Discovery of novel diarylpyrimidines as potent HIV-1 NNRTIs by investigating the chemical space of a less explored "hydrophobic channel"

Article abstract of DOI:10.1039/c7ob02828h

A new series of diarylpyrimidines (DAPYs) were designed, synthesized and evaluated as novel HIV-1 NNRTIs to further explore the chemical space surrounding the "hydrophobic channel" of the NNRTI binding pocket (NNIBP), guided by the comprehensive analysis

Full text of DOI:10.1039/c7ob02828h

View Article Online
View Journal
Organic &
Biomolecular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: X. Z. Zhou, T. Liu,
D. Kang, Z. Huo, G. Wu, D. Daelemans, E. De Clercq, C. Pannecouque, P. Zhan and X. Liu, Org. Biomol.
Chem., 2017, DOI: 10.1039/C7OB02828H.
This is an Accepted Manuscript, which has been through the
Volume 14 Number
1 7 January 2016 Pages 1–372
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Page 1 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
Discovery of Novel Diarylpyrimidines as Potent HIV-1 NNRTIs by
Investigating the Chemical Space of Less Explored “Hydrophobic
Channel”
Zhongxia Zhou,1 Tao Liu,1 Dongwei Kang,1 Zhipeng Huo,1 Gaochan Wu,1 Dirk
Daelemans,2 Erik De Clercq,2 Christophe Pannecouque,2,* Peng Zhan,1,* and
Xinyong Liu1,*
1Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry
of Education), School of Pharmaceutical Sciences, Shandong University, 44 West
Culture Road, 250012 Jinan, Shandong, PR China
2Rega Institute for Medical Research, K.U.Leuven, Minderbroedersstraat 10, B-3000
Leuven, Belgium
Abstract: In this article, guided by the comprehensive analysis of Xꢀray structural
biology data of HIVꢀ1 RT/NNRTIs complexes and molecular modeling, a new series
of diarylpyrimidines (DAPYs) was designed, synthesized and evaluated as novel
HIVꢀ1 NNRTIs, to further explore the chemical space surrounding the “hydrophobic
channel” of NNRTIs binding pocket (NNIBP). Encouragingly, most of the
synthesized DAPYs were found to be active against HIVꢀ1 wildꢀtype (WT) strain
with EC50 values ranging from 3 nM to 63 nM, and displayed significantly reduced
cytotoxicity compared with etravirine (ETV) and rilpivirine (RPV). Notably, two
most promising compounds Z10 (EC50 = 3 nM) and Z13 (EC50 = 3 nM) showed
equivalent potency against HIVꢀ1 WT to the reference drugs, efavirenz (EFV, EC50 =
3 nM) and ETV (EC50 = 3 nM). Among them, Z13 showed the most potent activities
against HIVꢀ1 mutant strains, including K103N (EC50 = 10 nM), E138K (EC50 = 22
nM) and RES056 (EC50 = 0.935 ꢁM), Z17 also showed doubleꢀdigit nanomolar
inhibitory activity toward Y181C, Y188L and F227L+V106A with EC50 values 27

Organic & Biomolecular Chemistry
Page 2 of 41
View Article Online
DOI: 10.1039/C7OB02828H
nM, 98 nM and 30 nM respectively. The structureꢀactivity relationships (SARs) and
molecular docking studies provided important clues for further molecular elaboration.
Collectively, this study provides useful information to guide lead optimization and
drug discovery via exploration of this scarcely investigated region.
Keywords: HIV; NNRTIs; hydrophobic channel; SBDD; antiꢀHIV activity
evaluation.
Introduction
HIVꢀ1 reverse transcriptase (RT) is an essential enzyme which plays a unique
role in viral replication cycle.1 As a result, inhibition of HIVꢀ1 RT will give rise to
viral load declines in patients and greatly contribute to antiꢀHIV therapy.
Nonꢀnucleoside RT inhibitors (NNRTIs) could occupy an allosteric site about 10 Å
away from HIVꢀ1 RT polymerase active site, inducing a hydrophobic pocket
(NNRTIs binding pocket, NNIBP) and then disrupting the key function of RT.2
Nowadays, NNRTIs present the most powerful and promising antiꢀHIV drugs and
have been successfully used as an essential component of highly active antiretroviral
therapy (HAART) regimens.

Page 3 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
Figure 1. Chemical structures of NNRTIs approved by the U.S. FDA.
Up to now, five NNRTIs have been approved by the U.S. Food and Drug
Administration (FDA) for antiꢀHIVꢀ1 therapy (Figure 1), including three
firstꢀgeneration inhibitors, nevirapine (NVP),2 delavirdine (DLV),3 and efavirenz
(EFV),2 and two secondꢀgeneration inhibitors, etravirine (ETV)4 and rilpivirine (RPV,
namely TMC278).5 Thereinto, EFV and RPV are also used as the important
components of two oneꢀaꢀday combination therapies Atripla and Complera
respectively, which both contains other two HIVꢀ1 nucleoside RT inhibitors (NRTIs),
emtricitabine and tenofovir.6,7
ETV and RPV belong to diarylpyrimidines (DAPYs) which represent the most
successful class of HIV NNRTIs, causing widespread interest due to their high
potency and moderate cytotoxicity.8 DAPYs adapt a Uꢀshaped conformation in the
NNIBP and show better compliance to common resistance mutations due to the
structural flexibility.9 The published cocrystal structures of DAPYs within HIVꢀ1 RT
indicated that the nature of torsional flexibility (wiggling) and reposition (jiggling) of
DAPYs allow their adaptation in different NNIBPs that caused by various resistant
RT mutations.10 However, hypersensitivity reactions or other adverse effects even
drug resistance still emerge from those patients receiving the secondꢀgeneration
NNRTIs regimens.8,11 Unlike the firstꢀgeneration NNRTIs that usually showed
significantly reduced activities against several commonly selected mutant strains such
as Y181C, Y188C, K103N and L100I,12 two mutant strains K103N and E138K are
most frequently selected by the secondꢀgeneration inhibitors.13 Despite the obvious
advantages for the secondꢀgeneration NNRTIs, it is still a challenge in seeking new
NNRTIs with high potency toward the wildꢀtype (WT) virus and numerous clinically
observed variants as well as good pharmacological properties.14,15
Large amounts of Xꢀray structural biology data of NNRTIs/RT complexes
revealed that several features should be considered in the design of novel NNRTIs
with improved resistance profiles, such as conformational flexibility and positional

Organic & Biomolecular Chemistry
Page 4 of 41
View Article Online
DOI: 10.1039/C7OB02828H
adaptability, targeting the highly conserved regions, and forming additional hydrogen
bonds with backbone atoms.16ꢀ18 For example, this highꢀresolution crystallographic
structure revealed that the cyanovinyl group in the left wing of RPV was positioned in
a cylindrical tunnel formed by highly conserved Phe227 and Trp229 (named as
tolerant region III), connecting the NNIBP to the nucleic acidꢀbinding cleft.10 The
improved activity of RPV against WT and mutant HIVꢀ1 strains compared to ETV
may involve a specific interaction of this cyanovinyl group with the indole ring of
W229.
Moreover, recent advance in structural biology study of HIV RT promoted the
identification of additional binding sites around the classical NNIBP, which afforded
a comprehensive understanding about the binding modes of NNRTIs within NNIBP
and also provided opportunities to seek more potent NNRTIs by occupying these
potential binding sites. Especially, the presence of two solventꢀaccessible regions,
namely, tolerant region I and tolerant region II, provided a broad chemical space for
the structural modification of NNRTIs (Figure 2). Consequently, there have been
intensive efforts from our lab to the modification of these two regions and many
highly active compounds were identified.19ꢀ22 Recently, as the continuous study of our
group on the discovery of novel NNRTIs, iterative modifications in the tolerant region
I and tolerant region II yielded a series of NNRTIs bearing a piperidineꢀsubstituted
thiophene[3,2ꢀd]pyrimidine scaffold, in which the most potent compound 25a,
featuring a cyanovinyl group, showed a broad spectrum of activity against WT and
common clinically relevant HIVꢀ1 mutant strains, moreover, 25a showed favorable
drugꢀlike pharmacokinetics and safety properties in mouse in vivo.23a,b Molecular
docking indicated that the cyanovinyl group of 25a is also positioned in this
hydrophobic tunnel and forms additional interactions with the key residues. These
interactions are expected to be kept despite mutations in the NNIBP, explaining the
reason why 25a displays high potency against a variety of drugꢀresistant strains.
However, the cyanovinyl group in RPV and 25a may be viewed as a “Michael
acceptor”, leading to potential offꢀtarget covalent modifications of proteins, nucleic

