Targeting the entrance channel of NNIBP: Discovery of diarylnicotinamide 1,4-disubstituted 1,2,3-triazoles as novel HIV-1 NNRTIs with high potency against wild-type and E138K mutant virus

Article abstract of DOI:10.1016/j.ejmech.2018.03.059

Inspired by our previous efforts on the modifications of diarylpyrimidines as HIV-1 non-nucleoside reverse transcriptase inhibitors (NNRTI) and reported crystallography study, novel diarylnicotinamide derivatives were designed with a “triazole tail” occupying the entrance channel in the NNRTI binding pocket of the reverse transcriptase to afford additional interactions. The newly designed compounds were then synthesized and evaluated for their anti-HIV activities in MT-4 cells. All the compounds showed excellent to good activity against wild-type HIV-1 strain with EC₅₀ of 0.02–1.77 μM. Evaluations of selected compounds against more drug-resistant strains showed these compounds had advantage of inhibiting E138K mutant virus which is a key drug-resistant mutant to the new generation of NNRTIs. Among this series, propionitrile (3b2, EC_50(IIIB) = 0.020 μM, EC_50(E138K) = 0.015 μM, CC₅₀ = 40.15 μM), pyrrolidin-1-ylmethanone (3b8, EC_50(IIIB) = 0.020 μM, EC_50(E138K) = 0.014 μM, CC₅₀ = 58.09 μM) and morpholinomethanone (3b9, EC_50(IIIB) = 0.020 μM, EC_50(E138K) = 0.027 μM, CC₅₀ = 180.90 μM) derivatives are the three most promising compounds which are equally potent to the marketed drug Etravirine against E138K mutant strain but with much lower cytotoxicity. Furthermore, detailed SAR, inhibitory activity against RT and docking study of the representative compounds are also discussed.

Full text of DOI:10.1016/j.ejmech.2018.03.059

Accepted Manuscript
Targeting the entrance channel of NNIBP: Discovery of diarylnicotinamide 1,4-
disubstituted 1,2,3-triazoles as novel HIV-1 NNRTIs with high potency against wild-
type and E138K mutant virus
Ye Tian, Zhaoqiang Liu, Jinghan Liu, Boshi Huang, Dongwei Kang, Heng Zhang, Erik
De Clercq, Dirk Daelemans, Christophe Pannecouque, Kuo-Hsiung Lee, Chin-Ho
Chen, Peng Zhan, Xinyong Liu
PII:
S0223-5234(18)30302-7
10.1016/j.ejmech.2018.03.059
EJMECH 10324
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 12 January 2018
Revised Date: 15 March 2018
Accepted Date: 20 March 2018
Please cite this article as: Y. Tian, Z. Liu, J. Liu, B. Huang, D. Kang, H. Zhang, E. De Clercq, D.
Daelemans, C. Pannecouque, K.-H. Lee, C.-H. Chen, P. Zhan, X. Liu, Targeting the entrance channel
of NNIBP: Discovery of diarylnicotinamide 1,4-disubstituted 1,2,3-triazoles as novel HIV-1 NNRTIs
with high potency against wild-type and E138K mutant virus, European Journal of Medicinal Chemistry
(2018), doi: 10.1016/j.ejmech.2018.03.059.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Targeting the entrance channel of NNIBP: Discovery of
diarylnicotinamide 1,4-disubstituted 1,2,3-triazoles as
novel HIV-1 NNRTIs with high potency against
wild-type and E138K mutant virus
1
1
2
1
1
Ye Tian , Zhaoqiang Liu , Jinghan Liu , Boshi Huang , Dongwei Kang ,
1
3
3
3
HengZhang , Erik De Clercq , Dirk Daelemans , Christophe Pannecouque ,
4
,5
6
∗ ,1
∗ ,1
Kuo-Hsiung Lee , Chin-Ho Chen , Peng Zhan , Xinyong Liu .
1
Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of
Education), School of Pharmaceutical Sciences, Shandong University, 44 West Culture Road,
2
50012, Jinan, Shandong, P.R. China;
2
Shool of life science and technology, China Pharmaceutical University, 639 Longmian Avenue,
2
10009, Nanjing, P.R. China;
3
Rega Institute for Medical Research, KU Leuven, Minderbroedersstraat 10, B-3000 Leuven,
Belgium;
4
Natural Products Research Laboratories, UNC Eshelman School of Pharmacy, University of
North Carolina, Chapel Hill, North Carolina 27599-7568, United States;
5
Chinese Medicine Research and Development Center, China Medical University and Hospital ,
Taichung, Taiwan;
6
Surgical Science, Department of Surgery, Duke University Medical Center, Durham, North
Carolina 27710, United States.
Corresponding authors. Tel: +86 531 88380270; fax: +86 531 88382731; E-mail address:
zhanpeng1982@sdu.edu.cn (Zhan P.); xinyongl@sdu.edu.cn (Liu X.Y.)
1

