Investigations on the 4-quinolone-3-carboxylic acid motif. 1. Synthesis and structure-activity relationship of a class of human immunodeficiency virus type 1 integrase inhibitors

Article abstract of DOI:10.1021/jm8003784

A set of 4-quinolone-3-carboxylic acids bearing different substituents on the condensed benzene ring was designed and synthesized as potential HIV-1 integrase inhibitors structurally related to elvitegravir. Some of the new compounds proved to be able to inhibit the strand transfer step of the virus integration process in the micromolar range. Docking studies and quantum mechanics calculations were used to rationalize these data.

Full text of DOI:10.1021/jm8003784

J. Med. Chem. 2008, 51, 5125–5129
5125
Investigations on the 4-Quinolone-3-carboxylic Acid Motif. 1. Synthesis and Structure-Activity
Relationship of a Class of Human Immunodeficiency Virus type 1 Integrase Inhibitors†
Serena Pasquini,‡ Claudia Mugnaini,‡ Cristina Tintori,‡ Maurizio Botta,‡ Alejandro Trejos,§ Riina K. Arvela,§ Mats Larhed,§
Myriam Witvrouw,| Martine Michiels,| Frauke Christ,| Zeger Debyser,| and Federico Corelli*,‡
Dipartimento Farmaco Chimico Tecnologico, UniVersita` degli Studi di Siena, Via A. Moro, 53100 Siena, Italy, Organic Pharmaceutical
Chemistry, Department of Medicinal Chemistry, Uppsala Biomedical Center, Uppsala UniVersity, P.O. Box 574, SE-751 23 Uppsala, Sweden,
Molecular Medicine, Katholieke UniVersiteit LeuVen and IRC KULAK, KapucijnenVoer 33, B-3000 LeuVen, Flanders, Belgium
ReceiVed April 2, 2008
A set of 4-quinolone-3-carboxylic acids bearing different substituents on the condensed benzene ring was
designed and synthesized as potential HIV-1 integrase inhibitors structurally related to elvitegravir. Some
of the new compounds proved to be able to inhibit the strand transfer step of the virus integration process
in the micromolar range. Docking studies and quantum mechanics calculations were used to rationalize
these data.
Introduction
OH and the influence of a protonatable group on the anti-IN
activity of the molecule, respectively. (ii) In most of the
compounds, the benzyl group of 1 and 2 was replaced with
potential bioisosteric moieties with the aim of exploring novel
structural motifs and establishing structure-activity relation-
ships. (iii) In none of the new compounds a methoxy group at
C-7 was introduced because this substituent, although important
for pharmacokinetic and in vivo activity profiles, does not affect
in vitro binding affinity, as demonstrated by the comparable
IC50 values of 2 and its 7-demethoxy derivative (7.2 nM vs 8.1
nM, respectively).2 Finally, quinone derivatives of general
structure 5 were also synthesized where a second carbonyl group
at C-5 might act as an additional chelating moiety, leading to
an improvement of the inhibitory potency against integrase.