Page 5 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
acids, or other biological entities.24a,b Compared to ETV, RPV and 25a demonstrated
greatly increased cytotoxicity (CC50 = 3.69 ꢁM24c, CC50 = 2.30 ꢁM23b). In addition,
fosdevirine (FDV) is also an cyanovinylꢀcontainng NNRTI, which was abandoned
after phase IIb clinical trials, and main metabolites resulting from the possible
participation in Michael addition with the cysteine of glutathione by the cyanovinyl
group of FDV were implicated as the source of seizures in patients.24c
Under the circumstances, we set out to find a replacement for the cyanovinyl
group in the left wing of DAPYs. According to the previously published cocrystal
structure of HIVꢀ1 RT/RPV complexes (PDB codes: 1TV6, 2ZD1, 1KLM, and
2B5J),10,25ꢀ27 it was clear that the larger sized 2,6ꢀdimethylꢀ4ꢀcyanovinyl aniline group
of RPV would be accommodated in the subꢀpocket formed by F227, Y188, L234 and
W229.10,28ꢀ30 One effective design strategy of eliminating drug resistance is to
maximize interactions between inhibitor and the RT, especially promote extensive
hydrophobic interactions with highly conserved residues and hydrogen bond
interactions with backbone atoms in the NNIBP. Based on this analysis, we
envisioned that diverse longer and larger alkyne and alkene (substituted by aromatic
rings, linear alkanes, cycloalkanes and heterocycles) replacements of 4ꢀcyanovinyl
could be expected to satisfy spatial orientation of the well tolerated cylindrical
hydrophobic tunnel (tolerant region III), and yield more optimal arylꢀaryl
hydrophobic interactions.

Organic & Biomolecular Chemistry
Page 6 of 41
View Article Online
DOI: 10.1039/C7OB02828H
Figure 2. Illustration of the chemical modification strategy and structures of the newly designed
DAPY derivatives in this study.
On the other hand, of particular note is that the newly introduced
piperidineꢀlinked NH in the right wing of 25a can form direct or waterꢀmediated
hydrogen bonds with the backbone of K101/K103. Besides, the
1ꢀ(4ꢀsulfonamidebenzyl)piperidine are directed toward the proteinꢀsolvent interface
(the P236 hairpin loop, tolerant region I) and are available to participate in additional
hydrogen bonds with the backbone of K104 and V106, which responsible for its
significantly improved drug resistance profiles.23 Moreover, the polar substituted
piperidinꢀ4ꢀylꢀamino moiety could contribute to improved physicochemical and
pharmacokinetic properties. Coincidentally, our previous replacement of the
p-cyanoaniline fragment in the DAPYs lead structure with substituted
piperidinꢀ4ꢀylꢀamino moiety has already led to the discovery of several series of
highly potent NNRTIs.23
Therefore, inspired by this privileged scaffold of highly active DAPYs, in part of

Page 7 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
the newly designed compounds, the pꢀcyanoaniline moiety (rightꢀring) was also
replaced with different substituted (varying in size, polar group substitution, and
electronic nature) benzylꢀlinked piperidinꢀ4ꢀylꢀamino moieties, in order to enhance
their potency against resistanceꢀassociated variants.
On the basis of the above analysis, herein, we describe our medicinal chemistry
efforts to the discovery of a series of novel DAPYs as potent HIVꢀ1 NNRTIs
featuring substituted alkynes or alkenes in the left wing and substituted benzylꢀlinked
piperidinꢀ4ꢀylꢀamino moiety in the right wing, by investigating the chemical space of
less explored tolerant region III. In antiꢀHIV evaluation assays, these novel NNRTIs
had potent broadꢀspectrum profiles against HIVꢀ1 WT and resistant mutant strains and
most of these compounds displayed low toxicities in cell culture, which confirmed the
feasibility of using substituted alkynes or alkenes as replaceable groups of cyanovinyl
to occupy the chemical space surrounding the left hydrophobic channel of NNIBP. In
addition, preliminary structureꢀactivity relationships (SARs) of the designed
compounds are discussed in detail for further optimization.
2. Results and discussion
2.1. Chemistry
The synthetic routes of novel DAPYs are outlined in Scheme 1 and Scheme 2.

Organic & Biomolecular Chemistry
Page 8 of 41
View Article Online
DOI: 10.1039/C7OB02828H
Reagents and conditions: (i) 4ꢀAminobenzonitrile, 180℃, 8h; (ii) POCl3, reflux, 0.5h; (iii)
K2CO3, DMF, 100℃, 10h; (iv) substituted alkynes (5a-k), Bis(triphenylphosphine)palladium
dichloride, cuprous iodide, 4ꢀiodoꢀ2,6ꢀdimethylphenol, triethylamine, THF, 60 ℃, 10h or
substituted alkenes (5l-m), 4ꢀiodoꢀ2,6ꢀdimethylphenol, Palladium acetate, Tol3P, sodium ethoxide,
DMA, 60℃, 10h; (v) 6a or 6f, K2CO3, DMF, r.t., 10h; (vi) NꢀBocꢀ4ꢀaminopiperidine, K2CO3,
DMF, 120°C, 12h; (vii): trifluoroacetic acid, DCM, r.t., 5h; (viii) benzyl chloride or benzyl
bromide, K2CO3, DMF, r.t.
The general synthetic steps adopted to obtain the target compounds Z1~14 was
straightforwardly outlined in scheme 1. These target compounds were synthesized
from the commercially available starting material 2ꢀ(methylthio)pyrimidinꢀ4(3H)ꢀone
(1). Firstly, 1 was reacted with 4ꢀaminobenzonitrile at 180℃ by melting method to
afford intermediate 4ꢀ((4ꢀoxoꢀ1,4ꢀdihydropyrimidinꢀ2ꢀyl)amino) benzonitrile (2),
which
followed
by
chlorination
with
POCl3
to
generate
4ꢀ((4ꢀchloropyrimidinꢀ2ꢀyl)amino)ꢀbenzonitrile
(3).31
Secondly,
4ꢀiodoꢀ2,6ꢀdimethylphenol (4) underwent Sonogashira coupling reaction with alkynyl
substituted intermediates cuprous iodide, catalyzed by triethylamine,
bis(triphenylphosphine)palladium(II) dichloride in anhydrous THF or acetonitrile and
oxygenꢀfree conditions to provide intermediate 6a-k.32 Intermediate 4 also underwent

Page 9 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
palladium acetate catalyzed heck coupling reaction with alkene substituted
intermediates to provide another class of intermediate 6l-m. Then, 3 were substituted
by different phenyl derivatives intermediate 6a-m at its Cꢀ4 position in the presence
of K2CO3 to produce Z1~14.
Scheme 2 was started from commercially available 2,4ꢀdichloropyrimidine (7),
successive nucleophilic substitution reactions with substituted phenols and
4ꢀ(tertꢀbutoxycarbonyl)ꢀaminopiperidine in the presence of K2CO3 and DMF afforded
the key intermediates 9.23 Deprotection with trifluoroacetic acid at room temperature
in dichloromethane afforded the corresponding analogues 10, which were then led to
the target compounds Z15~20 by nucleophilic substitution with substituted benzyl
chlorides (or bromines) in the presence of K2CO3. The synthesized compounds were
characterized by physicochemical and spectral means and the NMR and MS spectral
data are found in agreement with the assigned molecular structures.
2.2. Biological activity
The distinct synthesized DAPYs derivatives were evaluated for their antiꢀHIV
activities in MT4 cell against WT HIVꢀ1 strain (IIIB), a panel of NNRTIs resistant
strains with five single mutations (L100I, K103N, E138K, Y181C, Y188L) and two
double mutations (K103N/Y181C (RES056) and F227L/V106A) in RT respectively.
Besides, the inhibitory activity against HIVꢀ2 strain (ROD) was also evaluated to test
the selectivity of these NNRTIs, and rilpivirine (RPV), zidovudine (AZT), nevirapine
(NVP), efavirenz (EFV), etravirine (ETV) and delavidine (DLV) were selected as
reference drugs. The biological results are expressed as EC50, CC50 and SI (selectivity
index, i.e., CC50/EC50 ratio) and summarized in Table 1 and Table 2.
Table 1. Activity and cytotoxicity of Z1~20 against HIVꢀ1 (IIIB) and HIVꢀ2 (ROD) strains in
MTꢀ4 cells.

