Synthesis, cytotoxic activity, DNA binding and molecular docking studies of novel 9-anilinothiazolo[5,4-b]quinoline derivatives

Article abstract of DOI:10.1007/s00044-016-1718-4

Novel thiazolo[5,4-b]quinoline derivatives were prepared with or without a (2-(azacycloalkyl)ethyl)amino substituent at the 2-position. The effect of the substituent at 2-position on cytotoxic activity, DNA-intercalation and cytotoxic properties were evaluated. Substituents at 2-position bearing an aliphatic amine favored cytotoxicity, while removal of these substituents resulted in low or negligible cytotoxic properties. Additionally, the in silico predicted binding mode of the novel compounds into DNA correlated with the experimental intercalation data. These results suggest a strong influence of the substituent at 2-position on the DNA intercalation properties.

Full text of DOI:10.1007/s00044-016-1718-4

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-016-1718-4
ORIGINAL RESEARCH
Synthesis, cytotoxic activity, DNA binding and molecular docking
studies of novel 9-anilinothiazolo[5,4-b]quinoline derivatives
Francisco J. Reyes-Rangel1 A. Kémish López-Rodríguez1
●
●
Laura V. Pastrana-Cancino2 Marco. A. Loza-Mejía1 José D. Solano3
●
●
●
1
Rogelio Rodríguez-Sotres2 Alfonso Lira-Rocha
●
Received: 10 February 2016 / Accepted: 22 August 2016
© Springer Science+Business Media New York 2016
Abstract Novel thiazolo[5,4-b]quinoline derivatives were
prepared with or without a (2-(azacycloalkyl)ethyl)amino
substituent at the 2-position. The effect of the substituent at
2-position on cytotoxic activity, DNA-intercalation and
cytotoxic properties were evaluated. Substituents at 2-
position bearing an aliphatic amine favored cytotoxicity,
while removal of these substituents resulted in low or
negligible cytotoxic properties. Additionally, the in silico
predicted binding mode of the novel compounds into DNA
correlated with the experimental intercalation data. These
results suggest a strong influence of the substituent at 2-
position on the DNA intercalation properties.
Introduction
Cancer is still a worldwide health problem, affecting both
developing and developed countries. While several strate-
gies have been proposed for the treatment of cancer, the
small-molecule approach is still an important alternative
(Hoelder et al., 2012). In this case, the 9-anilinoacridine
derivatives (1, Fig. 1) have proven to be clinically effica-
cious and extensively studied as antitumor agents targeting
DNA-topoisomerase II (Siu and Pommier, 2013; Lindsey
et al., 2014; Demecunynck et al., 2001). One of these
derivatives, Amsacrine 2 (m-AMSA), has been used for the
treatment of leukemia and lymphoma (Denny, 1995, 2002).
In order to overcome the mechanism of resistance and to
increase the selectivity of drugs towards cancer cells, bioi-
sosteric replacement of a benzene moiety in the acridine
nucleus has led to cytotoxic novel compounds, such as
thiazolo[5,4-b]quinoline 3 (Rodríguez-Loaiza et al., 2004;
●
●
Keywords Thiazolo[5,4-b]quinoline Cytotoxic activity
●
DNA intercalators Molecular docking
Electronic supplementary material The online version of this article
(doi:10.1007/s00044-016-1718-4) contains supplementary material,
which is available to authorized users.
* Alfonso Lira-Rocha
lira@unam.mx
1
Departamento de Farmacia, Facultad de Química, Universidad
Nacional Autónoma de México, Cd. Universitaria, 04510
Coyoacán, México, Mexico
2
Departamento de Bioquímica, Facultad de Química, Universidad
Nacional Autónoma de México, Cd. Universitaria, 04510
Coyoacán, México, Mexico
3
Departamento de Biología, Facultad de Química, Universidad
Fig. 1 Chemical structure of acridine 1, m-AMSA 2 and some tri-
cyclic templates
Nacional Autónoma de México, Cd. Universitaria, 04510
Coyoacán, México, Mexico

Med Chem Res
chain. This point to a strong influence of the substitution
pattern at 2-position on cytotoxicity.
In the present work, two new miniseries of thiazolo[5,4-
b]quinoline derivatives were prepared. In the first series, a
ring of five or six members was incorporated at the end of
the side chain (series 15) to anchor its β-nitrogen. In the
second series, the substituent at 2-position was removed by
the displacement of a sulfonyl group (series 16). The
cytotoxic activity of these novel compounds was evaluated
in some tumor cell lines. The results confirm the sig-
nificance of the substituent at 2-position. The primary target
of thiazolo[5,4-b]quinoline derivatives has not been con-
firmed experimentally, but the DNA is the biological target
of m-AMSA and others 9-anilinoacridines. Therefore, we
include here the intercalating properties of the new com-
pounds from experimental data and predictions of the
binding mode of these novel compounds, based on in silico
molecular docking studies.
Results and discussion
Chemistry
Novel compounds were obtained by a divergent synthesis
based on a methodology already reported with some mod-
ifications (Loza-Mejía et al., 2009) Scheme 1.
Compound 10 was prepared using CHCl3 instead of
DMSO, as a reaction solvent, and dimethyl sulfate instead
of methyl iodide as an alkylating agent. In our experience,
the displacement of a methylthio group present in a thiazole
ring can be accomplished directly by a nucleophilic aro-
matic substitution reaction, but the yield can be improved
by the previous transformation to the corresponding sulfo-
nyl derivative. We have previously reported this last reac-
tion with H2O2 in AcOH for 24 h at room temperature, but
in some cases further purification procedures were required,
with a decrease in the yield of the sulfonyl compound.
Among the reported oxidizing agents, hydrogen peroxide is
an environmentally friendly substance and its oxidant
properties can be enhanced by some Lewis acid metal cat-
alysts (He et al., 2012; Sato et al., 2001). Thus, the oxida-
tion of compounds 13–14 was accomplished by H2O2 and
catalytic amounts of Na2WO4 ·2H2O to afford the corre-
sponding sulfonyl derivative in good yield. This procedure
cuts down reaction times (30 min) and facilitates the work-
up reaction. From this step onwards, the synthesis becomes
divergent. The substitution of the sulfonyl group with the
corresponding amine was carried out in DMF to render 15
series compounds.
Fig. 2 Chemical structure of thiazolo[5,4-b]quinoline derivatives
previously reported (Loza-Mejía et al., 2009)
Loza-Mejía et al., 2008; González-Sánchez et al., 2011),
furo[2,3-b]quinoline 4 (Chen et al., 2004), and pyrazolo[3,4-
b]quinoline 5 derivatives among others (Chen et al., 2005).
In previous reports, we described the synthesis, cytotoxic
activity and DNA-topisomerase II inhibition properties of
several derivatives of thiazolo[5,4-b]quinoline as potential
anticancer agents (Loza-Mejía et al., 2009). In these reports,
a diethylaminoethylamino group (series 6) or a 3-(diethy-
lamino)propyl-amino (series 7) substituent at 2-position of
the thiazoloquinoline nucleus was found to correlate with
increased cytotoxicity regardless of the substitution pattern
in the 9-anilino ring (Fig. 2). In addition, compounds with a
two-methylene chain were less cytotoxic than their three-
methylene chain counterparts. The influence of the side
chain conformation at 2-position was probed with several
saturated heterocyclic analogs (series 8). These compounds
were less cytotoxic than the corresponding open chain
analogs, indicating a flexibility requirement at the lateral
On the other hand, several methodologies have been
reported for the elimination of a methylthio group attached
to a heterocyclic ring; for instance, Baldwin et al. (1980)

