Studies of anti-HIV transcription inhibitor quinolones: Identification of potent N1-vinyl derivatives

Article abstract of DOI:10.1002/cmdc.201000267

The 6-desfluoroquinolones (6-DFQs) are anti-HIV agents that target Tat-mediated transcription. This particular mechanism of action makes this class of compounds very attractive for further structural investigations. Identification of the pharmacophore req

Full text of DOI:10.1002/cmdc.201000267

MED
DOI: 10.1002/cmdc.201000267
Studies of Anti-HIV Transcription Inhibitor Quinolones:
Identification of Potent N1-Vinyl Derivatives
Oriana Tabarrini,*[a] Serena Massari,[a] Dirk Daelemans,[b] Francesco Meschini,[a]
Giuseppe Manfroni,[a] Laura Bottega,[c] Barbara Gatto,[c] Manlio Palumbo,[c]
Christophe Pannecouque,[b] and Violetta Cecchetti[a]
The 6-desfluoroquinolones (6-DFQs) are anti-HIV agents that
target Tat-mediated transcription. This particular mechanism of
action makes this class of compounds very attractive for fur-
ther structural investigations. Identification of the pharmaco-
phore required for inhibition will ultimately result in the design
of more selective analogues for use in combination therapy for
the treatment of HIV infections. We have focused on the pyri-
done ring of the quinolone nucleus present in these com-
pounds, designing new modifications to broaden the struc-
ture–activity relationship knowledge base. Herein, we present
novel and very potent anti-HIV quinolones, most notably those
bearing an amino or vinyl group at the N1 position. Attempts
were made to determine the structural parameters necessary
to impart potent anti-HIV activity to the vinyl derivatives.
Introduction
The 6-desfluoroquinolones (6-DFQs)[1–4] are a promising class of
anti-HIV compounds which interfere with the Tat-mediated
transcription process of the viral replicative cycle.[3,5,6] Tran-
scription from the 5’ long terminal repeat (LTR) promoter is a
crucial step in the HIV replication cycle and is required to re-
initiate viral replication from post-integration latency after in-
terruption of therapy. Therefore, transcription inhibitors have
the potential to control HIV replication, not only in acutely but
also in chronically infected cells. In addition, the use of HIV-1
transcription inhibitors is expected to limit the incidence of
drug resistance, since HIV-1 transcription regulation requires a
complex interplay of both viral and cellular components.
pound 3a), acetylamino (compound 4a), or vinyl (com-
pound 5a) groups, were placed at the N1 position (Figure 1).
The promising biological properties exhibited by 1-amino and
1-vinyl derivatives 3a and 5a prompted us to expand both of
these compound series. The N1 vinyl series was further extend-
ed by preparing 6-hydrogen-1-vinylquinolones 6a–c. For the
N1 amino series, which were more synthetically accessible, we
prepared additional 1,6-diaminoquinolones (3b and 3c), 1-
amino-6-hydrogen derivatives (7a–c), and 1-amino-6-hydroxy
derivatives (8a and 8b) (Figure 1).
Chemistry
Due to their novel mechanism of action and potent anti-HIV
activity, 6-DFQs have significant potential for anti-HIV combina-
tion therapy. Further investigation of these compounds is
needed in order to identify the pharmacophore responsible for
anti-HIV properties and/or toxicity, which could ultimately
enable the design of more selective analogues. Structure–activ-
ity relationship (SAR) studies conducted to date[1–4] show that
the highest antiretroviral activity was achieved when the es-
sential 4-oxo-1,4-dihydroquinoline-3-carboxylic nucleus of
these compounds was functionalized by a suitable 4-arylpiper-
azine substituent at the C7 position and an amino group or hy-
drogen atom at the C6 position. Although a methyl group
emerged with optimal activity in the N1 position, the absence
of a substituent at N1 also yielded acceptable anti-HIV activity.
In the present study, we focused our attention on the pyri-
done ring of the quinolone nucleus, initially maintaining the
amino group and the 4-(2-pyridinyl)piperazine at the C6 and
C7 positions, respectively, which are characteristic of 6-DFQ
WM5[1] (Figure 1). Specifically, the nitrogen atom was replaced
by a sulfur (compound 1a), the C2 position was functionalized
by insertion of a methyl group (compound 2a), and small moi-
eties with various electronic properties, such as amino (com-
Thiochromene derivative 1a was prepared as shown in
Scheme 1 through reaction of acrylate 9[7] with bubbling H2S
to yield synthon 10, which was then converted to intermediate
11 a by reaction with 1-(2-pyridinyl)piperazine. Subsequent re-
duction by ammonium formate (HCO2NH4) and 10% Pd/C
gave amino derivative 12a, although only in 15% yield. Final
acid hydrolysis yielded target acid 1a.
2-Methyl quinolone 2a was prepared as outlined in
Scheme 2 through a Gould–Jacob procedure. The reaction of
3-chloro-4-nitroaniline with diethyl 2-(1-chloroethylidene)malo-
[a] Prof. O. Tabarrini, Dr. S. Massari, Dr. F. Meschini, Dr. G. Manfroni,
Prof. V. Cecchetti
Dipartimento di Chimica e Tecnologia del Farmaco, Universitꢀ di Perugia
Via del Liceo 1, 06123 Perugia (Italy)
Fax: (+39)075-585-5115
E-mail: oriana.tabarrini@unipg.it
[b] Dr. D. Daelemans, Prof. C. Pannecouque
Rega Institute for Medical Research, Katholieke Universiteit Leuven
Minderbroedersstraat 10, 3000 Leuven (Belgium)
[c] Dr. L. Bottega, Prof. B. Gatto, Prof. M. Palumbo
Dipartimento di Scienze Farmaceutiche, Universitꢀ di Padova
Via Marzolo 5, 35131, Padova (Italy)
1880
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
Figure 1. Structures of compounds synthesized in this study, along with WM5.
nate 13[8] gave intermediate 14 which, following cyclization
with refluxing diphenylether, furnished the desired ethyl 7-
chloro-2-methyl-6-nitro-4-oxo-1,4-dihydroquinoline-3-carboxyl-
ate, along with 30% of its regioisomer (ethyl 5-chloro-2-
methyl-6-nitro-4-oxo-1,4-dihydroquinoline-3-carboxylate). The
mixture was difficult to separate and was, therefore, used di-
rectly in the successive methylation step to yield desired N-
methyl derivative 15 following purification. Basic hydrolysis
gave compound 16, which was functionalized at the C7 posi-
tion to provide compound 17a. Reduction of the 6-nitro
group afforded target acid 2a.
An initial attempt (Scheme 3) to obtain 1-amino derivatives
involved introduction of an amino group at the N1 position
Scheme 1. Reagents and conditions: a) H2S, THF, ꢀ308C, 6 h; b) 1-(2-pyridi-
nyl)piperazine, CH3CN, Et3N, reflux, 7 h; c) HCO2NH4, 10% Pd/C, MeOH/DMF
(2:1), 5 h; d) 6n HCl, EtOH, 508C, 3 h.
Scheme 3. Reagents and conditions: a) 1-(2-pyridinyl)piperazine, DMF, 608C,
7 h; b) H2, Raney Ni, DMF, 15 min; c) p-toluenesulfonic acid, H2O, 2-propanol,
708C, 24 h, or Pd(OCOMe)2, SnCl2, H2O, DMF, 708C, 24 h; d) 6n HCl, EtOH,
reflux, 24 h.
through the precursor N,N-dimethylhydrazone (18), according
to a previously reported procedure.[9] Nucleophilic reaction of
18[9] with 1-(2-pyridinyl)piperazine gave intermediate 19a,
which afforded compound 20a following catalytic nitro group
reduction. Unfortunately, none of the hydrolysis (p-toluensul-
fonic acid or 6n HCl) or reduction (Pd(OCOMe)2 or SnCl2) con-
ditions permitted restoration of the N1 amino moiety. The sole
Scheme 2. Reagents and conditions: a) Et3N, toluene, 858C, 3 h; b) Ph2O,
reflux, 4 h; c) MeI, K2CO3, DMF, 808C, 2 h; d) 4% NaOH, reflux, 5 h; e) 1-(2-
pyridinyl)piperazine, DMF, 808C, 7 h; f) H2, Raney Ni, DMF, 30 min.
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1881

O. Tabarrini et al.
MED
Scheme 4. Reagents and conditions: a) NaH, DMF, 08C!RT, 21 h; b) Ar-piperazine (for Ar, see Figure 1), DMF, 708C, 3.5 h; c) HCO2NH4, 10% Pd/C, MeOH/DMF,
1 h; d) 4% NaOH, reflux, 1 h; e) 48% HBr, reflux, 1–8 h; f) Ar-piperazine (for Ar, see Figure 1), DIPEA, DMSO, 1058C, 4 h; g) Ar-piperazine (for Ar, see Figure 1),
N-methyl-2-pyrrolidone, 1058C, 5 h; h) Ac2O, 808C, 5 h; i) H2, Raney Ni, DMF, 15 min; j) LiOH, dioxane/H2O, 24 h.
acid derivative (21a) obtained from 6n HCl hydrolysis was,
however, submitted for biological evaluation, due to the novel-
ty of the N1 substituent.
shown in Scheme 5. Acrylates 9[7] and 35[3] were added to 2-
chloroethylamine in an diethyl ether/ethanol mixture at room
temperature to give intermediates 36 and 37, respectively.
Treatment with cesium carbonate (Cs2CO3) in acetonitrile at
reflux directly afforded 1-vinylderivatives 38 and 39. The 6-ni-
troderivative 38 was then subjected to nucleophilic reaction
with 1-(2-pyridinyl)piperazine to give 40a, which was convert-
ed to 6-amino derivative 41a using SnCl2 in 8n HCl. This was
the sole condition which permitted the 6-nitro group to be re-
duced selectively, albeit in very low yield (5%), due to forma-
tion of the 6-amino-5-chloro analogue as the main compound.
Basic hydrolysis afforded the target acid 5a. The synthesis of
6-hydrogen derivatives 6a–c began from compound 39, which
was hydrolyzed to the corresponding acid (42), followed by
suitable C7 functionalization.
An alternative approach, which enabled the amino group to
be introduced at the N1 position, involved amination of syn-
thon 22,[10] using freshly prepared O-p-tolylsulfonylhydroxyl-
amine (TSH) 25,[11] in the presence of sodium hydride in dime-
thylformamide (DMF) (Scheme 4). The 1-amino derivative 26,[9]
obtained by this alternative route, was combined with selected
arylpiperazines to give compounds 29a–c. These were reduced
to corresponding 1,6-diaminoderivatives 30a–c, by employing
HCO2NH4 and 10% Pd/C. Finally basic hydrolysis afforded the
target acids 3a–c. In an analogous manner, beginning from
synthon 23,[12] 6-hydrogen analogue 7a–c were obtained from
intermediates 27 and 31; while, starting from compound 24,[13]
6-hydroxy target acids 8a and
8b were obtained from inter-
mediates 28 and 32.
The 1-acetylamino derivative
4a (Scheme 4) was prepared by
adding 1-amino derivative 29a
to acetic anhydride at 808C,
yielding the diacetyl intermedi-
ate 33a from which 6-amino
monoacetyl derivative 34a was
directly obtained by catalytic hy-
drogenation. Subsequent hydrol-
ysis by lithium hydroxide in a
water/dioxane mixture at room
temperature gave target acid
4a. It was not possible to direct-
ly obtain the monoacetyl deriva-
tive under milder conditions.
The 1-vinyl derivatives 5a and
6a–c were prepared through a
Scheme 5. Reagents and conditions: a) 2-chloroethylamine hydrochloride, Et2O/EtOH, 30 min–12 h; b) Cs2CO3,
CH3CN, reflux, 3 h; c) 1-(2-pyridinyl)piperazine, DMF, 808C, 1 h; d) SnCl2·2H2O, 8N HCl, 408C, 2 h; e) 4% NaOH,
cycloaracylation procedure, as reflux, 1 h; f) Ar-piperazine, (for Ar, see Figure 1), DIPEA, DMSO, 1058C, 4 h.
