HIV-1 integrase strand-transfer inhibitors: Design, synthesis and molecular modeling investigation

Article abstract of DOI:10.1016/j.ejmech.2010.12.012

This study is focused on a new series of benzylindole derivatives with various substituents at the benzene-fused ring, suggested by our 3D pharmacophore model developed for HIV-1 integrase inhibitors (INIs). All synthesized compounds proved to be active in the nanomolar range (6-35 nM) on the strand-transfer step (ST). In particular, derivative 4-[1-(4-fluorobenzyl)- 5,7-dimethoxy-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoic acid (8e), presenting the highest best-fit value on pharmacophore model, showed a potency comparable to that of clinical INSTIs GS 9137 (1) and MK-0518 (2). The binding mode of our molecules has been investigated using the recently published crystal structure of the complex of full-length integrase from the prototype foamy virus in complex with its cognate DNA (PFV-IN/DNA). The results highlighted the ability of derivative 8e to assume the same binding mode of MK-0518 and GS 9137.

Full text of DOI:10.1016/j.ejmech.2010.12.012

European Journal of Medicinal Chemistry 46 (2011) 756e764
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
HIV-1 integrase strand-transfer inhibitors: Design, synthesis and molecular
modeling investigation
,
a
,
a
,
,
,2
Laura De Luca a , Sara De Grazia , Stefania Ferro 1, Rosaria Gitto a 1, Frauke Christ b
,
*
*
,
,1
Zeger Debyser b 2, Alba Chimirri a
a Dipartimento Farmco-Chimico, University of Messina, Viale Annunziata, I-98168 Messina, Italy
b Molecular Virology and Gene Therapy, Katholieke Universiteit Leuven and IRC KULAK, Kapucijnenvoer 33, B-3000 Leuven, Flanders, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
This study is focused on a new series of benzylindole derivatives with various substituents at the
benzene-fused ring, suggested by our 3D pharmacophore model developed for HIV-1 integrase inhibitors
(INIs). All synthesized compounds proved to be active in the nanomolar range (6e35 nM) on the strand-
transfer step (ST). In particular, derivative 4-[1-(4-fluorobenzyl)-5,7-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoic acid (8e), presenting the highest best-fit value on pharmacophore model,
showed a potency comparable to that of clinical INSTIs GS 9137 (1) and MK-0518 (2). The binding mode
of our molecules has been investigated using the recently published crystal structure of the complex of
full-length integrase from the prototype foamy virus in complex with its cognate DNA (PFV-IN/DNA). The
results highlighted the ability of derivative 8e to assume the same binding mode of MK-0518 and GS
9137.
Received 5 November 2010
Received in revised form
7 December 2010
Accepted 11 December 2010
Available online 21 December 2010
Keywords:
HIV-1 integrase
INSTIs
PFV-IN/DNA
Benzylindoles
Molecular docking
Ó 2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
Although the HAART has brought about a substantial decrease in
the death rate, changing AIDS from a rapidly lethal disease into
It was recently reported that more than 33 million of people
were infected worldwide with human immunodeficiency virus
(HIV-1). The overall number of people living with HIV has increased
as a result of new infections and the beneficial effects of the more
widely available highly active anti-retroviral therapy (HAART),
which employs a combinational use of drugs [1].
The currently FDA approved anti-HIV drugs belong to seven
different groups: nucleoside reverse transcriptase inhibitors
(NRTIs), nucleotide reverse transcriptase inhibitors (NtRTIs), non-
nucleoside reverse transcriptase inhibitors (NNRTIs), protease
inhibitors (PIs), fusion inhibitors (FIs), co-receptor inhibitors (CRIs)
and integrase inhibitors (INIs) [2].
a chronic manageable condition, the retroviral infection can be only
temporarily controlled but not eradicated since HIV-1 becomes
almost undetectable in the plasma for more than two years, per-
sisting in reservoirs.
Furthermore the HAART efficacy has been limited by the emer-
gence of drug-resistant viral strains, drug-toxicity, the poor ability of
patients to adhere to the prescribed therapy and costs, so a refining
of the current therapies and the developing of new therapeutic
paradigms are still warranted [3]. Considering that HIV-IN plays
a key role in stable infection and that, apart the RAG recombinases
[4], there is no known human counterpart of HIV-IN it is under-
standable how this enzyme is an attractive therapeutic target.
HIV-1 IN, mechanistically and structurally, belongs to the
superfamily of polynucleotidyl transferases, which includes the
RNaseH and the transposases from Tn5. IN inserts a double-
stranded DNA copy of the viral RNA genome into the chromosomes
of an infected cell through a multistep process that involves two
separate reactions. In the first step, termed 30-processing, IN leads
to the formation of new 30-CA-OH. The second step, named strand-
transfer (ST), is a trans-esterification reaction involving a direct
nucleophilic attack of the 30-hydroxy group of the two newly pro-
cessed viral 30-DNA ends on the phosphodiester backbone of the
Abbreviations: INSTIs, integrase strand transfer inhibitors; PFV, prototype foamy
virus; DKAs, diketo acids; IFD, induced-fit docking.
* Corresponding authors. Fax: þ39 090 355 613.
E-mail addresses: ldeluca@unime.it (L. De Luca), sdegrazia@pharma.unime.it
(S. De Grazia), sferro@unime.it (S. Ferro), rgitto@unime.it (R. Gitto), frauke.
christ@med.kuleuven.be (F. Christ), zeger.debyser@med.kuleuven.be (Z. Debyser),
achimirri@unime.it (A. Chimirri).
1
Fax: þ39 090 355 613.
2
Fax: þ32 16 336 336.
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.12.012

L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
757
host target DNA. The integration reaction is completed by the
removal of the two unpaired nucleotides at the 50-end of the viral
DNA and by the repair of the single stranded gaps created between
the viral and the target DNA [5,6]. No energy source (e.g., ATP) is
needed for the integration reaction, and only divalent cations are
required for catalytic activity. In particular, the two metal cofactors
of HIV-IN are thought to both be Mg2þ ions under physiological
conditions [7e9].
Among numerous attempts to identify new drugs active against
HIV-IN, diketo acids (DKAs) and their analogs were the first
derivatives displaying potent anti-retroviral activity via IN inhibi-
tion and several synthetic and biological studies have been repor-
ted [10,11]. It is believed that their IN-binding mechanism is
connected to the presence of DKA pharmacophoric motif which
could be involved in a functional sequestration of one or both
divalent metal ions in the enzyme catalytic site to form a ligand-
M2þ-IN complex. This would subsequently block the transition
state of the IN-DNA complex by competing with the target DNA
substrate, acting as an “interfacial inhibitor” [12].
The global research efforts to identify drugs that inhibit HIV
integrase recently led to two new strand-transfer inhibitors
(INSTIs) GS 9137 (1) and MK-0518 (2) (Fig. 1), which proved to be
highly effective and well-tolerated [13] and to date compound 2 is
the first drug (Isentress) active against IN enzyme, approved by the
FDA [14].
Fig. 2. Alignment of compound
hydrogen-bond acceptor; Z1eZ2, hydrophobic aliphatic; Y, hydrophobic aromatic).
3
into the pharmacophore model [17] (A1eA4,
In recent years also our research group has been engaged in the
structure-function study of HIV-IN and in the development of new
inhibitors. We reported ligand-based approaches useful to generate
three-dimensional pharmacophore models for INSTIs which led to
identify a series of benzylindoles as novel and potent IN inhibitors
(e.g. compound 3, CHI-1043, Fig. 1) [15e17].
In addition, considering that the structure of full-length retro-
viral HIV-1 integrase, either separately or in complex with DNA was
lacking, we developed different approaches to build models of the
full-length integrase-DNA complex and used the obtained results
for the design of new potential INSTIs and the improvement of their
activity [18e22].