ACCEPTED MANUSCRIPT
ABSTRACT: Inspired by our previous efforts on the modifications of
diarylpyrimidines as HIV-1 non-nucleoside reverse transcriptase inhibitors (NNRTI)
and reported crystallography study, novel diarylnicotinamide derivatives were
designed with a “triazole tail” occupying the entrance channel in the NNRTI binding
pocket of the reverse transcriptase to afford additional interactions. The newly
designed compounds were then synthesized and evaluated for their anti-HIV activities
in MT-4 cells. All the compounds showed excellent to good activity against wild-type
HIV-1 strain with EC of 0.02-1.77 µM. Evaluations of selected compounds against
50
more drug-resistant strains showed these compounds had advantage of inhibiting
E138K mutant virus which is a key drug-resistant mutant to the new generation of
NNRTIs. Among this series, propionitrile (3b2, EC50(IIIB) = 0.020 µM, EC50(E138K)
=
0
.015 µM, CC = 40.15 µM), pyrrolidin-1-ylmethanone (3b8, EC
= 0.020 µM,
50(IIIB)
50
EC5
0(E138K)
= 0.014 µM, CC = 58.09 µM) and morpholinomethanone (3b9, EC
50
50(IIIB)
=
0.020 µM, EC50(E138K) = 0.027 µM, CC = 180.90 µM) derivatives are the three
50
most promising compounds which are equally potent to the marketed drug Etravirine
against E138K mutant strain but with much lower cytotoxicity. Furthermore, detailed
SAR, inhibitory activity against RT and docking study of the representative
compounds are also discussed.
KEYWORDS: HIV-1; NNRTIs; triazole; diarylnicotinamide; entrance channel; drug design.
2

ACCEPTED MANUSCRIPT
1
. Introduction
Human immunodeficiency virus (HIV) is the pathogenic factor for acquired
immune deficiency syndrome (AIDS) which severely harms human health. From the
WHO most updated report: by the end of 2016, there were still 36.7 million people
living with HIV globally [1]. Among them, 1.8 million became newly infected in
2
016. And in the same year, 1.0 million people died from AIDS related illnesses.
Fortunately, due to the antiretroviral treatment, especially the application of highly
active antiretroviral therapy (HAART) which combines multiple drugs that act on
different viral targets, AIDS-related deaths have fallen by 39% since the peak in 2000.
Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are essential
components of HARRT for their excellent potencies and low toxicities. With superior
activity against wild-type (WT) and mutant HIV-1 strains, diarylpyrimidine (DAPY)
derivatives, as the second generation NNRTIs, have attracted considerable attentions
over the past few years. FDA approved drugs Etravirine (TMC125, ETR) and
Rilpivirine (TMC278, RPV) are the two representative compounds of this series (Fig.
1
). However, the unsatisfactory pharmacokinetic properties of these DAPY
derivatives [2-4] and adverse effects caused by them [5-7] demands efforts to identify
more NNRTI agents with higher potency and better drug-likeness [8, 9]. More
importantly, the low genetic barrier of viral mutations led to the rapid emergence of
drug resistance. From the study in the lab of Steegen, only one-third of patients
retained full susceptibility to Etravirine (36.5%) and Rilpivirine (27.3%) [10]. Among
3

ACCEPTED MANUSCRIPT
these mutations, E138K in RT is not only the most common NNRTI resistance
associated mutation found in patients who failed RPV-containing regimens [11] but
also the earliest mutation for ETR resistance in all subtypes in vitro. Besides, E138K
mutation also confers phenotypic resistance to Nevirapine (NVP) and Efavirenz
(EFV) [12, 13]. Therefore, the broad cross-resistance against nearly all of the
approved NNRTIs make the development of E138K must be avoided whenever
possible [14].
Figure 1. U.S. FDA approved DAPY NNRTIs.
The binding mode of ETR with the NNRTI binding pocket (NNIBP) in RT is
well depicted by the crystallography studies [15] (Fig. 2): the central pyrimidine ring
and the NH linker connecting to its 2-position form two important hydrogen bonds
with the main chain of K101; two aromatic wings of the molecule fit into two
adjacent grooves composed mainly of lipophilic residues; additionally, the NH and
2
Br groups on the central pyrimidine of ETR point to an entrance channel opening to
the solvent surface [16, 17] which formed by amino acids L100, K101, E138 and
V179.
4

ACCEPTED MANUSCRIPT
Figure 2. The design of a novel series of diarylnicotinamide 1,4-disubstituted
1,2,3-triazole derivatives.
Great efforts from our group [18-24] focused on the modifications of the central
ring of DAPY series trying to discover optimal moieties that can occupy the tolerant
space of the entrance channel. Among these different scaffolds, nicotinamide core
which can form new hydrogen bond with E138 endowed the molecules not only
excellent activities (e.g. L-6b11, EC50 (IIIB) = 0.027 uM, SI > 12,518) [20] but also a
port to connect more favorable groups. Thus, we designed a novel series of
nicotinamide derivatives based on the L-6b11 template which is the best compound
from our last series, hoping to acquire better potency especially against the E138K
strain. We kept the core and two phenyl wings as original while linked various
substituted benzyl to explore the chemical space of the entrance channel or other
groups (mainly hydrophilic groups) by suitable linker. Owing to the importance and
realizability of 1,2,3-triazole compounds, the triazole was used as a linker in the
structure to afford potential additional interactions with the surrounding residues [25]
(Fig. 2). Unique properties of 1,2,3-triazole such as rigidity and stability in vivo,
5