1882
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
nolones, including 7a and 7c, which are devoid of any cyto-
toxicity and therefore have correspondingly high selectivity
values. The benzothiazolpiperazinyl compound 7c emerged as
the most potent derivative, with selectivity index (SI) values
greater than 129 and 104, respectively, for HIV-1 and HIV-2.
To further explore the 1-amino series, the C6 position was
substituted with a hydroxyl group; the coupled presence of
the N1 amino group with this 6-
Biological Activity
The synthesized compounds were initially evaluated for anti-
HIV-1 (IIIB) and anti-HIV-2 (ROD) activity in MT-4 cells. The cyto-
toxicity of the compounds was determined in parallel. For
comparative purposes, the 6-DFQ WM5,[1] was evaluated in the
same cell lines (Table 1).
hydroxy moiety was unproduc-
Table 1. Anti-HIV-1 and anti-HIV-2 activity and cytotoxicity of quinolones in MT-4 cells.
tive, as shown by the inactivity
of both compounds 8a and 8b.
The N1 amino group cannot be
functionalized productively, as
demonstrated by the inactivity
of acetylated derivative 4a and
by the lack of antiviral activity of
aminomethylethylidene deriva-
tive 21a (obtained as a side
product in the synthesis of 1-
aminoquinolones, Scheme 3) at
concentrations below cytotoxic
levels.
Compd
EC50[a,c] [mm]
CC50[b,c] [mm]
SI[d]
HIV-1 (IIIB)
>138.05
HIV-2 (ROD)
IIIB
ROD[d]
1a
2a
3a
3b
3c
4a
5a
6a
6b
6c
7a
7b
7c
8a
8b
21a
WM5e
>138.05
>176.72
138.05ꢁ8.59
176.72ꢁ18.32
36.46ꢁ1.54
4.49ꢁ0.26
<1
<1
16
ꢃ3
<1
<1
9
18.5
31
10
>54
<1
>129
7
<1
<1
15
<1
<1
27
<1
<1
<1
8
>176.72
2.23ꢁ0.55
1.33ꢁ0.23
ꢂ1.70
>0.17
>4.49
>0.17
0.17ꢁ0.011
ꢂ218.46
>295.9
>295.9
0.033ꢁ0.01
0.038ꢁ0.00
ꢂ0.31
0.55ꢁ0.009
>0.05
3.01ꢁ1.55
>0.69
0.30ꢁ0.07
1.48ꢁ0.19
5.97ꢁ0.11
0.05ꢁ0.003
0.08ꢁ0.029
0.19ꢁ0.04
0.005ꢁ0.002
6.32ꢁ3.01
ꢃ5
11
<1
>114
<1
>104
ꢃ4
<1
<1
9
>342
>0.69
0.69ꢁ0.16
>296
185.32ꢁ9.80
20.96ꢁ1.69
1.11ꢁ0.02
2.21ꢁ1.05
Conversely, potent anti-HIV ac-
tivity was achieved by introduc-
tion of a vinyl group at the N1
position. This vinyl derivative ex-
hibited activity both when cou-
pled to a C6 amino group (5a)
or when present in 6-hydrogen
derivatives (6a–c). The antiviral
activity of this new series was
impressive and superior to both
2.30ꢁ0.30
2.84ꢁ1.82
28.44ꢁ2.77
ꢂ49.03
>20.96
>1.11
>20.96
>1.11
0.15ꢁ0.05
0.25ꢁ0.13
[a] EC50: concentration of compound required to achieve 50% protection of MT-4 cells from HIV induced cyto-
pathogenicity, as determined by the MTT method. [b] CC50: concentration of compound required to reduce the
viability of mock-infected cells by 50%, as determined by the MTT method. [c] All data represent the mean ꢁ
the standard deviation of at least two independent experiments. [d] SI: ratio of CC50/EC50. [e] Reference [1].
The recent observation that deletion of the N1 substituent
in anti-HIV 6-DFQs did not result in loss of activity,[3] in contrast
to the well-established SAR for antibacterial quinolone
agents,[14] led us to further modify this position by insertion of
a sulfur atom. Thiochromenone derivative 1a was completely
inactive against HIV in MT-4 cells. The different electronic distri-
bution, coupled with a slight deformation of the thiopyranone
ring, clearly results in an unsuitable replacement for the pyri-
done ring. Similarly, 2-methyl derivative 2a was devoid of any
biological activity, indicating that position 2 may play an im-
portant role in target recognition, which, in turn, seems to be
related to both cytotoxicity and antiviral activity.
the methyl and amino analogues. These results indicate that
vinyl substitution is responsible for the high levels of observed
biological activity. These 1-vinyl analogues exhibited EC50
values in the low micromolar range, and, although they had
pronounced cytotoxicity, retained positive SI values. Most im-
portantly, the presence of an N1 vinyl group results in the abili-
ty of these compounds to completely protect cells from HIV-1
infection. Satisfactory activity was also maintained in acutely
infected peripheral blood mononuclear cells (PBMCs) from
healthy donors (Table 2), where 1-vinyl derivatives typically ex-
Table 2. Anti-HIV-1 activity and cytotoxicity of 1-vinylquinolones in PBMC
Having established that the nitrogen atom at position 1 of
the quinolone nucleus is absolutely necessary and that the C2
position must be unsubstitued, we next focused our attention
on the effects of different substituents at N1. The results re-
ported in Table 1 show that the N1 amino group is important
for activity when coupled with 1-(2-pyridinyl)piperazine at the
C7 position; compound 3a has lower biological activity com-
pared to its 1-methyl counterpart (WM5[1]) in MT-4 cells but,
due to the lower cytotoxicity of 3a, the selectivity index of the
two compounds is similar. While the insertion of various C7 4-
arylpiperazines (compounds 3b and 3c) did not produce any
other active 1,6-diaminoquinolones, the deletion of the 6-
amino group yielded very selective 1-amino-6-hydrogenoqui-
cells.
Compd
HIV-1 (IIIB)
SI[d]
EC50[a,c] [mm]
CC50[b,c] [mm]
5a
6a
6b
6c
0.17ꢁ0.00
0.61ꢁ0.74
0.84ꢁ0.55
0.03ꢁ0.01
1.14ꢁ0.20
9.11ꢁ7.59
5.70ꢁ0.20
0.14ꢁ0.06
7
15
7
11
[a] EC50: concentration of compound required to achieve 50% inhibition
of p24 production in PBMC. [b] CC50: concentration of compound re-
quired to reduce the viability of mock-infected cells by 50%, as deter-
mined by the trypan blue exclusion method. [c] All data represent the
mean ꢁ the standard deviation of at least three independent experi-
ments. [d] SI: ratio of CC50/EC50.
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1883

O. Tabarrini et al.
MED
hibit slightly less cytotoxicity and lower activity than in MT-4
cells.
served in both latently HIV-1 infected OM-10.1 and U1 cell
lines. This inhibition was equivalent in both TNF-a and PMA
experimental stimulation conditions and took place at com-
pound concentrations far below the cytotoxicity level. Com-
pound 5a exhibited EC50 values ranging from 0.012 mm to
0.033 mm, which are of the same order as those observed in
MT-4 cells, although the lower toxicity in OM-10.1 and U1 cells
resulted in higher selectivity values (SI=183–357).
For 1-vinyl derivative 5a, a time-of-addition (TOA) experi-
ment was performed to identify the step being inhibited
within the replicative cycle. Compared to known inhibitors
dextran sulfate, azidothymidine (AZT), and ritonavir, for which
addition can be delayed for 0, 4, and more than 16 h post-in-
fection, addition of derivative 5a can be delayed for approxi-
mately 10 h post-infection without losing anti-HIV activity
(Figure 2). This finding clearly indicates that inhibition occurs
The inhibitory effect of 5a was next evaluated in chronically
infected HuT-78(IIIB) cells, which continuously produce HIV-
1(IIIB) from integrated proviral DNA. After removing the already
produced viruses, the cells were exposed to serial dilutions of
the compound, and the quantity of newly produced virus in
the presence of the inhibitor was evaluated in comparison to a
control. As shown in Table 3, compound 5a exhibited pro-
nounced antiviral activity far below the cytotoxicity level, with
an SI value greater than 170.
In an attempt to understand the features of the vinyl moiety
that impart to the compounds such potent biological activity,
N1 vinyl (5a), N1 amino (3a), and N1 methyl (WM5) analogues
were further analyzed. Significant efforts were made to identify
the molecular target responsible for the potent Tat-mediated
transcription inhibition by the 6-DFQs, but a unique unequivo-
cal target has not yet been identified. Among the many viral
and cellular targets involved in the viral transcription process,
direct interaction with the viral protein Tat, or with cellular
components such as the positive transcription elongation fac-
tor b (PTEF-b) and the histone acetyl transferase (HAT) p300/
CBP, was eliminated as a possible mechanism of action for
these compounds.[1,15] However, the ability of WM5 to interfere
with formation of the Tat/TAR complex by selectively binding
to the bulge of TAR has been reported.[16] Indeed, a fluores-
cence quenching assay (FQA) was used to determine the abili-
ty of 1-vinylquinolone 5a and 1-amino analogue 3a to inhibit
the complex formation between TAR RNA and a truncated Tat
peptide. As expected, both compounds were able to inhibit
formation of the Tat/TAR RNA complex with a Ki value in the
same order of magnitude as the 1-methyl analogue, WM5
(Table 4). This result indicates that modification at the N1 posi-
tion of the quinolone core does not impair the ability of these
compounds to recognize the Tat/TAR complex in vitro. The in-
creased activity observed for the 1-vinyl derivative may, there-
fore, arise from an alternative or additive mechanism of action.
Next, we investigated whether the different activities and cy-
totoxicities observed for the three series were related to their
respective abilities to chelate magnesium ions, as had already
Figure 2. Time-of-addition experiment (data are taken from a single repre-
sentative experiment). MT-4 cells were infected with HIV-1 (IIIB) at MOI=0.5,
with test compounds added at different times post-infection. Viral p24 pro-
duction was determined at 31 h post-infection. Control, *; DS8000
(100 mgmLꢀ1), &; AZT (0.5 mgmLꢀ1), ~; ritonavir (0.5 mgmLꢀ1), !; 5a
(0.1 mgmLꢀ1, 0.25 mm), ^; 5a (0.15 mgmLꢀ1, 0.38 mm), *.
at a post-integrational level (likely during the transcription pro-
cess), in agreement with previously reported data for the other
6-DFQs.[1–6] Transcription inhibitors are expected to inhibit viral
production in latently and chronically infected cells, in which
HIV-1 proviral DNA has already been integrated into the cellu-
lar genome of the host. Therefore, the anti-HIV activity of com-
pound 5a was also tested on both latently and chronically in-
fected cells (Table 3). Upon exposure to tumor necrosis factor a
(TNF-a) or phorbol 12-myristate 13-acetate (PMA), there was a
drastic (~1000-fold) increase in HIV-1 expression in both cell
lines. In the presence of varying concentrations of compound
5a, dose-dependent inhibition of p24 production was ob-
Table 3. Antiviral activity and cytotoxicity of 5a in latently (U1 and
OM10.1) and chronically (HuT-78) HIV-1-infected cells.
Cell line
U1
Stimulator
EC50[a,c] [mm]
CC50[b,c] [mm]
SI[d]
213
357
183
Table 4. Inhibition of Tat/TAR complex formation and magnesium bind-
ing constants (KMg).
TNF-a
PMA
TNF-a
PMA
0.015
0.012
0.033
0.028
0.15
3.19
4.29
6.05
[b]
Compd
Tat/TAR[a] [mm]
KMg [mꢀ1
]
OM10.1
6.28
224
>170
HuT-78 (IIIB)
none
>25.5
3a
5a
WM5
0.80ꢁ0.07
1.06ꢁ0.09
2.18ꢁ0.34[c]
4329ꢁ484
5855ꢁ1894
6650ꢁ380[d]
[a] EC50: concentration of compound required to achieve 50% reduction
of p24 production in HIV-1 infected cells. [b] CC50: concentration of com-
pound required to reduce the viability of cells by 50%, as determined by
the MTT method. [c] All data represent the mean ꢁ the standard devia-
tion of at least three independent experiments. [d] SI: ratio of CC50/EC50
[a] Data from FQA experiments performed with 10 nm fluorescein–Tat
(amino acids 48–57) in 50 mm Tris–HCl (pH 7.5). [b] Binding constants
(KMg) for quinolone–Mg2+ complex. [c] Reference [20]. [d] Reference [17].
