

(China). For array spotting, BioDot™ AD-1520 Array Printer (BIODOT Inc., USA)
was employed to printed samples and control EGCG as well as BXA (Figure S2) on
the surface. Each candidate compound was printed with a volume of 1.875 nL/dot
using BioDot 1520 array printer with the formation of 5 nonadjacent repetitions of
each sample in the same chip. During the SPR test, the surface was first primed three
times with HBS-EP running buffer (containing 10 mM HEPES, pH 7.0, 150 mM
NaCl, 3 mM EDTA and 0.005% (v/v) of P20 surfactant) at rate of 2 µL/s for 40 s and
one time with running buffer (1 × PBS with 5% DMSO) at rate of 2 µL/s for 40 s. The
PAN endonuclease (MyBiosource, San Diego, USA) was diluted separately with
running buffer at the concentration of 10 nM, 20 nM, 40 nM, 80 nM and 160 nM and
injected for 600 s at a flow rate of 0.5 µL/s using in each associating stage, followed
by running buffer for 360 s at a flow rate of 0.5 µL/s in each dissociating stage. The
raw sensor grams and measurements of the binding process of ligands and proteins
were recorded in real time. The response unit (RU) of surface resonance was
compared to determine the different binding affinity between each sample. The
equilibrium dissociation constant (KD) was performed using data analysis software of
the Screen LB 991 unlabeled microarray system according to a single-site binding
model (1:1 Langmuir binding) with mass transfer limitations for determination of the
binding kinetics.
4.16. Molecular docking
All calculations were performed using SYBYL 7.3 [39]. The protein structure of
H1N1 virus PAN endonuclease bound with BXA (PDB ID: 6FS6) [40] was applied for