Synthesis and evaluation of N-aryl pyrrolidinones as novel anti-HIV-1 agents. Part 1

Article abstract of DOI:10.1016/j.bmcl.2006.04.012

The synthesis and preliminary structure-activity relationship of a series of pyrrolidinones are described. These pyrrolidinones have been characterized as novel non-nucleoside reverse transcriptase inhibitors (NNRTIs) which are highly potent against wild-type and drug-resistant human immunodeficiency viruses (HIV-1).

Full text of DOI:10.1016/j.bmcl.2006.04.012

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3430–3433
Synthesis and evaluation of N-aryl pyrrolidinones as novel
anti-HIV-1 agents. Part 1
Baogen Wu,* Kelli Kuhen, Truc Ngoc Nguyen, David Ellis, Beth Anaclerio, Xiaohui He,
Kunyong Yang, Donald Karanewsky, Hong Yin, Karen Wolff, Kimberly Bieza,
Jeremy Caldwell and Yun He*
Genomics Institute of the Novartis Research Foundation (GNF), 10675 John Jay Hopkins Dr. San Diego, CA 92121, USA
Received 7 March 2006; revised 1 April 2006; accepted 3 April 2006
Available online 24 April 2006
Abstract—The synthesis and preliminary structure–activity relationship of a series of pyrrolidinones are described. These pyrrolid-
inones have been characterized as novel non-nucleoside reverse transcriptase inhibitors (NNRTIs) which are highly potent against
wild-type and drug-resistant human immunodeficiency viruses (HIV-1).
ꢀ 2006 Elsevier Ltd. All rights reserved.
Highly active antiretroviral therapy (HAART) combina-
O
tion regimens have dramatically decreased the morbidity
and mortality among patients with HIV infections.1
Four main classes of antiretroviral drugs are currently
available for the treatment of HIV infection. Non-nucle-
oside reverse transcriptase inhibitors (NNRTIs) have be-
come the key components in the combination regimens.
NNRTIs bind to an allosteric site of reverse transcriptase
(RT) in a noncompetitive manner which causes distor-
tion of the three-dimensional structure of the enzyme
and inhibits its catalytic function.2 Currently, three
NNRTIs (see Fig. 1) have been approved for the treat-
ment of HIV infection, namely nevirapine (1),3 delavir-
dine (2),4 and efavirenz (3).5 Due to the low genetic
barrier to resistance for these marketed NNRTIs, the ra-
pid emergence of resistance to NNRTIs, and cross resis-
tance within this class, has become a major drawback in
the clinic for NNRTIs. As the reverse transcriptase is
essential to the life cycle of the virus, many research
activities in this field have been focused on searching
for novel NNRTIs with high potency against wild-type
(WT) and drug-resistant viruses.
O
H
N
HN
N
S NH
O
N
F3C
Cl
N
O
N
N
N
N
N
O
H
O
H
3, efavirenz
1, nevirapine
2, delavirdine
Figure 1. Currently approved NNRTIs.
were screened and several lead scaffolds were discovered
with anti-HIV activity. Among these lead scaffolds, a
series of N-aryl pyrrolidinones represented by
4
(Fig. 2) were identified as novel NNRTIs. For the
convenience of discussion, the rings of this compound
are designated as ring A, B, C, and D as described in
Figure 2. This compound showed an EC50 of 125 nM
against WT HIV reporter virus in the 293T target cells.6
It also has an IC50 of 2 lM in a RT inhibition enzymatic
assay. Efavirenz showed an IC50 of 0.8 lM in the same
ELISA. A systematic SAR study is then undertaken and
the preliminary results are reported herein.
We have initiated a program to identify novel small
molecule inhibitors of HIV infection by screening our
in-house library using a cell-based HIV infection
inhibition assay.6 Approximately 230,000 compounds
O
NH2
C N
D
B
MeO
O
A
Keywords: Antiviral; HIV; NNRTI; Pyrrolidinone.