Organic & Biomolecular Chemistry
Page 10 of 41
View Article Online
DOI: 10.1039/C7OB02828H
EC50 (µM)a
SIc
Compds
R1
R2
CC50 (µM)b
IIIB
ROD
IIIB
ROD
<1
Z1
Z2
ꢀ
ꢀ
0.005±0.002
>12.116
12.116±6.16
12.476±7.450
2466
0.038±0.006
0.224±0.029
0.063±0.017
0.046±0.017
0.014±0.006
0.013±0.006
0.016±0.007
0.007±0.002
0.003
>3.29
>7.353
>25.035
>8.883
>5.385
>5.005
>0.844
>3.81
329
33
<1
<1
<1
≤9
<1
<1
<1
<1
<1
<1
<1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Z3
Z4
7.353±1.256
25.035±2.312
8.883±8.289
5.385±0.598
5.005±0.514
0.844±0.108
12.758±5.520
3.555±1.063
18.528±5.886
8.510±1.644
395
192
382
385
53
Z5
Z6
Z7
Z8
Z9
1863
1110
703
165
Z10
Z11
Z12
>3.555
>18.528
>8.510
0.026±0.006
0.052±0.011

Page 11 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
ꢀ
Z13
Z14
Z15
Z16
Z17
Z18
Z19
Z20
0.003±0.002
0.006±0.001
0.024±0.003
0.269
>3.476
>2.068
>1.063
>0.261
>0.732
>0.198
>0.630
>0.766
3.476±1.391
2.068±0.857
1.063±0.130
0.261±0.095
0.732±0.135
0.198±0.055
0.630±0.248
0.766±0.072
1010
339
44
<1
<1
<1
<1
<1
<1
<1
<1
ꢀ
1
0.006
118
32
0.006±0.002
0.007±0.002
0.005
93
SO2NH2
152
RPV
AZT
NVP
EFV
ETV
DLV
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
0.001
>3.98
3.984
>7.48
3989
>1043
>127
>2372
782
<1
ꢀ
0.007±0.006
0.118±0.067
0.003±0002
0.003
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
>15.02
>6.33
ꢀ
ꢀ
2.2±0.015
>43.81
ꢀ
0.023±0.024
>1879
ꢀ
aEC50: concentration of compound required to achieve 50% protection of MTꢀ4 cell cultures
against HIVꢀ1ꢀinduced cytotoxicity, as determined by the MTT method.
bCC50: concentration required to reduce the viability of mockꢀinfected cell cultures by 50%, as
determined by the MTT method.
cSI: selectivity index, the ratio of CC50/EC50.
As shown in Table 1, the biological results clearly showed that all newly

Organic & Biomolecular Chemistry
Page 12 of 41
View Article Online
DOI: 10.1039/C7OB02828H
synthesized compounds were very effective to inhibit HIVꢀ1 IIIB replication from
singleꢀdigit to doubleꢀdigit nanomolar ranges. Notably, compounds Z10 and Z13
turned out to be the most potent HIVꢀ1 inhibitors with the same EC50 value of 3.0 nM
against WT HIVꢀ1 strain, which were superior to those reference drugs NVP (EC50 =
118 nM), DLV (EC50 = 23 nM), AZT (EC50 = 7 nM) and comparable to EFV (EC50 =
3 nM) and ETV (EC50 = 3 nM). Besides, all other compounds except Z3 (EC50 = 224
nM) and Z16 (EC50 = 269 nM) exhibited considerable potency against the WT strain
with EC50 values in the range of 5~63 nM, which were superior to NVP. In addition,
most compounds from Z1~14 displayed decreased cytotoxicity compared with RPV
and ETV.
Some hydrophobic groups were firstly involved in our new explorations with the
purpose of extending to the left hydrophobic channel of NNIBP. For this purpose, we
kept the 4ꢀcyanophenyl moiety constant in the right wing and used a small set of triple
bondꢀlinked phenyl or substitutedꢀphenyl groups as the left ring, gave seven
compounds Z2~8. As expected, most of these compounds except compound Z3
exhibited high potency against the WT HIVꢀ1 strain with low nanomolar EC50 values
ranging from 13 to 63 nM and SI values between 53 and 395.
The preliminary SARs of these novel NNRTIs indicated that the substituents in
the left phenyl ring have great influence on the antiꢀHIV activity. For instance,
compound Z4 (EC50 = 63 nM) with a carboxylic group at C4 position in the phenyl
ring displayed a marked decreased potency against the WT HIVꢀ1 strain compared to
the aminoꢀsubstituted compound Z6 (EC50 = 14 nM). In addition, the electronic
nature or steric demand of substituents at the C4 position in the phenyl ring may also
affected the activity. For example, through comparative analysis of Z3 (EC50 = 224
nM), Z4 (EC50 = 63 nM) and Z6 (EC50 = 14 nM) with paraꢀsubstituents, it seems that
introduction of some polar groups can enhance activities, such as amino group.
However, compound Z4 with a carboxy group greatly reduced the activity compared
to Z2 (EC50 = 38 nM) with no substituent at C4 position, which may be due to its
negative charged form in experimental conditions, influencing its cell membrane

Page 13 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
permeability. In particular, Z7 displayed prominent inhibitory activity with an EC50
value of 13 nM among the initial seven compounds, being more potent than the
reference drugs NVP and DLV. Another compound Z8 (EC50 = 16 nM) with a
methanamide group displayed equipotent activity with Z7, consisting with the
aforementioned SAR that additional polar groups in the benzene ring could enhance
the activity.
The substitutedꢀposition in the additional left phenyl ring has some influence on
the antiꢀHIV activity. For instance, compounds Z6 (EC50 = 14 nM) and Z7 (EC50 = 13
nM) with paraꢀsubstituted and metaꢀsubstituted −NH2 group respectively, showing
equipotent activity against HIVꢀ1 WT strain, but Z5 (EC50 = 46 nM) with
orthoꢀsubstituted −NH2 group reduced the activity about 3ꢀfold. These observations
suggest the proper substituted positions should be C3 and C4 in the additional left
phenyl ring while substituents at C2 position impact negatively on the activity, which
may disturb the expected hydrophobic interactions between the phenyl group and
hydrophobic tunnel.
Back to our initial ideas of designing compounds exploring the chemical space
surrounding the hydrophobic channel of NNIBP, a more hydrophobic group,
cyclopropyl group was introduced, leading to compound Z9 (EC50 = 7 nM) with a
5ꢀfold enhanced potency compared to the phenylꢀsubstituted compound Z2. Notably,
compound Z10 (EC50 = 3 nM) holding the ethanol group at the benzene alkynyl
moiety exhibited more potent inhibition than the initial seven phenylꢀsubstituted
compounds, the reference drugs AZT, NVP and DLV, and were equipotent with EFV
and ETV. These results suggest small lipophilic groups could maintain the interaction
with the left hydrophobic channel of NNIBP, and polar groups extending from these
lipophilic groups may increase the solvent effect.
In addition, styryl or substitutedꢀstyryl replacements for the cyanovinyl group
were also considered in our initial design, styrylꢀsubstituted compound Z11 (EC50
=
26 nM) showed a slight increased potency compared to the phenylethynylꢀsubstituted