ACCEPTED MANUSCRIPT
hydrogen bonding capability, and dipole moment are considered as decisive factors
for their improved biological activity. For example, some 1,2,3-triazoles synthesized
in our group showed a broad spectrum of antiviral activities [25-28]. Besides, the
application of “click chemistry” make it very easy to form the triazole ring.
The newly designed compounds were then synthesized, evaluated for their
anti-HIV activities against WT (IIIB and ROD) and multiple mutant strains. The
representative compounds were also tested for their ability to inhibit the RT of HIV-1.
Herein, the result of evaluation, SAR and modeling analysis will be discussed in this
article.
2
. Results and Discussion
2
.1 Chemistry
The straightforward synthetic route of target diarylnicotinamide 1,4-disubstituted
,2,3-triazole derivatives is depicted in Scheme 1. For the preparation of the designed
1
series, substituted haloalkanes (1a or 1b) were used as the starting materials which
reacted with sodium azide by nucleophilic substitution to form the appropriate azides
(
2a or 2b). Then the desired compounds were generated by copper(I)-catalyzed
azide-alkyne cycloaddition (CuAAC) “click chemistry” reaction between the azides
and 6-((4-cyanophenyl)amino)-4-(mesityloxy)-N-(prop-2-yn-1-yl)nicotinamide
L-6b5) which was obtained as one of the active compounds in our previous series
20]. These synthesized compounds were characterized by physicochemical and
(
[
1
13
spectral means and the MS, H-NMR, C-NMR spectral data were found in
agreement with the assigned molecular structures.
6

ACCEPTED MANUSCRIPT
Scheme 1. The synthetic route of diarylnicotinamide 1,4-disubstituted 1,2,3-triazole
derivatives.
Reagents and conditions: i: NaN ii: CuSO , VcNa, THF-H O, 65 °C.
3
4
2
2
.2 Anti-HIV activity evaluation
The newly synthesized diarylnicotinamide triazole derivatives were evaluated for
their activities against WT HIV-1 strain (IIIB), most common K103N+Y181C double
mutant HIV-1 strain (RES056) and HIV-2 strain (ROD) in MT-4 cells using the MTT
method [29, 30]. And the inhibition assay for a panel of more mutant strains L100I,
K103N, E138K, Y181C, Y188L and F227L+V106A were conducted as well for
selected compounds. Nevirapine (NVP), Delavirdine (DLV), Efavirenz (EFV),
Etravirine and Zidovudine (azidothymidine, AZT) were used as reference drugs. The
cytotoxicity of these compounds was determined in parallel. The results, expressed as
EC , CC , SI and RF are illustrated in Table 1 - Table 4.
50
50
Table 1. Activity against IIIB and RES056 HIV-1 strains, cytotoxicity and SI of the
title compounds in MT-4 cells.
7

ACCEPTED MANUSCRIPT
a
c
EC50 (µM)
SI
b
No.
R
CC50 (µM)
III
B
RES056
> 164.2
> 222.5
> 188.5
> 219.8
> 212.3
> 113.8
> 204.7
> 187.4
> 129.6
> 156.3
> 165.1
III
B
RES056
d
3
3
3
3
3
3
3
3
3
a1
a2
a3
a4
a5
a6
a7
a8
a9
2-Me
2-F
0.077 ± 0.051
0.095 ± 0.023
0.104 ± 0.022
0.039 ± 0.007
0.057 ± 0.019
0.106 ± 0.046
0.079 ± 0.009
0.104 ± 0.054
0.025 ± 0.008
0.037 ± 0.014
0.167 ± 0.031
≥ 164.2
> 222.5
≥ 188.5
> 219.8
> 212.3
≥ 2132
> 2348
≥ 1807
> 5589
> 3724
1074
< or X1
X1
< or X1
X1
2-Cl
2-CN
2-NO
3-Me
3-F
2
X1
113.84 ± 27.32
≥ 204.7
187.4
< 1
≥ 2608
1799
< or X1
< 1
3-Cl
3-CN
≥ 129.6
≥ 156.3
165.1
≥ 5213
≥ 4220
990
< or X1
< or X1
< 1
3
3
a10
a11
3-NO
2
4-Me
8

ACCEPTED MANUSCRIPT
3
3
3
a12
a13
a14
4-F
4-Cl
0.074 ± 0.003
0.194 ± 0.087
0.040 ± 0.015
0.037 ± 0.007
0.020 ± 0.005
0.029 ± 0.016
1.77 ± 0.23
> 153.1
≥ 153.1
≥ 2056
≥ 620
≥ 3498
678
< or X1
> 120.0
≥ 120.0
< or X1
4-NO
2
> 139.6
≥ 139.6
< or X1
3
b1
b2
b3
b4
b5
b6
b7
5.23 ± 2.93
2.31 ± 0.57
2.21 ± 1.77
> 13.8
25.24 ± 5.03
40.15 ± 16.62
19.37 ± 11.42
13.8 ± 7.03
9.53
5
17
9
3
2044
670
3
3
3
3
3
O
8
< 1
< 1
< 1
5
O
0.653 ± 0.31
0.847 ± 0.30
0.172 ± 0.007
> 9.53
15
> 21.47
21.47 ± 1.38
24.01 ± 2.71
25
5.33 ± 1.21
139
3
3
b8
b9
0.020 ± 0.005
0.020 ± 0.009
1.43 ± 0.31
58.09 ± 34.60
180.9 ± 36.74
2897
41
12
15.70 ± 7.16
9279
e
L-6b5
NVP
DLV
EFV
-
-
-
-
0.029 ± 0.005
0.26
6.10 ± 0.54
72.96 ± 26.85
> 15.02
36.8
2471
> 58
12
-
> 15.02
0.038
-
733
-
f
0.005
0.31 ± 0.16
> 6.34
> 1212
-
9

