Novel dimeric aryldiketo containing inhibitors of HIV-1 integrase: Effects of the phenyl substituent and the linker orientation

Article abstract of DOI:10.1016/j.bmc.2008.07.008

Aryl diketoacids (ADK) and their bioisosteres are among the most promising HIV-1 integrase (IN) inhibitors. Previously, we designed a series of ADK dimers as a new class of IN inhibitors that were hypothesized to target two divalent metal ions on the active site of IN. Herein we present a further structure-activity relationship (SAR) study with respect to the substituent effect of the ADK and the dimerization with conformationally constrained linkers such as piperazine, 4-amino-piperidine, piperidin-4-ol, and trans-cyclohexan-1,4-diamine. The substituents on the phenyl ring as well as the spatial orientation of the two diketo units were observed to play important roles in the IN inhibitory potency. The hydrophobic group was an optimal substitution at the 3-position of the aryl ring. The piperazine and 4-amino-piperidine linkers brought about the most potent analogs among the hydrophobic group or halogen substituted ADK dimers. The docking studies suggested that the bulky hydrophobic substitution at 3-phenyl ring and the linker of 4-amino-piperidine were beneficial for adopting an active conformation to achieve strong interactions with the active site Mg²⁺ and the key residue E152 within the catalytic core domain. This study is a significant extension of our previous report on the dimeric ADK-containing IN inhibitors, providing a new promising template for further lead optimization.

Full text of DOI:10.1016/j.bmc.2008.07.008

Bioorganic & Medicinal Chemistry 16 (2008) 7777–7787
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel dimeric aryldiketo containing inhibitors of HIV-1 integrase:
Effects of the phenyl substituent and the linker orientation
Li-Fan Zeng a, Xiao-Hua Jiang a, Tino Sanchez b, Hu-Shan Zhang a,
a,
Raveendra Dayam b, Nouri Neamati b, , Ya-Qiu Long
*
*
a State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences,
555 Zuchongzhi Road, Shanghai 201203, China
b Department of Pharmacology and Pharmaceutical Sciences, School of Pharmacy, University of Southern California, 1985 Zonal Avenue, Los Angeles, CA 90089, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Aryl diketoacids (ADK) and their bioisosteres are among the most promising HIV-1 integrase (IN)
inhibitors. Previously, we designed a series of ADK dimers as a new class of IN inhibitors that were
hypothesized to target two divalent metal ions on the active site of IN. Herein we present a further
structure–activity relationship (SAR) study with respect to the substituent effect of the ADK and the
dimerization with conformationally constrained linkers such as piperazine, 4-amino-piperidine,
piperidin-4-ol, and trans-cyclohexan-1,4-diamine. The substituents on the phenyl ring as well as the
spatial orientation of the two diketo units were observed to play important roles in the IN inhibitory
potency. The hydrophobic group was an optimal substitution at the 3-position of the aryl ring. The piper-
azine and 4-amino-piperidine linkers brought about the most potent analogs among the hydrophobic
group or halogen substituted ADK dimers. The docking studies suggested that the bulky hydrophobic
substitution at 3-phenyl ring and the linker of 4-amino-piperidine were beneficial for adopting an active
conformation to achieve strong interactions with the active site Mg2+ and the key residue E152 within the
catalytic core domain. This study is a significant extension of our previous report on the dimeric
ADK-containing IN inhibitors, providing a new promising template for further lead optimization.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 29 May 2008
Revised 30 June 2008
Accepted 1 July 2008
Available online 8 July 2008
Keywords:
HIV-1 integrase inhibitors
Aryl diketoacid
Strand transfer
Dimer
Conformationally constrained linker
Orientation
Substitution effect
1. Introduction
30-processed proviral DNA into host DNA occurs in a ‘strand trans-
fer’ reaction.
Human immunodeficiency virus type 1 (HIV-1) encodes three
enzymes that are required for viral replication: reverse transcrip-
tase, protease, and integrase (IN). Although drugs targeting reverse
transcriptase and protease are in wide use and have shown effec-
tiveness particularly when employed in combination,1 the search
for new anti-HIV-1 agents continues to develop drugs capable of
overcoming toxicity and resistance by targeting alternate sites in
the viral life cycle.2 IN has emerged as an attractive target, because
it is required for stable infection and has no cellular homologues.3
The function of IN is to catalyze integration of proviral DNA, result-
ing from the reverse transcription of viral RNA, into the host gen-
ome. This is achieved in a stepwise fashion by endonucleolytic
processing of proviral DNA (termed 30-processing) within a cyto-
plasmic preintegration complex, followed by translocation of the
complex into the nuclear compartment where integration of
IN is composed of three structural and functional domains, each
capable of forming a dimer.3 The N-terminal domain (residues
1–50) contains the conserved HHCC motif that binds one atom of
zinc. This region is involved in protein–protein interaction and
contributes to the specific recognition of viral DNA ends. The core
or catalytic domain (residues 50–212) contains the highly con-
served DDE motif (D64, D116, and E152) present in all retroviral
integrases. Mutation of any one of these three acidic residues abol-
ishes IN enzymatic activities and viral replication. The two D64 and
D116 residues form a coordination complex with a divalent metal
(Mg2+ or Mn2+). Because a second metal has been observed in an
ASV integrase crystal structure,4 and because of the two-metal
structure for polynucleotide transferases,5 it has been proposed
that a second metal (Mg2+ or Mn2+) can be coordinated between
D116 and E152 once IN binds its DNA substrate(s).6,7 It is therefore
likely that the metal(s) coordinate(s) IN and the phosphodiester
backbone of the DNA substrate(s) during the 30-processing and
strand transfer steps. The C-terminal domain (residues 212–280)
is the least conserved IN domain. It is involved in DNA binding
and IN oligomerization, which seems necessary for the integration
* Corresponding authors. Tel.: +1 323 442 2341; fax: +1 323 442 1390 (N.N.); tel./
fax: +86 21 50806876 (Y.-Q.L.).
E-mail addresses: neamati@usc.edu (N. Neamati), yqlong@mail.shcnc.ac.cn
(Y.-Q. Long).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.07.008

7778
L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
process. These three domains of IN are required for in vitro 30-end
processing and strand transfer.
Since the ADK dimer represents a new class of IN inhibitors by
targeting two divalent metal ions on the active site of IN, and the
emergence of viral strains resistant to clinically studied IN inhibi-
tors and the dynamic nature of the HIV-1 genome demand a con-
tinued effort toward the discovery of novel inhibitors, we were
intrigued to design second generation ADK dimer class of IN inhib-
itors focused on the orientation of various cyclic linkers. Pipera-
zine, 4-amino-piperidine, piperidin-4-ol, and trans-cyclohexan-
1,4-diamine were employed as the dimerization linker to explore
the geometric requirements for optimal IN inhibitory activity.
The substituent effect of the phenyl ring was examined as well,
since previous study indicated that the aromatic portion was
responsible for the strand transfer selectivity.13
In order to better understand the role of different substituents
and linker orientation on activity of the ADK dimer, first, positions
2, 3, and 4 of the phenyl moiety in the aryl diketoacid were initially
explored to determine optimal substituent. Second, the inhibitory
activity of the corresponding b-diketo ester or amide was evalu-
ated in order to establish a coherent SAR among these compounds.
Third, various cyclic linkers are investigated to determine the opti-
mal positioning of the two diketo subunits. These structural mod-
ifications led to compounds 1–13 (a,b) and 1,3,4,9–11 (c,d,e,f)
(Fig. 2). The present article is focused on the synthesis and SAR
analysis of cyclic diamino-linked ADK dimers bearing electron-
donating or electron-withdrawing group to address additional ele-
ments of structure–inhibitory activity.