Organic & Biomolecular Chemistry
Page 14 of 41
View Article Online
DOI: 10.1039/C7OB02828H
compound Z2 (EC50 = 38 nM) while compound Z12 (EC50 = 52 nM) with a methyl
group at C4 position in the left phenyl ring exhibited about 2ꢀfold decreased potency,
indicating the doubleꢀbond linkage seems favorable for steric orientation of additional
hydrophobic groups into the left hydrophobic channel of NNIBP.
Furthermore, it is noteworthy that −NH2 substitution at C3 and C4 positions in
benzene ring gave the 13 and 14 nM compounds Z6 and Z7, combining the analysis
mentioned above that the C2 substituent may disturb the hydrophobic effects, pyridyl
was subsequently considered as the suitable substituent replacement for these
aminoꢀsubstituted compounds, which led to compounds Z13 (EC50 = 3 nM) and Z14
(EC50 = 6 nM), displayed 12ꢀfold and 6ꢀfold increased potency compared to the
phenylꢀsubstituted compound Z2. Especially, the activity of Z13 against HIVꢀ1 WT
was superior to those of AZT, NVP and DLV, equipotent with EFV and ETV.
Regarding the initial seven compounds Z2~8, the cytotoxicity showed 1.2 to
6.3ꢀfold decrease compared with RPV (CC50 = 3.984 ꢁM), 2.3 to 11.4ꢀfold lower than
ETV (CC50 = 2.2 ꢁM). Moreover, the cytotoxicity is strongly dependent on the nature
of the substituents in the left phenyl ring. For instance, the paraꢀsubstituent groups in
the phenyl ring influenced cytotoxicity in MTꢀ4 cells as the following order of:
−COOH (Z4: CC50 = 25.035 ꢁM) > −CF3 (Z3: CC50 = 7.353 ꢁM) > −NH2 (Z6: CC50
= 5.385 ꢁM). The cytotoxicity is also dependent on the positions of the substituents in
the phenyl ring, as for the amino group, the cytotoxicity in MTꢀ4 cells as the
following order of: ortho− (Z5: CC50 = 8.883 ꢁM) > para− (Z6: CC50 = 5.385 ꢁM) >
meta− (Z7: CC50 = 5.005 ꢁM).
Next, we turned our attention to the SAR of R2 on the right wing. It was
particularly noteworthy that the length between substitutedꢀphenyl groups and N in
piperidyl greatly influenced the activities, for instance, compounds Z15 (EC50 = 24
nM) showed more than 3ꢀfold loss in activity compared with Z19 (EC50 = 7 nM).
Moreover, all compounds with a CH2 linkage between phenyl and N in piperidyl
exhibited enhanced potency compared to Z7 except compound Z16 with a cyan group

Page 15 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
at C4 position in phenyl ring. Preliminary SAR of the paraꢀsubstituent group in the
phenyl ring indicated the following order of activity toward WT HIVꢀ1: −SO2NH2
(Z17: EC50 = 6 nM) = −CONH2 (Z18: EC50 = 6 nM) > −CN (Z16: EC50 = 269 nM).
By comparing the inhibitory activities of Z18 and Z19, it seems that the C3 or C4
position of the substituents in the right phenyl ring would be favorable to enhance the
activity. It was particularly noteworthy that when the 3ꢀaminophenyl group of Z17
was replaced with phenyl group, gave compound Z20 (EC50 = 5 nM), the activity
could be retained.
All these compounds containing substitutedꢀpiperidinamine group in the right
wing exhibited apparent cytotoxicity with CC50 values ranging from 0.198 to 1.063
ꢁM, resulting the SI values ranging from 1 to 152. In addition, the cytotoxicity is
strongly dependent on the nature of the substituents at the paraꢀposition of the phenyl
ring. For instance, Z17 (CC50 = 0.732 ꢁM) with −SO2NH2 substitution showed
2.8ꢀfold comparing to Z16 (CC50 = 0.261 ꢁM) bearing a –CN group, 3.7ꢀfold
comparing to Z18 (CC50 = 0.198 ꢁM) with a –CONH2 group. Compared to Z18, Z19
with a –CONH2 group at metaꢀposition in the phenyl ring decreased the cytotoxicity
over 3ꢀfold, suggesting that the substitutedꢀposition may greatly effect on the
cytotoxicity.
None of these compounds showed activity against HIVꢀ2 (ROD) in MTꢀ4 cells,
indicating that the newly designed DAPYs are specific for HIVꢀ1 and belong to the
typical HIVꢀ1 NNRTIs.
Table 2. Activity of Z1~20 against HIVꢀ1 mutant strains.
EC50 (µM)a
Compds
L100I
K103N
Y181C
Y188L
E138K
F227L+V106A
RES056
Z1
0.054±0.008
0.025±0.003
0.06±0.002
0.318±0.02
0.046±0.006
≥1.37
>12.116

Organic & Biomolecular Chemistry
Page 16 of 41
View Article Online
DOI: 10.1039/C7OB02828H
Z2
Z3
1.01±0.029
>7.353
0.125±0.057
1.131±0.121
0.138±0.069
0.537±0.153
0.045±0.003
0.072±0.025
0.036±0.002
0.013±0.004
0.026
0.574±0.007
>7.353
1.513±0.076
>7.353
0.237±0.022
≥1.34
>12.476
>7.353
8.202±0.469
>7.353
Z4
0.289±0.018
>8.88
0.45±0.092
1.286±0.487
0.227±0.001
0.259±0.06
0.15±0.032
0.09±0.022
0.185±0.012
0.273±0.007
0.671±0.183
0.065±0.007
1.231±0.208
0.124±0.008
>0.261
0.523±0.023
>8.88
0.244±0.011
0.8±0.082
0.101±0.069
0.135±0.016
0.124
0.296±0.108
>8.88
1.494±0.25
>8.88
Z5
Z6
0.259±0.025
0.279±0.088
≥0.169
0.504±0.074
0.762±0.071
>0.844
0.778±0.002
0.594±0.013
>0.844
>5.385
Z7
>5.005
Z8
>0.844
Z9
0.163±0.089
0.258±0.01
0.322±0.093
0.846±0.27
0.076±0.018
≥1.53
0.368±0.119
0.219±0.028
1.65±0.494
1.326±0.154
0.154±0.002
>2.068
0.046±0.005
0.05±0.004
0.235±0.016
0.307±0.041
0.022±0.012
0.138±0.004
0.146±0.008
>0.261
5.3±3.553
1.684±0.085
>18.528
1.0±0.104
>3.55
Z10
Z11
Z12
Z13
Z14
Z15
Z16
Z17
Z18
Z19
Z20
RPV
0.059±0.003
0.189±0.011
0.01±0.002
0.073±0.012
0.029±0.002
>0.261
6.433±1.243
>8.51
≥2.267
0.209±0.067
>2.068
0.935±0.143
>2.068
>1.063
0.13±0.002
>0.261
>1.063
>1.063
>0.261
>0.261
>0.261
0.14±0.023
>0.198
0.023±0.002
0.026±0.002
0.143±0.015
0.024±0.002
0.001
0.027±0.006
0.032±0.001
0.109±0.012
0.103±0.014
0.005
0.098±0.012
>0.198
0.025
0.03±0.003
>0.198
>0.732
0.029±0.003
0.027
>0.198
>0.63
0.2±0.067
0.123±0.008
0.079±0.001
≥0.163
>0.63
>0.766
0.037±0.011
0.006
0.182±0.029
0.082±0.021
>0.766
0.002
0.011±0.008

