Virtual screening based identification of novel small-molecule inhibitors targeted to the HIV-1 capsid

Article abstract of DOI:10.1016/j.bmc.2010.11.045

The hydrophobic cavity of the C-terminal domain (CTD) of HIV-1 capsid has been recently validated as potential target for antiviral drugs by peptide-based inhibitors; however, there is no report yet of any small molecule compounds that target this hydrophobic cavity. In order to fill this gap and discover new classes of ant-HIV-1 inhibitors, we undertook a docking-based virtual screening and subsequent analog search, and medicinal chemistry approaches to identify small molecule inhibitors against this target. This article reports for the first time, to the best of our knowledge, identification of diverse classes of inhibitors that efficiently inhibited the formation of mature-like viral particles verified under electron microscope (EM) and showed potential as anti-HIV-1 agents in a viral infectivity assay against a wide range of laboratory-adapted as well as primary isolates in MT-2 cells and PBMC. In addition, the virions produced after the HIV-1 infected cells were treated with two of the most active compounds showed drastically reduced infectivity confirming the potential of these compounds as anti-HIV-1 agents. We have derived a comprehensive SAR from the antiviral data. The SAR analyses will be useful in further optimizing the leads to potential anti-HIV-1 agents.

Full text of DOI:10.1016/j.bmc.2010.11.045

Bioorganic & Medicinal Chemistry 19 (2011) 77–90
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Virtual screening based identification of novel small-molecule inhibitors
targeted to the HIV-1 capsid
Francesca Curreli a, Hongtao Zhang a, Xihui Zhang a, Ilya Pyatkin b, Zagorodnikov Victor b,
Andrea Altieri b, Asim K. Debnath a,
⇑
a Laboratory of Molecular Modeling & Drug Design, Lindsley F. Kimball Research Institute of the New York Blood Center, 310 E 67th Street, New York, NY 10065, USA
b Asinex, 20 Geroev Panfilovtzev, Building 1, Moscow 125480, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
The hydrophobic cavity of the C-terminal domain (CTD) of HIV-1 capsid has been recently validated as
potential target for antiviral drugs by peptide-based inhibitors; however, there is no report yet of any
small molecule compounds that target this hydrophobic cavity. In order to fill this gap and discover
new classes of ant-HIV-1 inhibitors, we undertook a docking-based virtual screening and subsequent
analog search, and medicinal chemistry approaches to identify small molecule inhibitors against this tar-
get. This article reports for the first time, to the best of our knowledge, identification of diverse classes of
inhibitors that efficiently inhibited the formation of mature-like viral particles verified under electron
microscope (EM) and showed potential as anti-HIV-1 agents in a viral infectivity assay against a wide
range of laboratory-adapted as well as primary isolates in MT-2 cells and PBMC. In addition, the virions
produced after the HIV-1 infected cells were treated with two of the most active compounds showed
drastically reduced infectivity confirming the potential of these compounds as anti-HIV-1 agents. We
have derived a comprehensive SAR from the antiviral data. The SAR analyses will be useful in further opti-
mizing the leads to potential anti-HIV-1 agents.
Received 30 September 2010
Revised 16 November 2010
Accepted 22 November 2010
Available online 25 November 2010
Keywords:
Human immunodeficiency virus type-1
(HIV-1)
N-Terminal domain (NTD)
C-Terminal domain (CTD)
Capsid (CA)
Peripheral blood mononuclear cell (PBMC)
[(Sodium 30-(1-(phenylamino)-carbonyl)-
3,4-tetrazolium-bis(4-methoxy-6-nitro)
bezenesulfonic acid hydrate)] (XTT)
Structure–activity relationship (SAR)
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
been modeled based on structural studies and image reconstruc-
tions by cryo-electron microscopy of mature virions and assem-
The HIV-1 genome is composed of three major genes, gag, pol,
and env. The gag gene encodes the Gag protein, the critical struc-
tural protein of HIV-1, whereas pol encodes the viral enzymes re-
verse transcriptase (RT), protease (PR), and integrase (IN), essential
for the HIV-1 replication. During HIV-1 assembly and morphogen-
esis, the structural protein, Gag, organizes into two completely
different arrangements, immature, and mature forms. In the imma-
ture form, Gag remains intact, whereas in the mature form Gag
is cleaved by the viral protease. The formation of this mature par-
ticle is essential for HIV-1 infectivity and the capsid protein ob-
tained from the Gag cleavage product plays a central role in
forming the conical core of the virus that surrounds the viral gen-
ome. The capsid protein (CA, p24) is a hydrophobic protein con-
sisting of two domains, the N-terminal domain (NTD, aa 1–145)
and the C-terminal domain (CTD, aa 146–231). These two domains
are connected with a 5-amino acid linker. Although the exact nat-
ure of the CA–CA contacts and their interactions in immature par-
ticles are not fully known, in mature particles the CA lattice has
bled VLPs.1–4 Recently, a pseudo-atomic model of the full-length
HIV-1 capsid hexameric structure has been reported, which pro-
vided structural insights on the mechanism of action of some
known assembly inhibitors.5 HIV-1 capsid plays a crucial role in
viral assembly, maturation, and also early post-entry steps.1–4,6
Mutations in both the NTD and CTD lead to defects in viral assem-
bly and release.7–18 In addition, capsid has been shown to be a
dominant determinant of HIV-1 infectivity in non-dividing cells.19
Taken together, it is evident that capsid plays an important role in
HIV-1 assembly and maturation, and has been recognized as a po-
tential target for developing a new generation of drugs for AIDS
therapy.20–22
Several peptides and small molecules that disrupt HIV-1
assembly have been reported in the literature.23–28 The first break-
through in identifying small-molecule inhibitors (CAP-1 and
CAP-2) targeted to the NTD of capsid was reported by Summers’s
group.26 These compounds affect virus stability and/or assembly.
Only CAP-1 showed dose-dependent HIV-1 inhibition in viral infec-
tivity assays, while CAP-2 was cytotoxic. Although the affinity of
CAP-1 for the NTD was low (Kd ꢀ800
lM) it serves as a lead com-
pound for more potent inhibitors against this target. Recently, the
⇑
Corresponding author. Tel.: +1 (212) 570 3373; fax: +1 (212) 570 3168.
E-mail address: adebnath@nybloodcenter.org (A.K. Debnath).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.11.045

78
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
structure of CAP-1 in complex with the NTD has been solved.29
A
1. A further search of analogs of these two hits from commercial
sources resulted in additional compounds. Then we initiated
medicinal chemistry on one of the active hits, 6 to derive a compre-
hensive SAR.
small-molecule inhibitor, PA-457, which blocks HIV-1 maturation
by inhibiting CA-SP1 cleavage, has been recently reported.28,30–35
PA-457 has shown low nM antiviral potency in cell culture, provid-
ing credence to the hypothesis that highly potent inhibitors to Gag
can be identified despite the fact that literally thousands of copies
of Gag are present in a virion. This compound is currently in Phase
2b clinical trials (http://www.myriadpharma.com/product-pipeline/
clinical/mpc-4326). Recently, a 12-mer peptide, CAI, was identified
by phage-display technique and shown to inhibit HIV-1 assembly
in vitro by targeting the CTD.36 CAI was the first peptide reported
to disrupt both immature- and mature-like particles in vitro. How-
ever, CAI could not inhibit HIV-1 in cell culture due to its lack of
cell permeability and hence could not consider as a viable drug
candidate.
2.1.1. Compounds 6 and 50 both disrupt the formation of
mature-like particles
We have characterized these compounds for their inhibitory ef-
fect on viral assembly and maturation. Spherical immature-like
particles were formed by using full length gag proteins. The effect
of inhibitors on immature- and mature-like particle assembly was
studied by performing assembly reactions in the presence of varied
doses of 6 and 50. These compounds failed to disrupt immature-
like particle formation even at molar equivalent dose (data not
shown). Tube-shaped mature-like particles were obtained from
purified CA as reported.3,42 The rationale behind using CA instead
of CANC to form the mature-like particles was to confirm that
6 and 50 target CA only.39 In vitro data indicate that 6 and 50 dis-
rupt the formation of mature-like particles (Fig. 2) in a dose-depen-
dent manner as observed with the hydrocarbon-stapled peptide,
NYAD-139, which was used as positive control. After incubation
with 0.5-fold molar equivalents of 6, the integrity of tube-like par-
ticles was greatly damaged. After incubation with one or five molar
equivalent of 6, the assembly of tube-shaped particles was com-
pletely blocked. In case of 50, three molar equivalents completely
disrupted the formation of tube-shaped particles.
Recently, we successfully applied the hydrocarbon stapling tech-
nique37 to convertCAI to a cell-penetrating peptide (NYAD-1), which
binds to the CTD. The stapling technique involves the introduction of
two a,a-disubstituted non-natural amino acids containing terminal
olefinic side chains at positions i and (i + 4) or (i + 7) and cyclized
using ruthenium catalyzed metathesis reaction to create the stapled
peptides. NYAD-1 disrupts the formation of both immature- and
mature-like virus particles in cell-free and cell-based in vitro assem-
bly systems. In addition, NYAD-1 displays potent anti-HIV-1 activity
in cell culture against a range of laboratory-adapted and primary
HIV-1 isolates. It binds to a hydrophobic cavity identified previously
in X-ray studies of CTD complexed with CAI38 with an improved
affinity (Kd ꢀ1
l
M) compared to CAI (Kd ꢀ15
l
M).39 In this study,
2.2. Antiviral activity and cytotoxicity of small molecule
compounds in cell-based assays
we have exploited the same hydrophobic cavity as a target to iden-
tify organic small molecule inhibitors using a docking-based virtual
screening approach followed by synthetic medicinal chemistry. We
haveidentified several small moleculeinhibitorswhichmayserve as
effective leads against this target for further optimization.
Since both 6 and 50 showed inhibition of mature-like particle
assembly in cell-free assay we decided to test all similar com-
pounds identified from database against a laboratory-adapted
strain HIV-1 IIIB in MT-2 cells. The inhibition of p24 production
by these molecules was measured over a range of concentrations
to determine the IC50 values (concentration of inhibitor required
to produce 50% reduction in p24 production). The results in Table 1
showed that the analogs of 6 showed very good activity ranging
2. Results
2.1. Virtual screening
The binding site of the first reported peptide, CAI, was mapped
in an NMR chemical shift perturbation study to a highly conserved
hydrophobic cavity formed by residues 169–191 of CTD of HIV-1
capsid.36,38 This was also confirmed by a high-resolution x-ray
structure analysis of the CAI structure in complex with the CTD38
(PDB code: 2buo), which revealed the molecular details of the
interaction of this peptide in the hydrophobic cavity. Our recent
NMR structure of the next generation peptide-based inhibitor,
NYAD-1, also demonstrated that it bound to the same region of
the CTD (aa 169–192).40 These studies have provided us with a
wealth of structural information on the binding mode of these
inhibitors in the hydrophobic cavity and surrounding area. We
wanted to exploit the same hydrophobic cavity to identify organic
small molecules that might disrupt the formation and maturation
of virus particles and inhibit virus infection. We have screened a
subset of 100,000 drug-like molecules from the ZINC41 database
using virtual screening based on high-throughput flexible docking
technique, which is becoming an emerging technology for struc-
ture-based rational lead discovery. We have selected 200 top-
scored compounds for visual analysis using 3D stereoscopic glasses
to identify compounds, which had the best possible interactions at
the hydrophobic cavity of the CTD and selected 50 compounds for
experiments. We have initially ordered 8 compounds from a ven-
dor and tested their antiviral activity against HIV-1 IIIB. Two of
the 8 compounds (6 and 50) belonging to two diverse scaffolds
showed antiviral potency and one compound (6) was potent and
had a selectivity index (SI) of >60 (SI = CC50/IC50; Table 3). Struc-
tures of selected few active compounds have been shown in Figure
from 1.06 to 12.4 lM. Similarly, Table 2 indicated the activity
Figure 1. Chemical structures of selected inhibitors identified by docking-based
virtual screening and similarity search techniques.

