Synthesis, antiviral activity and molecular modeling of oxoquinoline derivatives

Article abstract of DOI:10.1016/j.bmc.2009.06.037

In the present article, we describe the synthesis, anti-HIV1 profile and molecular modeling evaluation of 11 oxoquinoline derivatives. The structure-activity relationship analysis revealed some stereoelectronic properties such as LUMO energy, dipole momen

Full text of DOI:10.1016/j.bmc.2009.06.037

Bioorganic & Medicinal Chemistry 17 (2009) 5476–5481
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, antiviral activity and molecular modeling of oxoquinoline derivatives
d
Fernanda da C. Santos a, Paula Abreu b, Helena C. Castro b, Izabel C. P. P. Paixão b, , Claudio C. Cirne-Santos ,
*
Viveca Giongo b, Juliana E. Barbosa b, Bruno R. Simonetti b, Valéria Garrido b, Dumith Chequer Bou-Habib d,
David de O. Silva a, Pedro N. Batalha a, Jairo R. Temerozo d, Thiago M. Souza b, Christiane M. Nogueira a,
Anna C. Cunha a, Carlos R. Rodrigues c, Vitor F. Ferreira a, Maria C. B. V. de Souza a,
*
a Universidade Federal Fluminense, Instituto de Química, Departamento de Química Orgânica, Programa de Pós-Graduação em Química,
Outeiro de São João Batista, s/n, Centro, Niterói, CEP 24210-150, RJ, Brazil
b Universidade Federal Fluminense, Instituto de Biologia, Departamento de Biologia Celular e Molecular, Programa de Pós Graduação em Neurociências,
Outeiro de São João Batista s/n, Centro, Niterói, CEP 24020-150, RJ, Brazil
c Laboratório de Modelagem Molecular e QSAR (ModMolQSAR), Faculdade de Farmácia, Universidade Federal do Rio de Janeiro, Rio de Janeiro, CEP 21941-590, RJ, Brazil
d Laboratório de Pesquisas Sobre o Timo, Instituto Oswaldo Cruz/Fiocruz, Rio de Janeiro, RJ 21045-900, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
In the present article, we describe the synthesis, anti-HIV1 profile and molecular modeling evaluation of
11 oxoquinoline derivatives. The structure–activity relationship analysis revealed some stereoelectronic
properties such as LUMO energy, dipole moment, number of rotatable bonds, and of hydrogen bond
donors and acceptors correlated with the potency of compounds. We also describe the importance of sub-
stituents R2 and R3 for their biological activity. Compound 2j was identified as a lead compound for future
investigation due to its : (i) high activity against HIV-1, (ii) low cytotoxicity in PBMC, (iii) low toxic risks
based on in silico evaluation, (iv) a good theoretical oral bioavailability according to Lipinski ‘rule of five’,
(v) higher druglikeness and drug-score values than current antivirals AZT and efavirenz.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 18 April 2009
Revised 17 June 2009
Accepted 18 June 2009
Available online 24 June 2009
This paper is dedicated to late Professor
Octávio Augusto Ceva Antunes to
acknowledge his many contributions to
organic chemistry
Keywords:
Antiviral
HIV-1
Oxoquinoline
Structure–activity relationship (SAR)
1. Introduction
vaccine against HIV, efforts to provide drug therapies stand as a
great success.
Acquired immunodeficiency syndrome (AIDS) caused by the
human immunodeficiency virus (HIV) remains a health threat of
global significance. Although the highly active anti-retroviral ther-
apy (HAART) effectively reduced the mortality rate and prolonged
the life expectancy in developing and developed countries, it has
not eradicated HIV-1 from the infected tissues.1 The powerful drug
cocktails can suppress virus replication, often keeping blood levels
so low as to be undetectable by standard tests.2 Since viral replica-
tion remains active in cellular reservoirs, the long-term use of the
combined therapy is mandatory but in a restricted way due to the
increased prevalence of HIV-1 resistant strains, metabolic disor-
ders, and complex administration.3 Therefore, the search for new
antiretroviral agents is crucial, and compounds that inhibit differ-
ent steps of HIV-1 replication cycle are under development or clin-
ical investigation. In contrast to the failed attempts at developing a
Quinolone derivatives are a class of potential drugs that may
contribute to the control of the latent HIV-1 reservoir.4 Oxoquino-
line ribonucleoside 1 has been shown to inhibit HIV-1 replication
in vitro in both acutely and chronically HIV-infected cell lines by
interfering with Tat-mediated transcription.5
Due to the need of new antivirals to deal with the HIV-resis-
tance to antivirals, we have been testing the oxoquinoline deriva-
tives against HIV virus replication in order to search for
biologically active compounds.5 We have recently identified chlo-
roxoquinolinic ribonucleoside 6-chloro-1,4-dihydro-4-oxo-1-(D-
ribofuranosyl)-quinoline-3-carboxylic acid (1) as an inhibitor of
the RNA-dependent DNA polymerase (RDDP) activity of HIV1-
Reverse Transcriptase with uncompetitive and noncompetitive
modes of action with respect to dTTP incorporation and to
template/primer (TP) uptake, respectively. Interestingly, this
compound presents an extremely low cytotoxic effect compared
to AZT.6 Nucleoside analogs are prodrugs that require conversion
to their triphosphate forms by cellular kinases to target the
* Corresponding authors. Tel.: +55 21 26292361; fax: +55 21 26292135 (M.C.B.V.S.).