1884
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
out on Merck silica gel 60 (mesh 230–400). Melting points were de-
termined in capillary tubes (Bꢁchi Electrothermal Model 9100) and
are uncorrected. Elemental analyses were performed on a Fisons
elemental analyzer, Model EA1108CHN, and the data for C, H, and
been demonstrated for other antiviral quinolones modified at
the N1 and C7 positions.[17] The key role of the Mg2+ bridge
between the 4-keto-3-carboxyl moiety of quinolone and the
DNA phosphodiester backbone in the mechanism of action of
the antibacterial quinolones has been extensively studied.[18,19]
The binding data reported in Table 4 show that compounds
WM5, 3a, and 5a have comparable capacities to chelate mag-
nesium ions, independent of the N1 substituent, with no ap-
parent correlation between antiviral activity and magnesium
chelation. Therefore, we hypothesize that parameters other
than magnesium chelation are responsible for the mechanism
of action of these antiviral quinolone derivatives.
1
N are within ꢁ0.4% of the theoretical values. H NMR and 13C NMR
spectra were recorded at 200 MHz (Bruker Avance DPX-200) and
400 MHz (Bruker Avance DRX-400) using residual solvents such as
chloroform (d=7.26) or dimethylsulfoxide (d=2.48) as internal
standards. Chemical shifts are given in ppm, and spectral data are
consistent with the assigned structures. Reagents and solvents
were purchased from common commercial suppliers and were
used directly. After extraction, organic solutions were dried over
anhydrous Na2SO4, filtered, and concentrated with a Bꢁchi rotary
evaporator at reduced pressure. Yields are of purified products and
are not optimized. All starting materials were commercially avail-
able unless otherwise indicated.
Uptake experiments were also performed, in order to deter-
mine if the improved activity displayed by 1-vinyl derivatives
could be attributed to better cellular penetration. This study
showed that both the 1-vinyl (5a, %uptake=45ꢁ20) and 1-
amino (3a, %uptake=94ꢁ15) derivatives exhibit higher cellu-
lar penetration in Jurkat cells in comparison to their 1-methyl
analogue (WM5, %uptake=23[20]). These results increase the
potential of these compounds for clinical application; however,
there was no correlation between drug uptake and anti-HIV
activity.
Ethyl
7-chloro-6-nitro-4-oxo-4H-thiochromene-3-carboxylate
solution of 9[7] (1.50 g,
(10): H2S was bubbled through
a
4.15 mmol) in THF (8 mL), maintained at ꢀ308C, for 6 h. The reac-
tion mixture was then evaporated to dryness, water was added,
and the solution was basified by the addition of 2n aq NaOH. The
resulting yellow solid was filtered, washed with EtOH, dried, and
crystallized from EtOH to give 10 (0.90 g, 70%); mp: 180–1818C;
1H NMR (CDCl3): d=1.45 (t, J=7.0 Hz, 3H, CH2CH3), 4.45 (q, J=
7.0 Hz, 2H, CH2CH3), 7.80 (s, 1H, H-8), 8.75 (s, 1H, H-5), 9.00 ppm (s,
1H, H-2); GC-MS: m/z (%): 313 (13) [M+], 283 (13), 270 (13), 268
(33), 243 (38) 241 (100), 225 (23) 213 (16), 211 (40).
Conclusions
Ethyl 6-nitro-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-4H-thiochro-
mene-3-carboxylate (11a): A mixture of 10 (0.50 g, 1.59 mmol),
Et3N (0.33 mL, 2.4 mmol), and 1-(2-pyridinyl)piperazine (0.36 mL,
2.4 mmol) in CH3CN (30 mL) was heated at reflux for 7 h. After
cooling, the reaction mixture was evaporated to dryness, and the
residue was treated with water and extracted with CHCl3. The or-
ganic layers were evaporated to dryness and the residue crystal-
lized from EtOAc/cyclohexane to give 11 a (0.630 g, 89%); mp:
187–1908C; 1H NMR (CDCl3): d=1.45 (t, J=7.0 Hz, 3H, CH2CH3),
3.30–3.45 and 3.70–3.90 (m, each 4H, piperazine CH2), 4.45 (q, J=
7.0 Hz, 2H, CH2CH3), 6.60–6.75 (m, 2H, pyridine CH), 7.25 (s, 1H,
H-8), 7.50–7.65 (m, 1H, pyridine CH), 8.20–8.30 (m, 1H, pyridine
CH), 8.75 (s, 1H, H-5), 9.00 ppm (s, 1H, H-2).
In summary, antiviral evaluation of the compounds synthesized
in this study provided new insights into the SAR of quinolones
with anti-HIV activity. No substituents appear to be tolerated at
the C2 position, and a sulfur atom cannot be used in place of
the quinolone nitrogen atom; however, in addition to the
methyl group, either an amino or a vinyl moiety can be added
at the N1 position. The identification of 1-amino derivative 7c
and the entire series of 1-vinylquinolones 5a and 6a–c, charac-
terized by potent antiviral activity on acutely, chronically, and
latently infected cells, provides the true novelty of this study.
These compounds act at a post-integrational level, as deter-
mined by TOA experiments and confirmed by their high selec-
tivity on latently and chronically infected cells.
General procedure for reduction of the 6-nitro group (Meth-
od A): A mixture of the selected 6-nitro derivative (1 equiv),
HCO2NH4 (1 equiv) and a catalytic amount of 10% Pd/C in MeOH/
DMF (2:1), was maintained at RT for 1–5 h. The mixture was then
filtered over Celite, and the filtrate was evaporated to dryness to
afford a residue that, after treatment with EtOH/Et2O, afforded a
solid, which was filtered and dried to give the corresponding
6-amino derivative.
The higher potency of 1-vinyl derivatives, with respect to
the 1-amino or 1-methyl analogues, cannot be ascribed to
better cellular penetration or more efficient interaction with
the Tat/TAR complex, or to a differential ability to bind magne-
sium ions. Nevertheless, this study highlights the critical role of
the N1 substituent in modulating anti-HIV activity. Indeed, the
vinyl moiety was the most suitable substituent with respect to
potency, independent of the arrangement of groups around
the quinolone ring. Further optimization of the vinyl series is
required to decrease cytotoxicity, in order to create new quino-
lones for effective use in combination therapy.
Ethyl
6-amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-4H-thio-
chromene-3-carboxylate (12a): Compound 12a was prepared
from 11 a, using Method A (5 h), in 15% yield following crystalliza-
tion from DMF; mp: 213–2148C; 1H NMR (CDCl3): d=1.45 (t, J=
7.0 Hz, 3H, CH2CH3), 3.30–3.45 and 3.70–3.90 (m, each 4H, pipera-
zine CH2), 4.40 (br s, 2H, NH2), 4.45 (q, J=7.0 Hz, 2H, CH2CH3),
6.60–6.75 (m, 2H, pyridine CH) 7.25 (s, 1H, H-8), 7.50–7.65 (m, 1H,
pyridine CH), 8.00 (s, 1H, H-5), 8.20–8.30 (m, 1H, pyridine CH),
9.00 ppm (s, 1H, H-2).
Experimental Section
Chemistry
6-Amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-4H-thiochro-
mene-3-carboxylic acid hydrochloride (1a): A solution of 12a
(0.07 g, 1.46 mmol) in EtOH (1 mL) and 10% HCl (2 mL) was heated
at 508C for 3 h. After cooling, the precipitate was filtered, washed
All reactions were routinely evaluated by thin layer chromatogra-
phy (TLC) on silica gel 60F254 (Merck) and visualized using UV light
or iodine. Flash column chromatography separations were carried
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1885

O. Tabarrini et al.
MED
with EtOH, dried, and crystallized from EtOH to yield acid 1a
(0.03 g, 45%); mp: 188–1898C; 1H NMR ([D6]DMSO): d=3.20–3.25
(m, 4H, piperazine CH2), 3.75–4.00 (m, 4H, piperazine CH2), 6.75–
6.80 (m, 1H, pyridine CH), 7.25–7.30 (m, 1H, pyridine CH), 7.55 (s,
1H, H-8), 7.75–7.90 (m, 2H, pyridine CH and H-5), 8.20–8.25 (m, 1H,
pyridine CH), 9.50 ppm (s, 1H, H-2); 13C NMR ([D6]DMSO): d=46.3,
49.1, 110.4, 113.4, 117.6, 119.5, 127.0, 127.4, 138.0, 144.3, 145.1,
147.1, 151.8, 159.2, 165.0, 180.1, 108.0 ppm; Anal. calcd for
C19H19ClN4O3S: C 54.48, H 4.57, N 13.37, found: C 54.25, H 4.60, N
13.20.
prepared from 16 by Method C (808C, 7 h) using 1-(2-pyridinyl)pi-
perazine to give intermediate 17a in 70% yield; mp: 249–2508C;
1H NMR ([D6]DMSO): d=2.90 (s, 3H, CH3), 3.20–3.40 (m, 4H, CH2 pi-
perazine), 3.50–3.70 (m, 4H, CH2 piperazine), 3.80 (s, 3H, NCH3),
6.50–6.60 (m, 1H, pyridine CH), 6.70–6.80 (m, 1H, pyridine CH),
7.20 (s, 1H, H-8), 7.35–7.50 (s, 1H, pyridine CH), 7.95–8.05 (s, 1H,
pyridine CH), 8.50 (s, 1H, H-5), 15.60 ppm (br s, 1H, COOH).
General procedure for reduction of 6-nitro group (Method D): A
stirred solution of the selected 6-nitro derivative in DMF was hy-
drogenated over a catalytic amount of Raney nickel at RT and at-
mospheric pressure until no starting material was detected by TLC
(15–30 min). The mixture was then filtered over Celite, and the fil-
trate was evaporated to dryness to afford a residue which, follow-
ing treatment with EtOH/Et2O, yielded a solid that was filtered and
dried to give the corresponding 6-amino derivative.
Diethyl 2-{1-[(3-chloro-4-nitrophenyl)amino]ethylidene}malonate
(14): A mixture of 3-chloro-4-nitroaniline (0.09 g, 0.54 mmol), dieth-
yl 2-(1-chloroethylidene)malonate 13[8] (0.12 g, 0.54 mmol) and
Et3N (0.06 g, 0.65 mmol) in toluene (2 mL), was heated at 858C for
3 h. After cooling, the reaction mixture was poured into ice/water
and extracted with CH2Cl2. The organic layers were evaporated to
dryness, yielding a solid which was triturated with cyclohexane to
6-Amino-1,2-dimethyl-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-
1,4-dihydroquinoline-3-carboxylic acid (2a): Compound 2a was
prepared from 17a, using Method D (30 min), in 18% yield; mp:
315–3168C; 1H NMR ([D6]DMSO): d=3.05–3.20 (m, 7H, piperazine
CH2 and CH3), 3.65–3.85 (m, 4H, piperazine CH2), 3.95 (s, 3H, NCH3),
5.45 (s, 2H, NH2), 6.60–6.70 (m, 1H, pyridine CH), 6.85–6.95 (m, 1H,
pyridine CH), 7.30 (s, 1H, H-8), 7.50–7.60 (m, 2H, pyridine CH and
H-5), 8.10–8.20 (m, 1H, pyridine CH), 17.60 ppm (br s, 1H, COOH);
13C NMR ([D6]DMSO): d=19.0, 36.9, 45.2, 49.9, 106.8, 107.0, 107.5,
107.6, 113.5, 120.9, 133.7, 138.0, 141.9, 146.2, 148.0, 157.8, 159.4,
168.3, 176.4 ppm; Anal. calcd for C21H23N5O3: C 64.11, H 5.89, N
17.80, found: C 64.35, H 6.10, N 17.83.