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
79
Figure 2. Inhibition of in vitro assembly by 6 and 50. Negatively stained EM images of mature-like particles resulting from in vitro assembly of CA proteins in the presence of
no inhibitors (control), 0.5-, 1.0-M equiv of NYAD-1 (positive control), 6 and 50. Two higher doses were used in case of NYAD-1 (5.0-M equiv) and for 50 (3.0-M equiv) to
reach complete inhibition.
ranges of the 50 type analogs tested. Although most of the com-
pounds showed low to moderate activity two compounds (50
and 55) showed very good activity profile.
produce 50% cytotoxicity) values for MT-2 and PBMC were pre-
sented in Tables 1–3.
The encouraging data in the primary antiviral screening along
with inhibition of mature-like particles by 6 and 50 motivated us
to select these compounds to assess their antiviral activities in a
wide variety of laboratory-adapted belonging to subtype B but
different co-receptor use (X4, R5 and R5X4). The activities were
assessed in two cell lines (MT-2 and PBMC). All three compounds
2.3. Virions released from HIV-1 infected cells after treatment
with the inhibitors are less infectious
Antiviral assays have shown a decrease in HIV-1 viral particle
release by small molecules represented by reduction in p24 pro-
duction (Tables 1–3). In addition, 6 and 50 effectively inhibited
mature-like particles and prevented assembly in vitro ( Fig. 3).
Therefore, we reason that the effect of these molecules on newly
generated virions is much greater than can be envisioned by the
antiviral data. In other words, we believe that these molecules
may possibly interfere with viral uncoating and viral assembly/
maturation inducing irreparable damage to the viral particles new-
ly produced and released in the supernatant, therefore, making
them less infectious. To support our hypothesis we evaluated not
only the quantity but also the quality of the virus progenies re-
leased in the supernatant using two of the most active compounds,
6 and 55. We first quantified the p24/viral particles released in the
supernatant following AMD3100 (positive control), 6 and/or 55
treatment then we treated cells with the identical amount of
p24-normalized supernatants. However, it is pertinent to note that
at higher doses of AMD3100, 6 and 55 we could detect very little to
no p24 in the supernatants. Therefore, we used the maximum
quantities of supernatants possible to infect MT-2 cells for measur-
ing the infectivity of the virus progenies released in the superna-
tant. As expected, we detected no viral infectivity for samples
showed excellent activity profiles (IC50 ꢀ1.06 to 7.09
lM; Table 3).
In addition, we also tested these compounds against one RT-resis-
tant and one protease-resistant strain. Remarkably, they all
showed very similar activities as laboratory strains indicating their
possible potential for further optimization as potential anti-HIV-1
agents with broad-spectrum activity.
We also tested these three molecules against a panel of primary
isolates in PBMC representing diverse subtypes from A to O and di-
verse coreceptor use. The compounds showed excellent activity in
all the primary isolates tested (0.73–5.46
lM) except group O
(6.07–10.95 M). Compounds 50 and 55 showed potent activity
l
against one primary isolate (92UG037 under subtype A). All three
compounds showed broad-spectrum activity (Table 3), which is
very encouraging. This provides additional confidence that some
of these lead molecules will be amenable for further modification
to increase potency and selectivity.
The cytotoxicity of these compounds was assessed by the XTT
method in either MT-2 cells or in both MT-2 cells and PBMC. Cyto-
toxicity assays were performed in parallel with the HIV-1 inhibi-
tion assays. The CC50 (concentration of inhibitor required to
treated with AMD3100 and with 5 lM of 6 and 4 lM of 55

80
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
Table 1
Antiviral activity of 1 and its analogs against HIV-1 IIIB in MT-2 cells
R5
H
N
R4
R3
O
N
O
R1
R2
No.
R1
R2
R3
R4
R5
lM ( SD)
IC50
CC50
∗
∗
∗
∗
O
O
O
1
2
H
H
H
H
CH3
CH3
4.02 0.28
>67.13
∗
∗
CH3
F
4.23 0.39
>64.7
3
4
H
H
H
H
CH3
CH3
7.25 0.54
>16.5
>64
∗
∗
∗
O
O
>66.1
S
∗
5
H
H
CH3
2.67 0.5
9.2 1.0
CH3
∗
∗
O
6
7
8
CH3
CH3
CH3
H
H
H
H
H
H
1.06 0.05
>61.45
ND
Cl
∗
∗
∗
O
O
>16.7
N
∗
2.16 0.07
>55.4
Br
∗
∗
O
9
CH3
CH3
H
H
H
H
9.28 0.68
>27.5
>67.1
ND
∗
∗
O
O
10
CF3
∗
N
11
12
CH3
H
H
H
H
>24.5
>25.6
ND
ND
N
CH3
∗
∗
O
CH3
F

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
81
Table 1 (continued)
No.
R1
R2
R3
R4
R5
lM ( SD)
IC50
CC50
∗
∗
O
O
13
CH3
H
H
9.2 0.7
70.7 5.4
Cl
Cl
∗
∗
14
15
CH3
H
H
H
H
>23.6
ND
OH
Cl
∗
∗
∗
O
O
CH3
7.1 0.4
>20.4
N
Cl
∗
16
CH3
H
H
>25.2
ND
N
∗
∗
O
17
18
CH3
H
H
H
H
8.2 1.2
>23.5
ND
Cl
F
∗
∗
∗
O
O
CH3
>21.9
Cl
Cl
OH
∗
19
20
CH3
H
H
H
H
>24.8
>24.1
ND
ND
O
CH3
∗
∗
∗
O
O
CH3
N
H3C
CH3
∗
21
22
23
CH3
CH3
CH3
H
H
H
H
H
H
>23.6
ND
Cl
OH
∗
∗
O
O
N
N
>26.6
ND
∗
∗
4.7 0.6
>24.5
F
F
(continued on next page)

82
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
Table 1 (continued)
No.
R1
R2
R3
R4
R5
lM ( SD)
IC50
CC50
ND
∗
∗
O
O
24
CH3
H
H
>22.1
H2N
O
∗
S
O
∗
25
26
CH3
H
H
H
H
>38.9
ND
Cl
N
∗
CH3
9.27 0.3
44.5 5.5
N
∗
Cl
∗
∗
O
27
28
C2H5
C2H5
H
H
H
H
>13.9
ND
∗
∗
∗
∗
O
O
12.3 0.85
>127.4
S
29
OCH3
H
H
>13
ND
Br
∗
∗
O
30
31
OCH3
OCH3
H
H
H
H
7.87 1.0
>31
ND
CH3
∗
∗
∗
O
S
>25.7
∗
∗
32
33
34
CH3
CH3
CH3
H
H
H
H
H
H
11.7 1.2
11.8 0.4
>12.9
>63.4
>64
∗
∗
S
S
S
∗
ND
∗
∗
S
35
36
CH3
H
H
H
5.0 0.2
12.7 1.0
O
CH3
∗
∗
S
CH3
CH3
CH3
8.6 1.1
>96