Page 17 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
AZT
NVP
EFV
ETV
DLV
0.005±0.004
0.705±0.451
0.057±0.045
0.009±0.006
3.472±3.077
0.006±0.004
3.551±2.413
0.097±0.055
0.003±0.001
2.444±1.796
0.005±0.003
3.785±2.578
0.005±0.002
0.014±0.006
1.916±1.44
0.004±0.003
4.416±2.469
0.089±0.056
0.015±0.006
0.542±0.396
0.007±0.004
0.158±0.112
0.004±0.001
0.011±0.007
0.069±0.056
0.004±0.004
3.538±2.75
0.072±0.052
0.016±0.007
2.616±2.971
0.01±0.01
2.221±0.736
0.105±0.052
0.035±0.02
>37.02
aEC50: concentration of compound required to achieve 50% protection of MTꢀ4 cell cultures against
HIVꢀ1ꢀinduced cytotoxicity, as determined by the MTT method.
The activity of the newly synthesized compounds was further evaluated against a
panel of clinical relevant HIVꢀ1 mutant strains, including L100I, K103N, Y181C,
Y188L, E138K, F227L+V106A, and RES056, in MTꢀ4 cells. The results were
summarized in Table 2. Generally speaking, near all of the compounds that displayed
potent activity against WT HIVꢀ1 strains also showed potent activity against mutant
HIVꢀ1 strains. In the case of the L100I mutant HIVꢀ1 strain, compounds Z1 and Z13
provided doubleꢀdigit nanomolar activity, Z1 exhibited potency of 54 nM, being
comparable to EFV, showing far more potent than NVP and DLV. In addition, more
than half of these compounds showed highly potent activities against K103N (the
most common mutation emerging in HIV patients treated with EFV), particularly,
Z13 (EC50 = 10 nM) was notably more active than EFV (EC50 = 97 nM). Furthermore,
compounds Z17 showed the highest potency with doubleꢀdigit nanomolar inhibitory
activity toward Y181C and Y188L with EC50 values of 27 nM, 98 nM, respectively.
Intriguingly, eight compounds displayed doubleꢀdigit inhibitory activity toward
E138K, a major mutation conferring resistance to the newꢀgeneration drug RPV.13
Among them, Z17 also showed the greatest potency (EC50 = 30 nM) against the
double mutation F227L+V106A, being slightly inferior to ETV (EC50 = 16 nM). As
for the doubleꢀmutant strain RES056, only five compounds showed activity at
micromolar levels.

Organic & Biomolecular Chemistry
Page 18 of 41
View Article Online
DOI: 10.1039/C7OB02828H
2.3. Inhibition of HIV-1 RT
To validate the binding target of these novel NNRTIs, all compounds were further
tested for their ability to inhibit recombinant WT and K103N/Y181C doubleꢀmutant
HIVꢀ1 RT enzymes; the results are shown in Table 3. In the enzymatic assay, all the
tested compounds exhibited potent inhibitory activities toward WT HIVꢀ1 RT with
IC50 values ranging from 0.013 ꢁM to 2.680 ꢁM, and most of these compounds also
showed potent inhibitory activities against doubleꢀmutant K103N/Y181C with IC50
values ranging from 0.19 ꢁM to 2.73 ꢁM, except Z14.
Table 3. Inhibitory activity of compounds against HIVꢀ1 RT (WT and K103N/Y181C double
mutant).
IC50 (µM)a
Compds
K103N/Y181C double
mutant
WT
Z1
Z2
0.024±0.018
0.049±0.008
0.071±0.004
0.013
0.876±0.010
2.235±0.098
ꢀ
Z3
Z4
0.187±0.040
ꢀ
Z5
0.131±0.011
0.049±0.003
0.016
Z6
0.990±0.403
2.496±0.041
ꢀ
Z7
Z8
0.077±0.004
0.022±0.002
0.018
Z9
0.348±0.061
2.729±0.541
2.303±1.696
2.039±1.985
1.758±0.193
210.639±10.605
1.137±0.269
ꢀ
Z10
Z11
Z12
Z13
Z14
Z15
Z16
Z17
Z18
Z19
Z20
0.029
0.021±0.003
0.034
0.036±0.003
0.212±0.050
2.680±0.773
0.186±0.004
0.182±0.009
0.238±0.023
0.126±0.011
0.589±0.235
ꢀ
ꢀ
0.734±0.270

Page 19 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
RPV
0.012±0.002
0.037±0.010
aIC50: Inhibitory concentration of test compound required to inhibit biotin deoxyuridine
triphosphate (biotinꢀdUTP) incorporation into HIVꢀ1 (WT, K103N/Y181C double mutant) RT by
50%.
Generally, compounds with the 4ꢀaminobenzonitrile group at right wing showed
an order of magnitude higher inhibitory potency than the compounds containing
substituted benzylꢀlinked piperidinꢀ4ꢀylꢀamino moieties in the WT RT inhibition
assay. Among them, Z4 with 4ꢀethynylbenzoic acid substituent in the phenyl ring
displayed the most potent activity toward WT RT with an IC50 value of 13 nM, being
comparable to that of RPV (IC50 = 12 nM), while the inhibitory activity against
doubleꢀmutant RT was inferior to RPV. Z7 also showed potent inhibitory activity
against WT RT with an IC50 value of 16 nM.
As expected, compounds with the para-substituents in the phenyl ring of left
wing exhibited the most potent activities toward WT RT by comparing the results of
Z3, Z4 and Z6, which may be explained that the paraꢀsubstituents could deeply insert
into the cylindrical hydrophobic tunnel of the binding site and possess better affinity
to RT.
It is noteworthy that the antiviral activity of the best compound in this study (Z4,
EC50 = 3 nM) was inconsistent with its enzymeꢀinhibitory potency to some extent.
The difference is considered to be due to templateꢀspecific variation in the
relationship between HIVꢀRTꢀRNA binding affinity and polymerase processivity and
have been observed in most NNRTI series.33 Besides, the differences between the two
assay systems can be attributed at least in part to differences in the physiochemical
properties (especially the intrinsic solubility).
2.4. Molecular modeling analysis
In order to better understand the SAR data and binding mode of this series of
compounds, by means of software SYBYLꢀX 2.0, the most potent compound Z4 in
the enzymatic assay was conducted further docking studies utilizing the structures of

Organic & Biomolecular Chemistry
Page 20 of 41
View Article Online
DOI: 10.1039/C7OB02828H
cocrystal of WT RT (PDB code: 2ZD1) and the K103N/Y181C doubleꢀmutant RT
(PDB: 3BGR) with the same analogue.
The docking simulations of Z4 with HIVꢀ1 WT RT indicated that the binding
mode resembles the approved drug RPV (Figure 3). In particular, Z4 showed similar
binding orientations, horseshoe conformational shape, and both formed hydrogen
bonds between pyrimidine central ring and the secondary amine linker of K101.
Despite the wellꢀsuperimposition between Z4 and RPV in molecular docking, the left
benzene ring of Z4 was shift 1.4 Å away from benzene ring of RPV (Figure 3a).
Above all, detailed computational analysis of the pharmacophoric interactions
revealed that the benzene ring of Z4 insert deeply into a cylindrical open hydrophobic
tunnel and reach the interface of protein/solvent (Figure 3b), which verified our
initial design. In particular, the left benzene ring occupied the subꢀpocket formed by
hydrophobic aromatic amino acid residues Tyr181, Tyr188, Phe227, and Trp229,
exhibiting πꢀπ interaction with these residues, moreover, the carboxyl group is
directed to the protein−solvent interface and can develop extensive interactions with
surrounding lipophilic amino acids, thereby improving the stability of the
RT−inhibitor complex. Therefore, the extensive interactions and additional hydrogen
bond involving the left wing may explain that the approximate enzyme activities
between Z4 and RPV.
When docking into the K103N/Y181C mutant RT, Z4 loses the original
Uꢀshaped conformation (Figure 3c), causing the decreased activity. However, the
distorted structure stretches into a novel channel and formed hydrogen bond with
Gln91, compensating the affinity loss with subꢀpocket in NNIBP to some extent, and
this may be the reason why Z4 possessed the most potent activity against WT and
mutant HIVꢀ1 strains.