Med Chem Res
Scheme 1 a CS2/TEA/CHCl3/(CH3)2SO4; b (CH3)2SO4/K2CO3/Me2CO; c 1. t-BuOK/THF 2. C6H5–NCS; d PPA/POCl3 130 °C; e NH2–C6H4–R;
f Na2WO4/AcOH/H2O2; g NaBH4/EtOH; h NH2–CH2CH2–N(CH2CH2)2 or NH2–CH2CH2–N(CH2CH2)2CH2/DMF
reported the elimination of the mentioned group attached at
2-position of a thiazole ring using Zn/HCl 3N. Other
authors have reported this transformation by treating furan
derivatives with Raney Ni in EtOH (Herrera et al., 2006;
Yin et al., 2008). In the case of series 16, the removal of this
substituent from 13 with Zn/HCl was unsuccessful. Pre-
vious reports concerning the hydrogenolytic removal of
halogen at 2-position of the thiazole ring prompted us to
apply this methodology by using sodium borohydride as a
reducing agent (Kerdesky and Seif, 1995); however, more
efficient desulfurization processes have been accomplished
by the previous transformation of the thioether group to the
corresponding sulfone derivative (Zumbrunn, 1998). Hence,
the corresponding sulfonyl derivative 14 was treated with
NaBH4 in THF, but the yields were about 40 % in all cases.
Yields improved with the use of absolute EtOH as solvent
(aprox 70 %). However, all of these three strategies failed to
remove the methylthio group from compounds 11 and 12.
The structures of novel compounds were determined by
an quasi-molecular ion peak M+ at m/z 307.0774 was in
good agreement with the molecular formula C17H13N3SO.
In the IR spectrum, the typical bands for a sulfonyl group
(1317 and 1146 cm−1) were absent. The NH band was
observed at 3120 cm−1, as well as the aromatic bands at
1595, 1571, 1543, 1503, 1476 and the vibration C–S at 764
cm−1. In the 1H NMR spectrum, the simple signal observed
at 9.40 ppm was assigned to the Ar–NH–Ar proton while
the other simple signal at 9.22 ppm was attributed to the
proton at 2-position. This chemical shift is in agreement
with a previous report of Dercitin spectral data, an acridine
alkaloid fused to a thiazole ring (Gunawardana et al., 1988).
The proton-proton coupling signals for H-2 were absent in
the NOESY experiment, providing evidence for the absence
of spatial proximity of other protons. Quinoline proton
signals appeared at 8.29 and 7.87 ppm and corresponded to
H-8 and H-5, respectively, while H-6 and H-7 protons
signals were observed at 7.75 and 7.49 ppm, respectively.
The multiple signals in the range of 6.60–6.58 ppm were
assigned to H-2′, H-4′, H-6′, and the signal at 7.11 ppm was
assigned to H-5′. The 13C NMR spectrum was used to
1
IR, H NMR and HRMS. For example, compound 9-[(3-
methoxyphenyl)amino]thiazolo[5,4-b]quinoline 16d with

Med Chem Res
verify the integrity of the molecule. The spectrum revealed
17 carbon-resonances, interpreted from DEPT experiment
data as seven quaternary, nine methine, and one methyl
carbons. The presence of the thiazole ring was supported by
the resonance signals at 160.7 (C2), 141.1 (C3a) and 133.1
(C9a) ppm. In addition, the absence of a signal at 14.5 ppm,
normally present in the spectrum of several methylthio
derivatives (Loza-Mejía et al., 2008), indicated the suc-
cessful removal of this substituent at 2-position. The
assignment of all signals is reported in the experimental
section.
activity data of previously reported compounds and novel
compounds are shown in Table 1.
By comparison of their activity profiles, compounds
15a–15c are slightly less active than compounds 6a–6c,
whereas there is no any significant difference between the
activity of compounds 6a–6c and 15d–15f. Compounds
15d–15f are slightly more active than compounds 15a–15c.
Apparently, a strainless tertiary amine at the β-nitrogen
atom of the side chain has a positive effect on cytotoxicity.
However, according to the results of molecular docking
study, the most cytotoxic compounds had high binding
score in their protonated state (vide infra). The theoretical
pKa values of compounds 6a, 15a and 15d (non-chlorinated
members of the corresponding miniseries) were calculated
using web calculator software. (Marvin 5.4.0.1, 2010, http://
www.chemaxon.com), and the pka value for the most active
compound 6a was 9.16, whereas for compounds 15a and
15d were 8.55 and 8.43, respectively (>90 % protonated
state for all three molecules, at pH 7.0). Thus, the basic
properties of the side chain may influence the efficacy of the
cytotoxic activity of these compounds and further studies
are in course in order to evaluate this possibility.
On the other hand, the halo derivatives of both series
showed activity improvements. According to these facts, the
cytotoxic activity is also modulated by the substitution
pattern in the anilino ring (Loza-Mejía et al., 2008).
In the 16 series, the removal of the substituent at 2-
position reduces or eliminates the cytotoxic activity
(Table 1, compare series 13 with series 16), except for
compounds 16d, 16e, and 16h, which were active on K-562
cell line and only slightly active on HeLa cells.
In order to make a more graphical comparison of the
cytotoxic potency of compounds for a given cell line, a
cytotoxic index (CtxI) was obtained by dividing the IC50 of
Amsacrine by that of the compound, for the given cell line.
A CtxI above one is indicative of cytotoxic compounds
more potent than Amsacrine. The CtxI may also be used to
identify those cell lines with a higher sensitivity to the
cytotoxic effect of the compounds under analysis. The CtxI
values for series 6 were obtained by using the m-AMSA
value previously reported (Loza-Mejía et al., 2009).
As shown in Fig. 3, no compound is more cytotoxic than
m-AMSA against HeLa cells (panel A) with exception of
compound 15e. In contrast, some compounds of the series 6
and 15 (Fig. 3, panels B and D) are more cytotoxic than m-
AMSA, against SW-480 and K-562 cell lines. A common
feature of these compounds is a dialkylaminoalkylamino
chain at the 2-position.
Cytotoxic activity
The cell lines used in the present study were: one cervical
cancer line (HeLa), two colorectal cancer cell lines (SW-
480 and SW-620) and one leukemic cell line (K-562). The
cytotoxic assays were carried out using the methodology
already described (Quintero et al., 1999). The cytotoxic
Table 1 Cytotoxic activity of already reported compoundsa (series
6 and 13) and novel compounds (series 15 and 16) (IC50 μM)
Compound
HeLa
SW-480
SW-620
K-562
6aa
6ba
6ca
15.96
9.12
37.7
21.6
16.8
12.19
7.26
>200
80.26
79.45
22.17
72.2
8.01
24.11
5.69
19.48
21.25
10.62
7.38
ND
14.33
12.56
>200
110.69
>200
66.65
>200
28.68
29.1
17.78
12.20
>200
129.73
149.95
26.58
110.8
43.75
31.58
14.37
13.47
20.13
18.12
11.6
10.16
>200
69.37
123.86
25.34
>200
7.75
13aa
13ba
13ca
13da
13ea
13fa
15a
15b
15c
27.81
14.64
17.46
32.28
12.06
14.42
ND
15.42
26.51
26.7
15d
15e
12.75
18.42
ND
15f
16a
16b
16c
ND
ND
ND
ND
ND
ND
ND
ND
ND
16d
16e
27.2
73.9
73.6
23.6
41.8
ND
30.8
ND
ND
16f
ND
ND
ND
16g
16h
16i
ND
ND
ND
ND
85.8
56.4
83.2
4.5
ND
ND
ND
33.7
9.84
DNA binding (Ethidium bromide displacement)
m-AMSA
14.63
19.75
16.73
ND no cytotoxicity at the highest concentration tested (100 μM)
Data taken from Loza-Mejía et al., 2009
Apparent DNA intercalation constants were determined by
a conventional method based on the quenching of the
a