The HIV-1 integrase enzyme (IN)a, which is essential for
the replication of the virus and has no cellular counterparts, is
an intriguing target for the development of anti-AIDS drugs.1
Recently, Sato and co-workers2 reported that the 4-quinolone-
3-carboxylic acid antibiotics can be used as an alternative
scaffold to diketo acids (DKA) in order to identify new IN
inhibitors. These novel quinolone IN inhibitors, such as 1 (Chart
1), were structurally optimized into the highly potent compound
2 (GS-9137, elvitegravir), exhibiting potent inhibitory activity
against IN-catalyzed DNA strand transfer (IC50 ) 7.2 nM) and
antiviral activity in vitro (EC50 ) 0.9 nM).2 Despite the high
interest generated by these compounds,3,4 only scattered ex-
amples of quinolones used as the supporting scaffold for the
keto acid motif are described in the literature as IN inhibitors,5
resulting in a total lack of structure-activity relationships
information for this class of inhibitors. On the basis of these
considerations as well as on our interest in the development of
new anti-IN agents,6 we planned to explore the effect of
chemical modifications on the 4-quinolone-3-carboxylic acid
scaffold, with the aim of gaining new insight in the fundamental
structural requirements for anti-IN activity. Keeping in mind
the pharmacophoric groups of 1 and 2 and the synthetic
accessibility of the new molecules, the basic scaffold was
decorated at N-1 and C-6 to provide the final compounds 3 and
4, according to the following considerations: (i) Hydroxyalkyl
chains of variable length were introduced at N-1 to study the
effect of the spacer between the heterocyclic nitrogen and the
OH group; moreover, chains with a terminal alkoxy moiety or
basic nitrogen were used to evaluate the importance of a free
Results and Discussion
Chemistry. Easily accessible anilines 6a-d (Scheme 1) were
condensed with diethyl ethoxymethylenemalonate (EMME) to
give the intermediate enaminoesters that were directly subjected
to cyclization by heating in diphenyl ether to yield 7a-d. These
compounds were then alkylated using an appropriate alkyl halide
to provide derivatives 8a-j in low to good yield (19-85%).
After hydrolysis of 8a-j in 10% aq NaOH, the corresponding
quinolone-3-carboxylic acids 3a-j were obtained in yields
ranging from 17% to 100%.
With this first set of compounds in our hands, different
modifications at the C-6 position of the quinolone scaffold were
made, keeping the 2-hydroxyethyl chain at N-1 fixed. The 6-bromo
derivative 9 (Scheme 2), in turn prepared by alkylation of ethyl
6-bromo-4-oxo-1,4-dihydroquinoline-3-carboxylate, was subjected
to different metal-catalyzed coupling reactions under high density
microwave irradiation7 to give, after basic hydrolysis of the
intermediate products, the quinolone-3-carboxylic acids 3k-n and
4a,b. Compound 3o was prepared in 28% overall yield by
hydrolysis of the corresponding ester, which was in turn synthesized
by a Buchwald-Hartwig amination of ethyl 6-bromo-1-ethyl-4-
oxo-1,4-dihydroquinoline-3-carboxylate in order to assess the actual
relevance for biological activity of the hydroxyalkyl side chain at
N-1 by direct comparison between 3o and 3n.
†
Dedicated with respect and gratitude to Professor Marino Artico on
the occasion of his retirement.
* To whom correspondence should be addressed. Phone: +39 0577
234308. Fax: +39 0577 234333. E-mail: corelli@unisi.it.
‡
Dipartimento Farmaco Chimico Tecnologico, Universita` degli Studi
di Siena.
§
Organic Pharmaceutical Chemistry, Department of Medicinal Chem-
istry, Uppsala Biomedical Center, Uppsala University.
|
Molecular Medicine, Katholieke Universiteit Leuven and IRC KULAK.
a Abbreviations: IN, integrase; 3′-P, 3′-processing; ST, strand transfer;
DKA, ꢀ-diketoacid; EMME, diethyl ethoxymethylenemalonate; BINAP,
2,2′-bis(diphenylphosphino)-1′,1-binaphthyl; Herrmann’s palladacycle, trans-
di(µ-acetato)bis[o-tolylphosphino)benzyl]dipalladium(II); Xantphos, 4,5-
bis(diphenylphosphino)-9,9-dimethylxanthene; MW, microwave; CPE, cy-
tophatic effect; Tnp, transposase; vdW, van der Waals.