Page 21 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
Figure 3. Predicted binding modes of Z4 with the crystal structures of HIVꢀ1 WT RT (PDB:
2ZD1) and K103N/Y181C RT (PDB: 3BGR). (a, b) superimposition of Z4 (cyan) and RPV
(orange) with HIVꢀ1 WT RT; (c, d) superimposition of Z4 (cyan) and RPV (orange) with
K103N/Y181C RT. The hydrogen bonds between the inhibitor and amino acid residues are
indicated with dashed lines (yellow) and protein carbon atoms in pink. Nonpolar hydrogen atoms
are not shown for clarity.
3. Conclusion
In summary, structureꢀbased drug design (SBDD) guided by structural
information from Xꢀray crystallography, and computational studies were successfully
applied to the discovery of novel DAPY derivatives bearing substituted alkynes or
alkenes in the left wing and substituted benzylꢀlinked piperidinꢀ4ꢀylꢀamino moiety in
the right wing as potent HIVꢀ1 NNRTIs targeting the less explored “hydrophobic
channel” of NNIBP. Encouragingly, most of these newly obtained compounds were
found to be active against HIVꢀ1 WT strain with EC50 values ranging from 3 nM to 63

Organic & Biomolecular Chemistry
Page 22 of 41
View Article Online
DOI: 10.1039/C7OB02828H
nM. Notably, two most promising compounds Z10 (EC50 = 3 nM) and Z13 (EC50 = 3
nM) showed the highest activity against WT strains and were as potent as the
secondꢀgeneration NNRTI ETV (EC50 = 3 nM). For HIVꢀ1 mutant strains, Z13
showed most potent activities against K103N (EC50 = 10 nM), E138K (EC50 = 22 nM)
and RES056 (EC50 = 0.935 ꢁM), Z17 showed doubleꢀdigit nanomolar inhibitory
activity toward Y181C, Y188L and F227L+V106A with EC50 values 27 nM, 98 nM,
30 nM, respectively. Overall, these results indicated that the left wing of DAPY
derivatives can accommodate various substituents and are worthy for further
exploring. Moreover, most compounds from Z1~14 showed lower cytotoxicity
toward the normal MT4 cells than RPV and ETV.
In addition, the inhibition of HIV RT indicated most of these compounds could
effectively inhibit the activity of HIVꢀ1 RT at low micromolar concentrations,
especially Z4 (IC50 = 13 nM) which exhibited comparable inhibitory activity to RPV
(IC50 = 12 nM). Then, the molecular docking studies of representative inhibitor Z4
with RT were performed to analyze ligandꢀreceptor interactions, which revealed that
the robust activity of these inhibitors against WT can be attributed at least in part to
the πꢀπ stacking interaction with Y188. Furthermore, detailed computational analysis
revealed that the additional left phenyl ring of Z4, inserting deeply into a cylindrical
open hydrophobic tunnel and forming extensive interactions with the interface of
protein/solvent, is critical to the ligand’s high enzyme affinity, which will direct
continued design and optimization of inhibitors to improve the drug resistance profile.
To sum up, these studies provide a novel series of promising NNRTIs for the
further development of drug candidates for treatment of HIV infection, and also
provide useful information to guide lead optimization and drug discovery by
investigating the chemical space of less explored “hydrophobic channel” of NNIBP.
4. Experimental section
4.1. Chemistry

Page 23 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
1H NMR and 13C NMR spectra were recorded on a Bruker AVꢀ400 spectrometer
using solvents as indicated (DMSOꢀ 6). Chemical shifts were reported in δ values
d
(ppm) with tetramethylsilane as the internal reference, and J values were reported in
hertz (Hz). Mass spectra were performed on a LC Autos ampler Device: Standard
G1313A instrument. Melting points (mp) were determined on a micromelting point
apparatus (Tian Jin Analytical Instrument Factory, Nankai, Tianjin, China). TLC was
performed on Silica Gel GF254 for TLC (Merck) and spots were visualized by
irradiation with UV light (λ=254 nm). Flash column chromatography was performed
on column packed with Silica Gel60 (200ꢀ300 mesh). Thin layer chromatography was
performed on preꢀcoated HUANGHAI_HSGF254, 0.15ꢀ0.2 mm TLCꢀplates. Solvents
were of reagent grade and were purified and dried by standard methods when
necessary. Concentration of the reaction solutions involved the use of rotary
evaporator at reduced pressure.
4.1.1. 4-((6-oxo-1,6-dihydropyrimidin-2-yl)amino)benzonitrile (2).
2ꢀ(Methylthio)pyrimidinꢀ4(3H)ꢀone (1) (3 g, 0.021 mol) was reacted with
4ꢀaminobenzonitrile (2.99 g, 0.025 mol) at 180 by melting method for 8h. After
cooling, CH3CN (20 mL) was added to dissolve the mixture by ultrasound treatment.
Thereafter, the reaction mixture was filtered and washed with CH3CN until no
residual 4ꢀaminobenzonitrile was detected by TLC. It was then dried in vacuo to give
the product 4ꢀ((4ꢀoxoꢀ1,4ꢀdihydropyrimidinꢀ2ꢀyl)amino)benzonitrile (2) as a light
yellow solid, yield 73.6%, ESIꢀMS: m/z 213.3 [M+H]+, C11H8N4O (212.12).
4.1.2. 4-((4-chloropyrimidin-2-yl)amino)benzonitrile (3).
A mixture of 4ꢀ((4ꢀoxoꢀ1,4ꢀdihydropyrimidinꢀ2ꢀyl)amino)benzonitrile (2) (0.80 g,
0.0038 mol) in POCl3 (5 mL) was refluxed for 0.5 h. Then, the mixture was slowly
poured into icy water (50 mL) vigorously stirring. Subsequently, the mixture was
filtered to give a filtrate cake, which was then again added into water (10 mL) and
was neutralized to pH = 7 with sodium hydroxide. The product was obtained by
filtration
and
dried
under
vacuo
to
provide

Organic & Biomolecular Chemistry
Page 24 of 41
View Article Online
DOI: 10.1039/C7OB02828H
4ꢀ((4ꢀchloropyrimidinꢀ2ꢀyl)amino)benzonitrile (3) as pale yellow solids with the yield
of 71.3%; 1H NMR (400 MHz, DMSOꢀ 6) δ 10.58 (s, 1H), 8.55 (d, = 5.2 Hz, 1H,
= 5.2 Hz, 1H, C6ꢀpyrimidineꢀH);
d
J
C6ꢀpyrimidineꢀH), 7.87 (dd, 4H, PhꢀH), 7.13 (d,
J
ESIꢀMS: m/z 231.2 [M+H]+, C11H7ClN4 (230.04).
4.1.3. General procedure for synthesis of intermediates 6a-k.
5a-k (1.0 equiv) was added to a solution of 4ꢀiodoꢀ2,6ꢀdimethylphenol (1.2 equiv),
cuprous iodide (0.12 equiv), bis(triphenylphosphine) palladium(II) dichloride (0.060
equiv) and trimethylamine (0.70 equiv) in anhydrous THF or acetonitrile. The mixture
was refluxed at 60 while stirring under nitrogen atmosphere and monitored by TLC
until its completion. After cooling, the solvent was filtered using diatomaceous earth
and then evaporated under reduce pressure. The residue was purified by flash column
chromatography using ethyl acetate/petroleum ether as eluent to obtain intermediates
6a-k.
4.1.4. General procedure for synthesis of target compounds (Z1~12).
In a schlenkꢀtype flask, 6a-k (1.0 equiv), intermediates 3 (1.0 equiv) and K2CO3
(1.2 equiv) were dissolved in DMF stirring for 10h at 100 . The products were
gained by vacuum distillation from reaction liquid. Purification on silica gel gave the
target compounds Z1~12, which were recrystallized by using acetate/petroleum ether
system.
4.1.4.1.
4-((4-(4-iodo-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzonitrile (Z1).
White crystal, purified by column chromatography using ethyl acetate/petroleum
ether (1:5) as an eluent, 81mg, yield 58%, mp: 228ꢀ230 ; 1H NMR (400 MHz,
DMSOꢀ
d
6) δ 10.15 (s, 1H, NH), 8.46 (d,
J = 5.6 Hz, 1H, pyridineꢀH), 7.94ꢀ7.19 (m,
6H, PhꢀH), 6.66 (d,
J
= 5.6 Hz, 1H, pyridineꢀH), 2.03 (s, 6H, 2CH3). 13C NMR (100
MHz, DMSO-d6) δ 168.63, 160.89, 159.56, 150.09, 144.98, 137.62, 133.80, 133.02,
118.73, 102.97, 99.38, 91.05, 40.65, 16.00. ESIꢀMS: m/z 443.4 [M+H]+. C19H15IN4O