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
83
Table 1 (continued)
No.
R1
R2
R3
R4
R5
lM ( SD)
IC50
CC50
∗
∗
∗
S
37
38
OCH3
OCH3
H
H
H
H
7.9 0.4
25 1.0
CH3
∗
∗
S
6.1 0.7
>48
ND
∗
O
39
40
H
H
CH3
H
H
>14.6
CH3
CH3
H3C
∗
∗
O
CH3
>15
ND
O
H3C
∗
∗
∗
O
O
41
42
H
H
CH3
CH3
H
H
>16.8
ND
∗
∗
12.4 0.6
>51.2
F
∗
∗
O
43
44
H
H
CH3
H
>16.2
ND
CH3
∗
H
CH3
3.59 0.36
>60
Cl
∗
∗
∗
∗
O
O
45
46
CH3
CH3
H
H
H
H
3.09 0.55
6.51 0.94
>67.13
>65
CH3
∗
∗
O
47
48
CH3
CH3
H
H
H
H
6.4 0.17
5.8 0.36
>19.8
>117
CH3
∗
∗
O
O
ND, not determined.
(Fig. 3). We calculated the TCID50 considering only those doses
which showed measurable infectivity. Infectivity efficiency of virus
progenies was expressed as percentage of infectivity with respect
to the untreated control. We detected about 60% of infectivity in
fore, these molecules interfere with viral assembly and/or
maturation.
2.4. Structure–activity relationship (SAR) analysis
the virus produced by the cells treated with 2.5
M of 55. Taken together these results suggest that 6 and 55-
treated virus-producing cells yield less infectious viruses; there-
lM of 6 and
2
l
Molecular docking-based virtual screening identified two
compounds (6, Table 1) and (50, Table 2) with diverse chemical

84
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
Table 2
23, 35, 44, and 45); however, the activity was dictated in parallel
by the substitutions in the phenyl ring as well as in the quinolinone
ring. In terms of phenyl ring substituents, 4-Cl, 4-Br and 4-CH3
showed better activity; however, phenyl group with a bulkier
group attached to it such as 4-iPr, 4-OCH3 (39 and 40) showed re-
duced activity. Surprisingly, two compound with furan at R1 and a
3-piperidinyl-4-chloro-phenyl group at R2 and a 4-Cl phenyl at R1
and a bulkier group imidazo[1,2-a]pyridine showed reasonable
activity; however, these compounds also showed enhanced toxic-
ity. The quinolinone ring seems very critical for antiviral activity
as evidenced in the docked pose (Figure 4) where –NH-moiety of
quinolinone ring forms a hydrogen bond with Asn183 located at
the hydrophobic cavity of CTD. Similarly, the methyl group at posi-
tion R3 is also important, which is positioned near a narrow hydro-
phobic pocket. The SAR analysis also confirms that a bulkier group
such as OCH3 (29–31, 37 and 38) and C2H5 (27 and 28) showed
much reduced activity. It appears that with furyl substituent at
R1 and phenyl (and its analogs) at R2, CH3 at position R5 showed
equally potent activity (1–3, 5, 36, and 44).
The SAR analyses of compounds in Table 2 revealed that a rela-
tively hydrophobic bulkier group is necessary for activity (50, 51,
54, and 55). However, relatively weak activity of 49 and 52 could
not be explained based on this analysis. Compound 55, which
shows excellent antiviral potency, has the most bulker R group
and the end portion is very hydrophobic. It is interesting to note
that in the Glide-based docking simulation, the biphenyl moieties
of the best pose of 55 make substantial hydrophobic contact with
the hydrophobic residues in the cavity (Val165, Thr186 and
Antiviral activity of compound 49 and its analogs against HIV-1 IIIB in MT-2 cells
N
OH
R
CH3
N
H
N
O
New No.
R
lM ( SD)
IC50
CC50
49
50
33.5a
>134.41
NH
∗
F
NH
∗
6.69 0.3
6.8 0.25
>128.08
<7.6
CH3
H3C
NH
51
∗
H3C
52
53
33.23a
30.0a
42.10a
∗
HN
>34.15b
O
N
O
∗
Cl
54
55
6.9 0.7
24.04a
Leu211) and also make a p–p interaction with the amino acid res-
idue Y169 (Fig. 5). These interactions are missing in other relatively
weak active molecules.
N
∗
N
3. Discussion and conclusions
1.78 0.17
>37.50b
N
∗
N
The HIV-1 capsid has been identified as a potential target for
discovering new inhibitors to fight against AIDS.28,34,36,38,43 Several
groups reported identification of small molecule inhibitors against
capsid.28,34,34,43,44 However, the target of the most active com-
pound, PA-457 (now termed as MPC-4326), which is now in Phase
2b clinical trials (http://www.myriadpharma.com/product-pipe-
line/clinical/mpc-4326), is the CA-SP1 cleavage site. Summers’
group used a computational based docking technique (DOCK 4.0)
to screen public domain libraries using a ‘b-hairpin cleft’ on the
surface of the N-terminal domain of the HIV-1 capsid and identi-
fied two compounds, CAP-1 and CAP-2. CAP-1 showed virtually
no inhibition of virus production; however, the particles produced
after treatment was poorly infectious and CAP-2 was toxic.26,29
Since then there was no report of identification of any small mol-
ecule inhibitors against HIV capsid, especially targeted to the
CTD of capsid.
In order to fill this gap and to identify new classes of anti-HIV-1
agents, we have used a similar docking-based virtual screening
technique used by Summer’s group; however, instead of NTD of
HIV-1 capsid, we selected the hydrophobic cavity within the CTD
of HIV-1 capsid, which had been shown to be large enough to
accommodate a 12 residue peptide (CAI) and it’s subsequent mod-
ified stapled peptide, NYAD-1.36,38–40 The screening method identi-
fied several compounds belonging to two diverse structural classes
HO
56
57
58
>30.6b
>32b
>30.60a
>63.7a
>66.1a
N
∗
N
N
∗
H
H3C
H3C
∗
>16.5b
N
a
50% antiviral activity or toxicity at this dose.
0–30% antiviral activity or toxicity at this dose.
b
structures, which showed consistent antiviral activity against a
broad-range of laboratory-adapted and primary isolates. We fur-
ther searched several commercial databases and identified similar
analogs, synthesized a series of analogs of 6 and tested them
against HIV-1 IIIB. The inhibitory data indicated a trend in SAR,
as described below, which may be used as a guiding tool to opti-
mize these two classes of inhibitors.
The SAR analyses of 6 and its analogs (Table 1) reveal that R1
and R2 positions require hydrophobic moieties (6, 8, and 45–47).
Thiophene (32–38) containing compounds, although an isostere
of furan, showed somewhat reduced activity profile. It appears that
R1 position can also accommodate groups like phenyl (44 and 45),
however, hydrophilic moieties such as pyridine (7), piperazine (11)
and pyrazole (22) in R2 position showed detrimental effect on anti-
viral activity. The R2 position showed trend towards more elon-
gated, bulkier hydrophobic group, such as phenyl (1, 2, 5, 6, 8,
(6 and 50) and inhibited HIV-1 IIIB in low lM potency in an anti-
viral assay. A subsequent similarity based search on different
databases further identified a large number of compounds repre-
senting those two diverse structures, many of which also showed
low
much better antiviral activity (IC50 ꢀ1.06 to 2.16
or CAP-2. We have tested 6, 50 and 55 against a wide variety of
lM potency in an antiviral assay. These compounds showed
lM) than CAP-1