Quinone derivative 5a, 5b, and 5c were obtained in satisfac-
tory yield of 30-44% by reacting 1-ethyl-4,5,8-trioxo-1,4,5,8-
tetrahydroquinolin-3-carboxylic acid (10)8 with 3-chloro-2-
10.1021/jm8003784 CCC: $40.75
2008 American Chemical Society
Published on Web 07/30/2008

5126 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 16
Brief Articles
Chart 1. Structure of Elvitegravir and New 4-Quinolones Inhibiting HIV-1 Integrase
Scheme 1. Synthesis of 6-Substituted 4-Quinolone-3-carboxylic Acids 3a-ja
a Reagents and conditions: (i) (1) EMME, 120 °C, 1 h; (2) Ph2O, reflux, 2 h. (ii) R4Y, K2CO3, DMF, 90 °C, 12 h. (iii) 10% aq NaOH, reflux, 2 h.
fluoroaniline,3-chloro-4-fluoroaniline,andthiophenol,respectively,
according to a known procedure (see Supporting Information).8
Inhibition of HIV-1 IN Activity. Compounds 3c-e, structur-
ally related to 2 and characterized by a hydroxyalkyl chain of
variable length at N-1 and a 2,4-dichlorobenzyl moiety at C-6,
showed anti-IN activity in the low micromolar range, being at least
20 times less potent than the reference compound 2 in the strand
transfer (ST) assay1 (Table 1). The introduction of a methoxyethyl
or a dimethylaminoethyl chain at N-1 resulted in 3g and 3f,
respectively, for which a further significant decrease in activity
was observed. This result highlights the fundamental role played
by the free terminal hydroxy group on the N-1 chain. The
replacement of the benzyl group of 3c-e with a bioisosteric
benzoyl group resulted in 3h, 3i, and 3j, respectively, which,
independently by the length of the chain at N-1, proved to be
completely inactive in the ST assay (IC50 > 100 µM). When the
2,4-dichlorobenzyl moiety of 3c was replaced with a 3-chloro-4-
fluorophenoxy group (3a), a loss of activity in the ST assay was
observed as well; the inhibitory potency was partially restored with
a thiophenyl group even if unsubstituted (3b, IC50 ) 18.5 µM).
Removal of the spacer between the two aromatic rings gave 4a
and 4b, characterized by a phenyl or a pyridyl group directly linked
to C-6, which proved to be completely inactive in inhibiting the
enzyme. Similarly, by increasing the distance between the two
aromatics by the use of rigid systems such as double bond (3k),

Brief Articles
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 16 5127
described by Barreca et al.12 In its best docking pose (conforma-
tion with the highest fitness score and belonging to the most
populated cluster), compound 2 (Figure 1A) presented the
carboxylate group, hydrogen-bonded to H329, chelating the
metal ion between E326 and D97 and the ꢀ-ketone oxygen
together with the fluorine atom coordinating the other metal
ion. Moreover, the hydroxy group of the N-1 chain made two
H-bonding interactions with S100, while the C-7 methoxy group
did not seem to be involved in any interaction. Finally, the 2-F,
3-Cl substituted benzyl group was perpendicular to the quinolone
ring, establishing favorable vdW interactions with E190 and
D188. Interestingly, the binding mode proposed for compound
2 is in perfect agreement with its mutation profile (see validation
of docking studies in Supporting Information) as well as with
the pharmacophoric model for quinolones recently reported.3
To facilitate the analysis of the biological data, quinolones 3a-o,
4a,b, 5a-c, and 10 were divided into two main groups on the
Scheme 2. Synthesis of 6-Substituted 4-Quinolone-3-carboxylic
Acids 3k-n and 4a,b by Metal-Catalyzed, Microwave-Assisted
Reactionsa
a Reagents and conditions: (i) (for 3k, 4a, 4b) appropriate arylboronic
acid, Pd(PPh3)2Cl2, 2 M Na2CO3, EtOH/DME, MW, 120 °C, 30 min; (for
3l) aniline, Mo(CO)6, Herrmann’s palladacycle, [(t-Bu)3PH]BF4, 2 M
Na2CO3, H2O, MW, 150 °C, 15 min; (for 3m) benzamide, Pd(dba)2,
Xantphos, Cs2CO3, NMP, dioxane, MW, 140 °C, 1 h; (for 3n) 3-chloro-
2-fluoroaniline, Pd(OAc)2, BINAP, Cs2CO3, toluene, MW, 150 °C, 15 min.