Med Chem Res
Table 2 Apparent constants for ethidium bromide displacement from
DNA of already reported compoundsa (series 13) and novel compounds
(series 15 and 16)
b
c
Compound
Qmax
Q50
Qmax/Q50
6aa
70.86
4.17
16.86
13.62
1.29
4.20
0.31
2.00
0.84
1.79
0.39
0.68
2.60
1.36
1.44
3.07
1.48
1.37
1.24
1.30
2.19
2.15
2.09
6.49
13aa
13ba
13ca
13da
13ea
13fa
15a
15b
15c
2.60
1.84
2.19
6.65
3.71
3.73
9.48
7.84
11.54
15.08
37.18
53.56
8.10
39.24
50.62
77.39
24.88
50.05
30.42
35.10
20.64
20.66
23.90
19.03
47.79
15d
15e
33.87
22.17
28.22
15.93
9.42
15f
16a
16b
16d
16e
11.11
9.09
16f
m-AMSA
a
7.36
Data taken from Loza-Mejía et al., 2008
Maximum quenching
b
c
Fig. 3 Cytotoxicity index for some compounds of series 6, 15 and 16
Concentration to give 50 % quenching of fluorescence of bound
ethidium bromide (µM). Values are the mean of three experiments
fluorescence of the ethidium bromide-DNA complex, as
previously described (McConnaughie and Jenkins, 1995).
DNA has been considered as the target of several antitumor
agents. In a previous report, we determined the DNA
intercalating properties of several thiazolo[5,4-b]quinoline
derivatives (Loza-Mejía et al., 2008).
According to data in Table 2, compounds with an ethy-
lendiamine group (series 15 and compound 6a) displace
more ethidium bromide than 2-methylthio derivatives
(Higher Qmax). However, an inverse tendency was observed
for the binding affinity values (Q50), i.e. less active com-
pounds bound to DNA tighter than more active ones. It is
clear that the compounds with a chlorine atom displaced
ethidium bromide from more sites (Higher Qmax) than the
compounds without this atom; and this trend roughly cor-
relates with the cytotoxic data.
On the other hand, it is interesting that the Qmax values
for the compounds without a methylthio group are higher
than those of 2-methyltio derivatives and the Q50 values for
the first ones also are higher. This implies that multiples
sites in the DNA are occupied by the compounds of series
16 with low binding affinity (low “selectivity”) and this
could explain their low or null cytotoxic activity. Only the
compounds 16d and 16e showed significant activity.
These two compounds showed a good apparent efficiency
index (ratio Qmax/Q50). It is true that compounds with an
ethylendiamine group also exhibit high Q50 values, and thus
low intercalation efficiency index, and yet these compounds
exhibit high cytotoxicity. These data indicates that inter-
calating properties are not enough to predict the cytotoxicity
of these compounds, and clearly additional properties ought
to be considered (Xiao et al., 2005).
Docking studies
In order to determine the binding mode of action of the
novel compounds, rigid/flexible molecular docking studies
with DNA-Nα-(9-acridinoyl)-tetraarginine complex crystal
(PDB code:1G3X) (Malinina et al., 2002) were performed.
Antitumor activity of DNA-intercalators is generally asso-
ciated with strong DNA-binding and long drug residence
times at individual sites. The binding energy values for
novel compounds and some previously reported are listed in
the Table 3. The tricyclic system was intercalated between
the base pairs of DNA, and the anilino group was always
oriented into the minor groove. In all of the cases analyzed