But the in silico models could be not realistic, even if they gave us
some right information regarding the mechanism of action and the
binding modes of INSTIs. So, when finally in 2010 Hare et al.
reported a crystal structure of full-length integrase from the
prototype foamy virus in complex with its cognate DNA and the
INSTIs 1 and 2 [23] we were particularly interested to compare their
results with ours. This novel structural information is definitely
a breakthrough in the field of IN research, since the disclosed
complexes can be considered the best surrogate structures for
reliably modeling the binding of these HIV-1 INIs.
2. Results and discussion
DKA derivatives and their analogs which belong to the INSTIs
represent the major leads in the development of anti-HIV-1 IN
drugs. In fact, both 2 (Fig. 1) that is the only one drug approved by
FDA and other potent IN inhibitors (i.e. 1) are structurally related to
the DKA family.
Our previously reported three-dimensional pharmacophore
models allowed us the discovery of potent INSTIs (i.e. CHI-1043, 3
Fig. 1). The more recent of our pharmacophore models consisted of
seven features, four hydrogen-bond acceptors (A1eA4), two
hydrophobic aliphatic regions (Z1 and Z2) and one aromatic feature
(Y). The overlapping of compound 3 onto our model is shown in
Fig. 2. The b-diketo acid moiety maps the three hydrogen-bond
acceptors (A1eA3), the p-fluorobenzyl ring overlaps the aromatic
feature Y, and the 4-methoxy group occupies the aliphatic region Z1
whereas zone Z2 remained clear and taken together with our
preliminary results this suggested that the introduction of an
additional methoxy group at C-7 position should result in an
improvement of inhibitory activity [17].
Taking into account these findings, in this work we focused our
efforts to occupy all the pharmacophore model features by
choosing to insert two methoxy groups in different positions of the
benzene-fused ring of the indole nucleus and keeping unchanged
the N-benzyl moiety of 3 (Fig. 2). So the synthesis of a new series of
dimethoxy-p-fluoro-benzylindoles have been performed and the
compounds obtained have been tested as IN inhibitors and anti-HIV
agents.
Therefore, in the present study we report the rational design
and expanded SAR investigation of new 4-[1-(4-fluorobenzyl)-
dimethoxy-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoic
deriva-
tives useful in clarifying the features important for ST inhibitors
whereas the crystallographic data of PFV-IN complex were used to
obtain more details about the binding mode of our new compounds
in comparison with other well-known INSTIs.
F
O
OH
OH
N
O
N
O
N
O
H
N
N
O
F
N
N
O
Cl
O
HN
COOH
OMe
F
OH
O
O
GS 9137 (elvitegravir) (1)
MK-0518 (raltegravir) (2)
Fig. 1. HIV-1 integrase strand-transfer inhibitors.
CHI-1043 (3)

758
L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
2.1. Chemistry
[28]. Moreover, their inhibitory effect on the HIV-1 induced cyto-
pathic effect (CPE) in human lymphocyte MT-4 cell culture was
determined by the MT-4/MTT-assay (Table 2) [29]. The biological
results were compared with the data previously reported for indole
derivative 3 and reference compounds 1e2.
The synthetic route employed to obtain the title compounds is
depicted in Scheme 1.
Dimethoxyindoles 4aef were 3-acetylated by a Vilsmeier Haack
reaction and N-alkylated by treatment with 4-fluorobenzyl bromide,
in the presence of a catalytic amount of sodium hydride to give
intermediates 6aef. Successively diketo esters 7aef, obtained by
coupling with diethyl oxalate, were converted in the final derivatives
8aef by hydrolysis in basic medium. The preparation of commer-
cially unavailable indoles 4a, 4e, and 4f is shown in Schemes 2e4.
The introduction of a nitro group on 2,3-dimethoxybenzaldehyde
afforded a mixture of 5- and 6-nitro derivatives which were sepa-
rated by flash chromatography on silica gel eluting with 30% EtOAc/
cyclohexane [24]. The 2,3-dimethoxy-6-nitrobenzaldehyde was
The biological assays revealed that the DKAs 8 exhibited
inhibitory potency higher than the corresponding esters 7 with
exception of ester derivative 7a which was more active than the
corresponding DKA 8a. The analysis of biological results suggests
that the introduction of an additional methoxy group at C-5 or C-7
position increases the IN-inhibitory effects. In fact, the disubsti-
tuted compounds 8a and 8c are more potent than the 4-mono-
substituted derivative 3. Moreover we found that the simultaneous
presence of the methoxy substituent at 5 and 7 positions afforded
the most potent INSTI of this series, the 4-[1-(4-fluorobenzyl)-5,7-
dimethoxy-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoic acid 8e; it
proved to be active at 6 nM concentration that is comparable to 2
(7 nM) and better than 1 (15 nM). On the contrary, the introduction
of an additional methoxy group at C-6 seems to negatively influ-
ence the activity of this class of INSTIs (see compounds 8b, 8d and
8f).
converted into the corresponding 2,3-dimethoxy-6,b-dinitrostyrene
by refluxing with nitromethane in the presence of glacial acetic acid
and ammonium acetate. Finally 4,5-dimethoxyindole (4a) was
obtained by a Henry’s reductive cyclization [25] (Scheme 2).
5,7-Dimethoxyindole (4e) was synthesized according with
Condie’s procedure, obtaining high yields. A solution of 3,5-dime-
thoxybenzaldehyde in nitromethane was refluxed to afford 3,5-
In the biological test against HIV-1 replication (Table 2), all DKAs
dimethoxy-
b
-nitrostyrene which was converted in the nitro
8aef showed EC50 values between 1.25 mM and 14.66 mM; they also
derivative and then into the desired product by reductive cycliza-
tion (Scheme 3) [26].
proved to be better than corresponding esters 7aef. Compounds 8a
and 8c displayed SI values (104 and 100, respectively) better than
“lead compound” 3 (SI value of 70).
A different synthetic pathway, according to a modification of
Rege’s procedure, was followed to obtain the dimethoxy substitu-
tion at the 6-C and 7-C positions of the indole nucleus (Scheme 4).
The treatment of commercially available 3-methoxy-4-hydrox-
ybenzaldehyde with acetic anhydride provided the O-acyl protected
intermediate which, after nitration, was hydrolyzed under basic
conditions. Methylation of the hydroxyl group with dimethyl sulfate
followed by reaction with nitromethane and reductive cyclization as
with the preparation of 4e, gave 6,7-dimethoxyindole (4f) [27].
2.3. Computational studies
In order to explain the IN-inhibitory effects of the benzylindole
derivatives we performed molecular modeling studies. All the
scores for the ranking of the compounds in the pharmacophore
model have been calculated (Supplementary data). Compound 8e,
that was the most active compound of the series, showed the best
alignment (Fig. 3) in our pharmacophore model [17] further con-
firming its prediction power. Derivative 8e (see Fig. 3) was able to
perfectly occupy all the features of our model showing a high best-
fit value (6.1). The two methoxy groups at C-5 and C-7 positions
2.2. Biological activity
To determine the susceptibility of the HIV-1 integrase enzyme
(Table 1) toward synthesized compounds 7aef and 8aef, we used
enzyme-linked immunosorbent assays as described previously
occupied both the hydrophobic areas Z1 and Z2; the b-diketo acid
moiety fitted the three hydrogen-bond acceptors (A1eA3) whereas
F
H
N
H
ii
N
N
i
RR'
RR'
RR'
CH3
CH3
O
O
6a-f
5a-f
4a-f
iii
R R'
a
b
c
4,5-(OCH3)2
4,6-(OCH3)2
4,7-(OCH3)2
5,6-(OCH3)2
5,7-(OCH3)2
F
F
d
e
N
iv
N
RR'
RR'
f
6,7-(OCH3)2
COOEt
COOH
O
OH
O
HO
8a-f
7a-f
Scheme 1. Reagents and conditions: (i) POCl3, CH3CON(CH3)2, RT, 12 h; (ii) 4-fluorobenzyl bromide, NaH, DMF, mw: 5 min at continuous temperature (50 ꢀC), 100 W; (iii) diethyl
oxalate, dry CH3ONa, THF, two separate steps under the same conditions mw: 2 min at continuous temperature (50 ꢀC), 250 W; (iv) 2 N NaOH, MeOH, RT, 1.5 h.