(ii) 10% aq NaOH, reflux, 2 h.
amide (3l), or inverse amide (3m), a loss of activity was observed,
thus confirming the importance of a methylene or sulfur spacer.
Most of the tested compounds displayed in the overall assay9
IC50 values lower than those found in the ST test, thus raising the
question whether their activity might be more likely due to the
inhibition of DNA binding or 3′-processing1 (3′-P) steps. Accord-
ingly, some selected compounds were subjected to biological
evaluation in the 3′-P assay in order to assess their actual
mechanism of IN inhibition. However, none of the compounds
proved to be more active in this test than in either the overall or
the ST assays. At the moment, no explanation can be given for
this result except that the three reactions (DNA binding, 3′-P, ST)
occur in a coupled mode in the overall integration assay. Finally,
the results obtained for the class of quinone derivatives (10, 5a-c),
all characterized by the presence of an ethyl chain at N-1, deserve
some attention. Starting from the unsubstituted compound 10 (IC50
) 58.7 µM), the introduction of a substituted phenylamino (5b,
IC50 ) 2.1 µM) or phenylthio moiety (5c, IC50 ) 1.6 µM) at C-6
gave compounds with IC50 values in the low micromolar range.
This positive effect on the inhibitory activity could be ascribed to
the introduction on the molecule of an additional chelating moiety.
Moreover, 10 and 5b,c exhibited IC50 values of the same order of
magnitude in both the ST and overall assays, thus suggesting a
completely selective mechanism of action for this class of
compounds.
basis of their IC50 value in the ST assay: compounds with IC50
<
100 µM were classified as active, while those endowed with an
IC50 > 100 µM were considered inactive. The active compounds
3b-f displayed a binding mode similar to that of 2, showing the
same important interactions with the exception of that involving
the fluorine atom on the benzyl moiety, which, in our compounds,
is replaced by a chlorine or a hydrogen atom. The lower activity
of 3f and 3g with respect to 3c could be explained by the lack of
the hydrogen bond between the side chain OH and S100. On the
other hand, the inactive compounds (3a, 3k-o, and 4a,b), because
of their enhanced conformational rigidity, were not able to assume
a conformation similar to that of the active compounds (Figure
1B).
It is worth noting that compounds 3h-j, with a carbonyl
substituent at C-6, showed a peculiar behavior, binding the
enzyme in a similar way as the active compounds although
showing no activity in the ST assay. To explain this finding,
theoretical quantum mechanics calculations were performed on
these molecules with the aim to measure the partial charge on
each atom. These calculations suggest for 3 h-j an electron-
withdrawing effect on the chelating system due to the C-6
carbonyl substituent, which results in a poor interaction of the
molecules with the divalent metal ions of the active site. As an
example, results obtained for 3c and 3h are comparatively shown
in Figure 2S of Supporting Information.
In Vitro HIV-1 Assay and Drug Susceptibility Assay. The
antiviral activity of the new compounds on the HIV-induced
cytopathic effect (CPE) in human lymphocyte MT-4 cell culture
was determined by the MT-4/MTT-assay. All compounds
proved to be inactive in inhibiting HIV-1 replication at
subcytotoxic concentration (Table 1). It should be noted that 2
proved to be the most cytotoxic compound under the test
With respect to the quinone series (compounds 5a-c and
10), docking results were unable to completely explain their
activity profile. In agreement with biological results, 5c showed
a similar orientation to that described above for the active
compounds. Conversely, a different docking mode was observed
for compounds 5a,b and 10 (Figure 1C). In fact, these
compounds presented the carboxylate group, hydrogen-bonded
to K164, chelating the metal ion between D188 and D97 and
the two carbonyl oxygens coordinating the ion between D97
and E326. Moreover, for compounds 5a and 5b, the anilino
moiety and the quinone ring established favorable interaction
with W125. This type of binding mode was not able to explain
the inactivity of 5a with respect to 5b because both of these
two compounds are expected to interact in the same way with
the enzyme. Further studies are ongoing to clarify the struc-
ture-activity relationship for the quinone series.
conditions. However, due to its high antiviral potency (EC50
)
0.00037 µM), it elicited a very favorable selectivity index
(3108).