Among an impressive number of synthetic IN inhibitors re-
ported, aryldiketo (ADK) containing compounds are the most
promising that potently block integration in extracellular assays
and exhibit good antiviral activity against HIV-infected cells.8–10
ADK-based inhibitors show selective inhibition of the strand trans-
fer reaction relative to the 30-processing step and may function by
competing with host DNA in binding to the IN proviral DNA com-
plex.11,12 However, the variation on the aromatic portion was
shown to alter the mode of action leading to reduced strand trans-
fer selectivity.13 Several IN inhibitors containing the aryl diketo
motif have been tested in the clinic (Fig. 1).14 The first FDA-ap-
proved IN inhibitor as anti-HIV drug, MK-0518 is also derived from
the aryl diketo scaffold.15
Based on the X-ray crystal structure of the IN core domain
complexed with an ADK-containing inhibitor, 5CITEP,16 we de-
signed and synthesized a series of novel dimeric diketo contain-
ing inhibitors of IN with the notion that such dimeric
compounds may simultaneously bind to two divalent metal ions
on the active site of IN.17 The two diketo subunits separated by
uniquely designed linkers can potentially chelate two metal ions
that are either provided from one IN active site or two active
sites juxtaposed together in a higher order tetramer. Indeed,
these aryl diketoacid dimers are highly potent against purified
IN with varied selectivity for strand transfer, and that some of
the analogues exert potent inhibition on the cytopathic effect
of HIV-1 in infected CEM cells.17 The modeling studies revealed
that these dimeric diketo containing inhibitors of IN function
by targeting two divalent metal ions on the active site of IN. Fur-
thermore, the dimerization linker was observed to modulate the
activity of the resulting dimeric aryldiketo containing IN inhibi-
tors by means of securing the distance and orientation of the
two separated diketo moieties. Among the linear and cyclic
diamino linkers studied, the piperazine appeared to be the best
linker for the amide-linked ADK dimers.
2. Synthesis
The monomer was conveniently synthesized by using our previ-
ously developed methodology.18 As shown in Scheme 1, differently
substituted acetophenones were oxalylated by dimethyl oxalate in
the presence of sodium tert-butoxide, affording b-diketo esters
1–12a in high yields. Hydrolysis of the methyl esters by aqueous
sodium hydroxide produced b-diketo acids 1–12b with various
substituent on the phenyl ring.
Figure 1. The structures of selected ADK-based clinically studied IN inhibitors.
Figure 2. The structures of aryl diketoacid derivatives with variations on the substitution of the aromatic ring, and the linker.

L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
7779
Scheme 1. Reagents and conditions: (i) t-BuONa, (CH3OOC)2, THF, 0 °C–rt; (ii) dioxane, 1 N NaOH, rt; (iii) HOAT, HATU, DIPEA, diamine, DMF or a—SOCl2, CH2Cl2, a drop of
DMF, 0 °C–rt; b—pyridine, diamine or 4-hydroxyl-piperidine, CH2Cl2.
Except for the piperidin-4-amino-linked dimer which required
a separate double coupling (Scheme 2), the synthesis of ADK di-
mers employing cyclohexane-1,4-diamine, piperazine, or piperi-
din-4-ol as the linker was accomplished by coupling the
monomer 1–12b with the cyclic diamine in the presence of HOBt
and EDCI or SOCl2 and cat. DMF, affording the corresponding
amide-linked dimers 1–12c,e,f (Scheme 1).
As depicted in Scheme 2, the similar coupling method was
applied to the synthesis of piperidin-4-amino-linked dimer
1–12d in a stepwise fashion. Since the linker of 4-amino-piperidine
was prepared in an N1-Boc protected form (compound 14), the cou-
pling was conducted twice separated by a deprotection reaction.
The linkers of 4-amino-piperidine and 4-hydroxypiperidine
were prepared from the common intermediate of tert-butyl
4-oxo-piperidine-1-carboxylate 15 via
a different reduction
approach, as shown in Scheme 3. The N-protection of the commer-
cially available piperidine-4,4-diol hydrochloride with Boc readily
yielded the key intermediate 15, which was converted to tert-butyl
4-amino-piperidine-1-carboxylate 14 by palladium-catalyzed
hydrogenation in the ammonia saturated methanol, or tert-butyl
4-hydroxypiperidine-1-carboxylate by the sodium boron hydride
reduction.
It is worth describing separately the synthesis of the aryl diketo
acid 13b bearing a bulky hydrophobic moiety (Scheme 4). The
Scheme 2. Reagents and conditions: (i) HOAT, HATU, DIPEA, DMF or a—SOCl2, CH2Cl2, a drop of DMF, 0 °C–rt; b—pyridine, alkylamine, CH2Cl2, rt; (ii) TFA, CH2Cl2, rt.
Scheme 3.
Scheme 4. Reagents and conditions: (i) HBTU, HOBT, DIPEA, CH2Cl2, 0 °C–rt; (ii) t-BuONa, (CH3OOC)2, THF–DME, 0 °C–rt; (iii) THF/MeOH, 1 N NaOH, rt.

7780
L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
4-pentylbicyclo[2.2.2]octane-1-carboxylic acid was coupled with
3-amino-acetophenone to give the precursor of N-(3-acetylphe-
nyl)-4-pentylbicyclo[2.2.2]octane-1-carboxamide 18, which under-
went the oxalylation and hydrolysis as described previously to
afford the desired 2,4-dioxo-4-(3-(4-pentylbicyclo[2.2.2]octane-1-
carboxamido)phenyl)butanoic acid 13b.
current study is complementary to this report in exploring a wider
range of the substituent structure on the phenyl portion in the con-
text of ADK dimers.
In this study, of the 3-, 4-, or 3,4-disubstituted phenyl diketoac-
ids, the best inhibitory activity was exhibited by compounds 1b,
2b, and 6a, which carry a bulky and hydrophobic group at 3-posi-
tion of the phenyl ring. The 3-benzyloxy substituent on the phenyl
moiety displayed highest affinity-enhancing effect (compounds 1b
3. Results and discussion
and 2b, IC50 < 0.4 l
M for strand transfer reaction). The 3-(20-fluo-
robenzyl) group and 3-benzyloxycarbonylamino group are favored
for the strand transfer inhibition for the same reason (IC50 = 0.6
and 7 lM for compounds 6a and 5a, respectively). Furthermore,
3.1. Hydrophobic group of the phenyl substitution is beneficial
for the interaction of the ADK compound with the IN enzyme
the hydrophobic substituent only at 3-position slightly increased
The substitution of the phenyl portion within the ADK structure
was examined first. As shown in Table 1, all aryl diketo containing
compounds showed preferential inhibition of strand transfer ver-
sus 30-processing (compounds 1–13a,b), and the methyl ester form
was significantly less potent than the free acid congener (com-
pounds 1a vs 1b, 2a vs 2b, 3a vs 3b), in which the carboxyl hydro-
xyl group might be involved in the binding with the host DNA site
within IN. This observation is consistent with previous find-
ings.19,20 Although ADK inhibitors are generally characterized by
high strand transfer selectivity, when the phenyl ring was substi-
tuted by an additional 2,4-diketobutyric acid group at the 3- or
4-position, the resulting bis-diketo acid analogues displayed en-
hanced 30-processing inhibitory potencies, such that selectivity to-
ward strand transfer was remarkably decreased.20 Literature study
has examined the aspects of both the left and right portions of the
b-diketopropyl central chain, with focus on the substitution of the
whole aromatic moiety with phenyl, indolyl, or quinolyl ring.20 Our
the 30-processing inhibitory potency, resulting in some loss of
strand transfer selectivity (IC50 = 19 and 26 l
M against 30-process-
ing for compounds 1b and 6a, respectively). However, an addi-
tional hydrophobic group substituted at 4-position restored the
strand transfer selectivity to some extent with retention of the sig-
nificant strand transfer inhibitory potency (3-benzyloxy-4-meth-
oxy 2b, IC50 = 55 and <0.4 lM
for 30-processing and strand
transfer, respectively; 3-methyl-4-chloro 8b, IC50 = 94 and 4.7 lM
for 30-processing and strand transfer, respectively). But introduc-
tion of a bulky hydrophobic group such as dibenzylamino at 3-po-
sition caused a significant loss of strand transfer inhibitory activity
(7a, IC50 > 100 lM), which might be due to steric hindrance. Prob-
ably for a similar reason, the bulky hydrophobic 4-pentylbicy-
clo[2.2.2]octane-1-carboxamido group (13b) substituted at 3-
position of the phenyl ring conferred only moderate activity
against strand transfer reaction (13b, IC50 = 14 lM). The following
Table 1
Inhibitory potencies of the aryl diketoacid derivatives against the HIV-1 integrasea
Compound
R
X
Inhibition of integrase catalytic activities
30-Processing IC50
(lM)
Integration IC50 (lM)
1a
1b
2a
2b
3a
3b
3-OBn
OCH3
OH
OCH3
OH
OCH3
OH
>100
14
<0.4
16
<0.4
>100
58
19
5
3-OBn, 4-OMe
2-F
>100
55
6
>100
>100
6
4a
4g
4-F
OCH3
>100
>100
>100
100
5a
6a
7a
8b
3-NHCbz
3-(20-Fluorobenzyl)
3-NBn2
3-Me, 4-Cl
2-Cl
4-CF3
3-OMe
4-NO2
OCH3
OCH3
OCH3
OH
OH
OH
99
26 11
>100
7
0.6
>100
4.7
17
33 12
94
6
9b
>100
>100
>100
>100
5
10b
11b
12b
OH
OH
27
45
6
4
13b
OH
OH
>100
>100
14
1
Reference compoundb
H
24.2, 25
a
Values were determined as described in Section 6.