Page 25 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
(442.03).
4.1.4.2.
4-((4-(2,6-dimethyl-4-(phenylethynyl)phenoxy)pyrimidin-2-yl)amino)benzonitrile
(Z2).
1
Light yellow powder, 38mg, yield: 53%, mp: 198ꢀ201 ; H NMR (400 MHz,
DMSOꢀ
(m, 4H, PhꢀH), 7.51 (d,
= 5.6 Hz, 1H, C5ꢀpyrimidineꢀH), 2.09 (s, 6H, 2CH3); 13C NMR (100 MHz,
d
6) δ 10.16 (s, 1H, NH), 8.48 (d,
J
= 5.6 Hz, 1H, C6ꢀpyrimidineꢀH), 7.55ꢀ7.66
J
= 8.7 Hz, 2H, PhꢀH), 7.45 (d,
J
= 5.9 Hz, 5H, PhꢀH), 6.66 (d,
J
DMSO-d6) δ 168.80, 160.89, 159.63, 150.42, 145.03, 137.64, 133.79, 133.04, 132.25,
131.78 , 129.26, 122.74, 120.17 , 119.84, 118.75, 103.00 , 99.36, 89.41, 89.26, 16.30;
ESIꢀMS: m/z 417.5 [M+1]+. C27H20N4O (416.16).
4.1.4.3.
4-((4-(2,6-dimethyl-4-((4-(trifluoromethyl)phenyl)ethynyl)phenoxy)pyrimidin-2-yl)ami
no)benzonitrile (Z3).
1
Light yellow crystal, 25mg, yield: 38%, mp: 188ꢀ191 ; H NMR (400 MHz,
DMSOꢀ
1H, C6ꢀpyrimidineꢀH), 7.76ꢀ7.95 (m, 4H, PhꢀH), 7.50ꢀ7.62 (m, 4H, PhꢀH), 7.13 (d,
= 5.6 Hz, 1H, C6ꢀpyrimidineꢀH), 6.67 (d, = 5.6 Hz, 1H, C5ꢀpyrimidineꢀH), 2.10 (s,
6H, 2CH3); 13C NMR (100 MHz, DMSO-d6) δ 168.76, 160.55, 159.62, 150.86,
145.01, 144.49, 133.55, 133.04, 132.54, 131.93, 129.27, 127.10, 126.14 (q, = 86.8
d6) δ 10.15 (s, 1H, NH), 8.55 (d, J = 5.2 Hz, 1H, PhꢀH), 8.49 (d, J = 5.6 Hz,
J
J
J
Hz), 125.79, 119.83, 119.47, 119.21, 118.74, 113.58, 103.85, 103.01, 99.37, 91.86,
87.99, 16.30; ESIꢀMS: m/z 485.6 [M+H]+, 507.5 [M+Na]+. C28H19F3N4O (484.15).
4.1.4.4.
4-((4-((2-((4-cyanophenyl)amino)pyrimidin-4-yl)oxy)-3,5-dimethylphenyl)ethynyl)ben
zoic acid (Z4).
Light yellow powder, purified by column chromatography using
methanol/methylene chloride (1:50) as an eluent, 40 mg, yield: 67%, mp: 273ꢀ281 ;

Organic & Biomolecular Chemistry
Page 26 of 41
View Article Online
DOI: 10.1039/C7OB02828H
1H NMR (400 MHz, DMSOꢀ
d
6) δ 13.10 (s, 1H, carbonylꢀH), 10.16 (s, 1H, NH), 8.49
= 5.7 Hz, 1H, C6ꢀpyrimidineꢀH), 7.99 (d, 2H, PhꢀH), 7.69 (d, 2H, PhꢀH), 7.62 (d,
= 8.5 Hz, 2H, PhꢀH), 7.48ꢀ7.53 (m, 4H, PhꢀH), 6.66 (d, = 5.6 Hz, 1H,
(d,
J
J
J
C5ꢀpyrimidineꢀH), 2.10 (s, 6H, 2CH3); 13C NMR (100 MHz, DMSO-d6) δ 168.77,
167.15, 160.90, 159.62, 150.76, 145.02, 133.04, 132.44, 131.94, 131.06, 130.06,
127.09, 119.84, 119.68, 118.74, 103.01, 99.37, 91.96, 88.73, 16.31; ESIꢀMS: m/z
459.4 [MꢀH]﹣. C28H20N4O3 (460.15).
4.1.4.5.
4-((4-(4-((2-aminophenyl)ethynyl)-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzon
itrile (Z5).
Light yellow powder, purified by column chromatography using ethyl
1
acetate/petroleum ether (1:3) as an eluent, 21mg, yield: 28%, mp: 268ꢀ270
NMR (400 MHz, DMSOꢀ 6) δ 10.16 (s, 1H, NH), 8.48 (d,
C6ꢀpyrimidineꢀH), 7.73 (s, 2H, PhꢀH), 7.67 (d, = 8.6 Hz, 2H, PhꢀH), 7.55 (d,
Hz, 1H, PhꢀH), 7.41ꢀ7.47 (m, 3H, PhꢀH), 7.11 (td, = 8.1 Hz, 1H, PhꢀH), 7.01 (td,
7.4 Hz, 1H), 6.92 (d, 1H, PhꢀH), 6.67 (d, = 5.6 Hz, 1H, C5ꢀpyrimidineꢀH), 5.29 (s,
℃
; H
d
J
= 5.6 Hz, 1H,
= 7.9
J
J
J
J =
J
2H, NH2), 2.14 (s, 6H, 2CH3); 13C NMR (100 MHz, DMSO-d6) δ 169.04, 160.76,
159.69, 149.28, 145.09, 137.67, 133.08, 131.33, 130.32, 129.15, 125.95, 121.97,
120.46, 119.83, 118.68, 111.76, 102.86, 99.43, 99.09, 16.76; ESIꢀMS: m/z 432.5
[M+H]+. C27H21N5O (431.17).
4.1.4.6.
4-((4-(4-((4-aminophenyl)ethynyl)-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzon
itrile (Z6).
Brown powder, 43mg, yield: 63%, mp: 195ꢀ198
δ 10.15 (s, 1H, NH), 8.47 (d, = 5.6 Hz, 1H, C6ꢀpyrimidineꢀH), 7.62 (d,
2H, PhꢀH), 7.51 (d, = 8.9 Hz, 2H, PhꢀH), 7.35 (s, 2H, PhꢀH), 7.22 (d, = 8.5 Hz, 2H,
PhꢀH), 6.58 (d, = 8.5 Hz, 2H, PhꢀH), 6.63 (d, = 5.7 Hz, 1H, C5ꢀpyrimidineꢀH),
5.59 (s, 2H, NH2), 2.07 (s, 6H, 2CH3); 13C NMR (100 MHz, DMSO-d6) δ 168.89,
℃
; 1H NMR (400 MHz, DMSOꢀd6
= 8.6 Hz,
)
J
J
J
J
J
J

Page 27 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
160.82, 159.64, 149.92, 149.63, 145.05, 133.04, 133.00, 131.71, 131.49, 121.37,
119.85, 118.74, 114.16, 108.67, 102.97, 99.33, 91.34, 86.35, 16.30; ESIꢀMS: m/z
432.6 [M+H]+. C27H21N5O (431.17).
4.1.4.7.
4-((4-(4-((3-aminophenyl)ethynyl)-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzon
itrile (Z7).
Light yellow powder, purified by column chromatography using
methanol/methylene chloride (1:50) as an eluent, 45mg, yield: 70%, mp: 180ꢀ183 ;
1HNMR (400 MHz, DMSOꢀ
d
6) δ 10.15 (s, 1H
= 8.5 Hz, 2H, PhꢀH), 7.51 (d,
= 7.8 Hz, 1H, PhꢀH), 6.76 (t, = 1.9 Hz, 1H, PhꢀH), 6.70
，
NH), 8.48 (d,
J = 5.7 Hz, 1H,
C6ꢀpyrimidineꢀH), 7.63 (d,
7.42 (s, 2H, PhꢀH), 7.07 (t,
J
J
= 8.8 Hz, 2H, PhꢀH),
J
J
(dt,
2H
J
= 7.5, 1.2 Hz, 1H, PhꢀH), 6.60ꢀ6.67 (m, 2H, C5ꢀpyrimidineꢀH, PhꢀH), 5.26 (s,
NH2), 2.09 (s, 6H, 2CH3); 13C NMR (100 MHz, DMSOꢀ
6) δ 168.83, 160.85,
，
d
159.64, 150.19, 149.30, 145.05, 133.05, 132.14, 131.68, 129.69, 122.96, 120.50,
119.85, 119.32, 118.75, 116.63, 115.09, 102.99, 99.37, 90.37, 87.83, 16.30; ESIꢀMS:
m/z 432.5 [M+H]+. C27H21N5O (431.17).
4.1.4.8.
N-(3-((4-((2-((4-cyanophenyl)amino)pyrimidin-4-yl)oxy)-3,5-dimethylphenyl)ethynyl)
phenyl)acetamide (Z8).
1
Light yellow powder, 36mg, yield: 41%, mp: 251ꢀ252
DMSOꢀ
C6ꢀpyrimidineꢀH), 7.91 (t,
7.47ꢀ7.57 (m, 5H, PhꢀH), 7.36 (t,
PhꢀH), 6.65 (d, = 5.6 Hz, 1H, C5ꢀpyrimidineꢀH), 2.09 (s, 6H, 2CH3), 2.07 (s, 3H,
CH3); 13C NMR (100 MHz, DMSOꢀ
6) δ 169.03, 168.81, 160.86, 159.64, 150.42,
145.03, 140.07, 133.06, 132.28, 131.77, 129.69, 126.45, 123.00, 121.84, 120.09,
℃; H NMR (400 MHz,
d
6) δ 10.14 (s, 1H, NH), 10.06 (s, 1H, NH), 8.48 (d,
= 1.8 Hz, 1H, PhꢀH), 7.63 (d, = 8.6 Hz, 2H, PhꢀH),
= 7.9 Hz, 1H, PhꢀH), 7.23 (dt, = 7.6, 1.3 Hz, 1H,
J = 5.6 Hz, 1H,
J
J
J
J
J
d