L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
759
H
N
NO2
NO2
H
i
iii
H
MeO
NO2
MeO
MeO
MeO
OMe
OMe O
OMe
OMe
O
4a
Scheme 2. Reagents and conditions: (i) glacial CH3COOH 14 ꢀC 10 min, fuming HNO3, RT 1 h; (ii) CH3NO2, CH3COONH4, glacial CH3COOH, Δ, 1.5 h; (iii) iron powder, glacial CH3COOH,
silica gel, toluene, Δ, 1.5 h.
the A4 region is occupied by the oxygen atom of the methoxy group
in position 5 and p-fluorobenzyl ring matched the aromatic feature
Y. With the aim to explore the main interactions with the target
HIV-IN enzyme, we had already undertaken docking studies, when
very interesting new data have been published [23]. In particular it
has been demonstrated that the prototype foamy virus (PFV) IN was
sensitive to HIV-1 INSTIs and the crystal structures of compounds 1
and 2 in complex with the full-length PFV IN and its DNA were
solved [23].
superimposed with the X-ray position of 1 (orange) and 2 (cyan),
showing a similar binding mode (Fig. 5).
In particular the benzyl ring of the three compounds is placed in
the same pocket in contact with viral DNA and all of three
compounds are able to interact with Mg ions. The methoxy group at
C-5 of 8e is orientated in the same position of the methyloxadiazole
moiety and isopropyl fragment of 1 and 2 respectively. In turn the
7-methoxy substituent of 8e corresponds to the position of
methoxy group of compound 1.
These novel crystallographic data suggest an induced-fit mech-
anism of inhibition, where the halobenzyl groups of INSTIs fit into
a pocket created by displacement of the terminal 30-adenosine.
These observations are in agreement with our induced-fit docking
(IFD) procedure in which the 30-terminal adenine nucleotide
underwent a dramatic conformational movement in order to allow
the insertion of the substituted benzyl moiety between the two viral
DNA strands [20].
Therefore, with the aim to gain more information about the
structure similarity of HIV-IN and PFV-IN, we decided to superim-
pose these two proteins using the core domain of PFV-IN complex
with MK-0518 (2, Fig. 1) (pdb code 3L2T) [23] and the core domain
of HIV-IN previously modeled by us (pdb code 1WKN) [18]. The
results of this study are reported in Fig. 4.
In Fig. 5B it is shown the best docked conformation of compound
8e after optimization of the complex using Macromodel software
(see Experimental section for details) together with our simplified
3D pharmacophore model. The p-fluorobenzyl group (Y) interacts
with the residues A17 and C16 of viral DNA and in particular the
fluorine atom seems to establish Van der Waals interactions with
conserved Gln 215, thus confirming the key role of the halogen
atoms in INSTIs [31]. The diketo acid moiety (A1eA3) is able to
interact with the two divalent ions in the active site whereas
5-methoxy (Z2) and 7-methoxy (Z1) groups are involved in hydro-
phobic interactions with Tyr 212 ad Pro 214 respectively.
3. Conclusion
As it is possible to note in the Fig. 4, there is a high similarity for
the two 3D structures of integrases; the only significant difference
is the orientation of the loop (residues 209e218 for PFV-IN and
140e149 for HIV-IN).
In summary, most of the compounds synthesized exhibit high
inhibitory potency against the IN enzyme both on over-all activity
and ST step. In particular, the highest potency has been found for
DKA derivative 8e, which inhibited the ST step of IN at 6 nM
concentration that is comparable to 2 (7 nM) and better than 1
(15 nM). Molecular modeling study, based on pharmacophore
models, as well as docking results revealed that the investigated
compounds show high binding similarity with clinical INSTIs 1 and
2. Moreover, the key role of some structural requirements for
binding between IN and STIs has been demonstrated.
It has been previously reported the importance of the flexibility
of this HIV-IN loop and the role of its residues associated with
resistance to IN inhibitors. So, we concentrated our attention on the
position of IN-PFV Tyr 212 residue (magenta stick style), corre-
sponding to Tyr 143 of HIV-IN (green stick line), that is involved in
stacking interactions with the inhibitor MK-0518 (2). In addition,
the important part of conserved PFV-IN Pro 214 corresponding to
HIV-IN Pro 145 highlighted the importance of this loop orientation
for the binding mode of the INIs. So, considering crucial the
orientation of the loop and the position of the side-chain of its
residues for the interaction of INIs we decided to use the X-ray
structure of integrase with DNA and two Mg ions from PFV-IN/MK-
0518 complex (pdb code 3L2T) for docking studies of our
compounds thus employing a realistic complex. Docking studies
were performed using the GOLD program [30]. After the validation
of the docking protocol (see Experimental section), compounds
7aef and 8aef were docked in the same binding site and the result
of the most active compound of the series 8e (yellow) was
4. Experimental section
4.1. Chemistry
All the microwave-assisted reactions were carried out in a CEM
Focused Microwave Synthesis System, Model Discover, working at
the power necessary for refluxing under atmospheric conditions
(i.e. 250e300 W). Melting points were determined on a BUCHI
Melting Point B-545 apparatus and are uncorrected. Elemental
analyses (C, H, N) were carried out on a Carlo Erba Model 1106
Elemental Analyzer and the results are within ꢁ0.4% of the
OMe
OMe
OMe
OMe
H
N
NO2
iii
i
ii
H
MeO
NO2
MeO
NO2
MeO
MeO
O
4e
Scheme 3. Reagents and conditions: (i) CH3NO2, CH3COONH4, glacial CH3COOH, Δ, 1.5 h; (ii) Cu(NO3), (CH3CO)2O, 60 ꢀC, 2 h; (iii) iron powder, 80% aqueous CH3COOH , 80ꢀC-15 min,
RT-1 h.

760
L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
O
O
O
H
ii (a,b)
i
H
H
HO
OAc
HO
NO2
OMe
OMe
OMe
iii
O
OMe
NO2
H
v
iv
MeO
N
H
MeO
NO2
MeO
NO2
OMe
OMe
4f
Scheme 4. Reagents and conditions: (i) Ac2O, NEt3, DMAP, rt, 3 h; (ii) a) fuming HNO3, -10 ꢀC, 4 h; b) 2 M KOH, Δ, 10 min; (iii) Me2SO4, absolute EtOH, 12 N NaOH, 50 ꢀC, 3h; (iv)
CH3NO2, CH3COONH4, glacial CH3COOH, Δ, 1.5 h; (v) iron powder, glacial CH3COOH, silica gel, toluene, Δ, 1.5 h.
theoretical values. Merck silica gel 60 F254 plates were used for
analytical TLC; column chromatography was performed on Merck
silica gel 60 (230e400 mesh) and Flash Chromatography (FC) on
a Biotage SP1 EXP. 1H NMR spectra were recorded in CDCl3 with
TMS as internal standard or [D6]DMSO on a Varian Gemini-300
4.1.1.1. 3-Acetyl-4,5-dimethoxy-1H-indole (5a). White solid (166 mg,
76%): mp: 166e168 ꢀC; 1H NMR (CDCl3):
3H, OCH3), 3.95 (s, 3H, OCH3), 7.00 (d, J ¼ 8.6 Hz, 1H, ArH), 7.13 (d,
J ¼ 8.8 Hz, 1H, ArH), 7.83 (d, J ¼ 2.7 Hz, 1H, ArH), 9.11 ppm (bs, 1H,
NH); Anal. calcd for C12H13NO3: C 65.74, H 5.98, N 6.39, found: C
65.85, H 5.71, N 6.51.
d
¼ 2.64 (s, 3H, CH3), 3.93 (s,
spectrometer. Chemical shifts were expressed in
d (ppm) and
coupling constants (J) in hertz. All the exchangeable protons were
confirmed by addition of D2O.