Docking Studies. It has previously been reported that the
quinolone integrase inhibitor 2 selectively inhibits the ST step
via a mechanism similar to that of the DKA.2 Accordingly, the
activity of quinolones is likely due to sequestration of divalent
metal ions within IN active site in complex with viral DNA.10
Although the crystallographic structure of IN in complex with
viral DNA has not yet been resolved, it has been reported that
Tn5 transposase (Tnp), which belongs to the superfamily of
polynucleotidyl transferases as IN, can be considered as an
excellent surrogate model for studying the mechanism of action
of ST inhibitors.11,12 In this context, docking calculations on
our compounds together with 2 were performed on Tn5
Tnp-DNA complex by following the computational protocol
Conclusions
The results of this first attempt aimed at rationalizing the
influence on the anti-IN activity of the different decorating elements
introduced on the 4-quinolone-3-carboxylic acid scaffold allowed
us to highlight new aspects of structure-activity relationships for

5128 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 16
Brief Articles
Table 1. Inhibition of HIV-1 Integrase Enzymatic Activity, Replication of HIV-1 (IIIB), and Cytotoxicity in MT-4 cells
a
IC50 (µM)
b
c
compd
overall
3′-P
>100
ST
EC50 (µM)
CC50 (µM)
3a
3b
3c
3d
3e
3f
3g
3h
3i
86.9 ( 24.4
4.8 ( 2.6
0.07 ( 0.03
0.9 ( 0.4
0.05 ( 0.0
10.8 ( 5.4
5.6 ( 2.6
18.7 ( 11.7
73.9 ( 49.0
54.9 ( 33.9
71.9 ( 13.6
>100
>100
>9.9
9.9
22.7
5.3
3.6
4.2
5.2
4.2
21.1
17.5
17
18.5 ( 3.7
0.2 ( 0.0
1.3 ( 0.1
0.6 ( 0.3
24.1 ( 5.9
16.5 ( 0.7
>100
>22.7
>5.3
>3.6
16.6 ( 4.8
>4.2
>5.2
>4.2
>100
>100
>100
>21.1
>100
>17.5
3j
3k
3l
>100
>17
>100
>22.6
22.6
224
220.5
27.4
9.8
44.9
138.1
2.4
>100
>224
3m
3n
3o
4a
4b
5a
5b
5c
10
1
>100
>100
>100
>220.5
>27.4
>100
38.0 ( 0.83
>100
77.6 ( 19.0
9.7 ( 1.6
1.0 ( 0.1
4.6 ( 0.1
43.7 ( 18.2
>100
>100
>9.8
>44.9
>100
>100
>100
>100
>138.1
>2.4
>100
2.1 ( 0.5
1.6 ( 0.3
58.7 ( 5.7
0.0242 ( 0.0116d
0.007 ( 0.0007
>2.2
2.2
>30.0
30.0
42.8
>15d
1.15
>42.8
0.0763 ( 0.0037d
0.00037 ( 0.00004
2
0.002 ( 0.001
0.05 ( 0.01
a Concentration required to inhibit by 50% the in vitro integrase activities. b Effective concentration required to reduce HIV-1-induced cytopathic effect
by 50% in MT-4 cells. c Cytotoxic concentration to reduce MT-4 cell viability by 50%. d Taken from ref 2.
trans-2-phenylvinylboronic acid), Pd(PPh3)2Cl2 (8.8 mg, 0.0125
mmol), 2 M Na2CO3 (0.2 mL, 0.4 mmol, for 4a; 0.38 mL, 0.75
mmol, for 4b; 0.25 mL, 0.5 mmol, for 3k), 0.6 mL of EtOH, and
2.4 mL of DME. The vessel was sealed under air and exposed to
microwave heating for 30 min at 120 °C (for 4a,b) or 20 min. at
130 °C (for 3k). The reaction mixture was thereafter cooled down
to room temperature, diluted with DCM, filtered through a short
plug of neutral aluminum oxide, and purified by flash chromatog-
raphy on silica gel (99:1 to 95:5 DCM/MeOH). The ester so
obtained was hydrolyzed with 10% NaOH as described for 3a-j.