Med Chem Res
Table 3 Binding free energy values for already reported compounds
and novel compounds
Compound DNA binding energy (ΔG) DNA binding energy
from molecular docking
(ΔHf) from PM7
force field (kcal mol−1
)
calculations (kcal mol−1
)
6a
−10.03
−10.81
−10.31
−8.51
−8.79
−8.65
−10.52
−11.19
−10.20
−10.66
−10.79
−11.21
−7.81
−8.32
−8.01
−8.08
−7.66
−8.22
−7.87
−8.08
−7.98
−65.12
−66.72
−76.06
−39.04
−47.02
−45.61
−65.32
−71.28
−62.17
−63.43
−72.85
−73.65
−40.75
−47.89
−47.88
−42.42
−45.01
−42.21
−44.20
−44.38
−42.96
6b
6c
13a
13b
13c
15a
15b
15c
15d
15e
15f
16a
16b
16c
16d
16e
16f
16g
16h
16i
Fig. 4 Predicted docking geometries for compounds 15f (blue) and 6c
(red). The theoretical hydrogen bonds are shown in yellow (Color
figure online)
this substituent were shifted towards thymine ring, thus
decreasing their overlap (Fig. 6). Formation of a charge
transfer complex through π–π interactions is known to play
a major role in the stability of DNA-intercalator complexes,
and such electronic interaction depends on a good overlap.
This observation can explain why the presence of a
methylthio group at position 2 led to a reduction of the Q50
(increase in affinity). The reduction on the number of sites
occupied by these derivatives in the DNA (QMAX) relates
with the ability of the compound to displace ethidium
bromide from specific DNA sites, and cannot be directly
correlated with the theoretical calculations presented here.
here, the thiazole ring was stacked in between thymine
bases (T619 and T620), with the exception of compounds
13a and 13b, in which it was stacked in between adenine
bases (A605 and A606). The energy values corresponding
to the compounds with a side-chain at 2-position were the
lowest (higher binding affinity) and directly correlated with
their Qmax values and cytotoxicity.
The principal interactions observed for the compounds of
series 15 were π–π stacking, hydrogen bonding and salt
bridge formation. A salt bridge was consistently formed
between the tertiary amino group and phosphate oxygen
(O620 or O619) of the DNA backbone (Fig. 4). The
orientation adopted by the tricyclic system in the docked
complex renders an optimum overlap with the DNA base
pairs (Fig. 5).
On the other hand, compounds lacking the methylthio
group were expected to have higher affinity for DNA than
the compounds bearing this group. However, the experi-
mental data indicated an opposite trend, because on aver-
age, series 16 showed a 2-times higher Q50 value than series
13. From the analysis of the docked complexes, compounds
bearing a methylthio group were predicted to dock into
DNA with a better overlap of the thiazoloquinoline nucleus
and the DNA base pairs; whereas the compounds without
Molecular electrostatic potential (MEP) profiles
The above theoretical calculation depends on a force-field
energy estimate. To take into account the electronic effects
in the interaction energy, all-atom semiempirical QM cal-
culations of the DNA–DNA intercalator complex were
performed for selected compounds. The docking geometries
were first refined through a PM7 geometry optimization,
and then the enthalpy of formation of the complex was
calculated as the difference of the complex energy minus
the sum of the energies of the isolated molecules (in the
same conformation). The resulting energy values are
included in the Table 3. Direct comparison of the two
estimates for the binding energy cannot be done, because
the methods have a different level of theory, and because
the QM calculation does not provide information on the
entropic contribution to the complex stability. However, the

Med Chem Res
Fig. 5 Left, predicted binding geometries of compounds 15f (blue) and
6c (red) in DNA suggest intercalation as the dominant binding mode.
The anilino group was found at the minor groove. Right, top view of
the DNA-thiazoloquinoline intercalation complex showing the max-
imum overlap with base pairs (Color figure online)
Fig. 6 Left, predicted binding
geometries of compounds 16c
(cyan) y 13c (red) in the DNA,
showing intercalation as binding
mode. Right, compound 13c
(red) intercalates with maximum
overlap with base pairs (Color
figure online)
results indicate a destabilization effect of anchoring the
tertiary amine on the substituent at the 2-position (Table 3,
6c vs. 15f). In agreement with a more extensive overlap
between the intercalator and the neighboring bases, the
presence of the methylthio substituent at position 2
increases the stability of the complex. In addition, in all