4.1.1.2. 3-Acetyl-4,6-dimethoxy-1H-indole (5b). White solid (101 mg,
1
46%): mp: 119e121 ꢀC; H NMR ([D6]DMSO):
d
¼ 2.46 (s, 3H, CH3),
4.1.1. General procedure for the synthesis of 3-acetyl-dimethoxy-
1H-indoles (5aef)
3.75 (s, 3H, OCH3), 3.79 (s, 3H, OCH3), 6.27 (d, J ¼ 2.1 Hz, 1H, ArH),
6.51 (d, J ¼ 1.9 Hz, 1H, ArH), 7.84 (s, 1H, ArH), 11.65 ppm (bs, 1H, NH);
Anal. calcd for C12H13NO3: C 65.74, H 5.98, N 6.39, found: C 65.83, H
6.21, N 6.51.
Phosphoryl chloride (0.92 mL, 10 mmol) was added to ice cold
dimethylacetamide (2.79 mL, 30 mmol) with stirring and cooling in
ice. The suitable dimethoxyindole derivative (4aef) (177 mg,
1 mmol) was added and reaction mixture was stirred at room
temperature for 12 h, then poured over ice and basified with a 4 N
aqueous sodium hydroxide solution. The mixture was extracted
with ethyl acetate and dried over Na2SO4. After removal of the
solvent under reduced pressure, the residue was powdered by
treatment with diethyl ether and recrystallized from dichloro-
methane [31].
4.1.1.3. 3-Acetyl-4,7-dimethoxy-1H-indole (5c). Spectral data are in
accordance with the literature [17].
4.1.1.4. 3-Acetyl-5,6-dimethoxy-1H-indole (5d). White solid (176 mg,
81%): mp: 147e149 ꢀC; 1H NMR (CDCl3):
3H, OCH3), 3.93 (s, 3H, OCH3), 6.89 (s,1H, ArH), 7.74 (d, J ¼ 3.0 Hz,1H,
d
¼ 2.53 (s, 3H, CH3), 3.89 (s,
Table 2
Table 1
Activity in MT-4 cells.
Inhibition of HIV-1 integrase activity.
Cpd
HIV-1a EC50
(
m
M)
Cytotoxicityb CC50
(
m
M)
SIc
Cpd
Over-alla IC50
(
m
M)
STb C50
(mM)
7a
7b
7cd
7d
7e
7f
8a
8b
8cd
8d
8e
8f
3.96 ꢁ 0.63
2.48 ꢁ 0.95
4.09 ꢁ 1.29
>21.76
111 ꢁ 11
28
22
20
7a
7b
7c
7d
7e
7f
8a
8b
8cc
8d
8e
8f
3
0.9 ꢁ 0.36
2.79 ꢁ 0.5
0.015 ꢁ 0.003
1.29 ꢁ 0.64
0.66 ꢁ 0.01
7.69 ꢁ 1.43
34.9 ꢁ 5.2
55.4 ꢁ 2.35
81.1 ꢁ 5.68
21.76
1.24 ꢁ 0.06
4.4 ꢁ 1.56
<1
<1
11
104
6
100
6
7
9
70
>1387
>2650
>62.07
62.07
18.1 ꢁ 3.5
10.41 ꢁ 3.46
1.21 ꢁ 0.28
3.2 ꢁ 0.97
119 ꢁ 2.87
125.5 ꢁ 5.5
20.7 ꢁ 3.9
213.3 ꢁ 50.3
65.46 ꢁ 2.97
58.5 ꢁ 12.2
130.5 ꢁ 12.5
41.1 ꢁ 16.7
>18
7.41 ꢁ 2.11
1.45 ꢁ 0.46
0.47 ꢁ 0.2
17.41 ꢁ 10.05
0.047 ꢁ 0.03
0.28 ꢁ 0.06
0.03 ꢁ 0.01
0.119 ꢁ 0.009
0.006 ꢁ 0.003
6.37 ꢁ 2.25
0.14 ꢁ 0.11
0.007 ꢁ 0.0005
0.015 ꢁ 0.002
2.13 ꢁ 0.05
10.52 ꢁ 0.35
8.51 ꢁ 0.32
14.66 ꢁ 3.6
0.59 ꢁ 0.38
0.06 ꢁ 0.01
0.547 ꢁ 0.15
0.251 ꢁ 0.034
3.55 ꢁ 3.78
0.08 ꢁ 0.003
3
2
1
0.013 ꢁ 0.0005
2
1
0.009 ꢁ 0.0002
0.004 ꢁ 0.003
0.0008 ꢁ 0.00009
2.12 ꢁ 0.36
a
Effective concentration required to reduce HIV-1-induced cytopathic effect by
50% in MT-4 cells.
a
Concentration required to inhibit the in vitro overall integrase activity by 50%.
Concentration required to inhibit the in vitro strand-transfer step by 50%. All
b
b
Cytotoxic concentration to reduce MT-4 cell viability by 50%.
c
data represent average results ꢁ SD.
Selectivity index: ratio CC50/EC50. All data represent average results ꢁ SD.
c
d
Data from Ref. [17].
Data from reference [17].

L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
761
4.1.1.6. 3-Acetyl-6,7-dimethoxy-1H-indole (5f). White solid (212 mg,
97%): mp: 139e141 ꢀC; 1H NMR (CDCl3):
d
¼ 2.53 (s, 3H, CH3), 3.94
(s, 3H, OCH3), 4.00 (s, 3H, OCH3), 6.99 (d, J ¼ 8.8 Hz, 1H, ArH), 7.78
(d, J ¼ 1.6 Hz, 1H, ArH), 8.00 ppm (d, J ¼ 8.8 Hz, 1H, ArH); Anal. calcd
for C12H13NO3: C 65.74, H 5.98, N 6.39, found: C 65.63, H 5.87,
N 6.19.
4.1.2. Generalprocedure forthesynthesis of3-acetyl-dimethoxy-1-(4-
fluorobenzyl)-1H-indoles (6aef)
The appropriate 3-acetyl-dimethoxy-1H-indole derivative
(5aef) (219 mg, 1 mmol) was dissolved in DMF (1 mL) at 0 ꢀC and
dry sodium hydride (120 mg, 5 mmol) was added. The mixture was
stirred for 2 min. 4-fluorobenzyl bromide (283 mg, 1.5 mmol) was
added dropwise and the resulting solution was placed in a cylin-
drical quartz tube (diam. 2 cm). The reaction mixture was then
stirred and irradiated in a microwave oven at 100 W and at
continuous temperature (50 ꢀC) for 5 min. A saturated NaHCO3
solution was added. The reaction mixture was extracted with ethyl
acetate and dried over Na2SO4. After removal of the solvent under
reduced pressure, the residue was powdered by treatment with
diethyl ether and crystallized from a mixture of diethyl ether and
dichloromethane [21].