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
85
Table 3
Antiviral activity of 6, 50, and 55 in laboratory-adapted and primary HIV-1 isolates
HIV-1 virus
Subtype
Cell type
Coreceptor
IC50
50
(l
M
SD)x
6
55
Laboratory strains
IIIB
MN
SF2
RF
B
B
B
B
B
B
MT-2
MT-2
MT-2
MT-2
PBMC
PBMC
X4
X4
R5X4
X4
R5
1.06 0.05
2.88 0.3
2.43 0.37
1.64 0.05
4.22 0.78
6.17 0.5
6.69 0.3
7.09 0.72
5.72 1.74
3.37 0.18
4.9 0.57
6.6 0.6
1.78 0.17
6.01 0.32
3.59 0.25
4.01 0.24
3.22 0.7
BaL
89.6
R5X4
5
1.2
RT-resistant isolate
AZT-R
B
B
MT-2
MT-2
X4
X4
5.77 0.55
1.18 0.03
3.97 0.67
9.05 1.13
3.15 0.1
5.7 0.6
Protease-resistant isolate
HIV-1RF/L-323-12-3
Primary isolates
94UG103
92UG037
92US657
93IN101
92UG024
93BR020
RU570
A
A
B
C
D
F
G
PBMC
PBMC
PBMC
PBMC
PBMC
PBMC
PBMC
PBMC
R5X4
R5
R5
R5
X4
R5X4
R5
R5
3.93 0.25
-
3.6 0.93
1.12 0.2
2.7 0.22
3.36 0.21
4.9 1.01
1.93 0.12
3.9 0.2
1.24 0.08
0.73 0.13
4.57 1.55
2.87 0.14
5.46 0.4
2.7 0.48
2.57 0.45
4.18 1.93
2.84 0.28
6.07 0.21
3
5
0.2
0.48
2.74 0.42
8.95 0.9
BCF02
Group (0)
10.95 2.51
Small molecules
CC50
(lM) in MT-2
CC50
(l
M) in PBMC
6
50
55
>61.45
>128.08
>37.6
>61.45
>64
>37.6
Figure 3. The percentage of infectivity of the virions in the supernatant. p24
normalized viruses produced by HIV-1 IIIB infected MT-2 cells treated with
different doses of 6 and 55 or AMD3100 (1
calculate TCID50
lg/mL) were titered on MT-2 cells to
.
primary isolates representing different subtypes and coreceptor
use and they showed remarkable activity.
Figure 4. GLIDE docking pose of 6 in the hydrophobic cavity of the CTD of HIV-1
capsid. The quinolinone NH forms a H-bond with Asn183 indicated in green line.
The phenyl and furan group are located in the hydrophobic grove formed by
residues Val165, Tyr169, Leu211, and Met215.
Despite their potential HIV-1 inhibitory activity, we wanted to
confirm that these compounds target capsid and inhibit mature-
like particles. Indeed 6 and 50 have been shown to be effective
in preventing mature like particles as was previously shown by a
stapled peptide, NYAD-1, which was shown to bind to the hydro-
phobic cavity of the CTD by NMR. This data give confidence that
these inhibitors have broad anti-HIV-1 activity with promising
selectivity index and can be used for further optimization.
A comprehensive SAR analysis of the 6 and 50 type analogs pro-
vided useful information for further optimizing this set of mole-
cules. The docking-based analysis of the poses of these two
compounds also provided additional information on the critical
interaction necessary for the binding of these compounds to the
hydrophobic cavity of the CTD of capsid. Future systematic synthe-
sis effort will be geared to get a better understanding on that
series.
The data presented in this report provides confidence that the
hydrophobic cavity of the CTD of HIV-1 capsid is a potential target
to identify antiviral agents and these preliminary data and the SAR
study will be the basis of further optimization of these series of
inhibitors.
4. Materials and methods
4.1. Molecular modeling and virtual screening
All molecular modeling and virtual screening studies were per-
formed on a HP xw9300 workstation running Red Hat Enterprise

86
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
4.1.2. Docking of ligands onto the hydrophobic cavity of the
capsid CTD
We used the automated docking software GLIDE 8.0 (Schrödinger,
Portland, OR) that applies a two-stage scoring process to sort out
the best conformations and orientations of the ligand (defined as
pose) based on its interaction pattern with the receptor. GLIDE has
been used successfully in drug design.47–53 The starting point of
the docking simulation was the X-ray structure of the CTD of
HIV-1 capsid bound with a linear peptide CAI (PDB code: 2buo)
available from protein data bank (PDB) at the Research Collabora-
tory for Structural Bioinformatics (RCSB) which has been exten-
sively used in our research program. We decided to use this
specific structure since this is the only ligand (CAI)-bound X-ray
structure available for the CTD and provides information on the
flexibility of the binding site for ligand binding. The peptide CAI
was removed from the protein for docking study. Three-dimen-
sional coordinates of the ligands, their isomeric, ionization and
tautomeric states were generated using the LigPrep (including Ion-
izer) module within the Schrödinger Suite 2008 programs. The pro-
tein was prepared using the ‘protein preparation tool’ and the
structures were minimized with Macromodel software from Schrö-
dinger. A grid file encompassing the area in the cavity that contains
information on the properties of the associated receptor was cre-
ated. Conformational flexibility of the ligands was handled via an
exhaustive conformational search. Initially, we used Schrödinger’s
proprietary GlideScore scoring function in standard precision (SP)
mode. We selected 2000 top-scored compounds to dock again in
extra precision (XP) mode to score the optimized poses. For each
ligand 50 different poses were evaluated and 200 top-scored li-
gands were examined carefully using molecular visualization tech-
niques with 3D CrystalEyes2 stereo eyewear from SteroGraphics
Corporation (San Rafael, CA). The poses were selected based on
hydrophobic, hydrogen bond and charge–charge interactions. Spe-
cial attention was also paid to identify poses that may indicate
alternative binding modes of the molecules in the hydrophobic
cavity. Interaction sites of the active ligands with the CTD from
dock poses were mapped using LigandScout 2.0 (Inte:Ligand,
Austria).
Figure 5. GLIDE docking pose of 55 in the hydrophobic cavity of the CTD of HIV-1
capsid. Two phenyl groups of this molecule are positioned in a hydrophobic group
formed by residues Val165, Thr186, Leu211, and Tyr 169. The phenyl moiety of
Tyr169 forms
in red arrow).
p–p interaction with one of the phenyl ring of this molecule (shown
Linux 4.0 and equipped with AMD Opteron dual processors and
NVIDIA Quatro FX3400 video card.
4.1.1. Selection of chemical databases
We selected ZINC database41 for the docking-based virtual
screening study. This database contains ꢀ4.6 million pre-filtered
compounds in a ready-to-dock stage in 3D format, which can be
downloaded free of cost from Brian Shoichet’s lab at the University
of California at San Francisco (UCSF). The database contains a di-
verse set of compounds (diversity was assessed based on Tanimoto
similarity coefficients) from a large number of vendors as well as
compounds from PubChem collections, a component of NIH’s
Molecular Library Roadmap Initiative. The database was created
by filtering out molecules with formula weight >700, calculated
Log P >6 and <ꢁ4, the number of hydrogen-bond donors >6, the
number of hydrogen bond acceptors >11, and the number of rotat-
able bonds >15. The database contains only molecules having H, C,
N, O, F, S, P, Cl, Br, or I atoms. However, exceptions were made by
including a number of actual drugs that violate these constraints.
The constraints set in the filter are based on the current opinion
in the field and is a little more lenient than prescribed by the
Lipinski’s ‘rule of five’.45,46 We have used a subset of about
100,000 drug-like molecules in our study.
4.2. Chemistry
All reagents were purchased from commercial suppliers and
were used without further purification. Target compounds 1–11,
27–58 were purchased from commercial sources (Asinex Corp.,
USA and Chembridge Corp., USA). Compounds 12–26 were synthe-
sized according to the methods described below. The melting
points were measured on a Sanyo Gallenkamp melting point
Scheme 1. Reagents: (a) Ac2O, TEA, MeCN; (b) POCl3, DMF; (c) HCl, AcOH; (d) R1–CH2–NH2, isopropanol then NaBH4; (e). R2COOH, TBTU, DCM.