*
4
Corresponding authors. Tel.: +1 858 332 4725; e-mail addresses:
bwu@gnf.org; yhe@gnf.org
Figure 2. Structure of HTS hit 4.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.012

B. Wu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3430–3433
3431
The synthesis of hit 4 is shown in Scheme 1. Following
the literature protocol,7 condensation of aldehyde 5 with
nitromethane gave nitro olefin 6. Michael addition of
oxazolidone 7 to olefin 6 afforded 8 in 75% yield and
98% ee after a single re-crystallization. Reduction of
the nitro group followed by intramolecular cyclization
provided lactam 9 in 88% yield. Lactam 9 was coupled
with 3-iodo-nitrobenzene catalyzed by CuI8 in the pres-
ence of a diamine as ligand, and the resulting product 10
was subjected to hydrogenation to furnish the desired
product 4.
analogs 11a–c. Simple N-acyl analogs 12a and 12b were
made by treating lactam 9 with acyl chlorides in pyri-
dine. Following the N-arylation protocol for the prepa-
ration of 10, a series of N-aryl analogs 13a–z were
synthesized.
To further define the SAR around the aniline group as
well as substitutions on aromatic D ring, additional
amine derivatives of 4 were made for biological evalua-
tion (Scheme 3). Thus, reductive amination of 4 with
different aldehydes provided amines 14a–d. Direct cou-
pling of 4 with different acyl chlorides in pyridine affor-
ded amides 15a–d. Finally, treatment of 4 with
substituted sulfonyl chlorides in pyridine gave a series
of sulfonamides 16a–s. All final products were purified
either by normal phase silica gel column chromatogra-
phy or reverse-phase preparative LC–MS.9 All analogs
that passed QC by analytical LC–MS were tested in
the biological assays.
In order to explore the SAR of D ring, different groups
were attached to lactam 9 (Scheme 2). Direct alkylation
of 9 with substituted benzyl bromides gave N-alkylated
NO2
CHO
a
b
d
MeO
MeO
O
O
The antiviral activities of these analogs were evaluated
by a cell-based single cycle replication assay.6 In general,
this series of analogs did not show cytotoxicity up to
10 lM.10 Table 1 summarizes the antiviral activities of
analogs 4, 11a–c, 12a,b, and 13a–z. The antiviral activity
of 4 was confirmed to possess an EC50 of 125 nM, and
nevirapine (NVP) was included for comparison. As
shown in Table 1, replacement of the aromatic D ring
with simple alkyl groups gave analogs 11a–c which
exhibited no antiviral activity up to a 10 lM concentra-
tion. Likewise, simple acyl analogs 12a,b and the
N-des-aryl analog 13a are not active. Therefore, further
efforts were then focused on the modification of the aro-
matic D ring. To assess the contribution of the amino
group to the antiviral activity, analogs 13b–j were eval-
uated. It is clear that this amino group plays a critical
role in the antiviral activity of the pyrrolidinones. This
was supported by the fact that the simple removal of
the amino group resulted in analog 13b with 20-fold
reduction in the antiviral activity. At the same time,
the N-methyl analog 13d showed an approximately
40-fold loss in activity. This indicated that the amino
6
5
O
O
O
Bn
N
c
NH
O
NO2
MeO
O
MeO
O
9
8
O
N
O
N
NO2
NH2
e
MeO
MeO
O
O
10
4
Scheme 1. Reagents and conditions: (a) AcOH, CH3NO2, NH4OAc,
reflux 2 h, 45%; (b) (R)-3-acetyl-4-benzyl-oxazolidin-2-one (7), LDA,
THF, À78 ꢁC, then 6, 75%; (c) EtOH, AcOEt, Raney Nickel, H2, 88%;
(d) 3-iodo-nitrobenzene, CuI, K3PO4, 1,2-cyclohexanediamine, DMF,
110 ꢁC, 12 h, 92%; (e) 10% Pd/C, EtOH, H2, 86%.
O
N
NH2
O
N
HN R
O
O
N
NH
a
a
R
MeO
MeO
MeO
MeO
O
O
O
O
4
14a-d
9
c
11a-c
c
b
b
O
O
O
R
O
O
N
O
O
HN
HN S
O
R
N
N
N
R
R
MeO
MeO
MeO
MeO
O
O
O
O
12a,b
15a-d
13a-z
16a-s
Scheme 2. Reagents and conditions: (a) RX, NaH, DMF, 50–76%;
(b) RX, K3PO4, CuI, DMF, 1,2-cyclohexanediamine, 110 ꢁC, 75–92%;
(c) RCOCl, DMAP, Py, 66–74%.
Scheme 3. Reagents and conditions: (a) RCHO, DMF,
NaB(O2CCH3)3H, 62–90%; (b) RCOCl, Py, 82–91%; (c) RSO2Cl,
Py, 86–92%.