From Ref. 20.
b

L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
7781
molecular modeling indicated that the crowded substitution on the
aryl ring would hamper the interaction of the aromatic portion
with the binding cavity.
analogues, all newly synthesized analogues were docked into the
active site of IN using GOLD program (CCDC, Cambridge, UK).21
Interestingly, ADK monomers with
a free carboxylate group
Accordingly, the electron-withdrawing group or polar group
was deleterious to the IN inhibition. The replacement of the ben-
zyloxy group with a trifluoromethyl (10b) or a nitro group (12b)
at 4-position reduced strand transfer potency by approximately
100-folds. The halogen played a different role, among which the
fluoro group at 2- (3a,b) or 4-position (4a,b) was disfavored, while
the chloro group at 2-position gave a slight increase in strand
adopted a similar binding conformation. Herein taken 2b, one of
the most potent ADK analogues, as an example. As shown in Figure
3B, the diketoacid group of compound 2b is interacting with the
active site Mg2+. The b-keto (C@Oꢀ ꢀ ꢀMg2+ 2.15 Å) and enol (C–
OHꢀ ꢀ ꢀMg2+ 1.85 Å) oxygens formed strong coordinate bonds with
the active site Mg2+. The hydrophobic benzyloxy substituent on
3-position of the aryl ring occupies a cavity surrounded by E152,
N155, K156, and K159.
transfer inhibitory activity (9b, IC50 = 17
lM) compared to the
unsubstituted congener (reference compound, IC50 = 25
l
M).
For the cyclic diamine-linked dimers, the 4-amino-piperidine
linker containing compound 1d adopted an interesting binding
mode inside the active site of IN. One of the aryl b-keto units of
the dimeric ADK 1d established strong interactions with the active
site Mg2+ (Fig. 3C). Similar to compound 2b, the hydrophobic ben-
zyloxy substituent on 3-position of the aryl ring of compound 1d
occupies a cavity surrounded by E152, N155, K156, and K159.
Additionally, the 4-amino-piperidine linker and the second aryl
b-keto unit of dimeric ADK occupies an area surrounded by amino
acid residues E92, T93, N117, S119, and T122. Interestingly, inac-
tive compound 9e adopted a very different bound conformation in-
side the active site of IN. Due to the absence of bulky hydrophobic
substituent at 3-position of the phenyl ring, the compound occu-
pied a large cavity inside the active site near the Mg2+. Conse-
quently, contrary to active compounds 2b and 1d, compound 9e
3.2. The constrained cyclic linker potentiates the IN inhibition
of the ADK compounds
In order to investigate the orientation effect of the linker on the
ADK inhibition against IN, three cyclic diamines and the 4-
hydroxypiperidine were examined as linkers to construct the
ADK dimers. As shown in Table 2, the resulting amide-linked con-
strained ADK dimers are more potent than the corresponding
monomers. Of note, the blockade of the carboxy hydroxyl of the
ADK dimer by conversion to its amide was accompanied with loss
of potency, so the activity comparison was based on the ADK ester
monomer instead of the free acid. In this regard, all conformation-
ally constrained amide-linked ADK dimers gained an enhancement
of the IN inhibitory activity, but the potentiating effect on the 30-
processing inhibitory potency is not as significant as the bis-diketo
acid analogues which bear two free carboxyate groups.13,19 They
may function via a different mechanism.13,17
did not make an interaction with active site Mg2+
.
The predicted binding modes of the active and inactive mono-
meric and dimeric ADK analogues demonstrate that the b-diketo-
acid part interacts with the active site Mg2+, while the inactive
dimer lacks such interactions. The well tolerated and potency-
enhancing hydrophobic substituent on the 3-position of the aryl
ring occupies a cavity close to one of the catalytically important
residues E152. The bound conformation of compound 9e indicates
the importance of a bulky substitution on 3-position of the aryl
ring. The predicted binding modes also reveal that further bulky
substitutions on the aryl ring other than 3-position could hamper
potency of the compounds.
For the most active ADK compound in this series, that is, (Z)-4-(3-
(benzyloxy)phenyl)-2-hydroxy-4-oxobut-2-enoic acid (1b), the 4-
amino-piperidine (1d) displayed the best potency among the four
cyclic linkers (1c–f). Compared to the (Z)-methyl 4-(3-(benzyl-
oxy)phenyl)-2-hydroxy-4-oxobut-2-enoate (1a, IC50 = 14 and
>100 l
M for strand transfer and 30-processing, respectively), the 4-
amino-piperidine linker achieved more than 35-fold enhancement
in the strand transfer inhibitory potency and meanwhile renders
the 30-processing inhibition effective (1d, IC50 = <0.4 and 23
lM for
strand transfer and 30-processing, respectively). The second best lin-
5. Conclusions
ker was piperazine (1d), which conferred potent strand transfer
inhibition (IC50 = 1.8
inhibitory activity at moderate level (IC50 = 56
piperidine and cyclohexane-1,4-diamine with IC50 values of 3 and
M for strand transfer, respectively, displayed similar potency.
Examining the other conformationally restricted dimers
l
M) while maintaining the 30-processing
In this study, we investigated the substitution and orientation
effects of amide-linked ADK dimers as a novel class of IN inhibitors.
The hydrophobic group at 3-position of phenyl ring was important
for IN inhibition. The 4-amino-piperidine and piperazine behaved
as the best cyclic linkers for the amide-linked ADK dimers to
achieve potent inhibitory activity against 30-processing reaction.
Docking studies revealed that the conformationally constrained
linker of 4-amino-piperidine positioned the two ADK subunits
properly inside the binding pocket of IN to establish interactions
with the active site Mg2+ and the key residue E152 within the cat-
alytic core domain. This study is a significant extension of our pre-
vious report on the dimeric ADK-containing IN inhibitor, providing
promising new templates for further development of potent inhib-
itors directed toward different sites.
lM). The 4-hydroxy-
7
l
(3,4,9,10,11c–e), the order of the cyclic linker with respect to the
potentiating effect varied for variously substituted ADK. However,
the piperazine behaved as the best cyclic linker in most cases ex-
cept for 2-chloro substituted phenyl diketoacid analog. Since the
conformationally constrained linker positioned the two ADK units
into close proximity, the substituent on the phenyl ring might
bring about the repulsion with another functionality, which might
distort the orientation of the ADK unit. So the final spatial arrange-
ment of the two ADK moieties was decided by both the cyclic lin-
ker and the phenyl substitution. In the case of electron-donating
group or halogen substituted phenyl diketo acids, 4-amino-piperi-
dine and piperazine can achieve an optimal orientation for the two
ADK subunits into the IN binding cavity. These observations were
rationalized by docking studies as described below.
6. Experimental
6.1. General synthetic methods
Unless otherwise stated, the 1H NMR spectra were recorded on
a Varian 300-MHz or 400-MHz spectrometer. The data are reported
in parts per million relative to TMS and referenced to the solvent.
Elemental analyses were obtained using a Vario EL spectrometer.