cases the LUMO orbital was localized on the thiazoloqui-
noline core (Fig. 7).
The introduction of an electron-withdrawing group on the
anilino group decreases the LUMO energy values as demon-
strated in Table 4 for compounds 6b, 6c, 15b, 15c, 15e 15f.
This trend correlates with cytotoxic activity and it is in
excellent agreement with our previous results (Loza-Mejía
et al., 2009). The same electronic trend is observed for com-
pounds 13b, 13c, 16b and 16c, but the low or null
cytotoxic activity of these compounds denotes that the presence
of an aliphatic tertiary amine at 2-position is a relevant factor
for the cytotoxic activity.
Previous studies have provided evidence of electrostatic
factors as significant contributors to intercalation energy
(Medhi et al., 1999; Bondarev et al., 2000). To evaluate
how electrostatics may contribute to the observed biological
differences between compounds of series 15 and 16, elec-
trostatic potential maps of DNA and thiazoloquinoline
derivatives were calculated. A negatively charged surface is
formed by the oxygen atoms of A617, A618, T619 and
T620 (Fig. 8a). Compounds with a side chain at 2-position
of the tricyclic system were predicted to orient this sub-
stituent towards this zone (Fig. 8b), which is expected to
maximize electrostatic interactions between the cation and
the DNA phosphates, particularly T619 and T620. In the
case of compounds of series 16 the positively charged zone
is located on the tertiary amine of position 2 (Fig. 8c), in
accordance with the orientation of the intercalator in the
predicted binding mode.

Med Chem Res
Table 4 PM7 calculated energy of the frontier orbitals (eV) for DNA
bases and thiazolo[5,4-b]quinoline derivatives
Compound A606
A605
T620
T619
Compound
HOMO HOMO HOMO HOMO LUMO
6a
−8.5396 −8.5133 −8.5619 −8.8068 −0.5931
−8.5996 −8.5410 −8.7457 −8.7884 −0.6868
−8.4597 −8.5300 −8.8322 −8.8205 −0.6919
−8.3603 −8.5435 −8.7015 −8.8130 −0.4837
−8.3410 −8.6618 −8.5177 −8.8674 −0.8177
−8.5907 −8.5617 −8.6701 −8.7833 −0.8862
−8.6490 −8.5540 −8.7308 −8.7126 −0.7260
−8.5818 −8.5333 −8.8000 −8.6712 −0.7033
−8.4977 −8.2012 −8.5703 −8.8382 −0.7446
−8.6429 −8.5067 −8.9088 −8.8246 −0.8684
−8.5502 −8.5145 −8.7888 −8.7512 −0.7169
−8.3681 −8.5360 −8.7800 −8.7182 −0.6182
−8.5508 −8.5444 −8.7100 −8.8046 −0.6920
−8.5193 −8.5100 −8.7355 −8.7257 −0.7475
−8.5355 −8.5425 −8.7429 −8.7859 −0.8581
−8.4989 −8.5384 −8.7459 −8.8601 −0.7260
−8.5204 −8.4574 −8.6691 −8.8775 −0.4956
−8.5543 −8.4890 −8.7466 −8.8178 −0.7387
−8.5882 −8.5723 −8.7561 −8.7926 −1.0184
−8.4675 −8.4640 −8.6772 −8.8169 −0.4545
−8.6103 −8.5457 −8.7061 −8.7459 −0.6459
6b
6c
13a
13b
13c
15a
15b
15c
15d
15e
15f
16a
16b
16c
16d
16e
16f
16g
16h
16i
Fig. 7 A 3D view of the LUMO orbital for complex DNA-Compound
6c. The positive or the negative sign of the wave function is indicated
by the green or golden color, respectively (Color figure online)
the methylthio group improves the intercalation of the
derivatives into the DNA, but it does not render the com-
pounds cytotoxic. The present study benefits from recent
improvements in the semiempirical QM calculations to
predict reliable geometries for macromolecular complexes
and exploits the ability of these methods to estimate the
influence of electronic effects in the binding mode of small
molecules to biological macromolecules.
Experimental
Chemistry
Fig. 8 Electrostatic potential surface maps of DNA a, compound 15f b
and compound 16c c mapped on the total electron density calculated
with PM7. Red color denotes the negative electrostatic potential and
blue color denotes positive electrostatic potential (Color figure online)
All starting materials were commercially available research-
grade chemicals and used without further purification.
Reactions were monitored by analytical TLC on precoated
silica gel 60 F254 plates (Aldrich). Column chromatography
was carried out on silica gel 60 (70–230 mesh, Merck).
Melting points were determined on a Fisher-Jones apparatus
and are uncorrected. Infrared spectra were recorded on a
Nicolet FT-5SX spectrophotometer model. 1H NMR spectra
were recorded on a Varian VxR-300S spectrometer
(300 MHz). Chemical shifts are reported in ppm (δ) and the
Conclusion
The present study reveals a strong influence of the sub-
stitution pattern at the 2-position of the thiazolo[5,4-b]qui-
noline derivatives on the cytotoxic activity. This
substitution pattern also has a significant influence on the
intercalating properties of these compounds. The removal of