3432
B. Wu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3430–3433
Table 1. Antiviral activity of 6, 11, 12 and 13 against WT HIV-1 virus6
acceptor at the 4 position of D ring, including analogs
13v, 13x, and 13z, demonstrated good antiviral activity.
Shifting the H-bond acceptor to a different position
resulted in a loss of activity by more than 20-fold (13t,
13u, 13w, and 13y), supporting the notion that specific
H-bond interactions in this region play a critical role.
Compound
R
WT-EC50
(lM)
CC50
(lM)
4
3-Amino-phenyl
3-Cyano-phenyl
0.125
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
11a
11b
11c
12a
12b
13a
13b
13c
13d
13e
13f
3-Amino-phenyl
3-Nitro-phenyl
Methyl
>10
>10
Aniline derivatives were also evaluated (Table 2). As
indicated in Table 2, simple N-alkyl analogs of aniline
14a–d showed decreased activity, while N-acylation
was somewhat tolerated. Fortuitously, the sulfonamide
analogs displayed a range of antiviral activity (16a–s).
In this series, many aryl sulfonamides showed improved
antiviral activity. In general, substitutions at the 2 or 3
position were well tolerated and many analogs exhibited
improved antiviral activity as indicated by analogs 16a,
16e–j, and 16l–q which displayed EC50’s in the low
nanomolar range. Some heterocyclic sulfonamides (16r
and 16s) also showed good antiviral activity.
Phenyl
H
Phenyl
>10
>10
2.280
1.780
4.710
>10
3-Hydroxy-phenyl
3-Dimethylamino-phenyl
3,5-Diamino-phenyl
3-Amino-4-methyl-phenyl
3-Amino-2-methyl-phenyl
2-Amino-phenyl
4-Amino-phenyl
3-Amino-5-fluoro-phenyl
2-Cyano-phenyl
0.196
>10
>10
13g
13h
13i
1.860
0.080
>10
13j
13k
13l
3-Cyano-phenyl
4-Cyano-phenyl
0.520
0.660
>10
13m
13n
13o
13p
13q
13r
13s
13t
Analogs with low nanomolar EC50’s were tested against
a panel of mutant viruses. The results are presented in
Table 3. NVP and efavirenz (EFV) were included for
comparison. NNRTI-specific mutant viruses K103N,
Y181C, L100I, and K103N/L100I were selected for
testing based on their prevalence in patients failing the
2-Methyl-phenyl
3-Methyl-phenyl
4-Methyl-phenyl
2-Trifluoromethyl-phenyl
3-Trifluoromethyl-phenyl
4-Trifluoromethyl-phenyl
2-Pyridyl
0.419
3.260
>10
2.950
3.620
2.630
2.370
0.11
13u
13v
13w
13x
13y
13z
NVPa
3-Pyridyl
4-Pyridyl
Table 2. Antiviral activity of 14, 15 and 16 against WT HIV-1 virus6
Compound
R
WT-EC50 CC50
(lM)
3-Methanesulfinyl-phenyl
4-Methanesulfinyl-phenyl
3-Methanesulfonyl-phenyl
4-Methanesulfonyl-phenyl
>10
0.040
>10
(lM)
7.51
14a
14b
14c
14d
15a
15b
15c
15d
16a
16b
16c
16d
16e
16f
Tetrahydrofuran-3-yl-methyl 1.024
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
Furan-3-yl-methyl
3-Phenyl propyl
Butyl
1.963
7.902
6.960
0.502
0.124
0.502
5.786
0.025
0.256
0.800
0.104
0.180
0.050
a NVP, nevirapine.
3-Methoxy benzoyl
3-Dimethylamino-benzoyl
Thiophene-2-carboxyl
Acetyl
group likely makes an H-bond interaction with amino
acid residues inside the binding pocket and functions
as an H-bond donor. For comparison, the correspond-
ing phenol analog 13c showed much lower antiviral
activity. The diamino analog (13e) showed diminished
antiviral activity, probably due to the interference of
the other amino group on binding to the enzyme. Based
on this result, substituted aromatic D ring analogs with
an amino group were studied in detail. Surprisingly,
analogs 13f and 13g showed a dramatic difference in
their antiviral activity. The 2-Me analog 13g lost its anti-
viral activity completely, while its positional isomer 13f
possessed similar antiviral activity compared to 4. To
further explore the SAR in this region, analogs 13h–s
were tested. It was found that a substitution at the posi-
tion 2 abolished the antiviral activity as observed for
13h, 13k, 13n, and 13q. All these analogs were inactive
against HIV replication up to 10 lM concentration.