4.1.2.1. 3-Acetyl-4,5-dimethoxy-1-(4-fluorobenzyl)-1H-indole (6a). Oil
(255 mg, 78%); 1H NMR ([D6]DMSO):
d
¼ 2.66 (s, 3H, CH3), 3.91 (s, 3H,
Fig. 3. Alignment of compound 8e into pharmacophore model [17] (A1eA4, hydrogen-
bond acceptor; Z1eZ2, hydrophobic aliphatic; Y, hydrophobic aromatic).
OCH3), 3.95 (s, 3H, OCH3), 5.24 (s, 2H, CH2), 6.92e7.16 (m, 6H, ArH),
7.75 ppm (s, 1H, ArH); Anal. calcd for C19H18FNO3: C 69.71, H 5.54, N
4.28, found: C 69.63, H 5.61, N 4.47.
ArH), 7.89 (s, 1H, ArH), 8.71 ppm (bs, 1H, NH); Anal. calcd for
C12H13NO3: C 65.74, H 5.98, N 6.39, found: C 65.58, H 6.11, N 6.24.
4.1.2.2. 3-Acetyl-4,6-dimethoxy-1-(4-fluorobenzyl)-1H-indole (6b). White
solid (242 mg, 74%): mp: 102e104 ꢀC; 1H NMR (CDCl3):
d
¼ 2.64 (s, 3H,
4.1.1.5. 3-Acetyl-5,7-dimethoxy-1H-indole (5e). White solid (110 mg,
CH3), 3.78 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 5.21 (s, 2H, CH2), 6.30 (d,
J ¼ 1.9 Hz, 1H, ArH), 6.35 (d, J ¼ 1.9 Hz, 1H, ArH), 7.02e7.11 (m, 4H, ArH),
7.59 ppm (s, 1H, ArH); Anal. calcd for C19H18FNO3: C 69.71, H 5.54, N 4.28,
found: C 69.85, H 5.41, N 4.63.
50%): mp: 198e200 ꢀC; 1H NMR (CDCl3):
d
¼ 2.52 (s, 3H, CH3), 3.89
(s, 3H, OCH3), 3.92 (s, 3H, OCH3), 6.41 (d, J ¼ 2.1 Hz, 1H, ArH), 7.45 (d,
J ¼ 2.1 Hz, 1H, ArH), 7.76 (d, J ¼ 3.4 Hz 1H, ArH), 8.76 ppm (bs, 1H,
NH); Anal. calcd for C12H13NO3: C 65.74, H 5.98, N 6.39, found: C
65.58, H 6.04, N 6.22.
4.1.2.3. 3-Acetyl-4,7-dimethoxy-1-(4-fluorobenzyl)-1H-indole (6c).
Spectral data are in accordance with the literature [17].
4.1.2.4. 3-Acetyl-5,6-dimethoxy-1-(4-fluorobenzyl)-1H-indole(6d). White
solid (219 mg, 67%): mp: 204e206 ꢀC; 1H NMR (CDCl3):
d
¼ 2.53 (s, 3H,
CH3), 3.89 (s, 3H, OCH3), 4.03 (s, 3H, OCH3), 5.33 (s, 2H, CH2), 6.72 (s, 1H,
ArH), 7.06e7.18 (m, 4H, ArH), 7.64 (s, 1H, ArH), 7.95 ppm (s, 1H, ArH);
Anal. calcd for C19H18FNO3: C 69.71, H 5.54, N 4.28, found: C 69.58, H 5.38,
N 4.42.
4.1.2.5. 3-Acetyl-5,7-dimethoxy-1-(4-fluorobenzyl)-1H-indole (6e). White
solid (196 mg, 60%): mp: 135e137 ꢀC; 1H NMR (CDCl3):
d
¼ 2.47 (s, 3H,
CH3), 3.80 (s, 3H, OCH3), 3.89 (s, 3H, OCH3), 5.55 (s, 2H, CH2), 6.38 (d,
J ¼ 2.1 Hz, 1H, ArH), 6.99e7.09 (m, 4H, ArH), 7.51 (d, J ¼ 2.1 Hz, 1H, ArH),
7.58 ppm (s, 1H, ArH); Anal. calcd for C19H18FNO3: C 69.71, H 5.54, N
4.28, found: C 69.63, H 6.12, N 4.37.
4.1.2.6. 3-Acetyl-6,7-dimethoxy-1-(4-fluorobenzyl)-1H-indole (6f). White
solid (186 mg, 57%): mp: 122e124 ꢀC; 1H NMR (CDCl3):
d
¼ 2.48 (s, 3H,
CH3), 3.68 (s, 3H, OCH3), 3.91 (s, 3H, OCH3), 5.57 (s, 2H, CH2), 6.98e7.12
(m, 5H, ArH), 7.60 (s, 1H, ArH), 8.07 ppm (d, J ¼ 8.5 Hz, 1H, ArH); Anal.
calcd for C19H18FNO3: C 69.71, H 5.54, N 4.28, found: C 69.81, H 6.02, N
4.11.
4.1.3. General procedure for the synthesis of 4-[1-(4-fluorobenzyl)-
dimethoxy-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoates (7aef)
A mixture of suitable 3-acetyl-dimethoxy-1-(4-fluorobenzyl)-
1H-indole (6aef) (327 mg, 1 mmol) diethyl oxalate (219 mg,
1.5 mmol) and a catalytic amount of NaOCH3 in anhydrous THF
Fig. 4. Superimposed PFV-IN/2 complex (magenta) and HIV-IN (green). Compound
MK-0518 (2), the residue Tyr 143 of HIV-IN and the corresponding Tyr 212 of PFV-IN
are showed in stick line. This Figure was prepared using the PyMOL program [37]. (For
interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.).

762
L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
Fig. 5. (A) Docking pose of compound 8e (yellow), compared to the crystallized position of 1 (orange) and 2 (cyan) in the IN-DNA active site. The divalent metal ions are shown as
gray spheres, while the viral DNA is depicted in violet. (B) Docking pose of compound 8e after minimization procedures. The colored spheres represented the 3D pharmacophore
model developed by Catalyst. This Figure was prepared using the PyMOL program [37]. (For interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.).
(2 mL) was placed in a cylindrical quartz tube (diam. 2 cm), then
stirred and irradiated at continuous temperature, in a microwave
oven for two subsequent periods in the same conditions (250 W,
2 min, 50 ꢀC). The solvent was evaporated under reduced pressure,
the collected solid was washed with ether and crystallized from
ethanol/diethyl ether. [21]
4.1.3.6. Ethyl 4-[1-(4-fluorobenzyl)-6,7-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7f). Yellow solid (419 mg, 98%): mp:
244 ꢀC dec; 1H NMR ([D6]DMSO):
d
¼ 1.23 (t, J ¼ 7.1 Hz, 3H, CH3),
3.60 (s, 3H, OCH3), 3.79 (s, 3H, OCH3), 4.09 (q, J ¼ 7.1 Hz, 2H, CH2),
5.52 (s, 2H, CH2), 6.16 (s,1H, CH), 6.91 (d, J ¼ 8.8 Hz,1H, ArH), 7.12 (d,
J ¼ 7.1 Hz, 2H, ArH), 7.89 (d, J ¼ 6.3 Hz, 1H, ArH), 8.03 ppm (d,
J ¼ 9.1 Hz, 1H, ArH); Anal. calcd for C23H22FNO6: C 64.63, H 5.19, N
3.28, found: C 64.47, H 5.06, N 3.11.