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
87
apparatus (Sanyo Gallenkamp, UK). The LC/MS analyses for the
compounds were done at Surveyor MSQ (Thermo Finnigan, USA)
with APCI ionization. The 1H NMR spectra were recorded on MER-
CURY plus 400 MHz spectrometer (Varian, Germany). Chemical
shift values are given in ppm relative to tetramethylsilane (TMS),
with the residual solvent proton resonance as internal standard.
High-resolution mass spectra (HRMS) were recorded on a Bruker
APEX II ICR-FTMS.
After we have performed a retro synthetic analysis of 6 the
focus was on the synthetic Scheme 1 which was validated and
allowed us to further explore the chemical space of these quino-
lin-2-one derivatives. We synthesized S2–S4 by minor modifica-
tion of the published methods54,55 as described below. The
secondary amines, S5–S7, were synthesized by a reductive amina-
tion of the aldehydes S4.56 Then these three secondary amines
were subjected to a TBTU coupling reaction with appropriate acids
leading to the desired products, 12–26, with yields ranging from
12% to 78%.
idue. The mixture was stirred for 1 h, the precipitate was filtered,
washed with water (1 ꢂ 10 mL) and dried on the air to give the de-
sired target secondary amines.
4.2.4.1. 3-{[(Furan-2-ylmethyl)-amino]-methyl}-6-methyl-1H-
quinolin-2-one (S5). Compound S5 (119 mg, 72%, mp 129–
130 °C) was synthesized according to the general method described
above. 1 H NMR (400 MHz, DMSO-d6 ppm): 2.31 (s, 3H), 3.61 (s,
2H), 3.71 (s, 2H), 6.21 (s, 1H), 6.32 (s, 1H), 7.15 (d, 1H), 7.25 (s, 1H),
7.41 (s, 1H), 7.52 (s, 1H), 7.71 (s, 1H); m/z (APCI+) 269 (M+H+).
4.2.4.2.
6-Methyl-3-{[(tetrahydro-furan-2-ylmethyl)-amino]-
methyl}-1H-quinolin-2-one (S6). Compound S6 (290 mg, 50%,
mp110–111 °C) was synthesized according to the general procedure
described above starting from tetrahdrofuran-2-yl-methylamine
(315 mg, 3.11 mmol). 1H NMR (400 MHz, DMSO-d6 ppm): 1.50 (m,
1H), 1.86 (m, 3H), 2.35 (s, 3H), 2.55 (d, 2H), 3.58 (m, 3H), 3.70 (q,
1H), 3.89 (q, 1H), 7.18 (d, 1H), 7.26 (d, 1H), 7.41 (s, 1H), 7.72 (s, 1H),
11.60 (br, 1H); m/z (APCI+) 273 (M+H+).
4.2.1. N-Acetyl-p-toluidine (S2)54,55
TEA (24.8 g, 0.25 mol) was added at once to a solution of 25.0 g
(0.23 mol) p-toluidine in 50 mL of MeCN, followed by a dropwise
addition of acetic anhydride (25.2 g, 0.25 mol). The reaction mix-
ture was stirred for 3 h at 50 °C (TLC control). The mixture was
cooled to 0 °C and the resulting precipitate was filtered, washed
with water (2 ꢂ 25 mL) and dried on the air to afford N-acetyl-p-
toluidine (S2) (25.7 g, 73.4%). m/z (APCI+) 150 (M+H+). Mp
145–146 °C.
4.2.4.3. 3-[(4-Chloro-benzylamino)-methyl]-6-methyl-1H-quin-
olin-2-one (S7).
Compound S7 (375 mg, 40%, mp 118–119 °C)
was synthesized following the general procedure described above
using 4-chlorobenzylamine (425 mg, 3.0 mmol). 1H NMR (400
MHz, DMSO-d6 ppm): 2.31 (s, 3H), 3.58 (s, 2H), 3.69 (s, 2H), 7.20
(d, 1H), 7.28 (d, 1H), 7.38 (d, 4H), 7.42 (s, 1H), 7.79 (s, 1H), 11.70
(br, 1H). m/z (APCI+) 313 (M+H+).
4.2.2. 2-Chloro-6-methylquinoline-3-carbaldehyde (S3)54,55
POCl3 (111 mL, 1.19 mol) was added drop wise to DMF (40 mL,
0.51 mol) at 0 °C, followed by a portion wise addition of N-acetyl-
p-toluidine (S2) (25.7 g, 0.17 mol). The resulting mixture was stir-
red for 30 min at rt and the reaction mixture heated at 85 °C for
6 h. The resulting solution was cooled to rt and slowly poured into
ice-water. The resulting precipitate was filtered, washed with
water, dried and crystallized from ethyl acetate to give aldehyde
(S3) (23.6 g, 66.7%). 1H NMR (400 MHz, DMSO-d6 ppm): 2.48 (s,
3H), 7.79 (d, 1H), 7.91 (d, 1H), 8.01 (s, 1H), 8.81 (s, 1H), 10.35 (s,
1H). m/z (APCI+) 206 (M+H+). mp 124–125 °C.
4.3. General procedure to synthesize target compounds (12–26)
The suitable acid (0.5 mmol) was dissolved in DCM (8 mL) and
TBTU (193 mg, 0.6 mmol) followed by the addition of TEA
(71 mg, 0.7 mmol). The resulting solution was allowed to stir for
30 min at rt and the suitable secondary amine (S5–S7) (0.5 mmol)
was added to the reaction mixture. The stirring was continued for
8 h. The reaction mixture was washed with a solution of K2CO3
(25% in water, 2 ꢂ 10 mL), water (1 ꢂ 10 mL) and dried over
Na2SO4. The solvent was distilled off in vacuo and the residue
was crystallized from MeOH to afford the desired target com-
pounds (12–26).
4.2.3. 6-Methyl-2-oxo-1,2-dihydro-quinoline-3-carbaldehyde
(S4)56
4.3.1. 4-Fluoro-N-furan-2-ylmethyl-N-(6-methyl-2-oxo-1,2-dihy
dro-quinolin-3-ylmethyl)-benzamide (12)
A solution of 2-chloro-6-methyl-quinoline-3-carbaldehyde (S3)
(23.6 g, 0.11 mol) in a mixture of AcOH/water (90 mL:10 mL) was
heated at 85–90 °C with stirring. After 2 h a precipitation started
and the reaction mixture was additionally heated for 4 h at
85–90 °C. The resulting mixture was cooled to 0 °C leading to an
intensive precipitate formation. The resulting precipitate was fil-
tered, washed with water, 5% NaHCO3 and again with water, dried
on the air to afford the title compound (S4) (14.3 g, 68.9%); mp
201–202 °C. 1H NMR (400 MHz, DMSO-d6 ppm): 2.35 (s, 3H),
7.25 (d, 1H), 7.49 (d, 1H), 7.71 (s, 1H), 8.39 (s, 1H), 10.21 (s, 1H),
12.05 (br, 1H). m/z (APCI+) 188 (M+H+).
Compound 12 (119 mg, 72%; mp 129–130 °C) was synthesized
according to the general procedure described above, starting from
the 4-Fluoro-benzoic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm: 2.32 (s, 3H), 4.39 (br, 2H), 4.71 (br, 2H), 6.29 (br,
1H), 6.31 (br, 1H), 7.21 (d, 1H, J = 8,5 Hz), 7.31 (br, 1H,), 7.33 (m,
2H,), 7.48 (s, 1H), 7.52 (br, 2H), 7.61 (br, 2H), 11.7 (br, 1H). m/z
(APCI+) 391 (M+H+). HRMS (ESI) m/z calcd for C23H19FN2O3
(M+H+): 391.42; found 391.42.
4.3.2. 3,4-Dichloro-N-furan-2-ylmethyl-N-(6-methyl-2-oxo-1,2-
dihydro-quinolin-3-ylmethyl)-benzamide (13)
4.2.4. General procedure to synthesize intermediate compounds
(S5–S7)
Compound 13 (112 mg, 52%; mp 138–139 °C) was synthesized
according to the general procedure described above starting from
the 3,4-dichloro-benzoic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm): 2.35 (s, 3H), 3.81 (br, 3H), 4.39–4.42 (two broad-
ened signals, 2H, Z/E forms), 4.59–4.71 (two broadened signals,
2H, Z/E forms), 6.31 (br, 1H), 6.39 (br, 1H), 7.25 (d, 1H,
J = 8,5 Hz), 7.31 (d, 1H, J = 8,5 Hz), 7.42 (superposition of two broad
singlets, 2H), 7.55 (s, 1H), 7.61 (s, 1H), 7.75 (br, 2H), 11.6 (br, 1H).
m/z (APCI+) 441, 443 (M+H+). HRMS (ESI) m/z calcd for
The suitable amine (3.00 mmol) was added to a solution of 6-
methyl-2-oxo-1,2-dihydro-quinoline-3-carbaldehyde (S4) (400 mg,
2.13 mmol) in 20 mL of isopropanol. The reaction mixture was re-
fluxed for 7 h, cooled to 0 °C and the resulting precipitate was fil-
tered and washed with hexane (2 ꢂ 15 mL) to afford the crude
Shiff base. The product was suspended in 15 mL of isopropanol.
NaBH4 (100 mg, 2 mmol) was added to the resulting suspension
and the mixture stirred for 4 h at rt (TLC control). The reaction mix-
ture was evaporated in vacuo and 20 mL of water was added to res-
C
23H18Cl2N2O3 (M+H+): 442.32; found 442.32.