Med Chem Res
signals are described as singlet (s), doublet (d), triplet (t),
quartet (q), broad (br) and multiplet (m); coupling constants
are reported in Hz. EI-MS were carried out on a JEOL JMS-
AX505-HA apparatus. FAB-MS were carried out on a
JEOL Sx102 apparatus. Compounds 10–13 were prepared
according to procedures already described.12 (Loza-Mejía
et al., 2009)
305 (M+−84, 5 %), 154 (M+−235, 65 %), 84 (M+−305,
42 %).
N9-(3-chlorophenyl)-N2-(2-(pyrrolidin-1-yl)ethyl)thiazolo
[5,4-b]quinoline-2,9-diamine (15b)
Yellow light solid (135 mg, 53.1 %), mp 109–111 °C; IR
(KBr) cm−1, 3373 (–NH– arom.), 3186 (–NH– aliphat),
3066 (–CH insatd.), 2997 (–CH satd.), 845 (–C–Cl), 759
General preparation of 2-methylsulfanyl derivatives
(series 14)
1
(–C–S); H NMR (300 MHz, DMSO-d6): δ = 9.0 (1H, s,
ArN–H), 8.73 (1H, br, aliph. N–H), 8.19 (1H, d, J = 6 Hz,
H-5), 7.92 (1H, d, J = 6 Hz, H-8), 7.64 (1H, t, J = 6 Hz, H-
7), 7.525 (1H, t, J = 6 Hz, H-6), 7.21 (1H, t, J = 9 Hz, H-6′),
6.76–6.87 (3H, m, H-2′, H-3′, H-4′), 2.91 (2H, s,
NH–CH2CH2N(CH2 CH2)2), 2.75 (2H, s, NH–CH2CH2N
(CH2 CH2)2), 2.525 (4H, br, NH–CH2CH2N(CH2CH2)2),
1.79 (4H, br, NH–CH2CH2N(CH2CH2)2); HRMS (FAB)
calcd for [MH]+ C22H23N5ClS: 424.14, found 424.1365;
MS (FAB) m/z: 424(MH+, 43 %), 426(MH++2, 16 %, 37Cl
and 34S): 423 (M+, 5 %), 154 (M+−269, 100 %), 136
(M+−287, 66 %), 84 (M+−339, 17 %).
To a stirred suspension of 1.2 mmol of the 2-(methylthio)-
9anilinothiazolo[5,4-b]quinoline derivative with the desired
substitution pattern in the aniline ring, in 5 mL of glacial
acetic acid, 20 mg of Na2WO4•2H2O were added. The
resulting mixture was stirred for five minutes. Then 3 mL of
hydrogen peroxide 30 % were dropwise in ten minutes. The
reaction mixture was vigorously stirred for 30 min at room
temperature. The suspension was poured over 50 mL of
water under continuous stirring. The solids formed were
collected by filtration, washed with water for several times,
dried by suction and used without further purification.
N9-(4-chlorophenyl)-N2-(2-(pyrrolidin-1-yl)ethyl)thiazolo
[5,4-b]quinoline-2,9-diamine (15c)
General preparation of compounds 15a–15f
Yellow light solid (150 mg, 62.4 %), mp 85–86 °C; IR
(KBr) cm−1: 3236 (–NH– arom.), 3029 (–NH aliphat. ),
2956 (–CH insatd.), 2818 (–CH satd.), 1672 (–C=N), 1559,
1523, 1490 (arom.) 851 (C–Cl), 761 (–C–S); 1H NMR (300
MHz, DMSO-d6): δ = 8.74 (1H, s, ArN–H), 8.42 (1H, t, J
= 5.1 Hz, aliph. N–H), 8.13 (1H, dd, J = 1.2, 8.7 Hz, H-5),
7.85 (1H, dd, J = 9.6, 0.1 Hz, H-8), 7.56 (1H, ddd, J = 1.2,
6.9, 9.6 Hz, H-7), 7.44 (1H, ddd, J = 1.2, 6.9, 8.4 Hz, H-6),
7.15 (2H, d, J = 8.7 Hz, H-3′ and H-5′), 6.78 (2H, d, J =
8.7 Hz, H-2′ and H-6′), 2.87 (2H, s, NH–CH2CH2N
(CH2CH2)2), 2.72 (2H, s, NH–CH2CH2N(CH2CH2)2), 2.36
(4H, br, NH–CH2CH2N(CH2CH2)2), 1.61–1.64 (4H, m,
NH–CH2CH2N(CH2CH2)2). HRMS (FAB) calcd for [MH]+
C22H23N5ClS: 424.14, found 424.1364; MS (FAB) m/z:
424 (MH+, 100 %), 426 (MH++2, 38 %, 37Cl and 34S), 423
(M+, 14 %), 154 (M+−269, 66 %), 84 (M+−339, 76 %).
To a suspension of the corresponding 2-methylsulfanyl
derivative 14 (0.6 mmol) in DMF (5 mL) at room tem-
perature, the corresponding amine (1.6 mmol) was added
dropwise. The suspension was stirred for 1 h at the same
temperature and then poured over 50 mL of crushed ice/
water. The precipitates formed were collected by filtration,
washed with water and dried by suction. The residue was
purified by column chromatography on silica gel using
chloroform/methanol (9:1) as the eluent to afford the cor-
responding product.
N9-phenyl-N2-(2-(pyrrolidin-1-yl)ethyl)thiazolo[5,4-b]
quinoline-2,9-diamine (15a)
Yellow solid (120 mg, 53.8 %), mp 89–90 °C; IR (KBr)
cm−1: 3413 (–NH–arom.), 3112–3063 (–CH insatd.),
2962–2919 (–CH satd.), 1476, 1536, 1577 (arom.), 756
N9-phenyl-N2-(2-(piperidin-1-yl)ethyl)thiazolo[5,4-b]
quinoline-2,9-diamine (15d)
1
(C–S); H NMR (300 MHz, DMSO-d6): δ = 8.56 (1H, s,
Ar–N–H), 8.41 (1H, t, J = 6 Hz, aliph. N–H), 8.05 (1H, d, J
= 6 Hz, H-5), 7.85 (1H, d, J = 6 Hz, H-8), 7.55 (1H, t, J = 6
Hz, H-7), 7.41 (1H, t, J = 9 Hz, H-6), 7.14 (2H, t, J = 9 Hz,
H-2′, H-6′), 6.76–6.86 (3H, dd, J = 9, 1.2 Hz, H-3′, H-4′,
H-5′), 2.61 (2H, t, J = 6 Hz, NH–CH2CH2N(CH2CH2)2),
3.29–3.46 (6H, m, NH–CH2CH2N(CH2CH2)2), 1.66 (4H,
br, NH–CH2CH2N(CH2CH2)2). HRMS (FAB) calcd for
[MH]+ C22H24N5S: 390.18, found 390.1737; MS (FAB)
m/z, rel. abundance %: 390 (MH+, 100 %), 389 (M+, 10 %),
Yellow light solid (180 mg, 68.5 %), mp 80–82 °C; IR
(KBr) cm−1: 3239 (–NH–arom.), 3191 (–NH-aliphat.), 3052
(–CH insatd.), 2852 (–CH satd.), 1588 (–C=N), 1523, 1493,
1434 (aromatic), 759 (–C–S); 1H NMR (300 MHz, DMSO-
d6): δ = 8.61 (1H, s, ArN–H), 8.45 1H, br, aliph. N–H), 8.10
(1H, d, J = 6 Hz, H-5), 7.86 (1H, d, J = 9 Hz, H-8), 7.6 (1H,
t, J = 6 Hz, H-6), 7.42 (1H, t, J = 9 Hz, H-7), 7.16 (2H, t, J
= 9 Hz, H-2′, H-6′), 6.81 (3H, t, J = 9 Hz, H-3′, H-4′, H-5′),