Melting points (uncorrected) were determined on a Buchi-510
4. Docking studies
The above observations can be reasonably rationalized by
molecular modeling. To predict the binding conformation of ADK

7782
L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
Table 2
Inhibitory potencies of the aryl diketoacid dimers against the HIV-1 integrase catalytic activity
Compound
1c
R
X
Inhibition of integrase catalytic activities
30-Processing IC50
(l
M)
Strand transfer IC50
(lM)
3-OBn
>100
7
1d
23
6
<0.4
1.8
1e
56 25
91 40
>100
1f
3
1
3c
2-F
4-F
55 15
4c
>100
35.5 0.5
88 12
4d
>100
4e
>100
29
8
9c
2-Cl
92
99
9
2
30 10
31 12
>100
9d
9e
>100
>100
>100
>100
>100
10c
10e
11c
11d
4-CF3
41 16
25 14
3-OMe
26
7
53 15
11e
>100
26
8

L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
7783
Cl
Cl
O
O
O
OH
O
OH
HO
O
O
OH
H
N
O
O
OH
O
N
N
OH
O
N
O
N
O
NH
O
OH
O
N
Cl
N
O
HN
5-CITEP
2b
1d
9e
Figure 3. (A) The diketoacid bioisostere 5-CITEP bound to the core domain of HIV-1 integrase (1QS4ꢀPDB). (B) The predicted bound conformation of one of the highly active
aryl diketoacids (2b) inside the active site of HIV-1 integrase (1BISꢀPDB). (C) The predicted bound conformation of highly active dimeric aryl diketo analogue (1d) inside the
active site of HIV-1 integrase. (D) The predicted bound conformation of one of the inactive compounds (9e) inside the active site of HIV-1 integrase. The blue surface shows
the active site region of the HIV-1 integrase. Green surface represents catalytic triad DDE. The magenta surface represents active site Mg2+
.
capillary apparatus. IR spectra were recorded on Bio-Rad FTS-185
spectrometers. The MS and HRMS (ESI) spectra were obtained on
an APEXIII 7.0 T FTMS mass spectrometer. Degree of rotation was
measured by P-1030 (A012360639) automatic polarimeter. The
flash column chromatography was performed on silica gel H (10–
with the inhibitor in reaction buffer (50 mM NaCl, 1 mM HEPES,
pH 7.5, 50 M EDTA, 50 M dithiothreitol, 10% glycerol (w/v),
l
l
7.5 mM MnCl2, 0.1 mg/mL bovine serum albumin, 10 mM 2-
mercaptoethanol, 10% dimethyl sulfoxide, and 25 mM MOPS, pH
7.2) at 30 °C for 30 min. Then, 20 nM of the 50-end 32P-labeled lin-
ear oligonucleotide substrate was added, and incubation was con-
tinued for an additional 1 h. Reactions were quenched by the
40
lm). Anhydrous solvents were obtained by standard procedure.
Analytical HPLC for selected compounds were performed on two
systems, using Vydac C18 column (10 ꢁ 250 mm). UV detector,
254 nm. System A: Solvent system used was 0.05% TFA in methanol
and 0.05% TFA in water, gradient 30–90% over 20 min @ 2 mL/min.
System B: Solvent system used was 0.05% TFA in 95% acetonitrile
and 0.05% TFA in water, gradient 30–90% over 20 min @ 2 mL/min.
addition of an equal volume (16
formamide, 10 mM EDTA, 0.025% xylene cyanol and 0.025% bromo-
phenol blue). An aliquot (5 L) was electrophoresed on a denatur-
ing 20% polyacrylamide gel (0.09 M tris–borate, pH 8.3, 2 mM
EDTA, 20% acrylamide, 8 M urea).
lL) of loading dye (98% deionized
l
Gels were dried, exposed in a PhosphorImager cassette, and
analyzed using a Typhoon 8610 Variable Mode Imager (Amersham
Biosciences) and quantitated using ImageQuant 5.2. Percent inhibi-
tion (%I) was calculated using the following equation:
6.2. Materials, chemicals, and enzymes
All compounds were dissolved in DMSO and the stock solutions
were stored at ꢂ20 °C. The
c[
32P]ATP was purchased from either
%I ¼ 100 ꢁ ½1 ꢂ ðD ꢂ CÞ=ðN ꢂ CÞꢃ
Amersham Biosciences or ICN. The expression systems for the
wild-type IN and soluble mutant INF185KC280S were generous gifts
of Dr. Robert Craigie, Laboratory of Molecular Biology, NIDDK,
NIH, Bethesda, MD.
where C, N, and D are the fractions of 21-mer substrate converted to
19-mer (30-processing product) or strand transfer products for DNA
alone, DNA plus IN, and IN plus drug, respectively. The IC50 values
were determined by plotting the logarithm of drug concentration
versus percent inhibition to obtain concentration that produced
50% inhibition.
6.3. Preparation of oligonucleotide substrates
The oligonucleotides 21top, 50-GTGTGGAAAATCTCTAGCAGT-30
and 21bot, 50-ACTGCTAGAGATTTTCCACAC-30 were purchased from
Norris Cancer Center Core Facility (University of Southern Califor-
nia) and purified by UV shadowing on polyacrylamide gel. To ana-
lyze the extent of 30-processing and strand transfer using 50-end
labeled substrates, 21top was 50-end labeled using T4 polynucleo-
6.5. Docking studies
The structures of all the compounds (Tables 1 and 2) were built
and minimized using Catalyst (Accelrys, Inc.).22 All compounds
were modeled in their enol tautomeric form as this form is consid-
ered biologically relevant. Compounds with a free acid group were
modeled as carboxylates. The poling algorithm implemented with-
in Catalyst was used to generate conformations for all the com-
pounds.23 For each compound all feasible unique conformations
were generated over a 20 kcal/mol range of energies using the best
flexible conformation generation method in Catalyst.24,25 The sub-
unit B of the core domain X-ray structure of IN (PDB 1BIS) in which
all the active site amino acid residues were resolved was chosen for
docking purpose.26 A Mg2+ ion was placed in the active site be-
tween carboxylate oxygen atoms of amino acid residues D64 and
D116 considering the geometry of the Mg2+ ion that was present
tide kinase (Epicentre, Madison, WI) and
c[
32P]ATP (Amersham
Biosciences or ICN). The kinase was heat-inactivated and 21bot
was added in 1.5-molar excess. The mixture was heated at 95 °C,
allowed to cool slowly to room temperature, and run through a
spin 25 mini-column (USA Scientific) to separate annealed dou-
ble-stranded oligonucleotide from unincorporated material.
6.4. Integrase assays
To determine the extent of 30-processing and strand transfer,
wild-type IN was preincubated at a final concentration of 200 nM

7784
L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
in the subunit A of the IN in PDB 1BIS and subunit A in IN–5CITEP
complex X-ray structure (PDB 1QS4).16 All the water molecules
present in protein were removed, and hydrogen atoms were added
to the protein considering appropriate ionization states for both
the acidic and basic amino acid residues. Docking was performed
using version 3.2 of the GOLD: Genetic Optimization for Ligand
Docking (Cambridge Crystallographic Data Centre) software pack-
age.21,27,28 A 20 Å radius active site was defined considering the
carboxylate oxygen atom (OD1) of amino acid residue D64 as
the center of the active site. All the compounds were docked into
the active site of the IN. On the basis of the GOLD fitness score,
for each molecule, a bound conformation with high fitness score
was considered as the best bound conformation. All docking runs
were carried out using standard default settings with a population
size of 100, a maximum number of 100,000 operations, and a
mutation and crossover rate of 95. The fitness function that was
implemented in GOLD consisted basically of H-bonding, complex
energy, and ligand internal energy terms.
6.6.5. (Z)-tert-Butyl 4-(4-(4-fluorophenyl)-2-hydroxy-4-oxobut-
2-enamido)piperidine-1-carboxylate (4g)
Prepared by the same procedure as the preparation of 11g,
pale yellow solid, yield 96%. 1H NMR (400 MHz, CDCl3): d 7.95
(d, 1H, J = 7.8 Hz); 7.50 (m, 1H); 7.40 (m, 1H); 7.18 (s, 1H);
7.14–7.12 (m, 1H); 7.09 (s, 1H); 4.09–3.96 (m, 3H); 2.91 (m,
2H); 1.98–1.94 (m, 2H); 1.46 (s, 9H); 1.49–1.39 (m, 2H). EI-MS
(m/z, %): 392 (M+, 10.0). Analytical HPLC tR = 25.51 min, purity
100% (solvent system A); tR = 23.19 min, purity 99.9% (solvent
system B).
6.6.6. (Z)-4-(4-Chloro-3-methylphenyl)-2-hydroxy-4-oxobut-2-
enoic acid (8b)
White solid, yield 86.0%. 1H NMR (400 MHz, CDCl3 + CD3OD): d
7.89–7.79 (m, 1H); 7.70–7.67 (m, 1H); 7.40–7.28 (m, 1H); 7.10 (s,
1H); 2.38 (s, 3H). EI-MS (m/z, %): 242 (M+, 4.0); 240 (M+, 12.0);
197 (32.0); 195 (96.0); 155 (13.0); 153 (39.0); 69 (100.0). HR-EIMS
calcd for C11H9ClO4: 240.0189, found: 240.0185. Anal. Calcd for
C11H9ClO4: C 54.90, H 3.77. Found: C 55.18, H 4.30.
6.6. General procedure for the synthesis of monomeric aryl
diketo acids
6.6.7. (Z)-4-(2-Chlorophenyl)-2-hydroxy-4-oxobut-2-enoic acid
(9b)
To a stirred mixture of t-BuONa (2.5 equiv) and dimethyl oxa-
late (2.0 equiv) in anhydrous THF at 0 °C was added dropwise
aryl methyl ketone (1.0 equiv) in DME. The resulting orange-yel-
low mixture was stirred at room temperature for 1.5 h at most.