E-mail address: gqocica@vm.uff.br (M.C.B.V. de Souza).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.06.037

F.C. Santos et al. / Bioorg. Med. Chem. 17 (2009) 5476–5481
5477
transcriptase reverse enzyme, to become active even in uninfected
activities were observed for compounds 2f (EC50 = 2.5
lM), 2g
cells. On the contrary, this oxoquinoline derivative (1) inhibits HIV-
1 replication without requiring any phosphorylation reaction, what
may explain its low toxicity. Besides, this compound presented
additive effects with AZT and nevirapine and a synergistic effect
with the protease inhibitor atazanavir, classical drugs currently
used in antiretroviral therapy.6
(EC50 = 3.3 M), 2j (EC50 = 4.5 M), and 2k (EC50 = 2.8 M).
l
l
l
2.2.2. Cytotoxicity effect
The evaluation of oxoquinoline derivatives using results of
peripheral blood mononuclear cells (PBMCs) showed that none of
the compounds were cytotoxic (Table 1). The CC50 values revealed
In this work, based on the chloroxoquinolinic ribonucleoside
(1), we planned a new series of oxoquinoline derivatives (2a–k)
to identify a structure–activity relationship. Thus, we performed
changes in structural and electronic properties as the replacement
of the riboside ring by hydrogen (2a and 2h–i) and hydrophobic
group such as benzyl (2b–c) (subunit A—R3 position). We also
tested the ring-opening of riboside (acyclic chain, 2d–g and 2j–k)
(R3 position). We changed the chorine by the fluorine atom in
the R1 position (2e, 2g, 2k) in order to increase the electronegativ-
ity effect. In addition, we performed the substitution of the acidic
function by other different functional groups in the R2 position
(subunit B): ethyl ester (2a–b, 2d–e), hydrazide (2f–h) or amide
(2i–k) (Scheme 1).
that compounds 2f (CC50 = 2400
(CC50 = 2800 M) were less cytotoxic than other compounds and
also better than AZT (CC50 = 126 M) and 1 (CC50 = 1701 M) un-
lM), 2j (CC50 = 3200 lM), and 2k
l
l
l
der the same assay conditions.6 This reflects the low cytotoxicity
of these compounds, which is in agreement with their high selec-
tive index (SI) (Table 1). The SI calculated based on the ratio of
CC50 and EC50 showed 2f, 2j, and 2k with the best results (960,
711, and 1000, respectively). In fact, they were close to AZT
(SI = 2520) and better than the commercial antivirals DDC
(SI = 110) and DDI (SI = 10).12
2.3. Molecular modeling analysis of oxoquinolines derivatives
Structure–activity relationship (SAR) studies of oxoquinolines
revealed some important structural features apparently related
to the anti-HIV activity. According to the theoretical results, the
most active compounds (2f, 2g, 2j, and 2k) with low toxicity risks
in PBMC and high selective index showed low values of LUMO en-
ergy, which is related to a higher reactivity of the compound as
electrophile. Most of the active compounds also presented higher
number of hydrogen bond acceptor and donor groups as well as
of rotatable bonds (Table 2) with lower dipole moment values
(not shown). These features are also observed for compound 1 pre-
senting the HIV-1 Reverse Transcriptase NNRTI binding pocket as
target.13
Since complementarity is essential for the interaction of active
compounds with this target enzyme13, we analyzed the tridimen-
sional structure and the electrostatic potential map, which allow
the evaluation of size, shape, and superficial charge distribution
to be related to the antiviral potency. Interestingly, in R3, com-
pounds 2b–g, 2j, and 2k present a similar overall shape with elec-
tronegative regions corresponding to the oxygen atom. The
negative region in R2 is related to the electron pair on nitrogen
or the electron density on the phenyl ring (Fig. 1).
2. Results and discussion
2.1. Chemistry
Compounds 2a–c and 2h (Scheme 2) were prepared by standard
procedures.7–10 We have previously described acyclonucleosides
2d and 2e.11
Carboxamides 2i and 2i0 (Scheme 2) were prepared by nucleo-
philic substitution at the carbonyl ester using benzylamine as the
nucleophile and phenyl ether as the solvent, at 210 °C.9 Then, pre-
vious silylation of these oxoquinolines by heating with bis(tri-
methylsilyl)trifluoroacetamide (BSTFA)11 containing 1% of
trimethylchlorosilane (TMCS), in acetonitrile, under reflux, fol-
lowed by addition of equimolar amount of 1,3-dioxolane, chlorotri-
methylsilane, and potassium iodide, at room temperature, and
subsequent neutralization with solid sodium bicarbonate afforded
the new acyclonucleosides 2j and 2k (Scheme 2).
Acyclonucleosides 2d and 2e (Scheme 2) were submitted to
nucleophilic substitution reaction with hydrazine monohydrate
(80%)9 in refluxing ethanol affording acylhydrazides 2f and 2g in
good yields, 75 and 70%, respectively.
Since all active compounds present a substituent in the R3 posi-
tion, this structural feature seems to be important for their activity.
The comparison of 2f and 2j with 2h and 2i, respectively, differing
only in the R3 position, clearly reinforced this hypothesis. The char-
acteristic of the substituent is also important, since the active com-
pounds present a hydrogen bond donor and a hydrogen bond
acceptor in R3, in contrast to the less active compounds that pres-
ent hydrophobic groups such as benzyl (2b and 2c) (Fig. 1).