ACCEPTED MANUSCRIPT
0
.017 ±
ETR
AZT
-
-
0.004
2.46
733
-
-
f
0
.001
0
.011 ±
0.007
> 7.48
> 1053
f
0
.005
a
EC50: concentration of compound required to achieve 50% protection of MT-4 cell cultures against HIV-1-induced cytotoxicity,
as determined by the MTT method.
b
CC50: concentration required to reduce the viability of mock-infected cell cultures by 50%, as determined by the MTT method.
c
SI: selectivity index, the ratio of CC50/EC50
.
d
X1: stands for ≥ 1 or <1.
e
f
L-6b5: the activity data were quoted from our published paper[20].
The data were obtained from the same lab of the Rega Institute for Medical Research, KU Leuven, Belgium using the same
:
method[24, 31].
All the 23 compounds exhibited good anti-HIV-1 (WT) activity with
sub-micromolar EC ranging from 0.02 to 0.85 µM, except for compound 3b4 with
5
0
EC50 of 1.77 µM (Table 1). Regarding the substituted benzyl derivatives, i.e.
a1-3a14, they all had EC values below 200 nM. Among several different groups on
3
50
the benzyl ring, the cyano- and nitro- with higher polarity apparently showed their
advantage in potency. It also indicated that the 3-position was more favorable for
these two polar groups than 4- or 2- positions, as 3a9 (3-CN, EC = 25 nM) was
5
0
more potent than 3a4 (2-CN, EC = 39 nM) while 3a10 (3-NO , EC = 37 nM) was
50
2
50
better than 3a14 (4-NO , EC = 40 nM) and 3a5 (2-NO , EC = 57 nM). Fluorine
2
50
2
50
had a secondary effect on the potency, furthermore 3a12 (4-F, EC = 74 nM) was
5
0
slightly better than 3a7 (3-F, EC50 = 79 nM) and 3a2 (2-F, EC50 = 104 nM).
Additionally, the compounds with hydrophobic Me or Cl were less potent and
especially for the para-substituted ones. This sub-series compounds were not
cytotoxic with CC higher than 110 µM and most of them had SI higher than 1000.
5
0
10

ACCEPTED MANUSCRIPT
Unfortunately, none of these substituted benzyl derivatives showed potency against
the RES056 strain.
We also tried various tails attached to the triazole, like ketone, esters or amides
instead of the benzyl group. The esters (3b4, 3b5, 3b6) displayed EC from 0.65 to
5
0
1
.77 µM with relatively higher cytotoxicity (9.53 – 21.5 µM), while 3b7 with an
acetamide had better potency (EC = 172 nM). The rest of compounds all exhibited
5
0
very good potency against the WT HIV-1 virus with EC50 below 37 nM. More
importantly, some of them (3b1, 3b2, 3b3, 3b7, 3b8 and 3b9) showed 50% of
inhibition against double-mutant stain RES056 at low micromolar concentration. 3b2,
3
b8 and 3b9 containing a Cyanomethyl, 2-oxo-2-(pyrrolidin-1-yl)ethyl or
2
-morpholino-2-oxoethyl, respectively, are the three most active compounds with the
same EC against WT HIV-1 at 20 nM, which were more potent than the control
5
0
drugs NVP (EC = 0.26 µM) and DLV (EC = 0.26 µM). Notably, 3b8 had the best
5
0
50
SI (= 41) against RES056 strain whereas 3b9 had the best SI (= 9279) against HIV
IIIB strain.
Regarding the cytotoxicity of these compounds, it is apparent that 3a sub-series
with substituted benzyl were much less toxic with all the CC values higher than 100
5
0
µM. Whereas, the 3b sub-series had CC values ranging from 9.53 – 180.9 µM, and
5
0
most of them had CC below 40 µM. From the detailed information, there seemed to
5
0
be a rough trend that the cyclic (bulkier) group helped to reduce the cytotoxicity:
b1-7 < 3b8 < 3b9 ≈ 3a series (in CC ).
3
5
0
11

ACCEPTED MANUSCRIPT
With the comparison of L-6b5 (EC = 0.029 µM, CC = 72.96 µM), we could
5
0
50
conclude that longer and diverse groups connecting to the amide of the nicotinamide
are tolerant for this region. That may because these groups were supposed to stay at
the interface between the enzyme and the solvent. And some groups are helpful to
make the molecules slightly more active. That may indicate these “tail tip” fragments
indeed have subtle interactions with the amino acid residues nearby. And as expected,
none of 23 compounds inhibited the HIV-2 strain (Data is not showed in the article)
showing that they all act as classical NNRTIs.
Table 2. Activity against a panel of mutant HIV-1 stains of selected compounds in
MT-4 cells.
a
EC50 (µM)
No.
L100I
K103N
E138K
Y181C
Y188L
3.17 ± 0
F227L+V106A
0.78 ± 0.053
0.33 ± 0.21
0.59 ± 0.006
1.80 ± 0.89
0.24 ± 0.008
0.61 ± 0.007
7.17 ± 1.86
2.27 ± 0.14
0.21 ± 0.008
0.005 ± 0
3
3
3
3
3
3
b1
b2
b3
b7
b8
b9
1.95 ± 0.45
1.06 ± 0.22
1.00 ± 0.16
3.52 ± 0.37
0.96 ± 0.12
6.13 ± 0.17
0.83 ± 0.32
2.54 ± 0.31
0.10 ± 0.01
0.007 ± 0.002
0.41 ± 0.15
0.25 ± 0.028
0.26 ± 0.072
1.03 ± 0.026
0.27 ± 0.006
0.38 ± 0.048
6.78 ± 0.08
3.00 ± 0
0.039 ± 0.001
0.015 ± 0.003
0.026 ± 0.001
0.15 ± 0.005
0.014 ± 0.001
0.027 ± 0.001
0.23 ± 0.05
0.23 ± 0.030
0.089 ± 0.004
0.16 ± 0.014
0.83 ± 0.033
0.085 ± 0.020
0.24 ± 0.085
6.21 ± 0.08
2.46 ± 0.54
0.007 ± 0.001
0.016 ± 0
2.24 ± 0.070
3.20 ± 0.67
4.46 ± 0.12
1.48 ± 0.081
7.16 ± 0.46
9.61 ± 1.86
0.84 ± 0.01
0.34 ± 0.01
0.011 ± 0.002
NVP
DLV
EFV
ETR
0.051 ± 0.006
0.005 ± 0.001
0.014 ± 0.002
0.12± 0.004
0.003 ± 0
12