1-(2-Hydroxyethyl)-4-oxo-6-(phenylcarbamoyl)-1,4-dihydro-
quinoline-3-carboxylic Acid (3l). A 5 mL microwave vial was
charged with 9 (149.0 mg, 0.39 mmol), aniline (0.071 mL, 0.78
mmol), molybdenum hexacarbonyl (103.6 mg, 0.39 mmol), 2 M
Na2CO3 (0.59 mL, 1.2 mmol), trans-di(µ-acetato)bis[o-tolylphos-
phino)benzyl]dipalladium(II) (Herrmann’s palladacycle) (18.3 mg,
0.019 mmol), [(t-Bu)3PH]BF4 (22.8 mg, 0.078 mmol), and water
(1.1 mL). Subsequent to the addition of reagents and solvents, the
vial was capped under air with a Teflon septum. The mixture was
thereafter irradiated with microwaves to 150 °C for 15 min. After
cooling, the crude mixture was filtered through a plug of celite
and the mixture of partially hydrolyzed compounds so obtained
was treated with 10% NaOH (4 mL) under reflux for 2 h. After
cooling at room temperature, the reaction mixture was acidified to
pH 3 using conc. HCl. The resulting precipitate was filtered and
washed with water to give 3l (27.5 mg, 20% overall yield).
IN inhibiting quinolones. In particular, the substitution of the benzyl
group with aryl, styryl, aroylamino, and aniline groups results in
inactive compounds that are not able to adopt the bioactive
conformation; conversely, the phenylthio group is able to mimic
quite efficiently the benzyl group. On the other hand, the substitu-
tion of the benzyl group with the electron-withdrawing benzoyl
moiety gives compounds that can be accommodated in the same
orientation as the active molecules within the binding site but show
a reduced chelating ability. Finally, the replacement of the benzene
ring of the quinolone with a quinone moiety gives compounds that,
with the exception of 5c, adopt a different orientation in the binding
site and likely act through a pure ST inhibition mechanism.
Experimental Section
Synthesis of 6-Substituted Quinolones 7a-d: General Pro-
cedure. A mixture of the appropriate aniline 6a-d (1 mmol) and
diethyl ethoxymethylenemalonate (200 µL, 1 mmol) was heated at
120 °C for 1 h. The crystalline solid obtained after cooling at room
temperature was dissolved in diphenyl ether (3 mL), and the solution
was refluxed for 2 h. After cooling to room temperature, diethyl ether
(4 mL) was added and the precipitated solid was filtered and
recrystallized from DMF.
Synthesis of 1,6-Disubstituted Quinolones 8a-j and 9: Gen-
eral Procedure. A mixture of the appropriate quinolone derivative
7a-d (1 mmol) or ethyl 6-bromo-4-oxo-1,4-dihydroquinoline-3-
carboxylate (296 mg, 1 mmol), the appropriate alkyl halide (2.8 mmol),
and solid K2CO3 (386 mg, 2.8 mmol) in dry DMF (1 mL) was heated
under N2 atmosphere at 90 °C for 12 h and then poured into an
ice-water mixture. The precipitated N-alkylquinolone was collected
by filtration and recrystallized or extracted into DCM and purified by
flash chromatography on silica gel eluting with DCM/MeOH (98:2).
Synthesis of 4-Quinolone-3-carboxylic Acids 3a-j: General
Procedure. A suspension of the appropriate ester (1 mmol) in 10%
aq. NaOH (10 mL) was refluxed for 2 h. After cooling at room
temperature, the reaction mixture was acidified using conc. HCl.
The resulting precipitate was filtered and washed with water to give
the corresponding 4-quinolone-3-carboxylic acid.