88
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
4.3.3. 4-Chloro-N-furan-2-ylmethyl-3-hydroxy-N-(6-methyl-2-
oxo-1,2-dihydro-quinolin-3-ylmethyl)-benzamide (14)
the 6-methoxy-nicotinic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm: 2.41 (s, 3H), 3.91 (br 3H), 4.41 (br, 2H), 4.68 (br,
2H), 6.31 (br, 1H), 6.39 (br, 1H), 6.89 (d, 1H, J = 8.5 Hz), 7.22 (d,
1H, J = 8.5 Hz), 7.31 (d, 1H, J = 8.5 Hz), 7.51 (s, 1H), 7.59 (s, 1H),
7.61 (s, 1H), 7.81 (br, 1H), 8.41 (br, 1H)11.7 (br, 1H). m/z (APCI+)
404 (M+H+). HRMS (ESI) m/z calcd for C23H21N3O4 (M+H+):
404.45; found 404.45.
Compound 14 (277 mg, 22%; mp 160–161 °C) was synthesized
according to the general procedure described above starting from
the 4-chloro-3-hydroxy-benzoic acid and amine (S5). 1H NMR
(400 MHz, DMSO-d6 ppm: 2.31 (s, 3H), 4.32 (br, 2H), 4.65 (br,
2H), 6.31 (br, 1H), 6.39 (br, 1H), 6.99 (br, 2H), 7.19 (d, 1H,
J = 8.5 Hz), 7.31 (d, 1H, J = 8.5 Hz), 7.36 (s, 1H), 7.41 (s, 1H), 7.52
(superposition of two singlets, 2H), 10.3 (br, 1H), 11.7 (br, 1H).
m/z (APCI+) 423, 425 (M+H+). HRMS (ESI) m/z calcd for
4.3.9. 4-Dimethylamino-N-furan-2-ylmethyl-N-(6-methyl-2-
oxo-1,2-dihydro-quinolin-3-ylmethyl)-benzamide (20)
Compound 20 (96 mg, 53%; mp 147–148 °C) was synthesized
according to the general procedure described above starting from
the 4-dimethylamino-benzoic acid and amine (S5). 1H NMR
(400 MHz, DMSO-d6 ppm: 2.31 (s, 3H), 2.92 (s, 6H), 4.38 (s, 2H),
4.68 (s, 2H), 6.32 (br, 1H), 6.38 (br, 1H), 6.71 (d, 2H, J = 8.5 Hz),
7.21 (d, 1H, J = 8.5 Hz), 7.29 (d, 1H, J = 8.5 Hz), 7.39 (br, 2H,
J = 8.5 Hz), 7.41 (s, 1H), 7.61 (br, 2H), 7.62 (br, 1H), 11.5 (br, 1H).
m/z (APCI+) 416 (M+H+). HRMS (ESI) m/z calcd for C25H25N3O3
(M+H+): 416.50; found 416.50.
C
23H18Cl2N2O3 (M+H+): 423.87; found 423.87.
4.3.4. 4-Chloro-N-furan-2-ylmethyl-N-(6-methyl-2-oxo-1,2-
dihydro-quinolin-3-ylmethyl)-3-piperidin-1-yl-benzamide (15)
Compound 15 (102 mg, 48%; mp 150–151 °C) was synthesized
according to the general procedure described above starting from
the 4-chloro-3-piperidin-1-yl-benzoic acid and amine (S5). NMR
1H (400 MHz, DMSO-d6 ppm: 1.5 (br, 6H), 2.32 (s, 3H), 2.72–2.91
(two broad peak 4H, Z/E forms), 4.22–4.41 (two broadened signals,
2H, Z/E forms), 4.61–4.75 (two broadened signals, 2H, Z/E forms),
6.31 (br, 1H), 6.39 (br, 1H), 7.15 (br, 2H), 7.21 (d, 1H, J = 8.5 Hz),
7.31 (d, 1H, J = 8.5 Hz), 7.41 (br, 1H), 7.49 (s, 1H), 7.61 (s, 1H),
7.62 (s, 1H), 11.7 (br, 1H). m/z (APCI+) 490, 492 (M+H+). HRMS
(ESI) m/z calcd for C28H28ClN3O3 (M+H+): 491.01; found 491.01.
4.3.10. 3-Chloro-N-furan-2-ylmethyl-4-hydroxy-N-(6-methyl-2-
oxo-1,2-dihydro-quinolin-3-ylmethyl)-benzamide (21)
Compound 21 (24 mg, 12%; mp 180–182 °C) was synthesized
according to the general procedure described above starting from
the 3-chloro-4-hydroxy-benzoic acid and amine (S5). 1H NMR
(400 MHz, DMSO-d6 ppm: 2.35 (s, 3H), 4.35 (s, 2H), 4.62 (s, 2H),
6.31 (br, 1H), 6.41 (br, 1H), 6.99 (br d, 1H, J = 8.5 Hz), 7.22 (d, 1H,
J = 8.5 Hz), 7.31 (d, 1H, J = 8.5 Hz), 7.35 (s, 1H), 7.51 (s, 1H), 7.53
(br, 1H), 7.58 (s, 1H), 7.61 (s, 1H), 10.5 (br, 1H), 11.6 (br, 1H). m/z
(APCI+) 423, 424 (M+H+). HRMS (ESI) m/z calcd for C23H19ClN2O4
(M+H+): 423.88; found 423.88.
4.3.5. 4-Cyano-N-furan-2-ylmethyl-N-(6-methyl-2-oxo-1,2-
dihydro-quinolin-3-ylmethyl)-benzamide (16)
Compound 16 (127 mg, 63%; mp 220–221 °C) was synthesized
according to the general procedure described above starting from
the 4-cyano-benzoic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm: 2.32 (s, 3H), 4.22–4.51 (two broadened signals,
2H, Z/E forms), 4.51–4.78 (two broadened signals, 2H, Z/E forms),
6.31 (br, 1H), 6.41 (br, 1H), 7.21 (br, 1H), 7.31 (d, 1H, J = 8.5 Hz),
7.51 (s, 1H), 7.59 (s, 1H), 7.62 (br, 2H), 7.72 (br, 1H), 7.81–8.05
(two broadened signals, 2H), 11.7 (br, 1H). m/z (APCI+) 398
(M+H+). HRMS (ESI) m/z calcd for C24H19N3O3 (M+H+): 398.44;
found 398.44.
4.3.11. 1-Methyl-1H-pyrazole-4-carboxylic acid furan-2-ylmet
hyl-(6-methyl-2-oxo-1,2-dihydro-quinolin-3-ylmethyl)-amide
(22)
Compound 22 (26 mg, 12%; mp 230–231 °C) was synthesized
according to the general procedure described above starting from
the 1-methyl-1H-pyrazole-4-carboxylic acid and amine (S5). 1H
NMR (400 MHz, DMSO-d6 ppm: 2.31 (s, 3H), 3.81 (br 3H), 4.41
(br, 2H), 4.75 (br, 2H), 6.31 (br, 1H), 6.35 (br, 1H), 7.19 (d, 1H,
J = 8,5 Hz), 7.25 (d, 1H, J = 8,5 Hz), 7.41 (s, 1H), 7.46 (s, 1H), 7.52
(br, 1H), 7.62 (br, 1H), 8.05 (br, 1H), 11.7 (br, 1H). m/z (APCI+)
377 (M+H+). HRMS (ESI) m/z calcd for C21H20N4O3 (M+H+):
377.43; found 377.43.
4.3.6. 3-Chloro-4-fluoro-N-furan-2-ylmethyl-N-(6-methyl-2-
oxo-1,2-dihydro-quinolin-3-ylmethyl)-benzamide (17)
Compound 17 (121 mg, 62%; mp 132–134 °C) was synthesized
according to the general procedure described above starting from
the 3-chloro-4-fluoro-benzoic acid and amine (S5). 1H NMR
(400 MHz, DMSO-d6 ppm: 2.32 (s, 3H), 4.41 (br, 2H), 4.61 (br,
2H), 6.31 (br, 1H), 6.41 (br, 1H), 7.21 (d, 1H, J = 8.5 Hz), 7.35 (d,
1H, J = 8.5 Hz), 7.51 (br, 3H), 7.59 (s, 1H), 7.62 (s, 1H), 7.81 (br,
1H), 11.6 (br, 1H). m/z (APCI+) 425, 427 (M+H+). HRMS (ESI) m/z
calcd for C23H18ClFN2O3 (M+H+): 425.87; found 425.87.
4.3.12. 3,4-Difluoro-N-furan-2-ylmethyl-N-(6-methyl-2-oxo-
1,2-dihydro-quinolin-3-ylmethyl)-benzamide (23)
Compound 23 (99 mg, 54%; mp 142–114 °C) was synthesized
according to the general procedure described above starting from
the 3,4-difluoro-benzoic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm: 2.31 (s, 3H), 4.41 (br peak, 2H), 4.61 (br peak,
2H), 6.27 (br peak, 1H), 6.30 (br peak, 1H), 7.19 (d, 1H,
J = 8,5 Hz), 7.29 (d, 1H, J = 8,5 Hz), 7.39 (s, 1H), 7.48 (s, 1H), 7.52
(br peak, 1H), 7.62 (br peak, 2H), 11.6 (br peak, 1H). m/z (APCI+)
409 (M+H+). HRMS (ESI) m/z calcd for C23H18F2N2O3 (M+H+):
409.42; found 409.42.
4.3.7. 3,5-Dichloro-N-furan-2-ylmethyl-4-hydroxy-N-(6-
methyl-2-oxo-1,2-dihydro-quinolin-3-ylmethyl)-benzamide
(18)
Compound 18 (33 mg, 18%; mp 188–189 °C) was synthesized
according to the general procedure described above starting from
the 3,5-dichloro-4-hydroxy-benzoic acid and amine (S5). 1H NMR
(400 MHz, DMSO-d6 ppm: 2.31 (s, 3H), 4.41 (br, 2H), 4.61 (br,
2H), 6.35 (br, 1H), 6.41 (br, 1H), 7.21 (d, 1H, J = 8.5 Hz), 7.32 (d,
1H, J = 8.5 Hz), 7.51 (br, 1H), 7.55 (br, 2H), 7.61 (s, 1H), 7.67
(s, 1H), 10.05 (br, 1H), 11.6 (br, 1H). m/z (APCI+) 457, 459 (M+H+).
HRMS (ESI) m/z calcd for C23H18Cl2N2O4 (M+H+): 458.32; found
458.32.
4.3.13. N-Furan-2-ylmethyl-N-(6-methyl-2-oxo-1,2-dihydro-
quinolin-3-ylmethyl)-4-sulfamoyl-benzamide (24)
Compound 24 (26 mg, 13%; mp 248–249 °C) was synthesized
according to the general procedure described above starting from
the 4-sulfamoyl-benzoic acid and amine (S5). 1H NMR (400 MHz,
DMSO-d6 ppm: 2.35 (s, 3H), 4.29–4.38 (two br peaks, 2H, Z/E
forms), 4.41–4.71 (two br peaks, 2H Z/E forms), 6.29 (br, 1H),
6.37 (br, 1H), 7.21 (d, 1H, J = 8.5 Hz), 7.29 (d, 1H, J = 8.5 Hz), 7.49
(s, 1H), 7.52 (s, 1H), 7.61 (br, 2H), 7.68 (br, 1H), 7.88 (br, 2H). m/z
4.3.8. N-Furan-2-ylmethyl-6-methoxy-N-(6-methyl-2-oxo-1,2-
dihydro-quinolin-3-ylmethyl)-nicotinamide (19)
Compound 19 (94 mg, 52%; mp 173–175 °C) was synthesized
according to the general procedure described above starting from