ACCEPTED MANUSCRIPT
AZT
0.005 ± 0
0.008 ± 0
0.013 ± 0.004
0.007 ± 0.001
0.011 ± 0.002
0.005 ± 0
a
EC50: concentration of compound required to achieve 50% protection of MT-4 cell cultures against HIV-1-induced cytotoxicity,
as determined by the MTT method.
Table 3. Resistance folds for a panel of mutant HIV-1 stains of selected compounds in
MT-4 cells.
Resistance Folds a
No.
F227L+
V106A
21.1
16.5
20.3
10.5
12.0
30.5
27.6
59.7
42.0
1.2
L100I K103N
E138K
Y181C
Y188L
3
3
3
3
3
3
b1
b2
b3
b7
b8
b9
52.7
53.0
34.5
20.5
48.0
306.5
3.2
11.1
12.5
9.0
1.1
0.8
0.9
0.9
0.7
1.4
0.9
1.3
1.0
3.5
1.9
6.2
4.5
85.7
112.0
110.3
25.9
74.0
358.0
37.0
22.1
68.0
2.8
5.5
6.0
4.8
13.5
19.0
26.1
78.9
24.0
0.8
4.3
12.0
23.9
64.7
1.4
NVP
DLV
EFV
ETR
AZT
66.8
20.0
1.7
4.0
0.7
1.2
1.0
1.6
0.7
a
Resistance Fold: RF, ratio of EC50 against mutant strain/EC50 against WT strain.
Based on the results of the first round of screening, six compounds 3b1-3b3,
13

ACCEPTED MANUSCRIPT
3
b7-3b9 were selected for testing against six more different HIV-1 mutant strains due
to their good potencies against both WT and RES056 strains. (Table 2). All six
compounds showed same sensitivity order against the six mutant viruses as follows:
Y188L (least sensitive to these compounds) < L100I < F227L+V106A < K103N <
Y181L < E138K (most sensitive to these compounds). For example, these molecules
all had single digit micromolar EC values against Y188L strain; for K103N mutated
50
virus, six compounds exhibited EC50 ranging from 0.25 – 1.03 µM; all tested
compounds except for 3b7 with amide showed EC under 40 nM against E138K
50
strain which is better than the control drug NVP and DLV and comparable to ETR and
AZT. Regarding the resistance fold (RF, ratio of EC50 against mutant strain/EC50
against WT strain), compounds 3b2, 3b3, 3b7, 3b8 showed the values for E138K less
than 1 while ETR had the RF of 3.5 (Table 3). That means these compounds inhibited
the E138K mutant strain specifically with a better EC than against the WT virus.
5
0
These findings implicated a same inhibitory mechanism of these analogues, especially
verified the introduction of a longer triazole tail may help to bind tighter to the RT
containing E138K. Among them, compounds 3b2 and 3b8 which contain N atom as
hydrogen bond acceptor have the best activity against E138K strain. Besides, all six
triazole analogues had lower EC values than both NVP and DLV against K103N,
5
0
Y181C, F227L+V106A, meanwhile they have better potency against Y188L than
NVP.
Table 4. Selectivity against a panel of mutant HIV-1 stains of selected compounds in
MT-4 cells.
14

ACCEPTED MANUSCRIPT
SI a
No.
F227L+
V106A
32
L100I
K103N
E138K
Y181C
Y188L
3
3
3
3
3
3
b1
b2
b3
b7
b8
b9
13
38
62
158
73
651
2631
756
109
450
123
29
8
18
121
19
6
33
7
23
159
5
13
60
214
473
>2
4189
6800
>64
686
761
>2
39
245
30
25
297
NVP
DLV
EFV
ETR
AZT
>18
14
>2
>2
12
715
15
44
16
> 62
370
> 1444
> 52
773
> 901
> 1235
179
> 928
159
> 1058
> 19
111
> 704
> 30
143
> 581
> 1544
a
SI: selectivity index, the ratio of CC50/EC50
.
Generally, in terms of the potency against all six mutants, the inhibitory ability
order of tested compounds was similar as that against WT virus, i.e. 3b2 ≈ 3b8 > 3b3
3b1 ≈ 3b9 >3b7, only with few exceptions to certain mutant strains. 3b2 and 3b8
>
which showed very similar potency were the best two compounds against these
mutants. They showed better potencies than the first generation NNRTIs NVP and
DLV against most of the mutants and were equally potent with ETR and AZT against
E138K strain. But since the two compounds and 3b9 had better cytotoxic properties,
15