6-Benzamido-1-(2-hydroxyethyl)-4-oxo-1,4-dihydroquinoline-
3-carboxylic Acid (3m). A 5 mL microwave vial was charged with
9 (105.0 mg, 0.27 mmol), benzamide (99.8 mg, 0.82 mmol), Cs2CO3
(358.2 mg, 1.1 mmol), Pd(dba)2 (9.5 mg, 0.016 mmol), 4,5-
bis(diphenylphosphino)-9,9-dimethylxanthene (Xantphos) (19.0 mg,
0.033 mmol), and 10% NMP in dioxane (3.6 mL). Subsequent to
the addition of reagents and solvents, the vial was capped under
air with a Teflon septum. The mixture was irradiated with
microwaves at 140 °C for 1 h. After cooling, the crude mixture
was filtered through a plug of celite and hydrolyzed with 10%
NaOH as described for 3l.
Synthesis of Quinolones 3n and 3o by Buchwald-Hartwig
Reaction: General Procedure. A 5 mL microwave vial was
charged with 9 (149.0 mg, 0.39 mmol) or ethyl 6-bromo-1-ethyl-
4-oxo-1,4-dihydroquinoline-3-carboxylate (126.0 mg, 0.39 mmol),
3-chloro-2-fluoroaniline (142.0 mg, 0.97 mmol), Cs2CO3 (381.0 mg,
Synthesis of Quinolones 3k and 4a,b by Suzuki Coupling:
General Procedure. A 5 mL process vial was charged with 9 (95.5
mg, 0.25 mmol), the appropriate boronic acid (1.25 mmol for
phenylboronic acid and 4-pyridineboronic acid, 0.75 mmol for

Brief Articles
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 16 5129
and Excellent-Hit (LSHP-CT-2006-037257). We thank Prof.
Eiichi Kodama from Kyoto University for providing us with a
copy of the paper cited in the text as ref 4 before the Web access.
The authors from Uppsala University thank the Swedish
Research Council and Knut and Alice Wallenberg’s Foundation
for financial support, and Dr. Luke Odell for reading the
manuscript. The Leuven laboratory acknowledges B. Van
Remoortel, L. Desender, and N. J. Van der Veken for technical
help with protein purification and antiviral testing.
Supporting Information Available: Detailed biological and
molecular modeling protocols; analytical and spectral data for
compounds 3a-o, 4a,b, 5a-c, 7a-d, 8a-j, 9. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Pommier, Y.; Johnson, A. A.; Marchand, C. Integrase inhibitors to
treat HIV/AIDS. Nat. ReV. Drug DiscoVery 2005, 4, 236–248.
(2) Sato, M.; Motomura, T.; Aramaki, H.; Matsuda, T.; Yamashita, M.;
Ito, Y.; Kawakami, H.; Matsuzaki, Y.; Watanabe, W.; Yamataka, K.;
Ikeda, S.; Kodama, E.; Matsuoka, M.; Shinkai, H. Novel HIV-1
Integrase Inhibitors Derived from Quinolone Antibiotics. J. Med.
Chem. 2006, 49, 1506–1508.
(3) Dayam, R.; Al-Mawsawi, L. Q.; Zawahir, Z.; Witvrouw, M.; Debyser,
Z.; Neamati, N. Quinolone 3-carboxylic acid pharmacophore: Design
of second generation HIV-1 integrase inhibitors. J. Med. Chem. 2008,
51, 1136–1144.
(4) Shimura, K.; Kodama, E.; Sakagami, Y.; Matsuzaki, Y.; Watanabe,
W.; Yamataka, K.; Watanabe, Y.; Ohata, Y.; Doi, S.; Sato, M.; Kano,
M.; Ikeda, S.; Matsuoka, M. Broad Anti-Retroviral Activity and
Resistance Profile of a Novel Human Immunodeficiency Virus
Integrase Inhibitor, Elvitegravir (JTK-303/GS-9137). J. Virol. 2008,
82, 764–774.