The comparison among compounds 2d, 2f, and 2j as well as 2e,
2g, and 2k showed that the hydrogen bond donor groups (nitrogen
of amide or hydrazide) in the R2 position may be an important
2.2. Biological assays
2.2.1. Anti-HIV-1 effect
The oxoquinolines were also evaluated for their inhibitory ef-
fect on HIV virus replication. The EC50 values were calculated by
comparing p24 antigen concentrations in supernatants of treated
infected cells with those in supernatants of untreated infected cells
at day
7
postinfection. Significant dose-dependent antiviral
Subunit B
R2= CO2Et, COOH and CONHNH2
O
R2= CO2Et, CONHNH2 and CONHCH2Ph
O
N
HOOC
Cl
O
R2
Cl
R2
R1
N
R3
1
Change of riboside
ring by
hydrogen or
benzyl group
molecular
simplification of
riboside ring
N
HO
O
HO
Structural
modifications
O
Structural
modifications
different functional
groups at C-3
position
R3= H and Benzyl
R1 = Cl and F
different functional
groups at C-3
position
OH OH
2a-c; 2h-i
Subunit A
2d-g; 2j-k
Scheme 1. Rational drug design for compounds 2a–k.

5478
F.C. Santos et al. / Bioorg. Med. Chem. 17 (2009) 5476–5481
O
O
2h
O
Cl
NH2
N
H
IV
N
H
O
O
N
R1
CO2Et
R1
CO2Et
OH
III
O
R1
N
N
V
H
H
O
2a, R1=Cl
N
H
2a'
, R1= F
2d, R1=Cl
2e,
R1= F
2i, R1= Cl
2i',
I
R1= F
IV
O
III
Cl
R2
O
R1
CONHNH2
OH
O
O
N
R1
N
H
N
O
N
OH
2f, R1=Cl
2g,
2b, R2=CO2Et
2c,
O
II
R1= F
R2= COOH
2j, R1= Cl
2k
, R1= F
Scheme 2. General synthetic route. Reagents and conditions: Route I—benzylchloride, DMF, K2CO3, 80 °C, 24 h; Route II—(a) NaOH/EtOH 1,0 M, rt, 3 h; (b) HCl 35 %; Route
III—(a) BSTFA/TMSCl, CH3CN, reflux; (b) 1,3-dioxolane/TMSCl/KI, rt; (c) NaHCO3; Route IV—hydrazine monohydrate, 80%, EtOH, reflux, 1 h; Route V—benzylamine, phenyl
ether, 210 °C, 1 h.
Table 1
Anti-HIV-1 profile (% and EC50), cytotoxicity (CC50) and selective index (SI) of oxoquinoline derivatives 1 and 2a–k
#
R1
R2
R3
HIV-1 inhibition (%)
EC50
(l
M)
CC50
(lM)
SI
1
Cl
Cl
Cl
Cl
Cl
F
Cl
F
Cl
Cl
Cl
F
CO2H
b-
H
CH2Ph
CH2Ph
CH2OCH2CH2OH
CH2OCH2CH2OH
CH2OCH2CH2OH
CH2OCH2CH2OH
H
D
-Ribofuranosyl
99 1.3
58 1.2
24 1.1
29 2.0
45 3.3
30 1.4
99 2.1
96 1.3
29 0.5
47 0.7
98 0.9
90 1.0
94 0.2
1.5 0.2
43.0 1.2
104.0 4.5
86.0 3.3
50.0 1.1
83.3 2.3
2.5 4.4
1701 10.1
1658 9.8
1839 12.2
1890 14.6
1800 13.3
1500 12.9
2400 10.5
909 9.1
1233 12.5
1395 11.4
3200 11.9
2800 12.7
126 1.8
1134
39
18
22
36
2a
2b
2c
2d
2e
2f
2g
2h
2i
CO2CH2CH3
CO2CH2CH3
CO2H
CO2CH2CH3
CO2CH2CH3
CONHNH2
CONHNH2
CONHNH2
CONHCH2Ph
CONHCH2Ph
CONHCH2Ph
—
18
960
275
14
3.3 1.1
86.0 2.8
53.0 2.2
4.5 0.8
2.8 0.1
0.05 0.002
H
26
2j
2k
AZT
CH2OCH2CH2OH
CH2OCH2CH2OH
—
711
1000
2520
—
Table 2
Electronic properties and pharmacokinetic parameters important for good oral bioavailability of 2a–m, AZT and efavirenz: HOMO energy (EHOMO), LUMO energy (ELUMO), number
of hydrogen bond acceptor (nHba), number of hydrogen bond donor (nHbd), molecular weight (MW), calculated octanol/water partition coefficient (cLog P), number of rotatable
bonds (nrotb) and the potential to qualify for a drug (druglikeness and drug-score)
#
Electronic properties
Lipinski ‘rule of five’
MW
Drug likeness
Drug score
EHOMO
ELUMO
nHBA
nHBD
cLog P
nrotb
1
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
AZT
Efavirenz
ꢀ8.40
ꢀ8.78
ꢀ8.61
ꢀ8.68
ꢀ8.92
ꢀ8.9
1.61
1.74
1.76
1.70
1.62
1.62
1.46
1.46
1.57
1.64
1.53
1.52
—
8
4
4
4
6
6
7
7
5
4
6
6
9
3
4
1
0
1
1
1
4
4
4
2
2
2
2
1
355.73
251.67
341.79
313.74
325.75
309.29
311.72
295.27
237.65
312.76
386.83
370.38
267.25
315.68
0
7
3
4
4
7
7
5
5
1
2
7
7
3
1
0.27
ꢀ7.11
ꢀ5.15
2.90
ꢀ6.83
ꢀ8.64
ꢀ2.83
ꢀ4.64
ꢀ2.85
4.07
4.31
2.56
2.12
ꢀ19.90
0.7
1.85
3.09
2.20
1.02
0.47
ꢀ1.07
ꢀ1.62
ꢀ0.24
2.31
1.48
0.93
ꢀ0.86
4.24
0.44
0.34
0.60
0.42
0.26
0.36
0.27
0.38
0.76
0.71
0.45
0.33
0.22
ꢀ8.92
ꢀ8.98
ꢀ8.68
ꢀ8.76
ꢀ8.82
ꢀ8.83
—
—
—
feature to interact with the target, since they are present in all of
the most potent compounds, including 1. The distribution of
hydrogen bond donors and acceptors of the active derivatives is
similar to that of compound 1, which reinforced the importance
of this feature for the biological activity in contrast to the inactive
derivatives, which only present hydrogen bond acceptors (Fig. 1).