The reaction mixture was quenched with 1.0 mol/L aqueous
HCl and extracted with CH2Cl2. The combined organic layers
were washed with saturated aqueous NaHCO3 and brine, respec-
tively, dried over Na2SO4 and concentrated under vacuum. Purifi-
cation by silica gel flash chromatography provided the desired 4-
(substituted phenyl)-4-oxo-2-hydroxy-2-butenoic acid methyl es-
ters 1–13a.18 The subsequent hydrolysis of the resulting aryl
diketoacid methyl ester readily furnished the corresponding aryl
diketoacids 1–13b by 1 N NaOH in MeOH/THF (1:1) at room
temperature.
White solid, yield 85%. 1H NMR (400 MHz, CDCl3): d 7.70–7.68
(m, 1H); 7.51–7.46 (m, 2H); 7.42–7.38 (m, 1H); 7.11 (s, 1H). EI-
MS (m/z, %): 228 (M++2, 4.0); 226 (M+, 4.0); 191 (38.0); 183
(32.0); 181 (100.0); 141 (14.0); 139 (42.0). Anal. Calcd for
C10H7ClO4: C 53.00, H 3.11. Found: C 53.21, H 3.08.
6.6.8. (Z)-2-Hydroxy-4-oxo-4-(4-(trifluoromethyl)phenyl) but-
2-enoic acid (10b)
White needle solid, yield 87.0%. 1H NMR (400 MHz, CDCl3): d
8.11 (d, 2H, J = 8.0 Hz); 7.82 (d, 2H, J = 8.0 Hz); 7.20 (s, 1H). EI-MS
(m/z, %): 260 (M+, 3.0); 241 (4.0); 215 (100.0); 173 (45.0); 145
(37.0). HR-EIMS calcd for C11H7F3O4: 260.0296, found: 260.0301.
Anal. Calcd for C11H7 F3O4ꢀ0.2CH2Cl2ꢀ0.8CH3OH: C 47.60, H 3.53.
Found: C 47.33, H 3.34.
6.6.1. (Z)-4-(3-(Benzyloxy)phenyl)-2-hydroxy-4-oxobut-2-enoic
acid (1b)
6.6.9. (Z)-2-Hydroxy-4-(3-methoxyphenyl)-4-oxobut-2-enoic
acid (11b)
Pale yellow solid, yield 73.0%. 1H NMR (400 MHz, CDCl3): d
7.61–7.59 (m, 2H); 7.47–7.35 (m, 6H); 7.25–7.23 (m, 1H); 7.15 (s,
1H); 5.14 (s, 2H). EI-MS (m/z, %): 298 (M+, 3.0); 91 (100.0). Anal.
Calcd for C17H14O5: C 68.45, H 4.73. Found: C 68.52, H 4.38.
Pale yellow solid, yield (66.6%). 1H NMR (300 MHz, CDCl3): 7.59
(d, 1H, J = 8.0 Hz); 7.51 (s, 1H); 7.43 (d, 2H, J = 8.0 Hz); 7.18 (m,
1H); 7.15 (s, 1H); 3.89 (s, 3H). EI-MS (m/z): 222 (M+); Anal. Calcd
for C11H10O5: C 59.46, H 4.54. Found: C 59.06, H 4.32.
6.6.2. (Z)-4-(3-(Benzyloxy)-4-methoxyphenyl)-2-hydroxy-4-
oxobut-2-enoic acid (2b)
6.6.10. (Z)-2-Hydroxy-4-(4-nitrophenyl)-4-oxobut-2-enoic acid
(12b)
Pale yellow solid, yield 84%. 1H NMR (400 MHz, CDCl3 + CD3OD):
d 7.79–7.76 (m, 2H); 7.64–7.52 (m, 5H); 7.15 (d, 1H, J = 8.8 Hz); 7.01
(s, 1H); 5.44 (s, 2H); 4.16 (s, 3H). EI-MS (m/z, %): 328 (M+, 8.0); 91
(100.0). HR-EIMS calcd for C18H16O6: 328.0947, found: 498.0950.
Calcd for C18H16O6: C 65.85, H 4.91. Found: C 65.85, H 4.84.
Yellow solid, yield 75.0%. 1H NMR (400 MHz, CDCl3 + CD3OD): d
8.31 (d, 2H, J = 8.8 Hz); 8.12 (d, 2H, J = 8.8 Hz); 7.08 (s, 1H). EI-MS
(m/z, %): 237 (M+, 3.0); 192 (100.0); 150 (34.0). Anal. Calcd for
C10H7NO6ꢀ0.6H2Oꢀ0.4CH3OH: C 47.90, H 3.79, N 5.37. Found: C
47.97, H 3.60, N 4.90.
6.6.3. (Z)-4-(2-Fluorophenyl)-2-hydroxy-4-oxobut-2-enoic acid
(3b)
6.6.11. N-(tert-Butyloxycarbonyl)-4-piperidinol (16)
A stirred solution of the 4-oxo-piperidine-1-carboxylic acid tert-
butyl ester 15 (0.398 g, 2 mmol) in MeOH (15 mL) was cooled to
0 °C under nitrogen and then treated portionwise with sodium
borohydride (0.4 g, 10 mmol) over 0.5 h. The resulting solution
was stirred at room temperature for 5 h and then diluted with
H2O (20 mL). The resulting mixture was partitioned between ethyl
acetate (50 mL) and washed with brine (50 mL), and then the or-
ganic phase was dried over Na2SO4, filtered. After removal of the
solvent, the residue was purified by column chromatography
(petroleum/EtOAc = 3:1) to provide 0.378 g of 16 as white solid
in yield of 93.8%. 1H NMR (400 Hz, CDCl3): d 4.67 (s, 1H); 3.57–
3.67 (m, 3H); 2.94 (s, 2H); 1.63–1.69 (m, 2H); 1.40 (s, 9H); 1.18–
1.35 (m, 2H).
Pale yellow solid, yield 86%. 1H NMR (400 MHz, CDCl3): d 8.00–
7.95 (m, 1H); 7.61–7.55 (m, 1H); 7.32–7.27 (m, 1H); 7.21–7.18 (m,
2H). EI-MS (m/z, %): 210 (M+, 4.0); 165 (100.0); 123 (64.0). Anal.
Calcd for C10H7FO4: C 57.15, H 3.36. Found: C 57.21, H 3.36.
6.6.4. (Z)-4-(4-Fluorophenyl)-2-hydroxy-4-oxobut-2-enoic acid
(4b)
Pale solid, yield 88%. 1H NMR (300 MHz, CDCl3): d 8.04 (m, 2H);
7.21 (m, 2H); 7.13 (s, 1H). EI-MS (m/z, %): 210 (M+, 4.0); 165
(100.0); 123 (56.0). HR-EIMS calcd for C10H7FO4: 210.0328, found:
210.0324. Analytical HPLC tR = 22.79 min, purity 99.3% (solvent
system A); tR = 15.433 min, purity 95.4% (solvent system B).

L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
7785
6.6.12. 1,4-Hydroxypiperidine (17)
6.6.17. tert-Butyl 4-(4-(3-methoxyphenyl)-2-hydroxy-4-
At 0 °C, trifluoroacetic acid (1 mL) was added dropwise to the
solution of 16 (0.360 g,1.8 mmol) in dry CH2Cl2 (3 mL), and then
the solution was stirred at room temperature for 1 h, the reaction
was monitored by TLC. After removal of the solvent, the residue
was treated with 2 N NaOH aq solution, extracted with CH2Cl2
(3ꢁ 30 mL), washed with saturated brine (20 mL), then the result-
ing organic phase was dried over anhydrous Na2SO4, concentrated
to an oil residue 17 (0.175 g), which was used directly for the next
step.
oxobut-2-enamido)piperidine-1-carboxylate (11g)
HATU (183 mg, 0.48 mmol), HOAT (66 mg, 0.48 mmol), N-Boc-
4-amino-piperidine (80 mg, 0.40 mmol) and 4-(3-methoxy-
phenyl)-2,4-dioxo-butyric acid (94 mg, 0.42 mmol) were dissolved
in 4.0 mL THF at 0 °C and stirred for 10 min, then DIPEA (0.084 mL,
0.48 mmol) was added to the mixture. The reaction mixture was
stirred at 0 °C for 2 h, then warm to room temperature and stirred
overnight. The reaction mixture was concentrated in vacuo, and
the residue was purified by chromatography using CHCl3/CH3OH
(40:1) as eluent to give the product 11g as pale yellow solid
6.6.13. N-(3-Acetylphenyl)-4-pentylbicyclo[2.2.2]octane-1-
carboxamide (18)
(152 mg, 94.0%). 1H NMR (400 MHz, CDCl3):
d 7.59 (d, 1H,
J = 7.8 Hz); 7.50 (m, 1H); 7.40 (m, 1H); 7.18 (s, 1H); 7.14–7.12
(m, 1H); 7.09 (s, 1H); 4.09–3.96 (m, 3H); 3.87 (s, 3H); 2.91 (m,
2H); 1.98–1.94 (m, 2H); 1.46 (s, 9H); 1.49–1.39 (m, 2H). EI-MS
(m/z, %): 404 (M+, 9.0); 348 (2.0); 304 (2.0); 177 (42.0); 82
(100.0). HR-EIMS calcd for C21H28N2O6: 404.1958, found:
404.1947.