ACCEPTED MANUSCRIPT
they showed much higher selectivity (3b2: SI = 2631, 3b8: SI = 4189, 3b9: SI = 6800
against E138K; 3b2: SI = 450, 3b8: SI = 686, 3b9: SI = 761 against Y181C) than
ETR (SI = 179 against E138K and SI = 159 against Y181C) when against E138K and
Y181C mutant virus strains (Table 4).
Table 5. Activity against NL4-3 HIV-1 strain, cytotoxicity and SI of compounds 3b8,
3
b9 and RPV in TZM-bl cells.
No.
EC50 (nM)
a
CC (nM)
SI
5
0
3
3
b8
b9
24 ± 5
20 ± 5
0.9 ± 0.3
> 221
> 215
> 341
> 9.2
> 10.8
> 378.9
RPV
a
EC50: Concentration of compound that causes 50% inhibition of viral infection, presented as the mean ± standard deviation (SD)
and determined in at least triplicate against HIV-1 in TZM-bl cells.
b
CC50: Concentrations that cause cytotoxicity to 50% of cells; CytoTox-Glo cytotoxicity assays (Promega) were used, and values
were averaged from two independent tests.
c
SI: selectivity index, the ratio of CC50/EC50
.
To further verify the above results, two compounds 3b8 and 3b9 were selected
for their excellent potency and low cytotoxicity along with RPV as a control to test
their potency against wild-type HIV-1 (NL4-3) replication in TZM-bl cell lines. Their
anti-HIV potency (EC , as measured by a luciferase gene expression assay [32]) and
5
0
cytotoxicity (CC ) as well as selective index (SI) data are summarized in Table 5.
5
0
The data for 3b8 and 3b9 were consistent between the two independent assays.
.3 Inhibition of HIV-1 RT
Table 6. Inhibitory activity of compound 3b8, 3b9 and ETR against HIV-1 RT.
2
Compd.
3b8
2.70
3b9
1.57
ETR
0.75
a
IC (µM)
5
0
a
5
0% inhibitory concentration of tested compounds required to inhibit biotin deoxyuridine triphosphate (biotin-dUTP)
incorporation into the HIV-1 RT by 50%.
With the aim to further confirm the target of the newly designed
16

ACCEPTED MANUSCRIPT
diarylnicotinamide 1,4-disubstituted 1,2,3-triazole derivatives, the two most
promising compounds 3b8 and 3b9 were also tested in enzymatic assays against
HIV-1 RT [33, 34]. ETR was used as a reference drug in this assay. As shown in
Table 6, compound 3b8 and 3b9 exhibited moderate inhibitory activity of the enzyme
with IC value of 2.70 and 1.57 µM, respectively, which are about one to three time
5
0
less potent than that of ETR (0.75 µM). To a certain extent, the result reflected their
activity against the WT HIV-1 virus in cell-based assays, and also manifested that
these derivatives represented by compound 3b8 and 3b9 targeted HIV-1 RT, thus
acting as genuine NNRTIs.
2
.4 Molecular modeling analysis
To better explain the SAR and find the potential binding modes of this series of
molecules, the two representative compounds 3b8 and 3b9 were docked into the
NNRTIs binding pocket in WT and E138K mutated RT using the Surflex-Dock of
SYBYL-X 2.0. The proposed binding modes of these compounds to the NNIBP are
shown in Fig. 3 & Fig. 4.
In the binding mode with WT RT (Fig. 3), the main structures of 3b8, 3b9 and
ETR, the original ligand of 3MEC, overlapped very well, and the three compounds
shared a common “U” shape binding mode [35-37] with NNIBP in reverse
transcriptase of WT HIV-1 virus: i) the left substituted phenol of these structures
pointed to a hydrophobic pocket composed by V108, K181, K188, F227 and W229,
with a π-π stacking with residue K188; ii) the right 4-cyanoaniline interacted with
another pocket which contains residues V106, L234, H235, P236 and Y318; iii) the
17

ACCEPTED MANUSCRIPT
NH between the right aniline and the central pyridine core in three structures all had a
robust hydrogen bond with the key residue K101 in RT. The differences between two
novel compounds and ETR in the binding mode were mainly the parts pointing to the
entrance channel. Particularly, the amino group on the pyrimidine of ETR formed a
hydrogen bond with the carboxyl of residue E138; while the triazoles of two new
structures pointing to the solvent-exposed channel oriented the connected amides to
form a hydrogen bond with the side chain of K101. Furthermore, the pyrrolidine of
3
b8 and morpholine of 3b9 at the far end exposed to the solvent and approached to
hydrophobic I135 and V179, respectively. It is possible that the amphiphilic property
of these two moieties are beneficial to increase the binding affinity to RT.
18

ACCEPTED MANUSCRIPT
Figure 3. Predicted binding mode of compound 3b8 (yellow), 3b9 (orange) in the
allosteric site of HIV-1 WT RT (PDB code: 3MEC) with the comparison with the
original ligand of this crystal structure ETR (pink). The docking results are shown by
PyMOL.hydrogen bond interactions are indicated by dashed lines.
Since there was no available crystal complex of E138K RT with ETR, we
obtained the RT harboring E138K from the crystal complex with PETT-2 (PDB:
2
HNZ). The ligand PETT-2 of 2HNZ is from PETT series which was obtained by
simplify the structure of NNRTI tetrahydroimidazobenzodiazepinthione derivatives
19