1
give 14 (0.15 g, 77%); mp: 69–708C; H NMR (CDCl3): d=1.30–1.40
(m, 6H, CH2CH3), 2.20 (s, 3H, CH3), 4.15–4.35 (m, 4H, CH2CH3), 7.00–
7.10 (m, 1H, aromatic CH), 7.20–7.30 (m, 1H, aromatic CH), 7.90–
8.00 (m, 1H, aromatic CH), 11.40 ppm (br s, 1H, NH).
Ethyl 7-chloro-1,2-dimethyl-6-nitro-4-oxo-1,4-dihydroquinoline-
3-carboxylate (15): Compound 14 (2.0 g, 5.60 mmol) was added
portionwise to diphenylether (20 mL) and heated at reflux. After
4 h, the mixture reaction was cooled and poured into cyclohexane
to yield a precipitate, which was filtered and washed several times
with cyclohexane, to give the desired ethyl 7-chloro-2-methyl-6-
nitro-4-oxo-1,4-dihydroquinoline-3-carboxylate (0.50 g), with the
General procedure for coupling reaction (Method E): A mixture
of the appropriate synthon (1 equiv) and selected arylpiperazine
(3 equiv) in DMF was heated at 60–808C until no starting material
was detected by TLC (1–12 h). The reaction mixture was then
poured into ice/water, and the precipitate was filtered, washed
with water and EtOH, and crystallized from EtOH/DMF.
ethyl
5-chloro-2-methyl-6-nitro-4-oxo-1,4-dihydroquinoline-3-car-
boxylate as an isomeric impurity. The mixture was used without
further purification in the subsequent step.
A solution of the isomeric mixture (0.50 g, 1.60 mmol) in DMF
(5 mL) was treated with K2CO3 (0.66 g, 4.82 mmol) and MeI (0.68 g,
4.82 mmol). The mixture was heated to 808C for 2 h, then poured
into ice/water. The resulting precipitate was filtered, dried and pu-
rified by flash chromatography (EtOAc, 100%) to yield 15 (0.45 g,
25%); mp: 279–2808C; H NMR (CDCl3): d=1.40 (t, J=7.1 Hz, 3H,
CH2CH3), 2.50 (s, 3H, CH3), 3.80 (s, 3H, NCH3), 4.40 (q, J=7.1 Hz,
2H, CH2CH3), 7.65 (s, 1H, H-8), 8.95 ppm (s, 1H, H-5).
Ethyl 1-[(1-methylethylidene)amino]-6-nitro-4-oxo-7-(4-(pyridin-
2-yl)piperazin-1-yl)-1,4-dihydroquinoline-3-carboxylate
(19a):
Compound 19a was prepared from synthon 18[9] by Method E
(608C, 7 h) using 1-(2-pyridinyl)piperazine, in 100% yield; mp: 176–
1
1
1778C; H NMR (CDCl3): d=1.45 (t, J=7.0 Hz, 3H, CH2CH3), 2.95 (s,
6H, CH3), 3.30–3.40 (m, 4H, piperazine CH2), 3.70–3.80 (m, 4H, pi-
perazine CH2), 4.45 (q, J=7.0 Hz, 2H, CH2CH3), 6.70–6.80 (m, 2H,
pyridine CH), 7.50–7.60 (m, 2H, H-8 and pyridine CH), 8.20–8.30 (m,
1H, pyridine CH), 8.70 (s, 1H, H-2), 8.85 ppm (s, 1H, H-5).
General procedure for basic hydrolysis (Method B): A suspension
of the selected ester (0.3 equiv) in 4% aq NaOH (5 mL) was re-
fluxed until no starting material was detected by TLC (15 min to
8 h) and was worked up and purified as described for each com-
pound.
Ethyl
6-amino-1-[(1-methylethylidene)amino]-4-oxo-7-(4-(pyri-
din-2-yl)piperazin-1-yl)-1,4-dihydroquinoline-3-carboxylate
7-Chloro-1,2-dimethyl-6-nitro-4-oxo-1,4-dihydroquinoline-3-car-
boxylic acid (16): Compound 16 was prepared from 15 using
Method B (5 h). After cooling, the solution was acidified (pH 4)
with 2n HCl to afford a solid, which was filtered and washed with
water and Et2O to give 16 in 97% yield; mp: 266–2678C; H NMR
([D6]DMSO): d=2.80 (s, 3H, CH3), 3.90 (s, 3H, NCH3), 8.40 (s, 1H,
(20a): Compound 20a was prepared from 19a, using Method D
1
(15 min), in 68% yield; mp: 282–2838C; H NMR (CDCl3): d=1.55 (t,
J=7.1 Hz, 3H, CH2CH3), 3.05 (s, 6H, CH3), 3.25–3.35 (m, 4H, piper-
azine CH2), 3.80–3.90 (m, 4H, piperazine CH2), 4.15 (br s, 2H, NH2),
4.50 (q, J=7.1 Hz, 2H, CH2CH3), 6.60–6.65 (m, 1H, pyridine CH),
6.70–6.75 (m, 1H, pyridine CH), 7.50–7.55 (m, 1H, pyridine CH),
7.65 (s, 1H, H-8), 7.75 (s, 1H, H-5), 8.20–8.25 (m, 1H, pyridine CH),
8.65 ppm (s, 1H, H-2).
1
H-8), 8.80 (s, 1H, H-5), 14.50 ppm (br s, 1H, COOH).
General procedure for coupling reaction (Method C): A mixture
of the appropriate synthon (1 equiv), selected arylpiperazine
(3 equiv), and DIPEA (4 equiv) in DMSO was heated at 80–1058C
until no starting material was detected by TLC (4–7 h). The reaction
mixture was then poured into ice/water, and the precipitate was
filtered, washed with water and EtOH, and crystallized from EtOH/
DMF.
6-Amino-1-[(1-methylethylidene)amino]-4-oxo-7-(4-(pyridin-2-yl)-
piperazin-1-yl)-1,4-dihydroquinoline-3-carboxylic acid (21a): A
mixture of ester 20a (0.20 g, 0.44 mmol) in EtOH (1 mL) and 6n
HCl (1 mL) was refluxed for 24 h. After cooling, the reaction mix-
ture was neutralized with 2n aq NaOH to obtain a solid, which
was filtered and crystallized from EtOH to give 21a (0.02 g, 10%);
mp: 226–2278C; 1H NMR (CDCl3): d=3.00 (s, 6H, CH3), 3.20–3.30
(m, 4H, piperazine CH2), 3.80–3.90 (m, 4H, piperazine CH2), 4.35 (br
1,2-Dimethyl-6-nitro-4-oxo-7-[4-(2-pyridinyl)-1-piperazinyl]-1,4-
dihydro-3-quinolinecarboxylic acid (17a): Compound 17a was
1886
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
s, 2H, NH2), 6.70–6.80 (m, 2H, pyridine CH), 7.55–7.65 (m, 1H, pyri-
dine CH), 7.70 (s, 1H, H-8), 7.80 (s, 1H, H-5), 8.20–8.30 (m, 1H, pyri-
dine CH), 8.90 (s, 1H, H-2), 15.60 ppm (s, 1H, COOH); 13C NMR
([D6]DMSO): d=45.2, 46.0, 49.9, 106.0, 106.6, 107.1, 107.7, 113.6,
122.3, 133.5, 138.0, 140.0, 142.8, 146.6, 148.0, 158.2, 159.4, 166.8,
175.9 ppm; Anal. calcd for C22H24N6O3: C 62.84, H 5.75, N 19.99,
found: C 62.85, H 5.75, N 19.80.
for 1 h. After cooling, the solid was filtered, washed with water and
EtOH, and recrystallized from EtOH/DMF to give 3a in 73% yield;
1
mp: 228–2308C; H NMR ([D6]DMSO): d=3.00–3.10 (m, 4H, pipera-
zine CH2), 3.80–3.90 (m, 4H, piperazine CH2), 6.95 (t, J=6.3 Hz, 1H,
pyridine CH), 7.40 (d, J=9.1 Hz, 1H, pyridine CH), 7.55 (s, 1H, H-8),
7.65 (s, 1H, H-5), 7.90–8.00 (m, 1H, pyridine CH), 8.05–8.15 (m, 1H,
pyridine CH), 8.60 ppm (s, 1H, H-2); 13C NMR ([D6]DMSO): d=46.2,
49.5, 105.3, 106.8, 107.8, 113.4, 122.6, 134.5, 138.1, 142.4, 145.0,
145.2, 147.9, 159.4, 167.2, 175.7 ppm; Anal. calcd for C19H22Cl2N6O3:
C 50.34, H 4.89, N 18.54, found: C 50.52, H 4.95, N 18.41.
Ethyl 1-amino-7-chloro-6-nitro-4-oxo-1,4-dihydroquinoline-3-car-
[9]
boxylate (26): NaH (0.04 g, 1.68 mmol) was added to a suspen-
sion of 22[10] (0.50 g, 1.68 mmol) in DMF (3 mL), and the mixture
was maintained at RT for 3 h. The reaction mixture was then
cooled to 08C and a solution of freshly prepared TSH (25)[11]
(0.31 g, 1.68 mmol) in CH2Cl2 (3 mL) was added. The mixture was
maintained at this temperature for 10 min, then at RT for 21 h. The
resulting precipitate was removed and the filtrate evaporated to
dryness to afford a residue that was dissolved in CH2Cl2 and
washed with saturated aq NaHCO3 solution. The organic layers
were evaporated to dryness, yielding a solid which was washed
with Et2O/EtOH and dried to give 26 (0.22 g, 70%); mp: 240–
2428C (247–2488C);[9] 1H NMR ([D6]DMSO): d=1.25 (t, J=7.0 Hz,
3H, CH2CH3), 4.25 (q, J=7.0 Hz, 2H, CH2CH3), 6.60 (br s, 2H, NH2),
8.25 (s, 1H, H-8), 8.60 (s, 1H, H-5), 8.70 ppm (s, 1H, H-2).
1,6-Diamino-4-oxo-7-{4-[3-(trifluoromethyl)phenyl]piperazin-1-
yl}-1,4-dihydroquinoline-3-carboxylic acid (3b): Compound 3b
was prepared from synthon 26[9] by Method E (708C, 3.5 h), using
1-[3-(trifluoromethyl)phenyl]piperazine, to afford intermediate 29b
in 97% yield. This intermediate was used directly in Method A (1 h)
to give intermediate 30b in 81% yield. Compound 3b was finally
obtained by Method B (8 h). After cooling, the precipitate was sus-
pended in water and neutralized with 2n HCl to give a solid which
was filtered, washed with water and EtOH, and recrystallized from
EtOH/DMF to give 3b in 11% yield; mp: 271–2728C; 1H NMR
([D6]DMSO): d=3.10–3.20 (m, 4H, piperazine CH2), 3.40–3.50 (m,
4H, piperazine CH2), 5.50 (br s, 2H, NH2), 6.80 (br s, 2H, NNH2), 7.05
(d, J=7.5 Hz, 1H, aromatic CH), 7.25 (s, 1H, aromatic CH), 7.30–
7.35 (m, 1H, aromatic CH), 7.40–7.45 (m, 1H, aromatic CH), 7.55 (s,
1H, H-8), 7.60 (s, 1H, H-5), 8.60 (s, 1H, H-2), 16.00 ppm (s, 1H,
COOH); 13C NMR ([D6]DMSO): d=48.3, 49.4, 105.3, 106.9, 108.0,
111.4, 115.3, 119.2, 122.6, 126.6, 130.3, 130.5, 133.8, 141.8, 145.2,
147.5, 151.2, 167.3, 175.7 ppm; Anal. calcd for C21H20F3N5O3: C
56.37, H 4.51, N 15.65, found C 56.48, H 4.60, N 15.60.