F.C. Santos et al. / Bioorg. Med. Chem. 17 (2009) 5476–5481
5479
Figure 1. Structure of oxoquinoline derivatives. (A) Electrostatic potential map and (B) the most stable conformation of 2a–k. (C) Structure of compound 1 that was used to
design the oxoquinoline derivatives. (D) Structure alignment of the active compounds (2f, 2g, 2j, and 2k) showing the pharmacophoric groups in the substituents. (E)
Structure of the less active compounds (2b). Hydrophobic groups are shown in orange and hydrogen bond donors and acceptors in green and blue, respectively.
Interestingly, both halogen (chlorine and fluorine) groups could
be used in these antiviral structures, since none of them compro-
mised the activity (i.e., 2f, 2g, 2j, and 2k).
violating more than one of these rules may have problems with
bioavailability.14 Our results pointed that all compounds fulfill this
rule and present molecular weight (237.65–386.83), cLog P (ꢀ1.62
to 3.09), nHBA (4–7), nHBD (0–4) and number of rotatable bonds
(rotb) (1–7) similar to the commercial drugs AZT and efavirenz
(Fig. 2).
2.4. In silico analysis of ADMET profile
Finally, we evaluated oxoquinoline derivatives 2a–k as poten-
tial drugs by calculating druglikeness and drug-score. Druglike-
ness, which is related to the similarity with trade drugs (ꢀ9.42
to 4.31) was better than for the commercial drugs AZT (2.12) and
efavirenz (ꢀ19.90). An analogous result was observed for the
drug-score (0.13–0.76), (the current AZT = 0.33 and efavi-
renz = 0.22). Among the active compounds, 2j showed the best val-
ues of druglikeness and drug-score (4.31 and 0.71, respectively)
with a lower toxicity effect, which points it for further in vivo
exploration (Table 2).
The analysis of theoretical toxicity risks for the oxoquinoline
derivatives using Osiris program showed compounds 2a, 2d, 2i,
and 2j with low toxicity risks in contrast to AZT, the current anti-
viral used in HIV therapy, which presents reproductive and muta-
genic effects (Fig. 2). Apparently, the mutagenic risk in 2e and 2k is
associated to the fluoro substituent, since compounds 2d and 2j,
which present the same substitutions, but chlorine instead of fluo-
rine, showed no theoretical risks. Mutagenic risk also seems to be
associated with the hydrazide group, since 2f, 2g, and 2h, which
present this group, are toxic in contrast to 2i and 2j (Fig. 2).
As these compounds are considered for oral delivery, they were
submitted to the analysis of Lipinski ‘rule of five’. This rule deter-
mines if a chemical compound presents absorption and perme-
ation across membranes properties that would make it a likely
orally active drug in humans. According to Lipinski, molecules
3. Conclusion
In summary, a series of oxoquinoline derivatives with different
substituents were synthesized and exhibited a significant anti-HIV
Figure 2. In silico toxicity risk analysis for oxoquinoline derivatives 2a–k and commercial antivirals.

5480
F.C. Santos et al. / Bioorg. Med. Chem. 17 (2009) 5476–5481
activity, suggesting their potential as antivirus agents. Molecular
modeling analysis revealed stereoelectronic properties such as
LUMO energy, dipole moment and number of rotatable bonds,
hydrogen bond donors and acceptors as correlated with the com-
pound potency. Based on this structural evaluation, a competitive
mechanism of action is apparently not feasible, since these deriva-
tives are not nucleoside analogs, but similar to compound 1, which
presents uncompetitive and noncompetitive modes of action
against HIV1-reverse transcriptase.13 Therefore, a further analysis
to investigate these two feasible mechanisms of action should be
considered.