ACCEPTED MANUSCRIPT
through bond disconnection [38]. PETT-2 showed excellent potency against HIV-1 RT
(
(
IC of 5 nM) [39], and shared a very similar binding mode of NNIBP with ETR
50
Fig. 4a. Note: The 3D structure of PETT-2 in the NNIBP was extracted directly from
the PDB complex 2HNZ, which bearing a methyl on the pyridine connected to the
thiourea instead of a Cl atom of the correct structure but should not confer a big effect
of whole binding mode). The extracted ligand PETT-2 (gray) overlapped well with the
upper part of ETR (pink) which was docked into the E138K mutant RT (cyan): two
pyridine rings just occupied the position where the two phenyl wings of ETR were;
the thiourea also could form a hydrogen bond with carboxyl of K101. Comparison of
the mutated RT (E138K, from PDB 2HNZ, cyan) with WT RT (from PDB 3MEC,
green) showed that most parts of the two structures superposed on each other with a
perturbation in the positions of some of sidechains and only bearing one large change
at the mutation site (Fig. 4a), which consistent with the description in the original
study [40]: The WT E138 and mutated K138 residues overlapped as far as the CB
carbon atom, with the rest of the side chain of K103 swung away from the binding
pocket by a rotation giving a different orientation of amino and carboxyl groups;
besides, there was a movement in the K101 side chain which involved a rotation
about the CD–CE bond resulting in the amino group movement toward further away
from the NNRTI pocket. Above findings indicated the feasibility of using the 2HNZ
in our modeling study.
20

ACCEPTED MANUSCRIPT
(a)
21

ACCEPTED MANUSCRIPT
(b)
22

ACCEPTED MANUSCRIPT
(c)
Figure 4. (a) The structure of PETT-2, superimposition of docked ETR (pink) to the
PETT-2 (gray) extracted from PDB 2HNZ and the overlap of WT (green, from PDB
3
MEC) and E138K mutant (cyan, from PDB 2HNZ) RT. (b) The proposed binding
mode of 3b8 (yellow), 3b9 (orange) and ETR (pink) with E138K RT (cyan). (c) The
proposed binding mode of 3b8 (yellow), 3b9 (orange) and ETR (pink) with E138K
RT (cyan): view from the entrance channel. The docking results are shown by
PyMOL.hydrogen bond interactions are indicated by dashed lines.
23

ACCEPTED MANUSCRIPT
Then we docked 3b8 (yellow) and 3b9 (orange) into this mutated RT with ETR
pink) as a reference (Fig. 4b and Fig. 4c). There were some obvious adjustments of
(
the main structure of two target structures. It seemed like there was a rotation of the
whole parts deep in the pocket compared to ETR, but all the key interactions
remained including the hydrophobic interactions with the two sub-pocket and the
double hydrogen bonds with backbone of K101 (Fig. 4b). For the parts pointing to the
entrance channel, ETR lost a hydrogen bond with E138 due to the mutation, while
3
b8 and 3b9 kept the interactions with the residues located at the interface with
solvent which may involve K101, K138, T139, R172 and P176 (Fig. 4c). Especially,
for 3b8, the most potent compounds against E138K strain, there was two additional
hydrogen bonds with K138 and R172, respectively. These findings may explain the
reason that this series had good potency against the key drug resistant mutant to the
new generation HIV-1 NNRTIs E138K and confirmed our rational design as effective
strategy to overcome the drug-resistant issues [37].
3
. Conclusion
In general, a series of novel diarylnicotinamide triazole analogues were
rationally designed based on structure-guided approach, synthesized and evaluated for
their bioactivities against HIV-1 (IIIB, K103N+Y181C, L100I, K103N, E138K,
Y181C, Y188L and F227L+V106A) and HIV-2 (ROD) in MT-4 cells. Several
promising compounds (3b2, 3b8, 3b9) were discovered with very good potency
against WT and several mutant strains. Noteworthily, regarding the inhibitory activity
against E138K, the major drug resistant mutant to the new generation HIV-1 NNRTIs,
24

ACCEPTED MANUSCRIPT
3
b8 and 3b9 with triazole linker targeting the entrance channel of NNRTIs in RT are
equally potent to marketed drug ETR but with much lower cytotoxicity. The
inhibitory activity of representative compounds confirmed their target is HIV-1 RT.
More valuable information from the SAR and molecular modeling study should be
helpful to develop potential preclinical candidates for necessary back-up drugs of the
second generation NNRTIs. Ongoing study in our lab will be reported in due course.
4
. Experimental Section
.1 Chemistry
4
All melting points were determined on a micromelting point apparatus and are
1
13
uncorrected. H-NMR and C-NMR spectra were recorded on a Bruker AV-400
spectrometer using DMSO-d as solvent and tetramethylsilane (TMS) as internal
6
standard. Chemical shifts are reported in parts per million (δ), and signals are
expressed as s (singlet), d (doublet), t (triplet), q (quartet) or m (multiplet). Mass
spectra were taken on a LC Autosampler Device: Standard G1313A instrument. TLC
was performed on Silica Gel GF254 for TLC (Merck) and spots were visualized by
iodine vapors or by irradiation with UV light (254 nm). Flash column
chromatography was performed on column packed with Silica Gel 60 (200-300
mesh). Solvents were reagent grade and, when necessary, were purified and dried by
standard methods. Concentration of the reaction solutions involved the use of rotary
evaporator at reduced pressure.
25