Ethyl 1-amino-7-fluoro-4-oxo-1,4-dihydroquinoline-3-carboxylate
(27): Starting from intermediate 23,[12] and following the procedure
used for the synthesis of compound 26, compound 27 was ob-
1
tained in 24% yield; mp: 228–2308C; H NMR ([D6]DMSO): d=1.30
(t, J=7.0 Hz, 3H, CH2CH3), 4.25 (q, J=7.0 Hz, 2H, CH2CH3), 6.55 (br
s, 2H, NH2), 7.35 (dt, J=8.9 and 2.3 Hz, 1H, H-6), 7.75 (dd, J=10.9
and 2.3 Hz, 1H, H-8), 8.25 (dd, J=8.9 and 6.2 Hz, 1H, H-5),
8.60 ppm (s, 1H, H-2).
1,6-Diamino-7-[4-(1,3-benzothiazol-2-yl)piperazine-1-yl]-4-oxo-
1,4-dihydroquinoline-3-carboxylic acid dihydrochloride (3c):
Compound 3c was prepared from synthon 26[9] by Method E
(708C, 3.5 h), using 1-(1,3-benzothiazol-2-yl)piperazine,[21] to give in-
termediate 29c in 99% yield, followed by Method A (1 h) to yield
intermediate 30c in 66% yield. Compound 3c was finally obtained
by Method B (8 h). After cooling, the precipitate was filtered, sus-
pended in 2n HCl (2 mL), and refluxed for 1 h. After cooling, the
hydrochloride salt was filtered, washed with water and EtOH, and
recrystallized from EtOH/DMF to give 1c in 11% yield; mp: 299–
Ethyl 1-amino-7-fluoro-6-methoxy-4-oxo-1,4-dihydroquinoline-3-
carboxylate (28): Starting from intermediate 24,[13] and following
the procedure used for the synthesis of compound 26, compound
28 was obtained in 57% yield; mp: 265–2668C; 1H NMR
([D6]DMSO): d=1.25 (t, J=6.8 Hz, 3H, CH2CH3), 3.95 (s, 3H, OCH3),
4.20 (q, J=6.8 Hz, 2H, CH2CH3), 6.50 (br s, 2H, NH2) 7.70–7.85 (m,
2H, H-5 and H-8), 8.60 ppm (s, 1H, H-2).
Ethyl 1-amino-6-nitro-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1,4-
dihydroquinoline-3-carboxylate (29a): Compound 29a was pre-
pared from synthon 26[9] by Method E (708C, 3.5 h), using 1-(2-pyri-
dinyl)piperazine, in 87% yield; mp: 267–2688C; 1H NMR
([D6]DMSO): d=1.25 (t, J=7.0 Hz, 3H, CH2CH3), 3.20–3.30 (m, 4H,
piperazine CH2), 3.65–3.75 (m, 4H, piperazine CH2), 4.25 (q, J=
7.0 Hz, 2H, CH2CH3), 6.60 (br s, 2H, NH2), 6.65–6.75 (m, 1H, pyridine
CH), 6.90 (d, J=8.5 Hz, pyridine CH), 7.50–7.60 (m, 2H, H-8 and pyr-
idine CH), 8.10–8.20 (m, 1H, pyridine CH), 8.55 ppm (s, 1H, H-5),
8.60 ppm (s, 1H, H-2).
1
3008C (dec); H NMR ([D6]DMSO): d=3.10–3.20 (m, 4H, piperazine
CH2), 3.80–3.90 (m, 4H, piperazine CH2), 7.10 (t, J=8.0 Hz, 1H, ben-
zothiazole CH), 7.30 (t, J=8.0 Hz, 1H, benzothiazole CH), 7.50 (d,
J=8.0 Hz, 1H, benzothiazole CH), 7.60 (s, 1H, H-8), 7.65 (s, 1H, H-
5), 7.80 (d, J=8.0 Hz, 1H, benzothiazole CH), 8.60 (s, 1H, H-2),
16.00 ppm (s, 1H, COOH); 13C NMR (D6]DMSO): d=49.0, 49.6, 105.4,
107.6, 108.1, 118.2, 122.1, 122.5, 122.6, 126.8, 129.4, 134.9, 141.5,
145.3, 145.7, 153.3, 167.1, 168.5, 175.7 ppm; Anal. cacld for
C21H22Cl2N6O3S: C 49.51, H 4.35, N 16.50, found: C 49.53, H 4.47, N
16.38.
Ethyl 1,6-diamino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1,4-di-
hydroquinoline-3-carboxylate (30a): Compound 30a was pre-
pared from 29a, using Method A (1 h), in 64% yield; mp: 280–
2818C; 1H NMR ([D6]DMSO): d=1.25 (t, J=7.0 Hz, 3H, CH2CH3),
3.00–3.10 and 3.45–3.55 (m, each 4H, piperazine CH2), 4.20 (q, J=
7.0 Hz, 2H, CH2CH3), 5.25 (br s, 2H, NH2), 6.40 (br s, 2H, NNH2),
6.60–6.70 (m, 1H, pyridine CH), 6.90 (d, J=8.5 Hz, pyridine CH),
7.35–7.45 (m, 2H, H-5 and H-8) 7.50–7.60 and 8.10–8.20 (m, each
1H, pyridine CH), 8.50 ppm (s, 1H, H-2).
1-Amino-7-fluoro-4-oxo-1,4-dihydroquinoline-3-carboxylic acid
(31): Compound 31 was prepared from synthon 27 using Meth-
od B (1 h). After cooling, the solution was neutralized with 2n HCl
to give a solid, which was filtered and washed with water and
EtOH to afford 31 in 56% yield; mp: 215–2168C; 1H NMR
([D6]acetone): d=6.50 (br s, 2H, NH2), 7.45 (dt, J=2.4 and 8.9 Hz
1H, H-6), 8.05 (dd, J=2.4 an 10.6 1H, H-8), 8.50 (dd, J=5.8 and
8.9 Hz 1H, H-5), 9.00 (s, 1H, H-2), 14.80 ppm (s, 1H, COOH).
1,6-Diamino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1,4-dihydro-
quinoline-3-carboxylic acid dihydrochloride (3a): Compound 3a
was obtained from 30a using Method B (1 h). After cooling, the
precipitate was filtered, suspended in 2n HCl (2 mL), and refluxed
1-Amino-7-fluoro-6-hydroxy-4-oxo-1,4-dihydroquinoline-3-car-
boxylic acid (32): A mixture of compound 28 (0.60 g, 2.14 mmol)
in 48% HBr (10 mL) was refluxed for 36 h. After cooling, the precip-
itate was filtered, suspended in water, and neutralized by the addi-
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1887

O. Tabarrini et al.
MED
tion of 10% aq NaOH. The solid was filtered, washed with water
and EtOH, and dried to give acid 32 (0.12 g, 24%): mp: >3008C;
1H NMR (D6]DMSO): d=6.85 (br s, 2H, NH2), 7.85 (d, J=9.1 Hz, 1H,
H-5), 8.00 (d, J=12.3 Hz, 1H, H-8), 8.80 (s, 1H, H-2), 11.20 (br s, 1H,
OH), 13.50 ppm (s, 1H, COOH).
compound 8a (5 h), using 1-(2-pyridinyl)piperazine in place of 1-[3-
(trifluoromethyl)phenyl]piperazine, in 55% yield; mp: 270–2718C;
1H NMR ([D6]DMSO): d=3.40–3.50 (m, 8H, piperazine CH2), 6.85 (br
s, 2H, NH2), 7.10 (d, J=7.3 Hz, 1H, aromatic CH), 7.30 (s, 1H, aro-
matic CH), 7.35 (d, J=7.4 Hz, 1H, aromatic CH), 7.45–7.55 (m, 2H,
H-8 and aromatic CH), 7.65 (s, 1H, H-5), 8.75 (s, 1H, H-2), 10.60 (s,
1H, OH), 16.00 ppm (s, 1H, COOH); 13C NMR ([D6]DMSO): d=48.0,
49.3, 105.6, 105.7, 108.6, 111.5, 115.3, 119.4, 120.2, 124.8, 130.3,
130.4, 136.8, 146.3, 146.8, 149.6, 151.6, 167.0, 175.5 ppm; Anal.
cacd for C21H19F3N4O4: C 56.25, H 4.27, N 12.49, found: C 56.28, H
4.40, N 12.40.
1-Amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1,4-dihydroqui-
noline-3-carboxylic acid (7a): Compound 7a was prepared from
31 by Method C (1058C, 4 h), using 1-(2-pyridinyl)piperazine, in
1
30% yield; mp: 299–3008C; H NMR (D6]DMSO): d=3.70–3.80 (m,
8H, piperazine CH2), 6.60–6.70 (m, 3H, H-6 and NH2), 6.85 (d, J=
8.4 Hz, 1H, pyridine CH), 7.30–7.40 (m, 2H, H-8 and pyridine CH),
7.50–7.60 (m, 1H, pyridine CH), 8.10–8.20 (m, 2H, H-5 and pyridine
CH), 8.75 (s, 1H, H-2), 15.75 ppm (br s, 1H, COOH); 13C NMR
(D6]DMSO): d=44.4, 46.6, 98.6, 105.9, 107.7, 113.7, 115.1, 116.4,
127.1, 138.1, 144.0, 147.9, 148.9, 154.4, 159.0, 166.8, 176.2 ppm;
Anal. calcd for C19H19N5O3: C 62.46, H 5.24, N 19.17, found: C 62.70,
H 5.32, N.19.38.
Ethyl
1-(diacetylamino)-6-nitro-4-oxo-7-(4-(pyridin-2-yl)pipera-
zin-1-yl)-1,4-dihydroquinoline-3-carboxylate (33a): A mixture of
29a (0.40 g, 0.91 mmol) and acetic anhydride (3 mL) was heated at
808C for 5 h. After cooling, the reaction mixture was poured into
ice/water to yield a precipitate, which was filtered, washed with
water and EtOH, and dried to give 33a (0.41 g, 86%); mp: 177–
1788C; 1H NMR ([D6]DMSO): d=1.25 (t, J=7.0 Hz, 3H, CH2CH3),
2.40 (s, 6H, CH3), 3.20–3.30 (m, 4H, piperazine CH2), 3.50–3.60 (m,
4H, piperazine CH2), 4.20 (q, J=7.0 Hz, 2H, CH2CH3), 6.65 (t, J=
5.8 Hz, 1H, pyridine CH), 6.75–6.90 (m, 2H, H-8 and pyridine CH),
7.55 (t, J=7.5 Hz, 1H, pyridine CH), 8.10 (d, J=3.3 Hz, 1H, pyridine
CH), 8.55 (s, 1H, H-2), 8.90 ppm (s, 1H, H-5).
1-Amino-4-oxo-7-{4-[3-(trifluoromethyl)phenyl]piperazine-1-yl}-
1,4-dihydroquinoline-3-carboxylic acid (7b): Compound 7b was
prepared from synthon 31 by Method C (1058C, 4 h), using 1-[3-
(trifluoromethyl)phenyl]piperazine, in 69% yield; mp: 260–2618C;
1H NMR ([D6]DMSO): d=3.40–3.50 (m, 4H, piperazine CH2), 3.60–
3.70 (m, 4H, piperazine CH2), 6.70 (br s, 2H, NH2), 7.05–7.15 (d, J=
7.3 Hz, 1H, aromatic CH), 7.20–7.50 (m, 5H, H-6 and H-8 and aro-
matic CH), 8.15 (d, J=9.3 Hz, 1H, H-5), 8.75 (s, 1H, H-2), 15.90 ppm
(s, 1H, COOH); 13C NMR ([D6]DMSO): d=46.7, 47.5, 98.8, 105.9,
111.4, 115.2, 115.3, 116.5, 119.2, 124.8, 127.1, 130.3, 130.4, 143.9,
148.9, 151.2, 154.3, 166.8, 176.3 ppm; Anal. calcd for C21H19F3N4O3:
C 58.33, H 4.43, N 12.96, found: C 58.30, H 4.40, N. 13.12.