HBTU (0.759 g, 2.0 mmol), HOBT (0.270 g, 2.0 mmol), 3-amino-
acetophenone (0.135 g, 1.0 mmol) and 4-pentylbicyclo[2.2.2]octane-
1-1 carboxylic acid (0.336 g, 1.5 mmol) were dissolved in CH2Cl2
(10 mL) at 0 °C and stirred for 10 min, then DIPEA (0.873 mL,
5.0 mmol) was added to the mixture. The reaction mixture was
stirred at 0 °C for 2 h, warmed to room temperature and stirred
overnight. The reaction mixture was concentrated in vacuo, and
the residue was purified by chromatography using PE/EtOAc
(5:1) as eluent to give compound 18 as white solid (0.183 g,
36%). 1H NMR (400 MHz, CDCl3): d 8.03 (t, 1H, J = 1.8 Hz); 7.87
(ddd, 1H, J = 1.0, 2.2, 8.1 Hz); 7.67 (ddd, 1H, J = 1.0, 1.7, 7.7 Hz);
7.40 (t, 1H, J = 7.9 Hz); 2.59 (s, 3H); 1.87–1.83 (m, 6H); 1.47–1.44
(m, 6H); 1.33–1.08 (m, 8H); 0.88 (t, 3H, J = 7.1 Hz).
6.6.18. 4-(3-Methoxyphenyl)-N-{1-[4-(3-methoxyphenyl)-2,4-
dioxo-butyryl]-piperidin-4-yl}-2,4-dioxo-butyramide (11d)
To the solution of compound 11g (0.1 g, 0.247 mmol) in meth-
ylene chloride (4.0 mL) was added trifluoroacetic acid (2.0 mL).
The mixture was stirred at room temperature for 2 h. After the
removal of the solvent and trifluoroacetic acid under reduced
pressure, 2 N NaOH was added and extractive work up with
EtOAc. The organic layer was washed with saturated NaHCO3,
brine and dried over Na2SO4 and the solvent was removed
under vacuum to give a solid. The solid above, HATU (113 mg,
0.297 mmol), HOAT (41 mg, 0.297 mmol), 11b (75 mg, 0.247
mmol) and 4-(3-methoxyphenyl)-2,4-dioxo-butyric acid (58 mg,
0.26 mmol) were dissolved in 2.5 mL THF at 0 °C and stirred for
10 min, then DIPEA (0.053 mL, 0.297 mmol) was added to the
mixture. The reaction mixture was stirred at 0 °C for 2 h, then
warm to room temperature and stirred overnight. The reaction
mixture was concentrated in vacuo, and the residue was purified
by chromatography using CHCl3/CH3OH (40:1) as eluent to give
the product 11d as glassy solid (70 mg, yield 56.0%). 1H NMR
(300 MHz, CDCl3): d 7.57–7.54 (m, 1H), 7.47 (m, 2H), 7.44–7.36
(m, 3H), 7.24–7.22 (m, 1H), 7.16 (s, 1H) 7.09 (dt, J = 1.8, 15.9,
2H), 6.53 (s, 1H), 4.53 (m, 1H), 4.08 (m, 2H), 3.83 (s, 3H), 3.80
(s, 3H), 3.25 (t, J = 11.7, 1H), 2.94 (t, J = 11.7, 1H), 2.07–2.02 (m,
2H), 1.63–1.55 (m, 2H), 1.25–1.21 (m, 1H). ESI-MS m/z: 508.2
(M)+. Anal. (C27H28O6N2): C 63.77, H 5.55, N 5.51. Found: C
63.53, H 5.70, N 5.51.
6.6.14. Methyl2,4-dioxo-4-(3-(4-pentylbicyclo[2.2.2]octane-1-
carboxamido)phenyl)butanoate (13a)
Compound 13a was prepared according to the general method,
white solid (0.147 g, 79%). 1H NMR (400 MHz, CDCl3): d 15.20 (br,
1H); 8.03 (m, 1H); 7.93 (m, 1H); 7.71 (m, 1H); 7.44 (t, 1H,
J = 1.8 Hz); 7.37 (s, 1H); 7.06 (s, 1H); 3.94 (s, 3H); 1.88–1.84 (m,
6H); 1.49–1.42 (m, 6H); 1.33–1.01 (m, 8H); 0.88 (t, 3H,
J = 7.0 Hz). EI-MS (m/z, %): 427 (M+, 18.0); 368 (10.0); 341 (8.0);
179 (100).
6.6.15. 2,4-Dioxo-4-(3-(4-pentylbicyclo[2.2.2]octane-1-
carboxamido)phenyl)butanoic acid (13b)
Compound 13b was prepared as white solid (0.077 g, yield
79.0%) according to the general method. 1H NMR (400 MHz,
CDCl3): d 8.14 (s, 1H); 7.83 (d, 1H, J = 7.8 Hz); 7.73 (d, 1H,
J = 7.8 Hz); 7.46 (m, 2H); 7.15 (s, 1H); 1.89–1.85 (m, 6H); 1.49–
1.45 (m, 6H); 1.31–1.13 (m, 8H); 0.88 (t, 3H, J = 7.1 Hz). EI-MS
(m/z, %): 413 (M+); 341 (36.0); 179 (100.0). Anal. Calcd for
C24H31NO5: C 69.71, H 7.56, N 3.39. Found: C 68.01, H 5.21,
N 4.02.
6.6.19. 4-(3-Benzyloxy-phenyl)-N-{1-[4-(3-benzyloxy-phenyl)-
2,4-dioxo-butyryl]-piperidin-4-yl}-2,4-dioxo-butyramide (1d)
Compound 1d was prepared in a similar fashion as described
for 11d, yields 65.0%, yellow glassy solid. 1H NMR (400 MHz,
CDCl3): d 7.61–7.59 (m, 2H); 7.54–7.50 (m, 2H); 7.47–7.33 (m,
12H); 7.22–7.16 (m, 4H); 5.12 (s, 4H); 4.57–4.54 (m, 1H);
4.16–4.09 (m, 2H); 3.27 (m, 1H); 2.98–2.83 (m, 1H); 2.08 (m,
2H); 1.60 (m, 2H). EI-MS (m/z, %): 435 (2.0); 226 (16.0); 91
(100.0); 58 (90.0). Analytical HPLC tR = 25.68 min, purity 96.9%
(solvent system A), tR = 24.49 min, purity 100% (solvent system
B).
6.6.16. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(3-(benzyloxy)-
phenyl)butane-1,2,4-trione) (1c)
A solution of compound 1b (0.15 g, 0.5 mmol), EDCI (0.101 g,
0.525 mmol), and HOBt (0.071 g, 0.525 mmol) in dry DMF
(1.5 mL) was stirred at 0 °C for 15 min. To this solution was
added cyclohexane-1,4-diamine (0.026 g, 0.23 mmol) in dry
DMF. The reaction mixture was stirred for 2.5 h at room temper-
ature and then poured into ice-water and extracted with dichlo-
romethane. The combined organic layers were washed with
water and brine and dried over Na2SO4. The concentration pro-
vided the residue which was purified by chromatography using
CH2Cl2/CH3OH (60:1) as eluent to give compound 1c as white so-
lid in 68% yield. 1H NMR (400 MHz, CDCl3): d 15.6 (br, 2H); 7.61
(m, 4H); 7.45–7.35 (m, 12H); 7.21–7.19 (m, 4H); 7.07 (d, 2H,
J = 8.4 Hz); 5.13 (s, 4H); 3.88 (s, 2H); 2.13 (m, 4H); 1.45 (m,
4H). EI-MS (m/z): 674 (M+). Analytical HPLC tR = 38.40 min, purity
100% (solvent system A); tR = 33.25 min, purity 98.9% (solvent
system B).