J = 5.4 Hz, 2H, H-40); 13C NMR (75.0 MHz, DMSO-d6): d 174.9 (C-4);
163.6 (C@ONH); 148.4 (C-2); 139.2 (C-200); 137.5 (C-8a); 132.8 (C-
7); 130.2 (C-6); 128.3 (C-500); 128.0 (C-4a); 127.3 (C-300 and C-700);
126.8 (C-5); 124.7 (C-400 and C-600); 120.4 (C-8); 110.9 (C-3); 83.0
(C-10); 70.1 (C-30); 59.9 (C-40); 42.2 (C-100); HRMS (ESI) m/z calcd for
C20H20ClN2O4 [M+H]+ 387.1106, found 387.1111; 2k: 6-fluoro-1-
[(20-hydroxyethoxy)methyl]-N-[(phenyl)methyl]-1,4-dihydro-4-oxo-
quinoline-3-carboxamide (83% yield), mp 159–161 °C; IR (KBr) cmꢀ1
3498, 1602,1655;1H NMR(300.13 MHz, DMSO-d6, internalstandard:
Me4Si): d 9.06 (s, 1H, H-2); 8.16 (dd, J = 9.5; 4.4 Hz, 1H, H-5); 8.12 (dd,
J = 9.0; 2.9 Hz, 1H, H-8); 7.72 (ddd, J = 9.0; 7.8; 2.9 Hz, 1H, H-7); 7.45
(m, 5H, H-300–H-700); 5.93 (s, 2H, H-10); 4.73 (s, 2H, H-100); 3.73–3.76
(m, 4H, H-30 and H-40); 13C NMR (75.0 MHz, DMSO-d6): d 166.7 (C-
4); 160.1 (C@ONH); 150.7 (C-3); 149.4 (C-2); 138.5 (d; 1JC–F = 205.3;
C-6); 137.2 (C-8a); 139.9 (C-200); 129.7 (C-500); 128.4 (C-4a); 128.5
Our studies pointed substituents in R2 and R3 as important for
the biological activity, and compound 2j as the best candidate for
further investigation. This compound presents the best parameters
including: (i) high activity against HIV-1, (ii) low cytotoxicity risks
in PBMC, (iii) low toxic effect risks in in silico analysis, (iv) good
oral bioavailability according to the Lipinski ‘rule of five’, and (v)
better druglikeness and drug-score values than some commercial
antivirals such as AZT and efavirenz (competitive and non compet-
itive inhibitors, respectively). Overall, our results point to introduc-
ing structural changes at position 3 of the oxoquinoline ring of
compound 2j to increase its activity and verify the eventual effect
at C-3 position. Thus, exploring the introduction of different elec-
tron-donating groups (i.e., –CH3, –NH2, and –OCH3) attached to
the N-benzyl moiety of the carboxamide group may also enable
the evaluation of not only electronic, but also steric parameters
for improving the antiviral activity.
2
(C-300 and C-700); 122.8 (d; JC–F = 25.0 Hz; C-5); 128.4 (C-400 and C-
3
2
600); 122.1 (d; JC–F = 8.3 Hz; C-8); 111.8 (d; JC–F = 24.4 Hz; C-7);
85.5 (C-10); 71.6 (C-30); 61.9 (C-40); 44.1 (C-100); HRMS (ESI) m/z calcd
for C20H20FN2O4 [M+H+]: 371.1401 found 371.1407.
4.1.1.2. Synthesis of 1-[(20-hydroxyethoxy)methyl]-1,4-dihydro-
4-oxoquinoline-3-carbohydrazides. A solution of the appropri-
ate acyclonucleosides (2d or 2e—3.70 mmol) and 3.7 mL of 80%
hydrazine monohydrate in 10 mL of ethanol, was stirred at reflux
for 2 h. The reaction mixture was then concentrated under reduced
pressure giving a colored solid which was collected by filtration,
washed with cold ethyl acetate and dried under vacuum, leading
to the desired hydrazides 2f and 2g. Compound 2f: 6-chloro-1-
[(20-hydroxyethoxy)methyl]-1,4-dihydro-4-oxoquinoline-3-carbo-
hydrazide (75% yield), mp 208–210 °C; IR (KBr) cmꢀ1 3436, 3050,
1624, 1650; 1H NMR (300.13 MHz, DMSO-d6, internal standard:
Me4Si): d 10.45 (sl, 1H, C@ONHNH2); (8.97 (s, 1H, H-2); 8.25 (d,
J = 2.4 Hz, 1H, H-5); 7.95 (d, J = 9.0 Hz, 1H, H-8); 7.89 (dd, J = 9.0;
2.4 Hz, 1H, H-7); 5.86 (s, 2H, H-10); 3.50–3.55 (m, 4H, H-30 and
H-40); 3.39 (sl, 2H, C@ONHNH2); 13C NMR (75.0 MHz, DMSO-d6):
d 176.1 (C-4); 164.8 (C@ONHNH2); 139.0 (C-6); 149.5 (C-2);
129.5 (C-4a); 134.4 (C-8); 131.9 (C-8a); 126.4 (C-5); 122.0 (C-7);
112.1 (C-3); 84.5 (C-10); 71.6 (C-30); 61.4 (C-40) HRMS (ESI) m/z
calcd for C13H15ClN3O4 [M+H+]: 312.0745 found 312.0751; 2g: 6-
fluoro-1-[(20-hydroxyethoxy)methyl]-1,4-dihydro-4-oxoquino-
line-3-carbohydrazide (70% yield), mp 202–203 °C; IR (KBr) cmꢀ1
3399, 3048, 1606, 1650; 1H NMR (300.13 MHz, DMSO-d6, internal
standard: Me4Si): d 8.99 (s, 1H, H-2); 8.24 (dd, J = 7.0; 2.3 Hz, 1H,
H-5); 7.95 (dd, J = 9.1; 7.0 Hz, 1H, H-8); 7.88 (ddd, J = 9.1; 7.0;
2.3 Hz, 1H, H-7); 5.69 (s, 2H, H-10); 4.61 (t, J = 5.0 Hz, 2H,
C@ONHNH2); 3.52 (dd, J = 9.4; 5.3 Hz, 2H, H-30); 3.47 (dd, J = 9.4;