Ethyl 1-(acetylamino)-6-amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-
1-yl)-1,4-dihydroquinoline-3-carboxylate (34a): Compound 34a
was prepared from 33a, using Method D (15 min), in 37% yield;
mp: 279–2808C; 1H NMR ([D6]DMSO): d=1.25 (t, J=7.0 Hz, 3H,
CH2CH3), 2.15 (s, 3H, CH3), 2.90–3.10 (m, 4H, piperazine CH2), 3.60–
3.80 (m, 4H, piperazine CH2), 4.20 (q, J=7.0 Hz, 2H, CH2CH3), 5.30
(br s, 2H, NH2), 6.60–6.70 (m, 1H, pyridine CH), 6.80–7.00 (m, 2H,
H-8 and pyridine CH), 7.40–7.60 (m, 2H, H-5 and pyridine CH),
8.10–8.20 (m, 1H, pyridine CH), 8.35 (s, 1H, H-2), 11.50 ppm (br s,
1H, NH).
1-Amino-7-[4-(1,3-benzothiazol-2-yl)piperazin-1-yl]-4-oxo-1,4-di-
hydroquinoline-3-carboxylic acid (7c): Compound 7c was pre-
pared from synthon 31 by Method C (1058C, 4 h), using 1-(1,3-ben-
1
zothiazol-2-yl)piperazine,[21] in 40% yield; mp: 329–3308C; H NMR
([D6]DMSO): d=3.65–3.85 (m, 8H, piperazine CH2), 6.75 (br s, 2H,
NH2), 7.05–7.15 (m, 1H, benzothiazole CH), 7.25–7.55 (m, 4H, H-6,
H-8 and benzothiazole CH), 7.75–7.85 (m, 1H, benzothiazole CH),
8.20 (d, J=8.2 Hz, 1H, H-5), 8.70 (s, 1H, H-2), 15.90 ppm (s, 1H,
COOH); 13C NMR ([D6]DMSO): d=46.1, 47.3, 98.7, 105.9, 114.8,
116.6, 119.1, 121.9, 122.4, 126.9, 127.1, 130.1, 144.1, 148.9, 153.5,
154.4, 166.8, 168.4, 176.3 ppm: Anal. calcd for C21H19N5O3S: C 59.84,
H 4.54, N 16.62, found: C 60.00, H 4.45, N 16.78.
1-(Acetylamino)-6-amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-
1,4-dihydroquinoline-3-carboxylic acid (4a): A mixture of 34a
(0.04 g, 0.08 mmol) and LiOH (0.008 g, 0.35 mmol) in water/dioxane
(1:3, 1 mL) was stirred at RT for 24 h. The reaction mixture was
then concentrated in vacuo and neutralized with 2n HCl to yield a
precipitate which was filtered, washed with water and EtOH, and
1
dried to give 4a (0.025 g, 67%); mp: >3008C; H NMR ([D6]DMSO):
d=2.15 (s, 3H, CH3), 3.00–3.15 (m, 4H, piperazine CH2), 3.65–3.80
(m, 4H, piperazine CH2), 5.60 (br s, 2H, NH2), 6.60–6.70 (m, 1H, pyri-
dine CH), 6.85–6.95 (m, 1H, pyridine CH), 7.10 (s, 1H, H-8), 7.50–
7.65 (m, 2H, H-5 and pyridine CH), 8.10–8.20 (m, 1H, pyridine CH),
8.65 (s, 1H, H-2), 15.70 ppm (s, 1H, COOH); 13C NMR ([D6]DMSO):
d=21.2, 45.0, 49.8, 105.4, 106.5, 106.7, 107.6, 113.6, 122.0, 133.7,
138.0, 142.8, 146.6, 147.8, 148.0, 159.4, 166.6, 169.6, 176.7 ppm;
Anal. calcd for C21H22N6O4: C 59.71, H 5.25, N 19.89, found: C 59.90,
H 5,28, N 19.70.
1-Amino-6-hydroxy-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1,4-
dihydroquinoline-3-carboxylic acid (8a): A mixture of synthon 32
(0.05 g,
0.20 mmol) and 1-(2-pyridinyl)piperazine (0.20 g,
1.25 mmol) in N-methyl-2-pyrrolidone (0.5 mL) was heated at
1058C for 5 h. After cooling, the mixture was poured into ice/water
and acidified with 2n HCl (pH 6) to give a solid, which was filtered,
washed with water, dried, and crystallized from DMF/EtOH to
1
afford 8a (0.07 g, 91%); mp: 277–2788C (dec); H NMR ([D6]DMSO):
d=3.25–3.35 (m, 4H, piperazine CH2), 3.60–3.70 (m, 4H, piperazine
CH2), 6.70 (t, J=5.8 Hz, 1H, pyridine CH), 6.80 (br s, 2H, NH2), 7.00
(d, J=8.5 Hz, 1H, pyridine CH), 7.45 (s, 1H, H-8), 7.60 (t, J=7.5 Hz,
1H, pyridine CH), 7.65 (s, 1H, H-5), 8.15 (d, J=3.7 Hz, 1H, pyridine
CH), 8.60 (s, 1H, H-2), 10.50 (br s, 1H, OH), 16.00 ppm (br s, 1H,
COOH); 13C NMR ([D6]DMSO): d=47.1, 48.9, 105.4, 105.8, 107.0,
108.5, 114.5, 120.7, 137.7, 138.2, 146.5, 147.2, 148.0, 149.8, 159.2,
167.5, 175.2 ppm; Anal. calcd for C19H19N5O4: C 59.84, H 5.02, N
18.36, found: C 59.65, H 4.95, N. 18.19.
Ethyl
3-[(2-chloroethyl)amino]-2-(2,4-dichloro-5-nitrobenzoyl)-
prop-2-enoate (36): A mixture of 9[7] (1 g, 2.76 mmol) and 2-chloro-
ethylamine hydrochloride (0.38 g, 3.32 mmol) in EtOH/Et2O (1:3,
30 mL) was stirred at RT for 30 min. The mixture reaction was then
concentrated in vacuo to yield an oil, which was purified by flash
chromatography, eluting with MeOH/CHCl3 (3%), to give 36
(0.77 g, 70%); mp: 133–1348C; 1H NMR (CDCl3): d=1.05 (t, J=
7.1 Hz, 3H, CH2CH3), 3.70–3.80 (m, 4H, CH2), 4.05 (q, J=7.1 Hz, 2H,
CH2CH3), 7.55 (s, 1H, H-3’), 7.80 (s, 1H, H-6’), 8.20 (d, J=13.8, 1H,
H-3), 11.10 ppm (d, J=13.8 Hz, 1H, NH).
1-Amino-6-hydroxy-4-oxo-7-{4-[3-(trifluoromethyl)phenyl]pipera-
zin-1-yl}-1,4-dihydroquinoline-3-carboxylic acid (8b): Compound
8b was prepared following the procedure used for the synthesis of
Ethyl 3-[(2-chloroethyl)amino]-2-(2-chloro-4-fluorobenzoyl)prop-
2-enoate (37): Starting from synthon 35[3] and following the proce-
1888
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
dure used for the synthesis of compound 36 (12 h), compound 37
was obtained in 99% yield; mp: 131–1328C; H NMR (CDCl3): d=
1H, H-8), 7.30–7.40 (m, 1H, pyridine CH), 7.55–7.70 (m, 2H, H-5 and
CH=CH2), 7.85–8.10 (m, 2H, pyridine CH), 8.60 (s, 1H, H-2),
15.50 ppm (br s, 1H, COOH); 13C NMR ([D6]DMSO): d=46.4, 49.2,
105.8, 107.1, 107.2, 107.8, 113.3, 114.9, 122.0, 132.1, 135.7, 138.1,
142.5, 143.5, 145.3, 147.9, 159.4, 166.9, 177.2 ppm; Anal. calcd for
C21H21N5O3: C 64.44, H 5.41, N 17.89, found: C 64.19, H 5.70, N
17.51.
1
0.85 (t, J=7.1 Hz, 3H, CH2CH3), 1.00 (t, J=7.1 Hz, 3H, CH2CH3),
3.70–3.80 (m, 4H, CH2), 4.00–4.10 (m, 2H, CH2CH3), 7.00 (dt, J=8.2
and 2.4 Hz, 1H, H-5’), 7.15 (dd, J=5.8 and 2.4 Hz, 1H, H-6’), 7.20
(dd, J=8.0 and 5.8 Hz, 1H, H-3’), 8.25 (d, J=13.8 Hz, 1H, H-2), 9.50
(br s, 1H, NH), 11.10 ppm (br s, 1H, NH).
Ethyl 7-chloro-6-nitro-4-oxo-1-vinyl-1,4-dihydroquinoline-3-car-
boxylate (38): A mixture of 36 (0.55 g, 1.39 mmol) and Cs2CO3
(1.13 g, 3.47 mmol) in CH3CN (5 mL) was refluxed for 3 h. After
cooling, the reaction mixture was poured into ice/water to give a
precipitate, which was filtered and washed with water and Et2O,
yielding 38 (0.44 g, 98%); mp: 191–1928C; 1H NMR (CDCl3): d=
1.45 (t, J=7.1 Hz, 3H, CH2CH3), 4.45 (q, J=7.1 Hz, 2H, CH2CH3),
5.70 (dd, J=2.0 and 14.9 Hz, 1H, CH=CH2), 5.80 (dd, J=2.0 and
8.0 Hz, 1H, CH=CH2), 7.05 (dd, J=8.0 and 14.9 Hz, 1H, CH=CH2),
7.65 (s, 1H, H-8), 8.65 (s, 1H, H-2), 9.00 ppm (s, 1H, H-5).
7-Fluoro-4-oxo-1-vinyl-1,4-dihydroquinoline-3-carboxylic
acid
(42): Compound 42 was prepared from 39 following Method B
(1 h). After cooling, the solution was acidified (pH 5) with 2n HCl
to obtaining a precipitate, which was filtered and washed with
1
water and Et2O to give 42 in 84% yield; mp: 246–2478C; H NMR
([D6]DMSO): d=5.60 (d, J=7.6 Hz, 1H, CH=CH2), 6.00 (d, J=
14.4 Hz, 1H, CH=CH2), 7.50–7.60 (m, 2H, H-5 and CH=CH2), 7.80 (d,
J=10.8 Hz, 1H, H-8), 8.40–8.50 (m, 1H, H-6), 8.80 (s, 1H, H-2),
14.75 ppm (br s, 1H, COOH).
4-Oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1-vinyl-1,4-dihydroquino-
line-3-carboxylic acid (6a): Compound 6a was prepared from 42
by Method C (1058C, 4 h), using 1-(2-pyridinyl)piperazine, in 51%
yield; mp: 248–2498C; 1H NMR ([D6]DMSO): d=3.65–3.75 (m, 4H,
piperazine CH2), 3.85–3.95 (m, 4H, piperazine CH2), 5.60 (dd, J=2.0
and 8.0 Hz, 1H, CH=CH2), 5.70 (dd, J=2.0 and 15 Hz, 1H, CH=CH2),
6.85–6.95 (m, 2H, H-8 and pyridine CH), 7.25–7.35 (m, 1H, pyridine
CH), 7.40 (dd, J=1.7 and 9.2 Hz, 1H, H-6), 7.65–7.75 (dd, J=8.0
and 15.0 Hz, 1H, CH=CH2), 7.85–7.95 (m, 1H, pyridine CH), 8.10 (d,
J=4.8 Hz, 1H, pyridine CH), 8.20 (d, J=9.2 Hz, 1H, H-5), 8.65 (s,
1H, H-2), 15.90 ppm (br s, 1H, COOH); 13C NMR ([D6]DMSO): d=
44.8, 46.9, 98.9, 107.5, 107.8, 113.4, 114.8, 115.2, 115.7, 127.1, 135.7,
138.2, 141.7, 146.5, 147.8, 154.3, 159.2, 166.5, 177.3 ppm; Anal.
calcd for C21H20N4O3: C 67.01, H 5.36, N 14.88, found: C 67.12, H
5.57, N 14.93.