Organic & Biomolecular Chemistry
Page 28 of 41
View Article Online
DOI: 10.1039/C7OB02828H
119.84, 118.75, 103.00, 99.38, 89.40, 88.99, 24.52, 16.30; ESIꢀMS: m/z 474.4
[M+H]+, 496.5 [M+Na]+. C29H23N5O2 (473.19).
4.1.4.9.
4-((4-(4-(cyclopropylethynyl)-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzonitrile
(Z9).
Colorless crystal, purified by column chromatography using ethyl
1
acetate/petroleum ether (1:3) as an eluent, 51mg, yield: 79%, mp: 227ꢀ229 ; H
NMR (400 MHz, DMSOꢀ
C6ꢀpyrimidineꢀH), 7.62 (s, 2H, PhꢀH), 7.48ꢀ7.58 (m, 4H, PhꢀH), 6.65 (d,
1H, C5ꢀpyrimidineꢀH), 2.03 (s, 6H, 2CH3); 1.54ꢀ1.58 (m, 1H, CH), 0.75ꢀ0.92 (m, 4H,
CH2); 13C NMR (100 MHz, DMSOꢀ
6) δ 168.63, 160.89, 159.56, 150.09, 145.03,
d
6) δ 10.14 (s, 1H, NH), 8.46 (d,
J
= 5.6 Hz, 1H,
J
= 5.6 Hz,
d
137.63, 133.79, 133.02, 119.85, 118.73, 102.97, 99.38, 91.06, 16.00, 13.41, 0.52;
ESIꢀMS: m/z 381.5 [M+H]+. C24H20N4O (380.16).
4.1.4.10.
4-((4-(4-(4-hydroxybut-1-yn-1-yl)-2,6-dimethylphenoxy)pyrimidin-2-yl)amino)benzoni
trile (Z10).
1
Yellow crystal, 33mg, yield: 49%, mp: 181ꢀ183 ; H NMR (400 MHz, DMSOꢀ
d
6) δ
= 8.6 Hz, 2H,
= 5.6 Hz, 1H,
= 6.7 Hz, 2H, CH2), 2.58
= 6.9 Hz, 2H, CH2), 2.04 (s, 6H, 2CH3); 13C NMR (100MHz, DMSOꢀ
6) δ
10.13 (s, 1H, NH), 8.46 (d,
PhꢀH), 7.49 (d, = 8.8 Hz, 2H, PhꢀH), 7.27 (s, 2H, PhꢀH), 6.62 (d,
C5ꢀpyrimidineꢀH), 4.91 (t, = 5.6 Hz, 1H, ꢀOH), 3.62 (q,
(t,
J = 5.6 Hz, 1H, C6ꢀpyrimidineꢀH), 7.60 (d, J
J
J
J
J
J
d
168.83, 160.82, 159.62, 149.72, 145.03, 133.03, 132.14, 131.38, 121.12, 119.84,
118.71, 102.96, 99.32, 88.49, 80.99, 60.35, 23.76, 16.26; ESIꢀMS: m/z 385.5 [M+1]+,
402.5 [M+NH3]+. C23H20N4O2 (384.16).
4.1.4.13.
4-((4-(2,6-dimethyl-4-(pyridin-3-ylethynyl)phenoxy)pyrimidin-2-yl)amino)benzonitrile
(Z13).

Page 29 of 41
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02828H
Colorless crystal, purified by column chromatography using ethyl
1
acetate/petroleum ether (1:2) as an eluent, 48mg, yield: 89%, mp: 220ꢀ223 ; H
NMR (400 MHz, DMSOꢀ
pyridineꢀH), 8.61 (dd, = 4.9, 1.7 Hz, 1H, pyridineꢀH), 8.49 (d,
C6ꢀpyrimidineꢀH), 8.00 (dt, = 7.9, 1.9 Hz, 1H, pyridineꢀH), 7.61 (d,
PhꢀH), 7.46ꢀ7.54 (m, 5H, PhꢀH, pyridineꢀH), 6.67 (d, = 5.6 Hz, 1H,
C5ꢀpyrimidineꢀH), 2.10 (s, 6H, 2CH3); 13C NMR (100MHz, DMSOꢀ
6) δ 168.76,
d
6) δ 10.16 (s, 1H, NH), 8.78 (dd,
J
= 2.2, 0.9 Hz, 1H,
= 5.6 Hz, 1H,
= 8.6 Hz, 2H,
J
J
J
J
J
d
160.92, 159.62, 152.04, 150.76, 149.48, 145.01, 138.96, 133.05, 132.40, 131.91,
124.13, 119.90, 119.84, 119.59, 118.75, 103.01, 99.37, 92.28, 86.25, 16.31; ESIꢀMS:
m/z 418.5 [M+H]+. C26H19N5O (417.16).
4.1.4.14.
4-((4-(2,6-dimethyl-4-(pyridin-2-ylethynyl)phenoxy)pyrimidin-2-yl)amino)benzonitrile
(Z14).
Colorless crystal, purified by column chromatography using ethyl
1
acetate/petroleum ether (1:2) as an eluent, 52mg, yield: 86%, mp: 213ꢀ216 ; H
NMR (400 MHz, DMSOꢀ
d
6) δ 10.15 (s, 1H, NH), 8.63 (dd,
= 5.7 Hz, 1H, C6ꢀpyrimidineꢀH), 7.87 (td,
= 7.8, 1.1 Hz, 1H, pyridineꢀH), 7.62 (d,
J
= 4.9, 1.8, 1.0 Hz, 1H,
= 7.8, 1.8 Hz, 1H,
= 8.6 Hz, 2H, PhꢀH),
= 5.6 Hz, 1H,
C5ꢀpyrimidineꢀH), 2.11 (s, 6H, 2CH3); 13C NMR (100 MHz, DMSOꢀ
6) δ 168.76,
pyridineꢀH), 8.49 (d,
pyridineꢀH), 7.66 (dt,
J
J
J
J
7.49ꢀ7.55 (m, 4H, PhꢀH), 7.41ꢀ7.46 (m, 1H, pyridineꢀH), 6.67 (d,
J
d
160.91, 159.62, 150.91, 150.66, 145.02, 142.80, 137.28, 133.05, 132.58, 131.96,
127.75, 123.98, 119.83, 119.30, 118.73, 103.01, 99.40, 89.13, 88.34, 16.32; ESIꢀMS:
m/z 418.6 [M+H]+. C26H19N5O (417.16).
4.1.5. General procedure for synthesis of 6l-m.
5l-m (1.5 equiv) was added to
a
twoꢀneck bottle flask with
4ꢀiodoꢀ2,6ꢀdimethylphenol (1.0 equiv), Palladium acetate (0.10 equiv), Tol3P (0.3
equiv) and sodium ethoxide (1.5 equiv). After that, the solution of DMA was added
using constant pressure funnel under nitrogen atmosphere subsequently rise to 60