Analogs possessing substituents at the 3 or 4 position
showed low micromolar to sub-micromolar antiviral
activities (13i, 13j, 13l, 13m, 13o, 13p, 13r, and 13s). It
is most likely that a substituent at the 2 position changes
the conformation between the C and D rings which led
to the diminished antiviral activity.
3-Methoxyphenyl
4-Methoxyphenyl
3-Trifluoromethoxyphenyl
2,5-Dimethoxyphenyl
7.587
2-Methoxy-5-methyl-phenyl 0.033
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
2-Methyl-3-chloro-phenyl
2-Chlorophenyl
3-Chlorophenyl
2-Fluorophenyl
3-Fluorophenyl
4-Fluorophenyl
2,6-Difluorophenyl
2-Bromophenyl
3-Bromophenyl
3-Trifluorophenyl
2-Methylphenyl
3-Methylphenyl
0.020
0.042
0.025
0.064
0.043
0.102
0.048
0.042
0.071
0.067
0.024
0.046
16g
16h
16i
16j
16k
16l
16m
16n
16o
16p
16q
S
Cl
16r
0.046
>10
Br
Me
Cl
N
16s
NVPa
0.053
0.050
>10
>10
N
Br
Further exploration of the SAR in this region led to
interesting results. Analogs possessing an H-bond
a NVP, nevirapine.

B. Wu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3430–3433
Table 3. Antiviral activity of selected pyrrolidinones against mutant HIV-1 viruses
3433
Compound
WT (lM)
K103N (lM)
Y181C (lM)
L100I (lM)
K103N/L100I (lM)
16a
16e
16f
0.025
0.033
0.020
0.042
0.025
0.064
0.043
0.048
0.042
0.071
0.067
0.024
0.046
0.046
0.050
0.0005
0.656 (26)a
0.172 (5.2)
0.162 (8.1)
0.170 (4)
0.990 (40)
0.507 (15)
0.389 (19)
0.557 (13)
2.320 (93)
1.496 (21)
0.684 (16)
1.305 (27)
0.794 (19)
2.263 (32)
2.163 (32)
0.246 (10)
0.284 (6.2)
0.753 (16)
>10 (>200)
0.001 (2)
0.033 (1.3)
0.023 (0.7)
0.008 (0.4)
0.019 (0.45)
0.040 (1.6)
0.034 (0.53)
0.020 (0.46)
0.068 (1.4)
0.016 (0.38)
0.053 (0.74)
0.091 (1.3)
0.010 (0.42)
0.014 (0.3)
0.039 (0.84)
0.164 (3.2)
0.010 (20)
0.076 (3)
0.033 (1)
0.010 (0.5)
0.028 (0.67)
0.056 (2.2)
0.041 (0.64)
0.026 (0.6)
0.052 (1)
16g
16h
16i
0.689 (27)
0.269 (4.2)
0.261 (6)
16j
16l
0.451 (9.4)
0.292 (6.9)
0.923 (13)
1.157 (17)
0.117 (4.9)
0.158 (3.4)
0.542 (12)
5.053 (101)
0.032 (64)
16m
16n
16o
16p
16q
16r
0.022 (0.52)
0.083 (1.1)
0.148 (2.2)
0.017 (0.7)
0.019 (0.41)
0.095 (2)
NVPb
EFVc
4.386 (88)
2.526 (5000)
a The numbers in parentheses are the fold changes compared to wild type.
b NVP, nevirapine.
c EFV, efavirenz.
W. A.; Sardana, V. V.; Long, W. J.; Byrnes, V. W.; Emini,
E. A. Antimicrob. Agents Chemother. 1995, 39, 2602.
6. HEK 293T cells are routinely cultured in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with
10% FBS, 1· Pen/Strep/Glutamine. The protocol is as
follows: 293T cells are seeded in the 1536-well format at
700 cells/well (5 lL volume) using an Aquamax (Molec-
ular Devices) liquid dispenser. Cells are cultured at 37 ꢁC
under 5% CO2 for 24 h. 50 nL of each compound (serially
diluted in DMSO) are transferred using the PinTool
(GNF). After a 1 h at 37 ꢁC incubation, HIV reporter
virus is transferred to the cells using the Aquamax in a
volume of 2 lL corresponding to a multiplicity of infec-
tion (MOI) of approximately 1.0. The treated and infected
cells are cultured for an additional 48 h at 37 ꢁC. Lucif-
erase activity is monitored by addition of Bright-Glo
(Promega, Cat. # E263B and E264B) luciferase reagent
(5 lL/well, Aquamax) followed by plate reading on the
CLIPR apparatus (Molecular Devices) using a 20 s shuttle
speed.