F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
89
(APCI+) 452 (M+H+). HRMS (ESI) m/z calcd for C23H21N3O5S (M+H+):
452.51; found 452.51.
were cultured at 37 °C for 2 h. Nonadherent cells were collected
and resuspended at 5 ꢂ 106 cells/mL RPMI-1640 medium contain-
ing 10% FBS, 5 lg/mL PHA, and 100 U/mL IL-2 (Sigma–Aldrich), fol-
4.3.14. 4-Chloro-N-(6-methyl-2-oxo-1,2-dihydro-quinolin-3-
ylmethyl)-N-(tetrahydro-furan-2-lmethyl)-benzamide (25)
Compound 25 (51 mg, 25%; mp 223–225 °C) was synthesized
according to the general procedure described above starting from
the 4-chloro-benzoic acid and amine (S6). 1H NMR (400 MHz,
DMSO-d6 ppm: 1.15–1.95 (two broad peaks, 4H) , 2.35 (s, 3H),
3.35 (br, 1H), 3.70 (br, 2H), 4.09 (br, 2H) 4.55 (br, 2H) 7.21 (d,
1H), 7.29 (d, 1H), 7.35–7.54 (br 5H), 7.66 (br s, 1H), 11.65 (br,
1H). m/z (APCI+) 411, 413 (M+H+). HRMS (ESI) m/z calcd for
lowed by incubation at 37 °C for 3 days. The PHA-stimulated cells
(5 ꢂ 104 cells) were infected with primary HIV-1 isolates at 500
TCID50 (0.01 MOI) in the absence or presence of small molecule
inhibitors at graded concentrations. Culture media were changed
every 3 days and replaced with fresh media containing freshly pre-
pared inhibitor. The supernatants were collected 7 days post-infec-
tion and tested for p24 antigen by ELISA. The percent inhibition of
p24 production and IC50 values were calculated by the GraphPad
Prism software (GraphPad Software Inc.).
C
23H23ClN2O3 (M+H+): 411.91; found 411.911.
4.6. Cytotoxicity assay
4.3.15. 8-Methyl-imidazo[1,2-a]pyridine-2-carboxylic acid
(4-chloro-benzyl)-(6-methyl-2-oxo-1,2-dihydro-quinolin-3-
ylmethyl)-amide (26)
Cytotoxicity of small molecules in MT-2 cells and PBMC was
measured by the XTT [(sodium 30-(1-(phenylamino)-carbonyl)-
3,4-tetrazolium-bis(4-methoxy-6-nitro) bezenesulfonic acid
hydrate)] method as previously described.39 Briefly, for MT-2 cells,
Compound 26 (47 mg, 20%; mp 250–251 °C) was synthesized
according to the general procedure described above starting from
the 8-methyl-imidazo[1,2-a]pyridine-2-carboxylic acid and amine
(S7). 1H NMR (400 MHz, DMSO-d6 ppm: 2.35 (s, 6H), 4.30–4.80
(two br peaks, 2H, Z/E forms), 5.10–5.60 (two br peaks, 2H, Z/E
forms), 6.80 (br, 1H), 7.05 (br, 1H), 7.10–7.45 (br m, 7H), 7.61
(br, 1H), 8.35 (br, 2H), 11.62 (br, 1H). m/z (APCI+) 471, 473
(M+H+). HRMS (ESI) m/z calcd for C27H23ClN4O2 (M+H+): 471.97;
found 471.97.
100 ll of a small molecule at graded concentrations was added to
an equal volume of cells (1 ꢂ 105 cells/mL) in 96-well plates fol-
lowed by incubation at 37 °C for 4 days, which ran parallel to the
neutralization assay in MT-2 (except medium was added instead
of virus). In the case of PBMC, 5 ꢂ 105 cells/mL were used and
the cytotoxicity was measured after 7 days. After addition of XTT
(PolySciences, Inc.), the soluble intracellular formazan was quanti-
tated colorimetrically at 450 nm 4 h later with a reference at
620 nm. The percent of cytotoxicity and the CC50 values were cal-
culated as above.
4.4. Electron microscopy to study inhibition of in vitro assembly
by small molecules
In vitro assembly systems were set up as previously reported.39
Briefly, stock proteins were adjusted to the appropriate concentra-
4.7. Infectivity assay
In these assays we wanted to measure the infectivity potential
of the virions released by treating cells infected with HIV-1 by 6
and 55. Briefly, 5 ꢂ 104/mL MT-2 cells were infected with HIV-1
tion (25 lV for Gag proteins and 50 lV for CA proteins) with the
Na2HPO4 buffer at pH 8.0. After addition of 5% total Escherichia coli
RNA (RNA/protein = 1:20 by weight), incubation with or without
varied doses of NYAD-1 as positive control and 6 and 50 as test
samples for 30 min at 4 °C, the samples were dialyzed overnight
in Na2HPO4 buffer at pH 8.0 containing 100 mM of NaCl at 4 °C.
For assembly of CA mature-like particles, addition of 5% total
E. coli RNA was omitted. Negative staining was used to check the
assembly. Carbon-coated copper grids (200 mesh size; EM Sci-
IIIB (m.o.i. = 0.01) in the presence of 5
l
M, 2.5
lM and 1.3 lM of
small molecules 6 and 4
lM, 2
l
M and 1 M of 55. Control cells
l
were untreated or treated with 1ug/mL of AMD3100 (CXCR4 recep-
tor inhibitor). Following overnight incubation the medium was
completely removed and replaced with fresh medium. On the
fourth day post-infection the supernatants were collected and fil-
tered with a 0.45 lm PVDF membrane.
ences) were treated with 20
for 2 min. Twenty microliters of reaction solution were placed onto
the grid for 2 min. Spotted grids were then stained with 30 l of
ll of poly-L-lysine (1 mg/mL; Sigma)
One aliquot of the supernatants from each sample was mixed
1:1 with a solution of 5% Triton X-100 for p24 quantification, and
stored at 4 °C, another aliquot was immediately frozen at ꢁ80 °C
and used to compare the infectivity of the viral particles released
by 6 and 55 treated cells to the viral particles released by untreated
cells. P24 was quantified by sandwich-ELISA then the viral samples
were normalized for the p24 content and titered by infecting MT-2
cells to calculate the TCID50. Following overnight incubation 3/4
of the medium containing the inoculum was replaced with fresh
medium. On the fourth day post infection supernatants were col-
lected for p24 quantification by sandwich-ELISA and the TCID50
was calculated by the Spearman–Karber statistical method.
l
uranyl acetate solution for 2 min. Excess stain was removed, and
grids were air-dried. Specimens were examined with a TECNAI
G2 electron microscope at 80 Kv.
4.5. Measurement of antiviral activity
The inhibitory activity of small molecules on infection by labo-
ratory-adapted HIV-1 strains was determined as previously de-
scribed.39 In brief, 1 ꢂ 104 MT-2 cells were infected with HIV-1 at
100 TCID50 (50% tissue culture infective dose) (0.01MOI) in
200 ll RPMI 1640 medium containing 10% FBS in the presence or
Acknowledgments
absence of small molecules at graded concentrations overnight.
The culture supernatants were then removed and fresh media con-
taining freshly prepared test small molecules were added. On the
fourth day post-infection, 100 ll of culture supernatants were col-
lected from each well, mixed with equal volume of 5% Triton X-100
This study was supported by NIH Grant RO1 AI081604 (A.K.D.)
and the intramural fund from the New York Blood Center (A.K.D.).
We thank Yelena Oksov (electron microscopy) for her technical help.
and tested for p24 antigen by ELISA.
References and notes
The inhibitory activity of selected lead compounds (6, 50 and
55) on infection by primary HIV-1 isolates was determined as pre-
viously described.39 PBMCs were isolated from the blood of healthy
donors at the New York Blood Center by standard density gradient
centrifugation using Histopaque-1077 (Sigma–Aldrich). The cells
1. Briggs, J. A.; Wilk, T.; Welker, R.; Krausslich, H. G.; Fuller, S. D. EMBO J. 2003, 22,
1707.
2. Briggs, J. A.; Grunewald, K.; Glass, B.; Forster, F.; Krausslich, H. G.; Fuller, S. D.
Structure 2006, 14, 15.