5.3 Hz; H-40); 13C NMR (75.0 MHz, DMSO-d6): d 174.8 (C-4);
4. Experimental
4.1. Synthesis
4.1.1. General procedures
Melting points were determined through a Fisher-Johns appara-
tus and are uncorrected. The IR spectra were recorded on a Perkin-
Elmer FT-IR 1600 spectrometer as potassium bromide pellets and
frequencies are expressed in cmꢀ1. NMR spectra were recorded
on
a
Varian Unity Plus 300 spectrometer operating at
300.13 MHz (1H) and 75.0 MHz (13C), in the specified solvents.
Chemical shifts are reported in ppm (d) relative to tetramethylsil-
ane. Proton and carbon NMR spectra were typically obtained at
room temperature. The two-dimensional experiments were ac-
quired using standard Varian Associates automated programs for
data acquisition and processing.
4.1.1.1. Synthesis of 1-[(20-hydroxyethoxy)methyl]-N-[(phenyl)-
methyl]-1,4-dihydro-4-oxoquinoline-3-carboxamides. oxoquin-
olinecarboxamides (2i or 2i0—3 mmol) were refluxed in bis(trimeth-
ylsilyl)trifluoroacetamide (BSTFA) (6.75 mmol) containing 1% TMSCl,
in 10 mL of acetonitrile, under nitrogen atmosphere, for 4 h, followed
by addition of potassium iodide (3 mmol), 1,3-dioxolane (3 mmol),
and TMSCl (3 mmol). The resulting mixture was stirred for 24 h at
room temperature. Afterwards, it was poured into a mixture of ethyl
alcohol/water (20 mL:10 mL), followed by addition of solid sodium
bicarbonate (3 mmol) and subsequent stirring for 10 min, leading to
the crude products, which were purified by crystallization from a
mixture of ethyl alcohol/methylene chloride (1:1), giving the pure
acyclonucleosides 2j and 2k. Compound 2j: 6-chloro-1-[(20-hydroxy-
ethoxy)methyl]-N-[(phenyl)methyl]-1,4-dihydro-4-oxoquinoline-3-
carboxamide (96% yield), mp 205–206 °C; IR (KBr) cmꢀ1 3468, 1624,
1662; 1H NMR (300.13 MHz, DMSO-d6, internal standard: Me4Si): d
10.20 (t, J = 5.6 Hz, 1H, C@ONH); 9.80 (s, 1H, H-2); 8.23 (d,
J = 2.4 Hz, 1H, H-5); 7.95 (d, J = 9.0 Hz, 1H, H-8); 7.89 (dd, J = 9.0;
2.4 Hz, 1H, H-7); 7.45 (m, 5H, H-300–H-700); 5.95 (s, 2H, H-10); 4.65
(d, J = 5.6 Hz, 2H, H-100); 3.52 (d, J = 5.4 Hz, 2H, H-30); 3.46 (d,
1
165.2 (C@ONHNH2); 159.4 (d; JC–F = 245.2 Hz; C-6); 147.6 (C-2);
3
3
128.4 (d; JC–F = 7.1 Hz; C-4a); 121.1 (d; JC–F = 8.3 Hz; C-8); 135.4
4
2
(d; JC–F = 1.2 Hz; C-8a); 121.4 (d; JC–F = 25.0 Hz; C-5); 110.2 (d;
2JC–F = 22.6 Hz; C-7); 110.5 (C-3); 83.0 (C-10); 70.0 (C-30); 59.9 (C-
40); HRMS (ESI) m/z calcd for C13H15FN3O4 [M+H+]: 296.1041 found
296.1046.
4.2. Biological evaluation
4.2.1. Cells and virus
Peripheral blood mononuclear cells (PBMCs) from healthy hu-
man donors were obtained by density gradient centrifugation
(Hystopaque, Sigma) from buffy coat preparations. PBMCs were
resuspended in RPMI 1640 (LGC Bio, São Paulo, SP, Brazil) supple-
mented with 10% heat-inactivated fetal bovine serum (FBS, Hy-
clone, Logan, UT), penicillin (100 U/mL), streptomycin (100
mL), 2 mM glutamine and 10 mM HEPES, stimulated with 5
l
l
g/
g/
mL of phytohemagglutinin (PHA, Sigma) for two to three days,
and further maintained in culture medium containing 5 U/mL of

F.C. Santos et al. / Bioorg. Med. Chem. 17 (2009) 5476–5481
5481
recombinant human interleukin-2 (Sigma). The virus isolate Ba-L
(R5-tropic, subtype B15) was used for cell infections, and virus
stocks were prepared in PHA-activated PBMCs from normal
donors.