Med Chem Res
2.49 (2H, t, J = 3 Hz, NH–CH2CH2N(CH2CH2)2CH2),
3.30–3.38 (6H, m, NH–CH2CH2N(CH2CH2)2CH2), 1.46–1.58
(4H, br, NH–CH2CH2N(CH2CH2)2CH2), 1.33–1.43 (2H, br,
NH–CH2CH2N(CH2CH2)2CH2). HRMS (FAB) calcd for
[MH]+ C23H26N5S: 404.19, found 404.1911; MS (EI) m/z: 403
(M+, 4 %), 401 (M+−2, 2 %), 305 (M+−98, 10 %), 292
(M+−111, 34 %), 111 (M+−292, 68 %), 98 (M+−305, 100 %).
General preparation of compounds 16a–16i
To stirred suspension of the corresponding 2-
a
methylsulfanyl derivative 14 (0.62 mmol) in absolute etha-
nol (6 mL) at room temperature, 5 mg (0.14 mmol) of
NaBH4 were added. The reaction mixture was stirred for 24
h at room temperature. The final solution was acidified with
2 mL HCl 10 % (v/v). The precipitates formed were col-
lected by filtration, washed with a saturated aqueous
NaHCO3 solution and water. The residue was purified by
column chromatography on silica gel using dichlor-
omethane/methanol (9:1) as the eluent to afford the corre-
sponding product.
N9-(3-chlorophenyl)-N2-(2-(piperidin-1-yl)ethyl)thiazolo
[5,4-b]quinoline-2,9-diamine (15e)
Yellow light solid (153 mg, 58.5 %), mp 95–96 °C; IR
(KBr) cm−1: 3213 (–NH–arom.), 3058 (–NHR), 2933 (–CH
insatd.), 2852, 2804 (–CH satd.), 1594, 1666 (–C=N),
N-phenylthiazolo[5,4-b]quinolin-9-amine (16a)
1
1560, 1477, 1441 (arom.), 855 (–C–Cl), 759 (–C–S); H
NMR (300 MHz, DMSO-d6): δ = 8.84 (1H, s, ArN–H),
8.41 (1H, br, aliph. N–H), 8.11 (1H, d, J = 6 Hz, H-5),
7.875 (1H, d, J = 6 Hz, H-8), 7.58 (1H, 1H, J = 6.9 Hz, H-
6), 7.7 (1H, t, J = 6.9 Hz, H-7), 7.14 (1H, t, J = 9 Hz, H-6′),
6.70–6.79 (3H, m, H-2′, H-3′, H-4′), 2.49 (2H, br, NH–CH2
CH2N(CH2CH2)2CH2), 2.40 (2H, br, NH–CH2CH2N(CH2
CH2)2CH2), 2.27 (4H, br, NH–CH2CH2N(CH2CH2)2CH2),
1.44 (4H, br, NH–CH2CH2N(CH2CH2)2CH2), 1.35 (2H, br,
NH–CH2CH2N(CH2CH2)2CH2). HRMS (FAB) calcd for
[MH]+ C23H25N5ClS: 438.152, found 438.1522; MS (FAB)
m/z: 438 (MH+, 88 %), 440 (MH++2, 31 %, 37Cl and 34S),
437 (M+, 16 %), 436 (MH+−2, 15 %), 435 (M+−2, 4 %),
326 (M+−111, 6 %), 111 (M+−326, 50 %), 98 (M+−339,
100 %).
Yellowish solid (130 mg, 75 %), mp 135 °C; IR (KBr)
cm−1: 3413 (–NH–aromatic), 3212–2924 (–CH insatd.),
1710, 1536, 1577 (aromatic), 1495, 756 (C–S). No bands at
1
1317 and 1146, typical for sulfonyl group. H NMR (300
MHz, DMSO-d6): δ = 9.46 (1H, s, ArN–H), 9.20 (1H, s, H-
2), 8.34 (1H, dd, J = 8.3, 0.8 Hz, H-8), 7.98 (1H, ddd, J =
8.7, 1.4, 0.6 Hz, H-5), 7.76 (1H, ddd, J = 8.4, 6.7, 1.4 Hz,
H-6), 7.50 (1H, ddd, J = 8.5, 6.7, 1.3 Hz, H-7), 7.25-7.22
(2H, m, H, 3′, H-5′), 7.10–7.04 (2H, m, H-2′, H-6′),
7.04–6.96 (1H, m, H-4′). HRMS (ESI) calcd for [M]+
C16H11N3S: 277.07, found: 277.0672; MS (FAB) m/z: 277
(MH+, 100 %).
N-(3-chlorophenyl)thiazolo[5,4-b]quinolin-9-amine (16b)
N9-(4-chlorophenyl)-N2-(2-(piperidin-1-yl)ethyl)thiazolo
[5,4-b]quinoline-2,9-diamine (15f)
Yellow light solid (140 mg, 79 %), mp 160 °C; IR
(KBr) cm−1: 3402, 3354 (–NH aromatic), 3045, 1589, 1517,
1
1431, 830, 761; H NMR (300 MHZ, DMSO-d6): δ = 9.62
(1H, s, ArN–H), 9.2 (1H, s, H-2), 8.45 (1H, d, J = 8.4 Hz,
H-8), 7.94 (1H, dd, J = 8.4, 0.9 Hz, H-5), 7.79 (1H, ddd,
J = 8.1, 6.6, 1.2 Hz, H-6), 7.58 (1H, ddd, J = 8.1, 6.6,
2.1 Hz, H-7), 7.3 (1H, dd, J = 8.1, 8.1 Hz, H-5′), 7.18 (1H,
dd, J = 2.1, 2.1 Hz, H-2′), 7.06 (1H, m, H-6′), 7.09 (1H, m,
H-4′). HRMS (ESI) calcd for [M]+ C16H10N3ClS:
311.03, found: 311.0278; MS (FAB) m/z: 311 (MH+,
100 %).
Yellow light solid (148 mg, 67.6 %), mp 140–141 °C; IR
(KBr) cm−1: 3373 (–NH–arom.), 3057 (–NH aliphat.), 2928
(–CH insatd.), 2851, 2797 (–CH satd.), 1694, 1605 (–C=N),
1
1557, 1493, 1469 (arom.), 855 (–C–Cl), 759 (–C–S); H
NMR (300 MHz, DMSO-d6): δ = 8.8 (1H, s, ArN–H), 8.38
(1H, t, J = 6 Hz, aliph. N–H), 8.15 (1H, d, J = 9 Hz, H-5),
7.9 (1H, d, J = 9 Hz, H-8), 7.6 (1H, ddd, J = 1.2, 7.2, 8.1
Hz, H-7), 7.49 (1H, ddd, J = 1.2, 6.9, 8.1 Hz, H-6), 7.25
(2H, d, J = 9 Hz, H-3′, H-5′), 6.8 (2H, d, J = 9 Hz, H-2′,
H-6′), 2.53 (2H, q, J = 6.3 Hz, NH–CH2CH2N(CH2
CH2)2CH2), 2.39 (2H, t, J = 6 Hz, NH–CH2CH2N(CH2
CH2)2CH2), 2.28 (4H, br, NH–CH2CH2N(CH2CH2)2CH2),
1.475 (4H, br, NH–CH2CH2N(CH2CH2)2CH2), 1.39 (2H,
br, NH–CH2CH2N(CH2CH2)2CH2). HRMS (FAB) calcd
for [MH]+ C23H25N5ClS: 438.15, found 438.1522; MS
(FAB) m/z: 438 (MH+, 100 %), 440 (MH++2, 38 %, 37Cl
and 34S), 437 (M+, 18 %), 326 (M+−111, 6 %), 111
(M+−326, 40 %), 98 (M+−339, 88 %).
N-(4-chlorophenyl)thiazolo[5,4-b]quinolin-9-amine (16c)
Yellow solid, (135 mg, 76.8 %), mp 180 °C; IR (KBr) cm−1:
3399 (–NH arom.), 3051–2992 (–CH insatd.), 1580, 1513,
826 (C–Cl), 753; 1H NMR (300 MHz, DMSO-d6): δ = 9.54
(1H, s, ArN–H), 9.22 (1H, s, H-2), 8.68 (1H, d, J = 8.4 Hz,
H-8), 7.94 (1H, d, J = 8.4 Hz, H-5), 7.91 (1H, dd, J = 7.9
Hz, H-6), 7.6 (1H, dd, J = 7.6 Hz, H-7), 7.4 (2H, d, J = 8.8
Hz, H-3′, H-5′), 7.3 (2H, d, J = 8.4 Hz, H-2′,H-6′). HRMS