4.1.3.1. Ethyl 4-[1-(4-fluorobenzyl)-4,5-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7a). Yellow solid (410 mg, 96%): mp:
213 ꢀC dec; 1H NMR ([D6]DMSO):
d
¼ 1.20 (t, J ¼ 7.1 Hz, 3H, CH3),
4.1.4. General procedure for the synthesis of 4-[1-(4-fluorobenzyl)-
dimethoxy-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoic acids
(8aef)
A methanol solution (5 mL) of derivatives 7aef (427 mg,
1 mmol) was treated with 2 N NaOH (5 mL, 50 mmol) and stirred at
room temperature for 1.5 h. Then the reaction mixture was acidified
with conc. HCl to afford a solid that was collected and recrystallized
from ethanol/diethyl ether [21].
3.75 (s, 3H, OCH3), 3.77 (s, 3H, OCH3), 4.09 (q, J ¼ 7.1 Hz, 2H, CH2),
5.35 (s, 2H, CH2), 6.47 (s, 1H, CH), 6.94 (d, J ¼ 8.8 Hz, 1H, ArH),
7.11e7.17 (m, 3H, ArH), 7.25e7.30 (m, 2H, ArH), 7.80 ppm (s, 1H,
ArH); Anal. calcd for C23H22FNO6: C 64.63, H 5.19, N 3.28, found: C
64.72, H 5.22, N 3.39.
4.1.3.2. Ethyl 4-[1-(4-fluorobenzyl)-4,6-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7b). Yellow solid (419 mg, 98%): mp:
110e112 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 1.22 (t, J ¼ 7.1 Hz, 3H, CH3),
4.1.4.1. 4-[1-(4-Fluorobenzyl)-4,5-dimethoxy-1H-indol-3-yl]-2-hydroxy-
3.73 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 4.09 (q, J ¼ 7.1 Hz, 2H, CH2),
5.32 (s, 2H, CH2), 6.23 (s, 1H, CH), 6.53e6.61 (m, 1H, ArH), 7.11e7.29
(m, 4H, ArH), 7.51 (d, J ¼ 5.5 Hz, 1H, ArH), 8.50 ppm (s, 1H, ArH);
Anal. calcd for C23H22FNO6: C 64.63, H 5.19, N 3.28, found: C 64.48,
H 5.37, N 3.38.
4-oxobut-2-enoic acid (8a). Yellow solid (199 mg, 50%): mp:
120e122 ꢀC; 1H NMR ([D6]DMSO):
OCH3), 5.46 (s, 2H, CH2), 7.08e7.19 (m, 3H, ArH e CH), 7.29e7.39 (m,
d
¼ 3.76 (s, 3H, OCH3), 3.81 (s, 3H,
4H, ArH), 8.67 (s, 1H, ArH), 13.74 (bs, 1H, OH), 15.83 ppm (bs, 1H, OH);
13C NMR ([D6]DMSO):
115.4, 126.4, 129.7, 130.6, 131.8, 134.8, 146.5, 148.1, 159.9, 164.5, 187.6,
189.7 ppm; Anal. calcd for C21H18FNO6: C 63.16, H 4.54, N 3.51, found: C
63.31, H 4.27, N 3.33.
d
¼ 56.3, 56.7, 58.8, 104.8, 110.2, 111.1, 113.9,
4.1.3.3. Ethyl 4-[1-(4-fluorobenzyl)-4,7-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7c). Spectral data are in accordance
with the literature [17].
4.1.4.2. 4-[1-(4-Fluorobenzyl)-4,6-dimethoxy-1H-indol-3-yl]-2-hydroxy-
4.1.3.4. Ethyl 4-[1-(4-fluorobenzyl)-5,6-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7d). Yellow solid (384 mg, 90%): mp:
4-oxobut-2-enoic acid (8b). Yellow solid (199 mg, 50%): mp:
163e165 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 3.76 (s, 3H, OCH3), 3.86 (s, 3H,
OCH3), 5.45 (s, 2H, CH2), 6.41 (s, 1H, CH), 6.76 (s, 1H, ArH), 7.17e7.37 (m,
5H, ArH), 8.39 ppm (s, 1H, ArH); 13C NMR ([D6]DMSO):
244 ꢀC dec; 1H NMR ([D6]DMSO):
d
¼ 1.32 (t, J ¼ 7.1 Hz, 3H, CH3),
3.82 (s, 6H, OCH3), 4.19 (q, J ¼ 7.1 Hz, 2H, CH2), 5.47 (s, 2H, CH2), 6.29
(s, 1H, CH), 7.15 (s, 1H, ArH), 7.20e7.26 (m, 2H, ArH), 7.39e7.43 (m,
2H, ArH), 7.95e7.98 ppm (m, 2H, ArH); Anal. calcd for C23H22FNO6:
C 64.63, H 5.19, N 3.28, found: C 64.51, H 5.27, N 3.11.
d
¼ 55.8, 55.9,
58.8, 87.7, 88.2, 111.1, 113.9, 115.4, 117.8, 130.6, 131.8, 134.8, 138.3, 153.3,
154.7, 159.9, 164.5, 187.6, 189.7 ppm; Anal. calcd for C21H18FNO6: C
63.16, H 4.54, N 3.51, found: C 63.27, H 4.71, N 3.64.
4.1.3.5. Ethyl 4-[1-(4-fluorobenzyl)-5,7-dimethoxy-1H-indol-3-yl]-2-
hydroxy-4-oxobut-2-enoate (7e). Yellow solid (415 mg, 97%): mp:
4.1.4.3. 4-[1-(4-Fluorobenzyl)-4,7-dimethoxy-1H-indol-3-yl]-2-hydroxy-
4-oxobut-2-enoic acid (8c). Spectral data are in accordance with the
literature. [17]
110e112 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 1.23 (t, J ¼ 7.1 Hz, 3H, CH3),
3.73 (s, 3H, OCH3), 3.77 (s, 3H, OCH3), 4.09 (q, J ¼ 7.1 Hz, 2H, CH2),
5.51 (s, 2H, CH2), 6.17 (s, 1H, CH), 6.31 (d, J ¼ 1.9 Hz, 1H, ArH),
7.08e7.52 (m, 5H, ArH), 8.49 ppm (s, 1H, ArH); Anal. calcd for
C23H22FNO6: C, 64.63; H 5.19, N 3.28, found: C 64.51, H 5.33, N 3.42.
4.1.4.4. 4-[1-(4-Fluorobenzyl)-5,6-dimethoxy-1H-indol-3-yl]-2-hydroxy-
4-oxobut-2-enoic acid (8d). Yellow solid (183 mg, 46%): mp:
193e195 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 3.78 (s, 6H, OCH3), 5.46 (s, 2H,

L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
763
CH2), 6.93 (s, 1H, CH), 7.14e7.20 (m, 2H, ArH), 7.25 (s, 1H, ArH),
7.42e7.46 (m, 2H, ArH), 7.69 (s, 1H, ArH), 8.77 ppm (s, 1H, ArH). 13C
NMR ([D6]DMSO):
119.4, 129.7, 130.6, 131.8, 134.8, 145.9, 148.1, 159.9, 164.5, 187.6,
189.7 ppm; Anal. calcd for C21H18FNO6: C 63.16, H 4.54, N 3.51, found: C
63.27, H 4.39, N 3.40.
4.3. Docking studies
d
¼ 56.3, 56.8, 58.8, 96.6, 103.4, 111.1, 113.9, 115.4,
The crystal structure of PFV-IN with DNA and two Mg ions
complexed with the inhibitor 2 were retrieved from Protein Data
Bank (pdb code 3L2T) and used for our modeling studies. Hydrogen
atoms were added using the Schrodinger Maestro. The 2 structure
was extracted from X-ray complex, and the other structures of the
ligands were constructed using the Schrodinger Maestro [33], and
were then submitted to Polak-Ribiere conjugate gradient minimi-
zation (0.0005 Kj/Å mol convergence). The compounds were pre-
sented in their enolic tautomeric form, since it has been clearly
established that DKAs mainly exist in this form in solution, the
carboxylic moiety was considered as carboxylate and the enolic
oxygen as enolate given the influence of the two metal ions in the
binding site [34]. The ligands minimized in this way were docked
using GOLD 4.1.2 [30]. The region of interest used by Gold was
defined in order to contain the residues within 15 Å from the
original position of the ligand in the X-ray structures. CHEMPLP
scoring function was chosen as the fitness function, default
parameters were used and the ligands were submitted to 100
genetic algorithm runs.