90
F. Curreli et al. / Bioorg. Med. Chem. 19 (2011) 77–90
3. Li, S.; Hill, C. P.; Sundquist, W. I.; Finch, J. T. Nature 2000, 407, 409.
4. Ganser, B. K.; Li, S.; Klishko, V. Y.; Finch, J. T.; Sundquist, W. I. Science 1999, 283,
80.
5. Ganser-Pornillos, B. K.; Cheng, A.; Yeager, M. Cell 2007, 131, 70.
6. Ako-Adjei, D.; Johnson, M. C.; Vogt, V. M. J. Virol. 2005, 79, 13463.
7. von Schwedler, U. K.; Stray, K. M.; Garrus, J. E.; Sundquist, W. I. J. Virol. 2003, 77,
5439.
32. Smith, P. F.; Ogundele, A.; Forrest, A.; Wilton, J.; Salzwedel, K.; Doto, J.; Allaway,
G. P.; Martin, D. E. Antimicrob. Agents Chemother. 2007, 51, 3574.
33. Adamson, C. S.; Ablan, S. D.; Boeras, I.; Goila-Gaur, R.; Soheilian, F.; Nagashima,
K.; Li, F.; Salzwedel, K.; Sakalian, M.; Wild, C. T.; Freed, E. O. J. Virol. 2006, 80,
10957.
34. Zhou, J.; Yuan, X.; Dismuke, D.; Forshey, B. M.; Lundquist, C.; Lee, K. H.; Aiken,
C.; Chen, C. H. J. Virol. 2004, 78, 922.
8. Abdurahman, S.; Hoglund, S.; Goobar-Larsson, L.; Vahlne, A. J. Gen. Virol. 2004,
85, 2903.
9. Chien, A. I.; Liao, W. H.; Yang, D. M.; Wang, C. T. Virology 2006, 348, 84.
10. Douglas, C. C.; Thomas, D.; Lanman, J.; Prevelige, P. E., Jr. Biochemistry 2004, 43,
10435.
11. Forshey, B. M.; von Schwedler, U. K.; Sundquist, W. I.; Aiken, C. J. Virol. 2002, 76,
5667.
12. Ganser-Pornillos, B. K.; von Schwedler, U. K.; Stray, K. M.; Aiken, C.; Sundquist,
W. I. J. Virol. 2004, 78, 2545.
13. Joshi, A.; Nagashima, K.; Freed, E. O. J. Virol. 2006, 80, 7939.
14. del Alamo, M.; Mateu, M. G. J. Mol. Biol. 2005, 345, 893.
15. Scholz, I.; Arvidson, B.; Huseby, D.; Barklis, E. J. Virol. 2005, 79, 1470.
16. Chang, Y. F.; Wang, S. M.; Huang, K. J.; Wang, C. T. J. Mol. Biol. 2007, 370, 585.
17. Abdurahman, S.; Hoglund, S.; Hoglund, A.; Vahlne, A. Retrovirology 2007, 4, 19.
18. Abdurahman, S.; Youssefi, M.; Hoglund, S.; Vahlne, A. Retrovirology 2007, 4, 69.
19. Yamashita, M.; Emerman, M. J. Virol. 2004, 78, 5670.
20. Vogt, V. M. Nat. Struct. Mol. Biol. 2005, 12, 638.
21. Li, F.; Wild, C. Curr. Opin. Investig. Drugs 2005, 6, 148.
22. Li, J.; Tang, S.; Hewlett, I.; Yang, M. Infect. Disord. Drug Targets 2007, 7, 238.
23. Niedrig, M.; Gelderblom, H. R.; Pauli, G.; Marz, J.; Bickhard, H.; Wolf, H.;
Modrow, S. J. Gen. Virol. 1994, 75, 1469.
24. Hoglund, S.; Su, J.; Reneby, S. S.; Vegvari, A.; Hjerten, S.; Sintorn, I. M.; Foster,
H.; Wu, Y. P.; Nystrom, I.; Vahlne, A. Antimicrob. Agents Chemother. 2002, 46,
3597.
25. Garzon, M. T.; Lidon-Moya, M. C.; Barrera, F. N.; Prieto, A.; Gomez, J.; Mateu, M.
G.; Neira, J. L. Protein Sci. 2004, 13, 1512.
35. Zhou, J.; Chen, C. H.; Aiken, C. J. Virol. 2006, 80, 12095.
36. Sticht, J.; Humbert, M.; Findlow, S.; Bodem, J.; Muller, B.; Dietrich, U.; Werner,
J.; Krausslich, H. G. Nat. Struct. Mol. Biol. 2005, 12, 671.
37. Schafmeister, C. E.; Po, J.; Verdine, G. L. J. Am. Chem. Soc. 2000, 122, 5891.
38. Ternois, F.; Sticht, J.; Duquerroy, S.; Krausslich, H. G.; Rey, F. A. Nat. Struct. Mol.
Biol. 2005, 12, 678.
39. Zhang, H.; Bhattacharya, S.; Tong, X.; Waheed, A. A.; Hong, A.; Heck, S.; Curreli, F.;
Goger, M.; Cowburn, D.; Freed, E. O.; Debnath, A. K. J. Mol. Biol. 2008, 378, 565.
40. Bhattacharya, S.; Zhang, H.; Debnath, A. K.; Cowburn, D. J. Biol. Chem. 2008, 283,
16274.
41. Irwin, J. J.; Shoichet, B. K. J. Chem. Inf. Model. 2005, 45, 177.
42. Gross, I.; Hohenberg, H.; Krausslich, H. G. Eur. J. Biochem. 1997, 249, 592.
43. Yu, D.; Wild, C. T.; Martin, D. E.; Morris-Natschke, S. L.; Chen, C. H.; Allaway, G.
P.; Lee, K. H. Expert Opin. Investig. Drugs 2005, 14, 681.
44. Qian, K.; Kuo, R. Y.; Chen, C. H.; Huang, L.; Morris-Natschke, S. L.; Lee, K. H. J.
Med. Chem. 2010, 53, 3133.
45. Lipinski, C. A. J. Pharmacol. Toxicol. Methods 2000, 44, 235.
46. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev.
2001, 46, 3.
47. Podvinec, M.; Lim, S. P.; Schmidt, T.; Scarsi, M.; Wen, D.; Sonntag, L. S.;
Sanschagrin, P.; Shenkin, P. S.; Schwede, T. J. Med. Chem. 2010, 53, 1483.
48. Kim, Y. A.; Sharon, A.; Chu, C. K.; Rais, R. H.; Al Safarjalani, O. N.; Naguib, F. N.
M.; el Kouni, M. H. Biochem. Pharmacol. 2007, 73, 1558.
49. Siddiquee, K.; Zhang, S.; Guida, W. C.; Blaskovich, M. A.; Greedy, B.; Lawrence,
H. R.; Yip, M. L. R.; Jove, R.; McLaughlin, M. M.; Lawrence, N. J.; Sebti, S. M.;
Turkson, J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 7391.
26. Tang, C.; Loeliger, E.; Kinde, I.; Kyere, S.; Mayo, K.; Barklis, E.; Sun, Y.; Huang,
M.; Summers, M. F. J. Mol. Biol. 2003, 327, 1013.
27. Sakalian, M.; McMurtrey, C. P.; Deeg, F. J.; Maloy, C. W.; Li, F.; Wild, C. T.;
Salzwedel, K. J. Virol. 2006, 80, 5716.
28. Li, F.; Goila-Gaur, R.; Salzwedel, K.; Kilgore, N. R.; Reddick, M.; Matallana, C.;
Castillo, A.; Zoumplis, D.; Martin, D. E.; Orenstein, J. M.; Allaway, G. P.; Freed, E.
O.; Wild, C. T. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 13555.
29. Kelly, B. N.; Kyere, S.; Kinde, I.; Tang, C.; Howard, B. R.; Robinson, H.; Sundquist,
W. I.; Summers, M. F.; Hill, C. P. J. Mol. Biol. 2007, 373, 355.
30. Martin, D. E.; Blum, R.; Wilton, J.; Doto, J.; Galbraith, H.; Burgess, G. L.; Smith, P.
C.; Ballow, C. Antimicrob. Agents Chemother. 2007, 51, 3063.
31. Stoddart, C. A.; Joshi, P.; Sloan, B.; Bare, J. C.; Smith, P. C.; Allaway, G. P.; Wild, C.
T.; Martin, D. E. PLoS One 2007, 2, e1251.
50. Tripathy, R.; Ghose, A.; Singh, J.; Bacon, E. R.; Angeles, T. S.; Yang, S. X.; Albom,
M. S.; Aimone, L. D.; Herman, J. L.; Mallamo, J. P. Bioorg. Med. Chem. Lett. 2007,
17, 1793.
51. Lyne, P. D.; Lamb, M. L.; Saeh, J. C. J. Med. Chem. 2006, 49, 4805.
52. Bembenek, S. D.; Keith, J. M.; Letavic, M. A.; Apodaca, R.; Barbier, A. J.; Dvorak,
L.; Aluisio, L.; Miller, K. L.; Lovenberg, T. W.; Carruthers, N. I. Bioorg. Med. Chem.
2008, 16, 2968.
53. Ravindranathan, K. P.; Mandiyan, V.; Ekkati, A. R.; Bae, J. H.; Schlessinger, J.;
Jorgensen, W. L. J. Med. Chem. 2010, 53, 1662.
54. Srivastava, A.; Singh, M. K.; Singh, R. M. Indian J. Chem., Sect. B 2006, 45, 292.
55. Meth-Cohn, O.; Narine, B.; Tarnowski, B. J. Chem. Soc., Perkin Trans. 1 1981,
1520.
56. Pawarr, R. A.; Bajare, P. B.; Mundade, S. B. J. Indian Chem. Soc. 1990, 67, 685.