pound’s potential to qualify for a drug and is related to topological
descriptors, fingerprints of molecular druglikeness, structural keys
and other properties such as cLog P and molecular mass.18
The pharmacokinetic profile, important for a good oral biovaila-
bility of a compound, was also evaluated according to the Lipinski’s
‘rule-of-five’, which analyses features that a drug should present to
allow the absorption and permeation across the membranes and
states molecular weight 6500 daltons (Da), calculated octanol/
water partition coefficient (cLog P) 65, number of hydrogen-bond
acceptors (nHba) 610, and number of hydrogen-bond donors
(nHbd) 65 (Lipinski, 2001), as well as a fifth rule added later, which
infers the number of rotatable bonds 610.19
4.2.2. Evaluation of cytotoxic effects
4.2.2.1. Cytotoxicity in peripheral blood mononuclear cells
(PBMCs). The cytotoxic effect was determined colorimetrically
using XTT method.16 In brief, 50
lL of XTT were added to cells
and the optical density was measured at 450 nm 2 h later. The
50% cytotoxic concentration (CC50) was defined as the concentra-
tion (lM) that reduced cell viability by 50% when compared to un-
treated controls.
Acknowledgments
4.2.2.2. Cytotoxicity in vero cells. Vero cells were cultured in
Dulbecco’s modified Eagle’s medium (DMEM) supplemented with
5% fetal bovine serum (FBS; HyClone, Logan, Utah), 100 U/mL pen-
We thank the Conselho Nacional de Desenvolvimento Científico
e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal
Docente (CAPES) and Fundação de Amparo à Pesquisa do Estado do
Rio de Janeiro (FAPERJ) for the financial support. Thanks are also
due to HU/UFRJ hemoterapy service and NIH AIDS Reagent Pro-
gram. The technical assistance of Samara and Hania is gratefully
acknowledged. The authors also thank Professor Gilberto Alves
Romeiro (UFF) and Professor Marcos Eberlin (UNICAMP) for
recording the HRMS spectra.
icillin and 100
Monolayer of 104 vero cells in 96-multiwell plates was treated
with various concentrations of the compounds for 72 h. Then, 50
lg/mL streptomycin, at 37 °C in 5% CO2.
ll
of 1 mg/mL solution of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazolium bromide (MTT; Sigma) was added to evaluate cell via-
bility according to procedures described elsewhere.17 The 50%
cytotoxic concentration (CC50) was calculated by linear regression
analysis of the dose–response curves.
References and notes
4.2.3. Anti-hiv-1 activity
1. (a) Snyder, S.; D’Argenio, D. Z.; Weislow, O.; Bilello, J. A.; Drusano, G. L.
Antimicrob. Agents Chemother. 2000, 44, 1051; (b) Stebbing, J.; Bower, M.;
Mandalia, S.; Nelson, M.; Gazzard, B. Clin. Exp. Immunol. 2006, 1452, 271.
2. Stevenson, M. Sci. Am. 2008, 299, 78.
3. (a) Zhang, Z.; Fu, J.; Zhao, Q.; He, Y.; Jin, L.; Zhang, H.; Yao, J.; Zhang, L.; Wang, F.
J. Immunol. 2006, 176, 5644; (b) Rodriguez-Arenas, M. A.; Jarrin, I.; Del Amo, J.;
Iribarren, J. A.; Moreno, S.; Viciana, P.; Pena, A.; Gomez Sirvent, J. L.; Vidal, F.;
Lacruz, J.; Gutierrez, F.; Oteo, J. A.; Asencio, R.; Castilla, J.; Perez Hoyos, S. AIDS
Res. Hum. Retroviruses 2006, 22, 715; (c) Sabin, C. A.; Smith, C. J.; Youle, M.;
Lampe, F. C.; Bell, D. R.; Puradiredja, D.; Lipman, M. C. I.; Bhagani, S.; Phillips, A.
N.; Johnson, M. A. AIDS 2006, 20, 67.
4. (a) Baba, M.; Okamoto, M.; Makino, M.; Kimura, Y.; Ikeuchi, T.; Sakaguchi, T.;
Okamoto, T. Antimicrob. Agents Chemother. 1997, 41, 1250; (b) Cecchetti, V.;
Parolin, C.; Moro, S.; Pecere, T.; Filipponi, E.; Calistri, A.; Tabarrini, O.; Gatto, B.;
Palumbo, M.; Fravolini, A.; Palu, G. J. Med. Chem. 2000, 43, 3799.
5. (a) Canuto, C. V. B. S.; Gomes, C. R. B.; Marques, I. P.; Faro, L. V.; Santos, F. C.;
Frugulhetti, I. C. P. P.; Souza, T. M. L.; Cunha, A. C.; Romeiro, G. A.; Ferreira, V. F.;
Souza, M. C. B. V. Lett. Drug Des. Discovery 2007, 4, 404; (b) Souza, T. M. L.;
Cirne-Santos, C. C.; Rodrigues, D. Q.; Abreu, C. M.; Tanuri, A.; Ferreira, V. F.;
Marques, I. P.; Souza, M. C. B. V.; Fontes, C. F. L.; Frugulhetti, I. C. P. P.; Bou-
Habib, D. C. Curr. HIV Res. 2008, 6, 209.
6. Souza, T. M.; Souza, M. C. B. V.; Ferreira, V. F.; Canuto, C. V.; Marques, I. P.;
Fontes, C. F.; Frugulhetti, I. C. P. P. Arch. Virol. 2007, 152, 1417.