Test docking calculation was carried out using compound 2 with
the aim of comparing experimental and predicted binding modes
and validating the docking protocol.
The best docking pose found for 2 agreed well with its experi-
mental binding mode even if with root-mean square deviations
(RMSD) of 1.13. This value can be due to the rotation, in the docking
pose, of the oxadiazole ring where the two nitrogen atoms are
oriented in opposite site with respect to X-ray position (see Sup-
porting Information). In fact removing the methyloxadiazole group
resulted in the RMSD value becoming 0.66.
4.1.4.5. 4-[1-(4-Fluorobenzyl)-5,7-dimethoxy-1H-indol-3-yl]-2-hydroxy-
4-oxobut-2-enoic acid (8e). Yellow solid (183 mg, 46%): mp:
175e177 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 3.77 (s, 3H, OCH3), 3.81 (s, 3H,
OCH3), 5.54 (s, 2H, CH2), 6.45 (s,1H, CH), 6.93 (s,1H, ArH), 7.13e7.34 (m,
5H, ArH), 8.81 ppm (s, 1H, ArH). 13C NMR ([D6]DMSO):
d
¼ 55.2, 55.8,
59.0, 85.6, 94.1, 111.1, 113.9, 115.4, 121.9, 128.6, 130.6, 131.8, 134.8, 147.8,
156.9, 159.9, 164.5, 187.6, 189.7 ppm; Anal. calcd for C21H18FNO6: C
63.16, H 4.54, N 3.51, found: C 63.31, H 4.38, N 3.66.
4.1.4.6. 4-[1-(4-Fluorobenzyl)-6,7-dimethoxy-1H-indol-3-yl]-2-hydroxy-
4-oxobut-2-enoic acid (8f). Yellow solid (176 mg, 44%): mp:
169e171 ꢀC; 1H NMR ([D6]DMSO):
d
¼ 3.63 (s, 3H, OCH3), 3.82 (s, 3H,
OCH3), 5.55 (s, 2H, CH2), 6.94 (s, 1H, CH), 7.08e7.21 (s, 5H, ArH), 7.92 (d,
J ¼ 8.5 Hz, 1H, ArH), 8.83 (s, 1H, ArH), 13.78 (bs, 1H, OH), 15.17 ppm (bs,
1H, OH). 13C NMR ([D6]DMSO):
113.9, 115.4, 119.4, 130.0, 130.6, 131.8, 134.8, 137.5, 145.9, 159.9, 164.5,
187.6, 189.7 ppm; Anal. calcd for C21H18FNO6: C 63.16, H 4.54, N 3.51,
found: C 63.33, H 4.37, N 3.72.
d
¼ 56.3, 56.8, 59.0, 111.1, 112.4, 112.9,
4.2. Biological assays
4.2.1. Overall integrase assay using an enzyme-linked
immunosorbent assay (ELISA)
To determine the susceptibility of the HIV-1 integrase enzyme
toward different compounds we used enzyme-linked immunosor-
bent assays. These assays use an oligonucleotide substrate of which
one oligonucleotide (50-ACTGCTAGAGATTTTCCACACTGACTAAAAG-
GGTC-30) is labeled with biotin at the 30 end and the other oligo-
nucleotide is labeled with digoxigenin at the 50 end. For the overall
integration assay the second 50-digoxigenin labeled oligonucleotide
is (50-GACCCTTTTAGTCAGTGTGGAAAATCTCTAGCAGT-30). For the
Strand Transfer assay the second oligonucleotide lacks GT at the 30
end. The integrase enzyme was diluted in 750 mM NaCl, 10 mM Tris
4.4. Minimization process
All of PFV-IN/ligand complexes obtained by docking studies were
minimizedfreezing thetwocations andthe30 OHoxygen, using1000
iterations of SD and 1000 interaction of Polak-Ribiere Conjugate
Gradient. Minimization of complexes was performed using OPLS-
2005[35] force field and GB/SA model [36] as solvation treatment.
pH 7.6, 10% glycerol and 1 mM
reaction 4 l diluted integrase (corresponding to a concentration of
1.6 M) and 4 l of annealed oligonucleotides (7 nM) was added in
a final reaction volume of 40 l containing 10 mM MgCl2, 5 mM DTT,
20 mM HEPES pH 7.5, 5% PEG and 15% DMSO. The reaction was
carried out for 1 h at 37 ꢀC. Reaction products were denatured with
30 mM NaOH and detected by an immunosorbent assay on avidin
coated plates. [32]
b-mercapto ethanol. To perform the
Acknowledgments
m
m
m
This work was supported by THINC project (European
Commission HEALTH-F3-2008-201032).
m
Appendix. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2010.12.012.
4.2.2. In vitro anti-HIV and drug susceptibility assays
The inhibitory effect of antiviral drugs on the HIV-induced
cytopathic effect (CPE) in human lymphocyte MT-4 cell culture was
determined by the MT-4/MTT-assay [29].
References
[1] J.A. Este, T. Cihlar, Current status and challenges of antiretroviral research and
therapy, Antiviral. Res. 85 (2010) 25e33.
This assay is based on the reduction of the yellow colored 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) by
mitochondrial dehydrogenase of metabolically active cells to a blue
formazan derivative, which can be measured spectrophotometri-
cally. The 50% cell culture infective dose (CCID50) of the HIV(IIIB)
strainwas determined bytitrationof the virusstock usingMT-4cells.
For the drug-susceptibility assays, MT-4 cells were infected with
100e300 CCID50 of the virus stock in the presence of five-fold serial
dilutions of the antiviral drugs. The concentration of the various
compounds that achieved 50% protection against the CPE of the
different HIV strains, which is defined as the EC50, was determined.
In parallel the 50% cytotoxic concentration (CC50) was determined.
[2] Y. Mehellou, E. De Clercq, Twenty-six years of anti-HIV drug discovery: where
do we stand and where do we go? J. Med. Chem. 53 (2010) 521e538.
[3] M. Sechi, G. Rizzi, A. Bacchi, M. Carcelli, D. Rogolino, N. Pala, T.W. Sanchez,
L. Taheri, R. Dayam, N. Neamati, Design and synthesis of novel dihy-
droquinoline-3-carboxylic acids as HIV-1 integrase inhibitors, Bioorg. Med.
Chem. 17 (2009) 2925e2935.
[4] M. Melek, J.M. Jones, M.H. O’Dea, G. Pais, T.R. Burke Jr., Y. Pommier,
N. Neamati, M. Gellert, Effect of HIV integrase inhibitors on the RAG1/2
recombinase, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 134e137.
[5] T.K. Chiu, D.R. Davies, Structure and function of HIV-1 integrase, Curr. Top.
Med. Chem. 4 (2004) 965e977.
[6] V. Nair, G. Chi, R. Ptak, N. Neamati, HIV integrase inhibitors with nucleobase
scaffolds: discovery of a highly potent anti-HIV agent, J. Med. Chem. 49 (2006)
445e447.

764
L. De Luca et al. / European Journal of Medicinal Chemistry 46 (2011) 756e764
[7] J. d’Angelo, J.F. Mouscadet, D. Desmaele, F. Zouhiri, H. Leh, HIV-1 integrase: the
next target for AIDS therapy? Pathol. Biol. (Paris) 49 (2001) 237e246.