NNRTI-containing regimen. In general, with the excep-
tion of 16a, 16h and 16o, these compounds only suffered
moderate loss of activity against K103N, L100I and
K103N/L100I mutant viruses compared to NVP and
EFV. Most of the compounds lost significant activity
(10- to 90-fold) against the Y181C mutant virus. How-
ever, 16q was only 6-fold less potent against the
Y181C virus while remaining much of its activity against
the other mutant viruses. For comparison, NVP suffered
a greater loss in potency against 4 of 5 mutant viruses
while EFV exhibited a greater than 5000-fold loss in
potency against the K103N/L100I double mutant virus
(Table 3).
In summary, a novel class of NNRTI was discovered
with potent antiviral activity. The SAR of the D ring
region was established. This class of pyrrolidinones
demonstrated distinct antiviral activity profiles against
mutant viruses. In particular, it is worthy of note that
analogs in this class showed potent antiviral activity
against the K103N and K103N/L100I mutant viruses,
which are prevalent in patients failing NNRTI regimens.
Further investigations shall be focused on improving the
antiviral activities of this class of compounds against
WT virus as well as additional drug-resistant viruses.
7. Mulzer, J.; Zuhse, R.; Schiechen, R. Angew. Chem. Int.
Ed. Engl. 1992, 31, 870.
8. Klapars, A.; Hunang, X.; Buchwald, S. L. J. Am. Chem.
Soc. 2002, 124, 7421.
1
9. Analytic data of 16a: H NMR (CDCl3): d 7.70 (1H, br
s), 7.37 (1H, d, J = 7.8 Hz), 7.19–7.29 (4H, m), 7.00 (2H,
t, J = 7.6 Hz), 6.83 (1H, d, J = 8 Hz), 6.79 (1H, s), 6.79
(1H, d, J = 8 Hz), 4.76 (1H, br s), 4.11 (1H, t,
J = 8.4 Hz), 3.83 (3H, s), 3.78 (1H, t, J = 8.4 Hz), 3.60
(1H, m), 3.02 (1H, q, J = 8.1 Hz), 2.85 (1H, q,
J = 8.1 Hz), 1.61–1.88 (8H, m) ppm; LCMS m/z:
537.20 [M+H]+.
References and notes
1. Palella, F. J.; Delaney, K. M.; Moorman, A. C.; Loveless,
M. O.; Fuhrer, J.; Satten, G. A.; Aschiman, D. J.;
Holmberg, S. D. N. Engl. J. Med. 1998, 338, 853.
2. Esnouf, R.; Ren, J. S.; Ross, C.; Jones, Y.; Stammers, D.;
Stuart, D. Nat. Struct. Biol. 1995, 2, 303.
10. 293T cells are seeded in the 1536-well format at 700 cells/
well (5 lL volume) using an Aquamax (Molecular Devic-
es) liquid dispenser. Cells are cultured at 37 ꢁC under 5%
CO2 for 24 h. 50 nL of each compound (serially diluted in
DMSO) is transferred using the PinTool (GNF). The
treated and uninfected cells are cultured for an additional
48 h at 37ꢁC. Cell viability is assessed by addition of 1 lL
of Alamar Blue (Promega, Cat.# 00-100) diluted 1:1 in
DMEM. Cells are further incubated for 4 h at room
temperature and subsequent fluorescence intensity is read
using an Acquest (TREK systems) with a 50/50 beam
splitter.
3. Koup, R. A.; Merluzzi, V. J.; Sullivan, J. L. J. Infect. Dis.
1991, 163, 966.
4. Freimuth, W. W. Adv. Exp. Med. Biol. 1996, 394, 279.
5. Young, S. D.; Britcher, S. F.; Tran, L. O.; Payne, L. S.;
Lumma, W. C.; Lyle, T. A.; Huff, J. R.; Anderson, P. S.;
Olsen, D. B.; Carroll, S. S.; Pettibone, D. J.; O’Brien, J.
A.; Ball, R. G.; Balani, S. K.; Lin, J. H.; Chen, I.; Schleif,