ACCEPTED MANUSCRIPT
4
.1.1 General procedure for the synthesis of substituted (azidomethyl)benzene (2a)
and other nonaromatic substituted azidomethane (2b)
Method A
To a flask, appropriate benzyl bromide (1a, 0.50 g) and NaN (1.5 eq) was added
3
and then dissolved in 6 mL DMF : H O (5:1) as solvent. The mixture was stirred
2
under oil bath at 50 °C for 12 hours. After the reaction, the solvent was removed by
rotary evaporation, and 20 mL of water was added. The solution was then extracted by
ethyl acetate for three times. The organic phase was combined, dried and concentrated
to yield the crude which was used for the next step without further purification. This
method was applied in the preparation of 2a1-2a14.
Method B
To a flask, appropriate halide (1b2-1b4 and 1b7, 0.50 g) and NaN (2~5 eq) (plus
3
0
.1 eq of tetrabutylammonium, when the halide was bromoacetamide) was added and
then dissolved in 5~10 mL DMF as solvent. The mixture was stirred under oil bath at
0~80 °C for 12 hours. After the reaction, the solvent was removed by rotary
7
evaporation, and 15 mL of water was added. The solution was then extracted by ethyl
acetate for three times. The organic phase was combined, dried and concentrated to
yield the crude product which was used for the next step without further purification.
This method was applied in the preparation of 2b2-2b4 and 2b7.
Method C
To a flask, appropriate halide (1b5 and 1b6, 0.70 g) was dissolved in 15 mL of
acetone and then NaN (3~5 eq) (plus 0.1 eq of KI, when the halide was
3
26

ACCEPTED MANUSCRIPT
3
5
-Bromo-1-propanol) and 5 mL of H O was added. The mixture was stirred bath at
2
0 °C for 12 hours. After the reaction, the solvent was removed by rotary evaporation,
and 15 mL of water was added. The solution was then extracted by ethyl acetate for
three times. The organic phase was combined, dried and concentrated to yield the
crude product which was used for the next step without further purification. This
method was applied in the preparation of 2b5 and 2b6.
Method D
To a flask containing 10 mL anhydrous methylene chloride, bromoacetyl
bromide (1b8) or chloroacetyl chloride (1b9) (1eq) was added and the mixture was
stirred at -30 °C for 15min. Then morpholine (1.5 eq, for 2b8) or tetrahydropyrrole (2
eq, for 2b9) was added dropwise. After 5 hours of stirring at rt, the solution was
washed by saturated NH Cl aq, saturated Na CO aq, H O and brine successively. The
4
2
3
2
organic phase was dried by Na SO , filtered, and concentrated. To the residue, 5~10
2
4
mL DMF and NaN (1.8 eq) was added. The mixture was stirred under oil bath at
3
7
0~80 °C for 12 hours. After the reaction, the solvent was removed by rotary
evaporation, and 15 mL of water was added. The solution was then extracted by ethyl
acetate for three times. The organic phase was combined, dried and concentrated to
yield the crude product which was used for the next step without further purification.
This method was applied in the preparation of 2b8 and 2b9.
4
.1.2 General procedure for the synthesis of N-((1-(substituted
benzyl)-1H-1,2,3-triazol-4-yl)methyl)-6-((4-cyanophenyl)amino)-4-(mesityloxy)nicoti
namide (3a) and
27

ACCEPTED MANUSCRIPT
N-((1-substituted-1H-1,2,3-triazol-4-yl)methyl)-6-((4-cyanophenyl)amino)-4-(mesitylo
xy)nicotinamide (3b)
L-6b5 (0.15 g, 0.37 mmol), appropriate azide (3 eq), CuSO (0.0091 g, 0.036
4
mmol), and VcNa (0.0217 g, 0.11 mmol) was added to the flask together with the 10
mL of THF:H O (1:1) as solvent. After the mixture was stirred at 65 °C for 12 hours,
2
1
0 mL of H O was added to the flask. Then it was extracted by ethyl acetate,
2
concentrated and then the product (3a and 3b) was obtained by column
chromatography.
4
.1.2.1
N-((1-(2-methylbenzyl)-1H-1,2,3-triazol-4-yl)methyl)-6-((4-cyanophenyl)amino)-4(me
sityloxy)nicotinamide (3a1)
2-Methylbenzyl bromide (1a1) was used as starting material. White solid, yield:
1
9
2.0%, m.p.: 220-223 °C. H NMR (400 MHz, DMSO-d6, ppm) δ: 9.74 (s, 1H, NH),
.70 (s, 1H, pyridine-H), 8.66 (t, 1H, J = 5.6 Hz, NH), 7.88 (s, 1H, triazole-H), 7.85
8
(d, 2H, J = 8.8 Hz, Ph-H), 7.68 (d, 2H, J = 8.7 Hz, Ph-H), 7.25-7.07 (m, 4H, Ph-H),
7
.05 (s, 2H, Ph-H), 5.93 (s, 1H, pyridine-H), 5.58 (s, 2H, CH ), 4.59 (d, 2H, J = 5.5
2
1
3
Hz, CH ), 2.30 (s, 3H, CH ), 2.23 (s, 3H, CH ), 2.03 (s, 6H, 2 × CH ). C-NMR (100
2
3
3
3
MHz, DMSO-d6, ppm) δ: 163.60, 162.69, 158.62, 152.17, 147.27, 145.89, 145.66,
1
1
36.77, 135.85, 134.63, 133.54 (2 × C, Ph), 130.87, 130.57, 130.26, 129.14, 128.74,
26.65, 123.21 (triazole-CH), 120.05, 118.45 (2 × C, Ph), 112.21, 102.47, 94.41
(
pyridine-C3), 51.38 (CH ), 35.44 (CH ), 20.85 (CH ), 19.04 (CH ), 16.13 (2 × CH ).
2
2
3
3
3
+
+
ESI-MS: m/z 558.5 [M+H] , 580.5 [M+Na] . C H N O (557.25).
3
3
31
7
2
28