6.6.20. 4-(3-Benzyloxy-phenyl)-1-{4-[4-(3-benzyloxy-phenyl)-
2,4-dioxo-butyryl]-piperazin-1-yl}-butane-1,2,4-trione (1e)
Compound 1e was prepared in
scribed for 1c, as red solid (85–90%). 1H NMR (300 MHz,
CDCl3): 7.55–7.34 (m, 16H), 7.20–7.16 (m, 2H), 6.63 (s,
2H), 5.12 (s, 4H), 3.78 (s, 8H). ESI-MS m/z: 646.2 (M)+.
Anal. (C38H34O8N2): 70.58, 5.30, 4.33. Found:
70.42, H 5.10, N 4.18.
a similar fashion as de-
d
C
H
N
C

7786
L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
6.6.21. 4-(3-Benzyloxy-phenyl)-2,4-dioxo-butyric acid 1-[4-(3-
benzyloxy-phenyl)-2,4-dioxo-butyryl]-piperidin-4-yl ester (1f)
To a solution of 4-(3-(benzyloxy)phenyl)-2,4-dioxobutanoic
acid (1b) (0.297 g, 1.0 mol) in 5 mL dry CH2Cl2 were added
0.15 mL SOCl2 and 1 drop of DMF at room temperature for 1.5 h.
Then the excess SOCl2 was removed under reduced pressure at
20–30 °C. The residue was dissolved in 5 mL dry CH2Cl2 and then
was added to a solution of 1,4-hydroxypiperidine (17) (0.045 g,
0.45 mol) and 0.2 mL pyridine at 0 °C for 0.5 h. After being stirred
at 20–25 °C for 3 h, the reaction mixture was diluted with 20 mL
CH2Cl2 and then partitioned between dichloromethane (20 mL)
and brine (20 mL). The organic layer was separated and dried over
Na2SO4, filtered and concentrated under reduced pressure to afford
1f (0.227 g, 68.7%) as yellow solid. Chromatography (silica gel, elu-
tion with CH2Cl2/CH3OH = 30:1) provided a light yellow foam solid,
1H NMR (400 Hz, CDCl3): d 7.42–7.27 (m, 14H); 7.01–7.22 (m, 4H);
5.03 (s, 4H); 3.88 (m, 1H); 3.49 (m,2H); 2.94–3.09 (m, 4H); 1.71–
1.85 (m, 2H). MS (ESI) m/z: 700 (M+K)+
1H, J = 1.5 Hz), 7.65 (d, 1H, J = 1.4 Hz), 7.49–7.41 (m, 4H), 7.38–7.35
(m, 2H), 7.08 (s, 2H), 7.04 (d, 2H, J = 8.0 Hz), 3.86 (m, 2H), 2.13–2.11
(m, 4H), 1.49–1.41 (m, 4H). EI-MS (m/z): 530 (M+ꢂH); 532 (M++H).
HR-EIMS calcd for C26H24N2Cl2O6: 530.1011, found: 530.0996.
Analytical HPLC tR = 26.95 min, purity 98.6% (solvent system A);
tR = 24.77 min, purity 97.9% (solvent system B).
6.6.27. 4-(2-Chlorophenyl)-N-{1-[4-(2-chlorophenyl)-2,4-
dioxo-butyryl]-piperidin-4-yl}-2,4-dioxo-butyramide (9d)
Compound 9d was prepared in a similar fashion as described for
11d, red solid (85.7%). 1H NMR (300 MHz, CD3OD): d 7.60–7.56 (m,
2H), 7.43–7.30 (m, 6H), 6.56 (s, 2H), 4.53–4.45 (m, 1H), 4.22–4.13
(m, 1H), 4.03–3.90 (m, 1H), 3.15–3.06 (m, 1H), 2.92–2.84 (m, 1H),
1.93–1.88 (m, 2H), 1.65–1.57 (m, 2H). ESI-MS m/z: 540.2 (M+Na)+.
Anal. (C25H22Cl2O6N2) C 58.27, H 3.91, N 5.44. Found: C 58.05, H
3.73, N 5.63.
6.6.28. 1,10-(Piperazine-1,4-diyl)bis(4-(2-chlorophenyl)butane-
1,2,4-trione) (9e)
6.6.22. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(2-fluorophenyl)-
butane-1,2,4-trione) (3c)
Compound 9e was prepared in a similar fashion as described for
1c, glassy solid (158 mg, 90%). 1H NMR (400 MHz, CDCl3): d 7.70–
7.67 (m, 2H); 7.49–7.36 (m, 6H); 6.57 (s, 2H); 3.87–3.76 (m, 8H).
EI-MS (m/z): 502 (M+ꢂ1). Analytical HPLC tR = 25.99 min, purity
98% (solvent system A); tR = 22.56 min, purity 95.0% (solvent
system B).
Compound 3c was prepared in a similar fashion as described for
1c, as pale solid, yield 80.0%. 1H NMR (400 MHz, CDCl3): d 15.50
(br, 2H); 7.98–7.94 (m, 2H); 7.57–7.52 (m, 2H); 7.30–7.21 (m,
2H); 7.20–7.15 (m, 2H); 7.05 (d, 2H, J = 8.1 Hz); 3.86 (m, 2H),
2.12 (m, 4H); 1.42–1.39 (m, 4H). EI-MS (m/z): 498 (M+). HR-EIMS
calcd for C26H24N2F2O6: 498.1602, found: 498.1597. Analytical
HPLC tR = 26.12 min, 100% (solvent system A), tR = 24.82 min,
100% (solvent system B).
6.6.29. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(4-(trifluoromethyl)
phenyl)butane-1,2,4-trione) (10c)
Compound 10c was prepared in a similar fashion as described
for 1c, white solid, yield 90%. 1H NMR (400 MHz, CDCl3): d 8.22
(m, 2H); 8.10 (m, 4H); 7.76 (m, 6H); 3.90 (m, 2H); 2.13 (m, 4H);
1.47 (m, 4H). EI-MS (m/z, %): 598 (M+). HR-EIMS calcd for
6.6.23. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(4-fluorophenyl)-
butane-1,2,4-trione) (4c)
Compound 4c was prepared in a similar fashion as described
for 1c, white solid, yield 70.0%. 1H NMR (400 MHz, CDCl3): d 15.6
(br, 2H); 8.05–8.02 (m, 4H); 7.20–7.16 (m, 6H); 7.06 (d, 2H,
J = 7.83 Hz); 3.92 (m, 2H); 2.13 (m, 4H); 1.46 (m, 4H). EI-MS
(m/z): 498 (M+). HR-EIMS calcd for C26H24N2F2O6: 498.1602,
found: 498.1630. Analytical HPLC tR = 25.45 min, purity
99.3% (solvent system A); tR = 22.54 min, purity 98.6% (solvent
system B).
C28H24N2F6O6: 598.1539, found: 598.1538. Analytical HPLC
tR = 30.24 min, purity 99.8% (solvent system A), tR = 28.58 min,
purity 99.6% (solvent system B).
6.6.30. 1,10-(Piperazine-1,4-diyl)bis(4-(4-(trifluoromethyl)-
phenyl)butane-1,2,4-trione) (10e)
Compound 10e was prepared in a similar fashion as described
for 1c, white solid, yield 70%. 1H NMR (400 MHz, CDCl3): d 8.03
(m, 4H), 7.75 (d, 4H, J = 7.9 Hz), 6.71 (s, 1H), 6.70 (s, 1H), 3.78 (m,
8H); EI-MS (m/z, %): 570 (M+, 2.0); HR-EIMS Calcd. for
C26H20N2F6O6: 570.1226, found 570.1238. Analytical HPLC
tR = 25.96 min, purity 100% (solvent system A), tR = 25.14 min,
purity 100% (solvent system B).
6.6.24. 4-(4-Fluorophenyl)-N-(1-(4-(4-fluorophenyl)-2,4-
dioxobutanoyl)piperidin-4-yl)-2,4-dioxobutanamide
Compound 4d was prepared in a similar fashion as described for
11d. 1H NMR (300 MHz, CDCl3): d 8.05–8.00 (m, 2H), 7.96–7.91 (m,
2H), 7.20–7.11 (m, 4H), 6.52 (s, 1H), 4.54 (m, 1H), 4.16–4.08 (m,
2H), 3.30–3.25 (m, 1H), 2.97–2.87 (m, 1H), 2.11–2.10 (m, 2H),
1.64–1.52 (m, 2H). MS m/z: 508.2 (M)+. EI-MS (m/z, %): 388
(20.0); 165 (100.0); 138 (16.0); 123 (36.0). Analytical HPLC
tR = 26.05 min, purity 98.8% (solvent system A), tR = 21.40 min, pur-
ity 98.6% (solvent system B).