1H-benzylindole derivatives through fluorine substitutions, J. Med. Chem. 52
(2009) 569e573.
[8] W. Yang, T.A. Steitz, Recombining the structures of HIV integrase, RuvC and
RNase H, Structure 3 (1995) 131e134.
[22] M.L. Barreca, F. Ortuso, N. Iraci, L. De Luca, S. Alcaro, A. Chimirri, Tn5 trans-
posase as a useful platform to simulate HIV-1 integrase inhibitor binding
mode, Biochem. Biophys. Res. Commun. 363 (2007) 554e560.
[23] S. Hare, S.S. Gupta, E. Valkov, A. Engelman, P. Cherepanov, Retroviral intasome
assembly and inhibition of DNA strand transfer, Nature 464 (2010) 232e236.
[24] Y.I. Fukuyama, C. Kodama, M. Ochi, M. Kataoka, K. Shibata, Antimicrobial
indolequinones from the mid-intestinal gland of the muricid gastropod Dru-
pella fragum, Tetrahedron 54 (1998) 10007e10016.
[9] O. Delelis, K. Carayon, A. Saib, E. Deprez, J.F. Mouscadet, Integrase and inte-
gration: biochemical activities of HIV-1 integrase, Retrovirology 5 (2008) 114.
[10] D.J. Hazuda, P. Felock, M. Witmer, A. Wolfe, K. Stillmock, J.A. Grobler,
A. Espeseth, L. Gabryelski, W. Schleif, C. Blau, M.D. Miller, Inhibitors of strand
transfer that prevent integration and inhibit HIV-1 replication in cells, Science
287 (2000) 646e650.
[11] A.A.M. Johnson, C. Pommier, HIV-1 IN inhibitors:a decade of research and two
drugs in clinical trial, Curr. Top. Med. Chem. 4 (2004) 1059e1077.
[12] A.Bacchi,M.Biemmi,M.Carcelli,F.Carta,C.Compari,E.Fisicaro,D.Rogolino,M.Sechi,
M. Sippel, C.A. Sotriffer, T.W. Sanchez, N. Neamati, From ligand to complexes. Part 2.
Remarks on human immunodeficiency virus type 1 integrase inhibition by beta-
diketo acid metal complexes, J. Med. Chem. 51 (2008) 7253e7264.
[25] L. Henry, C.R. Hebd. Acad. Sci. 120 (1895) 1265e1268.
[26] G.C. Condie, M.F. Channon, A.J. Ivory, N. Kumar, D.St.C. Black, Regioselective
reactivity of some 5,7-dimethoxyindoles, Tetrahedron 61 (2005) 4989e5004.
[27] P.D. Rege, Y. Tian, E.J. Corey, Studies of new indole alkaloid coupling methods
for the synthesis of haplophytine, Org. Lett. 8 (2006) 3117e3120.
[28] K. Busschots, A. Voet, M. De Maeyer, J.C. Rain, S. Emiliani, R. Benarous,
L. Desender, Z. Debyser, F. Christ, Identification of the LEDGF/p75 binding site
in HIV-1 integrase, J. Mol. Biol. 365 (2007) 1480e1492.
[13] J. Ramacharan, A.M. Skalka, Strategies for identification of HIV-1integrase
inhibitors, Future Virol. 1 (2006) 717e731.
[14] www.fda.gov/bbs/topics/NEWS/2007/NEW01726.htmlin.
[29] R. Pauwels, J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J. Desmyter,
E. De Clercq, Rapid and automated tetrazolium-based colorimetric assay for the
detection of anti-HIV compounds, J. Virol. Methods 20 (1988) 309e321.
[30] GOLD 4.1.2; Cambridge Crystallographic Data Center: Cambridge CB2 1EZ, UK.
[31] S. Ferro, L. De Luca, M.L. Barreca, S. De Grazia, F. Christ, Z. Debyser, A. Chimirri,
New chloro, fluorobenzylindole derivatives as integrase strand-transfer
inhibitors (INSTIs) and their mode of action, Bioorg. Med. Chem. 18 (2010)
5510e5518.
[32] Y. Hwang, D. Rhodes, F. Bushman, Rapid microtiter assays for poxvirus top-
oisomerase, mammalian type IB topoisomerase and HIV-1 integrase: appli-
cation to inhibitor isolation, Nucleic Acids Res. 28 (2000) 4884e4892.
[33] Maestro, Version 8.0. Schrodinger, LLC, New York, NY, 2007.
[34] C. Maurin, F. Bailly, E. Buisine, H. Vezin, G. Mbemba, J.F. Mouscadet, P. Cotelle,
Spectroscopic studies of diketoacids-metal interactions. A probing tool for the
pharmacophoric intermetallic distance in the HIV-1 integrase active site,
J. Med. Chem. 47 (2004) 5583e5586.
[15] M.L. Barreca, A. Rao, L. De Luca, M. Zappala, C. Gurnari, P. Monforte, E. De
Clercq, B. Van Maele, Z. Debyser, M. Witvrouw, J.M. Briggs, A. Chimirri, Effi-
cient 3D database screening for novel HIV-1 IN inhibitors, J. Chem. Inf.
Comput. Sci. 44 (2004) 1450e1455.
[16] M.L. Barreca, S. Ferro, A. Rao, L. De Luca, M. Zappala, A.M. Monforte,
Z. Debyser, M. Witvrouw, A. Chimirri, Pharmacophore-based design of HIV-1
integrase strand-transfer inhibitors, J. Med. Chem. 48 (2005) 7084e7088.
[17] L. De Luca, M.L. Barreca, S. Ferro, N. Iraci, M. Michiels, F. Christ, Z. Debyser,
M. Witvrouw, A. Chimirri, A refined pharmacophore model for HIV-1 inte-
grase inhibitors: optimization of potency in the 1H-benzylindole series, Bio-
org. Med. Chem. Lett. 18 (2008) 2891e2895.
[18] L. De Luca, A. Pedretti, G. Vistoli, M.L. Barreca, L. Villa, P. Monforte, A. Chimirri,
Analysis of the full-length integrase-DNA complex by a modified approach for
DNA docking, Biochem. Biophys. Res. Commun. 310 (2003) 1083e1088.
[19] L. De Luca, G. Vistoli, A. Pedretti, M.L. Barreca, A. Chimirri, Molecular dynamics
studies of the full-length integrase-DNA complex, Biochem. Biophys. Res.
Commun. 336 (2005) 1010e1016.
[20] M.L. Barreca, N. Iraci, L. De Luca, A. Chimirri, Induced-fit docking approach
provides insight into the binding mode and mechanism of action of HIV-1
integrase inhibitors, ChemMedChem. 4 (2009) 1446e1456.
[35] W.L.M. Jorgensen, D.S. Maxwell, J. Tirado-Rives, Development and testing of
the OPLS all-atom force field on conformational energetics and properties of
organic liquids, J. Am. Chem. Soc. 118 (1996) 11225e11236.
[36] W.C. Still, A. Tempczyk, R.C. Hawley, T. Hendrickson, Semianalytical treatment
of solvation for molecular mechanics and dynamics, J. Am. Chem. Soc. 112
(1990) 6127e6129.
[21] S. Ferro, L. De Luca, M.L. Barreca, N. Iraci, S. De Grazia, F. Christ, M. Witvrouw,
Z. Debyser, A. Chimirri, Docking studies on a new human immunodeficiency
virus integrase-Mg-DNA complex: phenyl ring exploration and synthesis of
[37] W.L. DeLano, The PyMOL Molecular Graphics System. DeLano Scientific LLC,
San Carlos, CA, USA., Available from http://www.pymol.org/.