(5) Di Santo, R.; Costi, R.; Roux, A.; Artico, M.; Lavecchia, A.; Marinelli,
L.; Novellino, E.; Palmisano, L.; Andreotti, M.; Amici, R.; Galluzzo,
C. M.; Nencioni, L.; Palamara, A. T.; Pommier, Y; Marchand, C. Novel
bifunctional quinolonyl diketo acid derivatives as HIV-1 integrase
inhibitors: design, synthesis, biological activities, and mechanism of action.
J. Med. Chem. 2006, 49, 1939–1945, and references cited therein.
(6) Mugnaini, C.; Rajamaki, S.; Tintori, C.; Corelli, F.; Massa, S.;
Witvrouw, M.; Debyser, Z.; Veljkovic, V.; Botta, M. Toward novel
HIV-1 integrase binding inhibitors: molecular modeling, synthesis, and
biological studies. Bioorg. Med. Chem. Lett. 2007, 17, 5370–5373,
and references cited therein.
(7) Nilsson, P.; Olofsson, K.; Larhed, M. Microwave-assisted and metal-
catalyzed coupling reactions. Top. Curr. Chem. 2006, 266, 103–144.
(8) Link, V. H.; Bernauer, K.; Englert, G. Abko¨mmlinge der 3-Chino-
lincarbonsa¨ure mit Sauerstoffsubstitution in Stellung 4, 5 und 8:
Synthese, Reaktionen, NMR-Studien. HelV. Chim. Acta 1982, 65,
2645–2667.
(9) Debyser, Z.; Cherepanov, P.; Pluymers, W.; De Clercq, E. Assays for
the evaluation of HIV-1 integrase inhibitors. In Methods in Molecular
Biology; Schein, C. H., Ed.; Humana Press Inc.: Totowa, NJ, 2001;
pp 139-155.
(10) Richter, S. N.; Gatto, B.; Tabarrini, O.; Fravolini, A.; Palumbo, M.
Antiviral 6-aminoquinolones: molecular basis for potency and selectiv-
ity. Bioorg. Med. Chem. Lett. 2005, 15, 4247–4251.
Figure 1. Binding mode of the studied compounds within the Tnp
Tn5 binding site (PDB code 1MUS). (A) Binding mode of compound
2 (magenta). (B) Binding mode of compound 3n (yellow) superimposed
to 3c (blue). Interestingly, the anilino moiety of 3n is not able to
correctly overlay the benzyl group of 3c. (C) Binding mode of
compound 5b (pink). Metal ions are presented as green balls. Tn5 Tnp/
ligand interactions are shown as dashed lines (metal coordination in
yellow, hydrogen bonds in white).
1.17 mmol), Pd(OAc)2 (4.3 mg, 0.0195 mmol), BINAP (24.3 mg,
0.039 mmol), and toluene (4.15 mL). After the addition of reagents
and solvents, the vial was capped with a Teflon septum. The reaction
mixture was irradiated with microwaves at 150 °C for 15 min. The
crude mixture was filtered through a plug of celite and then purified
by flash chromatography on silica gel (hexane:AcOEt gradient 40:
60 to 20:80 for the precursor of 3n, and DCM:MeOH 98:2 to 94:6
for the precursor of 3o). The product so obtained was hydrolyzed
as described above to give the title compound.
(11) Ason, B.; Knauss, D. J.; Balke, A. M.; Merkel, G.; Skalka, A. M.;
Reznikoff, W. S. Targeting Tn5 transposase identifies human immu-
nodeficiency virus type 1 inhibitors. Antimicrob. Agents Chemother.
2005, 49, 2035–2043.
(12) Barreca, M. L.; De Luca, L.; Iraci, N.; Chimirri, A. Binding mode
prediction of strand transfer HIV-1 integrase inhibitors using Tn5
transposase as a plausible surrogate model for HIV-1 integrase. J. Med.
Chem. 2006, 49, 3994–3997.
Integrase Assays. The enzymatic integration reactions were
carried out as described previously with minor modifications.9
Acknowledgment. This project was supported by grants
from the European TRIoH Consortium (LSHB-2003-503480)
JM8003784