Ethyl 7-fluoro-4-oxo-1-vinyl-1,4-dihydroquinoline-3-carboxylate
(39): Compound 39 was obtained from synthon 37, following the
procedure used for the synthesis of compound 38, in 72% yield;
mp: 101–1028C; 1H NMR (CDCl3): d=1.40 (t, J=7.1 Hz, 3H,
CH2CH3), 4.50 (q, J=7.1 Hz, 2H, CH2CH3), 5.60 (dd, J=2.0 and
8.0 Hz, 1H, CH=CH2), 5.70 (dd, J=2.0 and 14.9 Hz, 1H, CH=CH2),
7.10 (dd, J=14.9 and 8.0 Hz, 1H, CH=CH2), 7.15 (dd, J=2.5 and
10 Hz, 1H, H-8), 7.20–7.25 (m, 1H, H-6), 8.50 (dd, J=5.4 and 2.4 Hz,
1H, H-5), 8.60 ppm (s, 1H, H-2).
Ethyl 6-nitro-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1-vinyl-1,4-
dihydroquinoline-3-carboxylate (40a): Compound 40a was pre-
pared from 38 by Method E (808C, 1 h), using 1-(2-pyridinyl)piper-
azine, in 61% yield; mp: 205–2068C; 1H NMR (CDCl3): d=1.45 (t,
J=7.1 Hz, 3H, CH2CH3), 3.30–3.40 (m, 4H, piperazine CH2), 3.70–
3.80 (m, 4H, piperazine CH2), 4.45 (q, J=7.1 Hz, 2H, CH2CH3), 5.60
(dd, J=2.0 and 8.0 Hz, 1H, CH=CH2), 5.70 (dd, J=2.0 and 15 Hz,
1H, CH=CH2), 6.65–6.75 (m, 2H, pyridine CH), 6.80 (s, 1H, H-8), 7.05
(dd, J=8.0 and 15 Hz, 1H, CH=CH2), 7.50–7.60 (m, 1H, pyridine
CH), 8.20–8.30 (m, 1H, pyridine CH), 8.55 (s, 1H, H-5), 8.90 ppm (s,
1H, H-2).
7-{4-[3-(trifluoromethyl)phenyl]piperazin-1-yl}-4-oxo-1-vinyl-1,4-
dihydroquinoline-3-carboxylic acid (6b): Compound 6b was pre-
pared from 42 by Method C (1058C, 4 h), using 1-[3-(trifluorome-
thyl)phenyl]piperazine, in 63% yield; mp: 240–2418C; 1H NMR
([D6]DMSO): d=3.35–3.45 (m, 4H, piperazine CH2), 3.60–3.70 (m,
4H, piperazine CH2), 5.60 (dd, J=2.0 and 8.0 Hz, 1H, CH=CH2), 5.90
(dd, J=2.0 and 15 Hz, 1H, CH=CH2), 6.90 (s, 1H, H-8), 7.10 (d, J=
7.4 Hz, 1H, aromatic CH), 7.20 (s, 1H, aromatic CH), 7.30 (d, J=
7.8 Hz, 1H, H-6), 7.40–7.50 (m, 2H, CH=CH2 and aromatic CH),
7.60–7.70 (m, 1H, aromatic CH), 8.20 (d, J=9.1 Hz, 1H, H-5), 8.65 (s,
1H, H-2), 15.90 ppm (br s, 1H, COOH); 13C NMR ([D6]DMSO): d=
46.5, 47.4, 98.7, 107.8, 111.3, 114.8, 115.2, 115.7, 119.2, 124.8, 127.4,
130.2, 130.5, 135.7, 141.7, 146.5, 151.2, 154.4, 166.5, 177.3 ppm;
Anal. calcd for C23H20F3N3O3: C 62.30, H 4.55, N 9.48, found: C
62.45, H 4.57, N 9.55.
Ethyl 6-amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1-vinyl-1,4-
dihydroquinoline-3-carboxylate (41a): A solution of SnCl2·2H2O
(2.50 g, 1.34 mmol) in 8n HCl (3 mL) was added portionwise to a
solution of 6-nitro derivative 40a (1.70 g, 3.78 mmol) in 8n HCl
(7 mL), maintained at 408C. After 2 h, the reaction mixture was
poured into ice/water and neutralized with 2n NaOH to obtain a
precipitate, which was filtered and purified by flash chromatogra-
phy (MeOH/CHCl3, 4%) to give 41a (0.14 g, 9%); mp: 2508C (dec);
1H NMR ([D6]DMSO): d=1.35 (t, J=7.1 Hz, 3H, CH2CH3), 3.00–3.15
(m, 4H, piperazine CH2), 3.60–3.75 (m, 4H, piperazine CH2), 4.20 (q,
J=7.1 Hz, 2H, CH2CH3), 5.25 (s, 2H, NH2), 5.40 (dd, J=2.0 and
8.0 Hz, 1H, CH=CH2), 5.70 (dd, J=2.0 and 15 Hz, 1H, CH=CH2), 6.65
(t, J=5.5 Hz, pyridine CH), 6.90 (d, J=8.5 Hz, pyridine CH), 7.10 (s,
1H, H-8), 7.45–7.65 (m, 3H, H-5, CH=CH2 and pyridine CH), 8.15 (d,
J=3.5 Hz, pyridine CH), 8.40 ppm (s, 1H, H-2).
7-[4-(1,3-Benzothiazol-2-yl)piperazin-1-yl]-4-oxo-1-vinyl-1,4-dihy-
droquinoline-3-carboxylic acid (6c): Compound 6c was prepared
from 42 by Method C (1058C, 4 h), using 1-(1,3-benzothiazol-2-yl)-
1
piperazine,[21] in 75% yield; mp: >3008C; H NMR ([D6]DMSO): d=
3.70–3.80 (m, 8H, piperazine CH2), 5.60 (d, J=8.0 Hz, 1H, CH=CH2),
5.90 (d, J=15 Hz, 1H, CH=CH2), 6.90 (s, 1H, H-8), 7.15 (t, J=7.6 Hz,
1H, benzothiazole CH), 7.30 (t, J=7.6 Hz, 1H, benzothiazole CH),
7.40 (d, J=8.9 Hz, 1H, H-6), 7.50 (d, J=7.6 Hz, 1H, benzothiazole
CH), 7.65 (dd, J=8.0 and 15.0 Hz, 1H, CH=CH2), 7.80 (d, J=7.5 Hz,
1H, benzothiazole CH), 8.20 (d, J=9.2 Hz, 1H, H-5), 8.65 (s, 1H, H-
2), 15.90 ppm (br s, 1H, COOH); 13C NMR ([D6]DMSO): d=46.2, 47.8,
99.0, 107.8, 114.8, 115.3, 115.8, 118.9, 121.8, 122.0, 126.6, 127.5,
130.3, 135.7, 141.7, 146.5, 153.4, 154.3, 166.5, 168.3, 177.3 ppm;
Anal. calcd for C23H20N4O3S: C 63.87, H 4.66, N 12.95, found: C
64.05, H 4.78, N 13.04.
6-Amino-4-oxo-7-(4-(pyridin-2-yl)piperazin-1-yl)-1-vinyl-1,4-dihy-
droquinoline-3-carboxylic acid (5a): Compound 5a was prepared
from 41a using Method B (1 h). After cooling, the solution was
neutralized with 2n HCl to obtain a precipitate, which was filtered,
washed with water and EtOH, and crystallized from DMF/EtOH to
give 5a in 42% yield; mp: 254–2558C; 1H NMR ([D6]DMSO): d=
3.10–3.25 (m, 4H, piperazine CH2), 3.85–4.00 (m, 4H, piperazine
CH2), 5.60 (dd, J=2.0 and 8.0 Hz, 1H, CH=CH2), 5.70 (dd, J=2.0
and 15 Hz, 1H, CH=CH2), 6.85–6.95 (m, 1H, pyridine CH), 7.20 (s,
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1889

O. Tabarrini et al.
MED
sulfate (DS) 8000 (interferes with the binding of virus to cells), AZT
(inhibits the reverse transcription process), and ritonavir (inhibits
the proteolytic cleavage of precursor proteins). The compounds
were added at concentrations 100-fold greater than their IC50
values, as determined in MT-4 cells using an MTT assay.
Biology
In vitro anti-HIV assays
Evaluation of the antiviral activity of the compounds against an
HIV-1 strain (IIIB) and an HIV-2 strain (ROD) in MT-4 cells was carried
out using an MTT assay, as previously described.[22,23] Briefly, stock
solutions (10
ꢂ final concentration) of test compounds were
Persistently infected HuT-78 Experiments
added in 25 mL volumes to two series of triplicate wells, so as to
allow simultaneous evaluation of their effects on mock- and HIV-
infected cells at the beginning of each experiment. Serial fivefold
dilutions of test compounds were made directly in flat-bottom 96-
well microtiter trays using a Biomek 3000 robot (Beckman Coulter,
Inc.). Untreated HIV- and mock-infected cell samples were included
as a control for each sample. HIV-1 (IIIB)[24] or HIV-2 (ROD)[25] stock
(50 mL) at 100–300 CCID50 (50% cell culture infectious dose) or cul-
ture medium was added to either the infected or mock-infected
wells of the microtiter tray. Mock-infected cells were used to evalu-
ate the effect of test compounds on uninfected cells, in order to
assess cytotoxicity. Exponentially growing MT-4 cells[26] were centri-
fuged for 5 min at 1000 rpm, and the supernatant was discarded.
The cells were resuspended at 6ꢂ105 cellsmLꢀ1, and 50 mL volumes
were transferred to the microtiter tray wells. Five days post-infec-
tion, the viability of mock- and HIV-infected cells was examined
spectrophotometrically using an MTT assay.
Chronically infected HuT-78 cells were washed with PBS and pellet-
ed by centrifugation. The supernatant was discarded, and washing
was repeated three times. The test compounds, at the required
concentrations, were transferred into a 48-well plate. To each well,
200000 cells were added (final volume per well=1 mL). The cul-
tures were allowed to grow for 44 h before the supernatant was
collected and virus-associated p24 concentrations were determined
by ELISA. Nevirapine (non-nucleoside reverse transcriptase inhibi-
tor; NNRTI) and ritonavir (protease inhibitor; PI) were included as
controls.
Latently infected OM-10.1 and U1 cells
Activity values of the compounds against chronic HIV-1 infection
were based on the inhibition of p24 antigen production in OM-
10.1 and U1 cells after stimulation with tumor necrosis factor a
(TNF-a, Roche Diagnostics) and phorbol 12-myristate 13-acetate
(PMA, Sigma–Aldrich). Briefly, OM-10.1 and U1 cells (5ꢂ105
cellsmLꢀ1) were incubated in the presence or absence of test com-
pounds for 2 h in 48-well plates. After this short incubation period,
the cells were stimulated with 1 ngmLꢀ1 of TNF-a or 0.02 mm of
PMA, followed by the transfer of 2ꢂ200 mL into 96-well plates for
toxicity evaluation. After a two-day incubation at 378C, the cell cul-
ture supernatants were collected from the 48-well plates and ex-
amined for their p24 antigen levels with the HIV-1 p24 ELISA kit
(Perkin–Elmer). Cytotoxicity of the compounds for both latently
HIV-1-infected OM-10.1 and U1 cell lines in the 96-well plates were
based on the MTT cell viability staining, as previously
described.[22,23]
The MTT assay is based on the reduction of yellow 3-(4,5-dime-
thylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Acros Or-
ganics) by mitochondrial dehydrogenase of metabolically active
cells to a blue–purple formazan, which can be measured spectro-
photometrically. Absorbances were read in an eight-channel, com-
puter-controlled photometer (Tecan Safire, MTX Lab Systems, Inc;
Virginia, USA) at two wavelengths (540 and 690 nm). All data were
calculated using the median optical density (OD) value of three
wells. The 50% cytotoxic concentration (CC50) was defined as the
concentration of test compound required to reduce absorbance
(OD540) of the mock-infected control sample by 50%. The concen-
tration achieving 50% protection from the cytopathic effect of the
virus in infected cells was defined as the 50% effective concentra-
tion (EC50).