6.6.31. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(3-methoxy-
phenyl)butane-1,2,4-trione) (11c)
Compound 11c was prepared in a similar fashion as de-
scribed for 1c, white solid, yield 64.0%. 1H NMR (400 MHz,
CDCl3): d 7.59 (m, 2H); 7.50 (m, 2H); 7.40 (t, 2H, J = 8.0 Hz);
7.19 (s, 2H); 7.15–7.14 (m, 1H); 7.13–7.12 (m, 1H); 7.08 (d,
2H, J = 8.4 Hz); 3.87 (m, 2H), 3.87 (s, 6H), 2.12 (m, 4H); 1.50–
1.42 (m, 4H). EI-MS (m/z): 522 (M+). HR-EIMS calcd for
6.6.25. 1,10-(Piperazine-1,4-diyl)bis(4-(4-fluorophenyl)butane-
1,2,4-trione)
Compound 4e was prepared in a similar fashion as described for
1c, pale yellow solid, yield 60.0%. 1H NMR (400 MHz, CDCl3): d
7.97–7.92 (m, 4H); 7.20–7.14 (m, 4H); 6.61 (s, 2H); 3.78 (m, 8H).
EI-MS (m/z, %): 470 (M+, 7.0); 306 (5.0); 278 (14.0); 165 (43.0);
123 (100.0). HR-EIMS calcd for C24H20N2F2O6: 470.1289, found:
470.1282. Analytical HPLC tR = 24.86 min, purity 99.0% (solvent
system A), tR = 21.20 min, 100% (solvent system B).
C28H30N2O8: 522.2002, found: 522.2003. Analytical HPLC
tR = 27.75 min, purity 98.8% (solvent system A); tR = 21.80 min,
purity 100% (solvent system B).
6.6.32. 1,10-(Piperazine-1,4-diyl)bis(4-(3-methoxyphenyl)-
butane-1,2,4-trione) (11e)
Compound 11e was prepared in a similar fashion as described
for 1c, white solid, yield 60.0%. 1H NMR (400 MHz, CDCl3): d
7.51–7.36 (m, 6H); 7.13–7.11 (m, 2H); 6.63 (s, 2H); 3.85 (s, 6H);
3.78 (s, 8H). EI-MS (m/z): 494 (M+). Analytical HPLC tR = 25.87 min,
purity 99.2% (solvent system A), tR = 26.42 min, purity 100%
(solvent system B).
6.6.26. 1,10-(Cyclohexane-1,4-diaminyl)bis(4-(2-chlorophenyl)
butane-1,2,4-trione) (9c)
Compound 9c was prepared in a similar fashion as described for
1c, white solid (113 mg, 85%). 1H NMR (400 MHz, CDCl3): d 7.66 (d,

L.-F. Zeng et al. / Bioorg. Med. Chem. 16 (2008) 7777–7787
7787
13. Marchand, C.; Zhang, X. C.; Pais, G. C. G.; Cowansage, K.; Neamati, N.; Burke, T.
R.; Pommier, Y. J. Biol. Chem. 2002, 277, 12596.
Acknowledgments
14. (a) Dayam, R.; Al-Mawsawi, L. Q.; Neamati, N. Drugs R&D 2007, 8, 155; (b)
Al-Mawsawi, L. Q.; Al-Safi, R. I.; Neamati, N. Expert Opin. Emerg. Drugs 2008,
13, 1.
15. Evering, T. H.; Markowitz, M. Expert Opin. Invest. Drugs 2008, 17, 413.
16. Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.; Yoshinaga, T.; Fujishita, T.;
Sugimoto, H.; Endo, T.; Murai, H.; Davies, D. R. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 13040.
17. Long, Y.-Q.; Jiang, X.-H.; Dayam, R.; Sanchez, T.; Shoemaker, R.; Sei, S.; Neamati,
N. J. Med. Chem. 2004, 47, 2561.
18. (a) Jiang, X.-H.; Song, L.-D.; Long, Y.-Q. J. Org. Chem. 2003, 68, 7555; (b)
Jiang, X.-H.; Long, Y.-Q. Chin. J. Chem. 2004, 22, 978.
National Natural Science Foundation of China (No. 30672528)
and Science and Technology Commission of Shanghai Municipality
(07QH14018) are greatly appreciated for the financial supports.
The work in NN’s laboratory was supported in part by funding from
the Campbell Foundation.
References and notes
1. Agrawal, L.; Lu, X.; Jin, Q.; Alkhatib, G. Curr. Pharm. Des. 2006, 12, 2031.
2. Cohen, J. Science 2002, 296, 2320.
3. Asante-Appiah, E.; Skalka, A. M. Antiviral Res. 1997, 36, 139.
4. Bujacz, G.; Alexandratos, J.; Wlodawer, A. J. Biol. Chem. 1997, 272, 18161.
5. Yang, W.; Steitz, T. A. Structure 1995, 3, 131.
19. Wai, J. S.; Egbertson, M. S.; Payne, L. S.; Fisher, T. E.; Embrey, M. W.; Tran, L. O.;
Melamed, J. Y.; Langford, H. M.; Guare, J. P., Jr.; Zhuang, L.; Grey, V. E.; Vacc, J.
P.; Holloway, M. K.; Naylor-Olsen, A. M.; Hazuda, D. J.; Felock, P. J.; Wolfe, A. L.;
Stillmock, K. A.; Schleif, W. A.; Gabryelski, L. J.; Young, S. D. J. Med. Chem. 2000,
43, 4923.
6. Grobler, J. A.; Stillmock, K.; Hu, B. H.; Witmer, M.; Felock, P.; Espeseth, A. S.;
Wolfe, A.; Egbertson, M.; Bourgeois, M.; Melamed, J.; Wai, J. S.; Young, S.; Vacca,
J.; Hazuda, D. J. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6661.
7. Marchand, C.; Johnson, A. A.; Karki, R. G.; Pais, G. C. G.; Zhang, X. C.; Cowansage,
K.; Patel, T. A.; Nicklaus, M. C.; Burke, T. R.; Pommier, Y. Mol. Pharmacol. 2003,
64, 600.
8. Pommier, Y.; Johnson, A. A.; Marchand, C. Nat. Rev. Drug Disc. 2005, 4, 236.
9. Pais, G. C. G.; Burke, T. R. Drugs Future 2002, 27, 1101.
10. Nair, V.; Chi, G. Rev. Med. Virol. 2007, 17, 277.
11. Hazuda, D. J.; Felock, P.; Witmer, M.; Wolfe, A.; Stillmock, K.; Grobler, J. A.;
Espeseth, A.; Gabryelski, L.; Schleif, W.; Blau, C.; Miller, M. D. Science 2000, 287,
646.
12. Anthony, N.; Egbertson, M.; Melamed, J. Y.; Young, S.; Hamill, T.; Cole, J. L.;
Hazuda, D. J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 11244.
20. Pais, G. C. G.; Zhang, X. C.; Marchand, C.; Neamati, N.; Cowansage, K.;
Svarovskaia, E. S.; Pathak, V. K.; Tang, Y.; Nicklaus, M.; Pommier, Y.; Burke, T. R.
J. Med. Chem. 2002, 45, 3184.
21. GOLD 3.2, The Cambridge Crystallographic Data Centre, Cambridge, UK, 2005.
22. Catalyst, Accelrys, Inc., San Diego, USA.
23. Smellie, A.; Dyda, F.; Hickman, A. B. J. Comput. Chem. 1995, 16, 171.
24. Smellie, A.; Kahn, S. D.; Teig, S. L. J. Chem. Inf. Comput. Sci. 1995, 35, 285.
25. Smellie, A.; Kahn, S. D.; Teig, S. L. J. Chem. Inf. Comput. Sci. 1995, 35, 295.
26. Goldgur, Y.; Dyda, F.; Hickman, A. B.; Jenkins, T. M.; Craigie, R.; Davies, D. R.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9150.
27. Verdonk, M. L.; Chessari, G.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Nissink,
J. W.; Taylor, R. D.; Taylor, R. J. Med. Chem. 2005, 48, 6504.
28. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
2003, 52, 609.