7. (a) Deffin, G. F.; Kendall, J. D. J. Am. Chem. Soc. 1948, 70, 893; (b) Snyder, H. R.;
Freier, H. E.; Kovacic, P.; Heyningen, E. M. V. J. Am. Chem. Soc. 1947, 69, 371.
8. Tsuchiya, S.; Hiratsuka, K.; Yasuda, N.; Fukuzaki, A. Worldwide Patent, WO
92,21,658, 1992, 92 pp.
The effect on HIV replication was analyzed using PBMC initially
exposed to viral suspensions containing 5–10 ng/mL of HIV-1
p24Ag for 2–3 h. Cells were washed, resuspended in complete med-
ium, plated in 96-well culture plates (2 ꢁ 105 cells/well) in tripli-
cates, and treated with different concentrations of the compounds.
After 7 days at 37 °C in 5% CO2, viral replication was assessed by
measuring the HIV-1 p24Ag in culture supernatants by ELISA cap-
ture assay (ZeptoMetrix Co., Buffalo, NY). Viral replication was eval-
uated by measuring the HIV-1 p24Ag in culture supernatants.
4.3. Molecular modeling, SAR studies and in silico ADMET
screening
Initially, the structures were submitted to a conformational
analysis procedure available in the SPARTAN’06 software package
(Wavefunction Inc. Irvine, CA, 2000), using the MMFF94 force field.
The most stable conformers were submitted to a full geometry
optimization process, using the AM1 semiempirical Hamiltonian,
and, subsequently, to a single-point energy ab initio calculation,
at the 6–311G* level, both methods available in SPARTAN’06. The ste-
reoelectronic properties of all structures were evaluated in order to
perform structure–activity relationship (SAR) studies. Values of
HOMO (Highest Occupied Molecular Orbital) and LUMO (Lowest
Unoccupied Molecular Orbital) energy, molecular weight, volume,
and surface area, HOMO and LUMO orbital coefficients distribu-
9. He, J. F.; Yun, L. H.; Yang, R. F.; Xiao, Z. Y.; Cheng, J. P.; Zhou, W. X.; Zhang, Y. X.
Bioorg. Med. Chem. Lett. 2005, 15, 2980.
10. (a) Al-Soud, Y. A.; Al-Masoudi, N. A. J. Braz. Chem. Soc. 2003, 14, 790; (b) Cunha,
A. C.; Figueiredo, J. M.; Tributino, J. L. M.; Miranda, A. L. P.; Castro, H. C.; Zingali,
R. B.; Fraga, C. A. M.; Souza, M. C. B. V.; Ferreira, V. F.; Barreiro, E. J. Bioorg. Med.
Chem. 2003, 11, 2051.
11. Lucero, B. A.; Gomes, C. R. B.; Frugulhetti, I. C. P. P.; Faro, L. V.; Alvarenga, L.; Souza,
M. C. B. V.; Souza, T. M. L.; Ferreira, V. F. Bioorg. Med. Chem. Lett. 2006, 16, 1010.
12. Silva, F. C.; Souza, M. C. B. V.; Frugulhetti, I. C. P. P.; Castro, H. C.; Souza, S. L. O.;
Souza, T. M. L.; Rodrigues, D. Q.; Souza, A. M. T.; Abreu, P. A.; Passamani, F.;
Rodrigues, C. R. Eur. J. Med. Chem. 2009, 44, 373.
13. Souza, T. M. L.; Rodrigues, D. Q.; Ferreira, V. F.; Marques, I. P.; Santos, F. C.;
Cunha, A. C.; Souza, M. C. B. V.; Frugulhetti, I. C. P. P.; Bou-Habib, D. C.; Fontes,
C. F. L. Curr. HIV Res. 2009, 7, 327.
14. Lipinski, A. C.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev.
1997, 23, 3.
15. Lima, R. G.; Van, W. J.; Saraiva, E. M. B.; Barral-Netto, M.; Galvao-Castro, B.;
Bou-Habib, D. C. J. Infect. Dis. 2002, 185, 1561.
tion, molecular dipole moment (l), and molecular electrostatic po-
tential (MEP) maps were calculated. MEP isoenergy surface maps
were generated in the range of ꢀ150.0 (deepest red color) to
+150.0 (deepest blue color) kJ/mol and superimposed onto a
molecular surface of constant electron density of 0.002 e/au3.12
In addition, descriptors such as cLog P (octanol/water partition
coefficient) and Log S (water solubility) were calculated using the
Osiris Property Explorer on-line system available at http://
www.organic-chemistry.org/prog/peo/.
The oxoquinoline derivatives were submitted to in silico AD-
MET (absorption, distribution, metabolism, excretion, and toxicity)
screening, using the Osiris program. Values of druglikeness are
based on the occurrence frequency of each fragment of the mole-
cule in commercial drugs while the drug-score evaluates the com-
16. Scudiero, D. A.; Shoemaker, R. H.; Paull, K. D.; Monks, A.; Tierney, S.; Nofziger,
T. H.; Currens, M. J.; Seniff, D.; Boyd, M. R. Cancer Res. 1988, 48, 4827.
17. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
18. Tekto, I. V. Drug Discovery Today 2005, 10, 1497.
19. Wenlock, M. C.; Austin, R. P.; Barton, P.; Davis, A. M.; Leeson, P. D. J. Med. Chem.
2003, 46, 1250.