Peripheral blood mononuclear cells (PBMCs) from healthy donors
were isolated by density centrifugation (Lymphoprep, Nycomed
Pharma, AS Diagnostics) and stimulated with phytohemagglutin
(PHA; Sigma–Aldrich) for three days. The activated cells (PHA-
stimulated blasts) were washed with phosphate-buffered saline
(PBS), and viral infections were induced as described by the AIDS
clinical trial group protocols.[27] Briefly, PBMCs (2ꢂ105/200 mL) were
plated in the presence of serial dilutions of test compound and
were infected with HIV stocks at 1000 CCID50 per mL. Four days
post-infection, 125 mL of the supernatant from the infected cul-
tures was removed and replaced with 150 mL of fresh medium con-
taining test compound at the appropriate concentration. Seven
days after plating the cells, p24 antigen was detected in the cul-
ture supernatant by an enzyme-linked immunosorbent assay
(ELISA; Perkin–Elmer, Brussels, Belgium).
Fluorescence quenching assay (FQA)
The effect of these new quinolones on the Tat/TAR complex was
evaluated using an FQA, a FRET-based competition assay, as previ-
ously described,[20] using the Tat-derived peptide labeled with the
donor fluorescein at its N terminus and the 29 nt wtTAR labeled at
its 3’ end with a dabcyl moiety (quencher). The peptide sequence
corresponds to the minimal amino acid sequence (48–57;
GRKKRRQRRR) necessary for TAR binding, with the sequence 5’-
GGCAGAUCUGAGCCUGGGAGCUCUCUGCC-3’. Quenching of the
Tat peptide fluorescence occurs upon incubation with increasing
concentration of the quenching nucleic acid; disruption of pep-
tide–RNA binding by competing quinolones is observed as an
increase in donor fluorescence intensity.
Titrations of labeled Tat/TAR in the presence of competitors were
performed in triplicate in a 96-well plate reader (Victor III, Perkin–
Elmer); the fluorescein-labeled peptide was excited at 490 nm and
emission was recorded at 535 nm. The plates were assembled with
190 mL of a solution containing 10 nm fluorescein–peptide in
10 mm Tris, 1 mm Mg(ClO4)2, 20 mm NaCl, and 0.01% Triton X-100
(for the ten-amino acid peptide), in the presence of 10 mm quino-
lone. Fluorescence intensities in the presence of increasing concen-
trations of dabcyl-TAR were measured and curves were fitted to
obtain Ki values. Preliminary determination of the spectral parame-
Time-of-addition (TOA) experiments
MT-4 cells were infected with HIV-1 (IIIB) at a multiplicity of infec-
tion (moi) of 0.5 to synchronize all steps of viral replication. The
test compounds were added at different times post-infection
(0–25 h).[28] Viral p24 Ag production was determined at 31 h post-
infection by ELISA (Perkin–Elmer, Brussels, Belgium). Reference
compounds with known modes of action were included: dextran
1890
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892

Anti-HIV Quinolones
ters of all quinolones allowed us to exclude optical interferences
with fluorescein–peptide signal within the conditions of the assay.
lone compounds were used to determine background fluorescence
for each sample.
Magnesium binding
Acknowledgements
Magnesium binding constants were determined by fluorometric ti-
trations on a Perkin–Elmer LS 50 instrument in 10 mm Tris and
20 mm NaCl (pH 7.5) at 258C, as previously described.[17] Briefly, the
fluorescence intensity of the quinolone solution was measured fol-
lowing addition of known, small amounts of a 0.3 m solution of
Mg(ClO4)2 in the same buffer. Each solution was filtered through a
0.45 mm filter before use to eliminate any particulate material that
would interfere with the fluorescence response. Due to the large
excess of Mg2+ (millimolar range), as compared to the quinolone
(micromolar range), only the 1:1 equilibrium was taken into ac-
count. The related association constant KMg for this equilibrium is
defined by Equation (1):
This work was supported in part by the MIUR (PRIN 2008, Roma,
Italy) (VC and BG), and grants from the Katholieke Universiteit
Leuven (Concerted Actions, GOA 05/19). We are grateful to Rob-
erto Bianconi, Kristien Erven, Cindy Heens, and Kris Uyttersprot
for excellent technical assistance.
Keywords: 1-vinylquinolones · anti-HIV quinolones · antiviral
agents · medicinal chemistry · transcription inhibitors
[1] V. Cecchetti, C. Parolin, S. Moro, T. Pecere, E. Filipponi, A. Calistri, O. Tab-
arrini, B. Gatto, M. Palumbo, A. Fravolini, G. Palꢃ, J. Med. Chem. 2000,
43, 3799–3802.
KMg ¼ ½QMꢄ=f½Qꢄ ꢅ ½Mꢄg
ð1Þ
[2] O. Tabarrini, M. Stevens, V. Cecchetti, S. Sabatini, M. Dell’Uomo, G. Man-
froni, M. Palumbo, C. Pannecouque, E. De Clercq, A. Fravolini, J. Med.
Chem. 2004, 47, 5567–5578.
[3] O. Tabarrini, S. Massari, D. Daelemans, M. Stevens, G. Manfroni, S. Sabati-
ni, J. Balzarini, V. Cecchetti, C. Pannecouque, A. Fravolini, J. Med. Chem.
2008, 51, 5454–5458.
[4] S. Massari, D. Daelemans, G. Manfroni, S. Sabatini, O. Tabarrini, C. Panne-
couque, V. Cecchetti, Bioorg. Med. Chem. 2009, 17, 667–674.
[5] M. Stevens, J. Balzarini, O. Tabarrini, G. Andrei, R. Snoeck, V. Cecchetti, A.
Fravolini, E. De Clercq, C. Pannecouque, J. Antimicrob. Chemother. 2005,
56, 847–855.
where Q is the quinolone, M is the magnesium ion, and QM repre-
sents the quinolone–magnesium complex. At any Mg2+ concentra-
tion, the observed fluorescence emission signal (F) is related to the
fluorescence of the quinolone in the absence of magnesium (F0),
its fluorescence when fully bound to magnesium ions obtained at
Mg2+ saturation (F ), and the total quinolone concentration (Ct,
1
where Ct =[Q]+[QM]), by Equation (2).
F Ct ¼ F0½QꢄþF ½QMꢄ
ð2Þ
1
For every Mg2+, therefore, we can experimentally measure n, the
fraction of binary quinolone–magnesium complex, defined as
[QM]/Ct [see Equations (3), (4) and (5)].
[6] D. Daelemans, C. Pannecouque, G. N. Pavlakis, O. Tabarrini, E. De Clercq,
Mol. Pharmacol. 2005, 67, 1574–1580.
[7] V. Cecchetti, S. Clementi, G. Cruciani, A. Fravolini, P. G. Pagella, A.
Savino, O. Tabarrini, J. Med. Chem. 1995, 38, 973–982.
[8] R. R. Poondra, P. M. Fischer, N. J. Turner, J. Org. Chem. 2004, 69, 6920–
6922.
[9] K. Grohe, H. Heitzer, Liebigs Ann. Chem. 1987, 871–879.
[10] H. Koga, A. Itoh, S. Murayama, S. Suzue, T. Irikura, J. Med. Chem. 1980,
23, 1358–1363.
[11] E. E. Glover, K. T. Rowbottom, J. Chem. Soc., Perkin Trans. I 1976, 367–
371.
[12] K. J. Shah, E. A. Coats, J. Med. Chem. 1977, 20, 1001–1006.
[13] D. H. Boschelli, Y. D. Wang, S. L. Johnson, D. M. Berger, (Wyeth Holdings
Corporation; Madison, NJ, USA), WO2003093241, 2003.
[14] L. A. Mitscher, P. Devasthale, R. Zavod, Structure–Activity Relationships.
In Quinolone Antimicrobial Agents, 2nd ed.; (Eds.: D. C. Hooper, J. S.
Wolfson,) American Society for Microbiology, Washington D. C., 1998,
pp. 3–51.
[15] S. Massari, D. Daelemans, M. L. Barreca, A. Knezevich, S. Sabatini, V. Cec-
chetti, A. Marcello, C. Pannecouque, O. Tabarrini, J. Med. Chem. 2010,
53, 641–648.
n¼ ðFꢀF0Þ=ðF ꢀF0Þ
ð3Þ
ð4Þ
ð5Þ
1
½QMꢄ¼ n ꢅ Ct
½Qꢄ¼ ð1ꢀnÞ ꢅ Ct
Assuming that [M] is equivalent to the total magnesium concentra-
tion (the amount of complex formed is micromolar or lower,
whereas the total metal ion is in the millimolar range), Equation (1)
becomes:
KMg ¼ n=f ð1ꢀnÞ ½Mꢄ g
or
ð1bÞ
n=ð1ꢀnÞ ¼ KMg ½Mꢄ
Hence, the slope of the linear plot n/(1ꢀn) versus [Mg2+] corre-
[16] S. Richter, C. Parolin, B. Gatto, C. Del Vecchio, E. Brocca-Cofano, A. Fravo-
lini, G. Palu, M. Palumbo, Antimicrob. Agents Chemother. 2004, 48,
1895–1899.
sponds to the association constant.
[17] S. Richter, B. Gatto, O. Tabarrini, A. Fravolini, M. Palumbo, Bioorg. Med.
Chem. Lett. 2005, 15, 4247–4251.
[18] M. Palumbo, B. Gatto, G. Zagotto, G. Palꢃ, Trends Microbiol. 1993, 1,
232–235.
[19] G. Palꢃ, S. Valisena, G. Ciarrocchi, B. Gatto, M. Palumbo, Proc. Natl. Acad.
Sci. USA 1992, 89, 9671–9675.
[20] G. Manfroni, B. Gatto, O. Tabarrini, S. Sabatini, V. Cecchetti, G. Giaretta,
C. Parolin, C. Del Vecchio, A. Calistri, M. Palumbo, A. Fravolini, Bioorg.
Med. Chem. Lett. 2009, 19, 714–717.
[21] M. Verderame, J. Med. Chem. 1972, 15, 693–694.
[22] R. Pauwels, J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J.
Desmyter, E. De Clercq, J. Virol. Methods 1988, 20, 309–321.
[23] C. Pannecouque, D. Daelemans, E De Clercq, Nat. Protoc. 2008, 3, 427–
434.
Uptake studies
Jurkat cells (2,5ꢂ106 per well) were incubated in Hank’s balanced
salt solution (HBSS, GIBCO) with quinolone compounds (100 mm).
After 3 h incubation at 378C, cells were separated from the extra-
cellular solution by centrifugation through a water-impermeable
silicone oil barrier (density 1029 gcmꢀ3) in a microcentrifuge tube.
The cell pellet was lysed overnight with 1 mL of 0.1 m glycine–HCl
buffer (pH 3.0), then the samples were centrifuged for 5 min at
5600 g. The amount of compound in the lysis supernatant was de-
termined by fluorescence, using the experimentally determined ti-
tration curve at pH 3 for each quinolone. Controls without quino-
ChemMedChem 2010, 5, 1880 – 1892
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1891

O. Tabarrini et al.
MED
[24] M. Popovic, M. G. Sarngadharan, E. Read, R. C. Gallo, Science 1984, 224,
497–500.
[25] F. Barre-Sinoussi, J. C. Chermann, F. Rey, M. T. Nugeyre, S. Chamaret, J.
Grest, C. Dauget, C. Axler-Blin, F. Vezinet-Brun, C. Rouzioux, W. Rozen-
baum, L. Montagnier, Science 1983, 220, 868–871.
[28] R. Pauwels, K. Andries, J. Desmyter, D. Schols, M. J. Kukla, H. J. Breslin, A.
Raymaeckers, J. Van Gelder, R. Woestenborghs, J. Heykants, Nature
1990, 343, 470–474.
[26] I. Miyoshi, H. Taguchi, I. Kobonishi, S. Yoshimoto, Y. Ohtsuki, Y. Shiraishi,
T. Akagi, Gann Monogr. Cancer Res. 1982, 28, 219–228.
[27] J. Balzarini, A. Karlsson, M. J. Pꢄrez- Pꢄrez, L. Vrang, J. Walbers, H. Zhang,
B. ꢅberg, A. M. Vandamme, M. J. Camarasa, E. De Clercq, Virology 1993,
192, 246–253.
Received: June 28, 2010
Revised: August 31, 2010
Published online on October 6, 2010
1892
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1880 – 1892