Med Chem Res
(ESI) calcd for [M]+ C16H10N3ClS: 311.03, found:
(300 MHz, DMSO-d6): δ = 9.6 (1H, s, Ar–NH–), 9.3 (1H, s,
H-2), 8.5 (1H, d, J = 8.4 Hz, H-8), 8.0 (1H, d, J = 8.4 Hz,
H-5), 7.8 (1H, dd, J = 7.2 Hz, H-6), 7.7 (2H, d, J = 8.8 Hz,
H-3′, H-5′), 7.6 (1H, dd, J = 7.2 Hz, H-7), 7.2 (2H, d, J =
8.4 Hz, H-2′, H-6′). HRMS (ESI) calcd for [M]+
C17H10N4S: 302.06, found: 302.0619; MS (EI) 302 M+.
311.0277; MS (FAB) m/z: 311 (MH+, 100 %).
N-(3-methoxyphenyl)thiazolo[5,4-b]quinolin-9-amine (16d)
Yellow solid (137 mg, 72 %), mp 151 °C; IR (KBr) cm−1:
3120 (–NH arom.), 3053–3013 (–CH insatd.), 2966 1595,
1571, 1543, 1503, 1476 (arom), 1266 (Ar–O), 764; 1H
NMR (300 MHz, DMSO-d6): δ = 9.4 (1H, s, –NH–Ar),
9.22 (1H, s, H-2), 8.29 (1H, d, J = 8.4 Hz, H-8), 7.97 (1H,
dd, J = 8.7, 0.9 Hz, H-5), 7.75 (1H, ddd, J = 8.4, 6.9, 1.2
Hz, H-6), 7.49 (1H, ddd, J = 8.4, 6.9, 1.2 Hz, H-7), 7.11
(1H, t, J = 8.7 Hz, H-5′), 6.60–6.58 3H, (m, 3H, H-2′, H-4′,
H-6′), 3.65 (3H, s, –OCH3). 13C NMR (75 MHz, DMSO-
d6): δ 160.7 (C-2), 141.1 (C-3a), 144.7 (C-4a), 128.7 (C-5),
130.2 (C-6), 124.7 (C-7), 124.4 (C-8), 119.5 (C-8a), 160
(C-9), 133.1 (C-9a), 148.1 (C-1′), 106.5 (C-2′), 153.2 (C-
3′), 108.2 (C-4′), 129.5 (C-5′), 113.1 (C-6′), 55.3, (–OCH3).
HRMS (ESI) calcd for [M]+ C17H13ON3S: 307.08, found:
307.078; MS (EI) m/z: 307 (M+, 100 %), 292 (M+−15,
78 %).
N-(3-methylphenyl)thiazolo[5,4-b]quinolin-9-amine (16h)
Yellow solid, (150 mg, 79 %), mp 132 °C; IR (KBr) cm−1:
3120 (NH arom.), 3068–2911 (–CH insatd.), 1595, 1571,
1
1543, 1503, 1476 (arom.), 753 (C-S); H NMR (300 MHz,
DMSO-d6): δ = 9.35 (1H, s,–NH–Ar), 9.18 (1H, s, H-2),
8.29 (1H, d, J = 8.4 Hz, H-8), 7.95 (1H, dd, J = 8.7, 0.9 Hz,
H-5), 7.73 (1H, ddd, J = 8.4, 6.9, 1.2 Hz, H-6), 7.47 (1H,
ddd, J = 8.4, 6.9, 1.2 Hz, H-7), 7.1 (1H, t, J = 8.6 Hz, H-5′),
6.86 (3H, m, H-2′, H-4′, H-6′), 2.21 (3H, s, Ar–CH3).
HRMS (ESI) calcd for [M]+ C17H13N3S: 291.03,
found: 291.0813; MS (EI) m/z: 291 (M.+, 100 %), 290
(M+−1, 75 %).
N-(4-methylphenyl)thiazolo[5,4-b]quinolin-9-amine (16i)
N-(4-methoxyphenyl)thiazolo[5,4-b]quinolin-9-amine (16e)
Yellow solid (145 mg, 76 %) mp 196 °C; IR (KBr) cm−1:
3120 (NH arom.), 3048–2941 (–CH arom.), 1595, 1571,
1543, 1503, 1476 (arom.), 739 (C–S); 1H NMR (300 MHz,
DMSO-d6): δ = 9.34 (1H, s, –NH–Ar), 9.14 (1H, s, H-2),
8.3 (1H, d, J = 8.4 Hz, H-8), 7.93 (1H, dd, J = 8.7, 0.9 Hz,
H-5), 7.72 (1H, ddd, J = 8.4, 6.9, 1.2 Hz, H-6), 7.45 (1H,
ddd, J = 8.4, 6.9, 1.2 Hz, H-7), 7.05 (2H, d, J = 8.6 Hz, H-
2′, H-6′), 6.97 (2H, d, J = 8.6 Hz, H-3′, H-5′), 2.26 (3H, s,
Ar–CH3). HRMS (ESI) calcd for [M]+ C17H13N3S: 291.03,
found: 291.0825; MS (EI) m/z: 291(M.+, 100 %), 290
(M+−1, 75 %).
Yellow solid (140 mg, 70 %), mp 147; IR (KBr) cm−1: 3120
(NH– arom.), 3053–2966 (–CH insatd.), 1595, 1571, 1543,
1503, 1476 (arom.), 1244 (Ar–O), 1041 (Ar–O–CH3), 752
(C–S); 1H NMR (300 MHz, DMSO-d6): δ = 9.29 (1H,
s,–NH–Ar), 9.07 (1H, s, H-2), 8.31 (1H, d, J = 8.4 Hz, H-8),
7.9 (1H, dd, J = 8.4 Hz, H-5), 7.7 (1H, ddd, J = 7.6 Hz, H-
6); 7.42 (1H, dd, J = 7.6 Hz, H-7), 7.06 (1H, d, J = 8.4 Hz,
H-2′), 6.85 (3H, d, J = 8.4 Hz, H-3′, H-5′, H-6′), 3.73 (3H,
s, –OCH3). HRMS (ESI) calcd for [M]+ C17H13ON3S:
307.08, found: 307.0772; MS (EI) m/z: 307(M.+, 100 %),
292 (M+−15, 80 %).
Biological
N-(3-cyanophenyl)thiazolo[5,4-b]quinolin-9-amine (16f)
Cytotoxic assay
Yellow solid (135 mg, 77.3 %), mp 160 °C; IR (KBr)
cm−1: 3329 (–NH arom.), 3056–2923 (–CH insat.), 2231
(–CN), 1578, 1548, 1533 (arom.), 757 (C–S); H NMR
The effects of the compounds were determined in one
cervical cell line (HeLa), two human colorectal cancer cell
lines (SW480 and SW620) and one myelogenous leukemia
human cell line (K-562). The cytotoxic assays were carried
out according to the microculture MTT method (Quintero
et al., 1999). The cells were harvested at 4.5–5.0 × 104
cells/mL/well and inoculated on 24 well microtiter plates.
Then the culture cells were inoculated alone and with the
compounds (which were dissolved in DMSO and added in a
maximum volume of 2 mL/mL/well). After 72 h incubation,
100 mg/mL of MTT (in PBS, pH 7.2) were added. By
adding 1 mL of DMSO to each well, followed by gentle
shaking, the formazan dye was dissolved. After cen-
trifugation the extinction coefficient was measured at 540
1
(300 MHz, DMSO-d6): δ = 9.7 (1H, s, ArN–H), 9.3 (1H, s,
H-2), 8.64 (1H, d, J = 8.7 Hz, H-8), 8.0 (1H, dd, J = 8.7,
0.9 Hz, H-5), 7.89 (1H, ddd, J = 8.4, 6.9, 1.2 Hz, H-6), 7.68
(1H, ddd, J = 8.4, 6.9, 1.2 Hz, H-7), 7.56–7.46 (4H, m, H-
2′, H-4′, H-5′, H-6′). HRMS (ESI) calcd for [M]+
C17H10N4S: 302.06, found: 302.0615; MS (EI) 302 M+.
N-(4-cyanophenyl)thiazolo[5,4-b]quinolin-9-amine (16g)
Yellow solid (142 mg, 80 %), mp 162 °C; IR (KBr) cm−1:
3320 (–NH–arom.), 3055–2923 (–CH insatd.), 2219 (–CN),
1
1607, 1578, 1524, 1405 (arom.), 753 (–C–S); H NMR

Med Chem Res
nm using a Beckman photometer model DUR-64. Cell
growth inhibition was determined by the formula % cell
growth inhibition = (1 − absorbance of treated cells/absor-
bance of untreated cells) × 100. The assays were carried out
in three independent experiments in quadruplicate.
based on the lowest binding free energy and the largest
cluster size.
Ligands equilibrium geometries. All calculations were
performed with SPARTAN'08® software (2008, WAVE-
FUNCTION, Inc., Irvine, CA). Molecules were built by
assembling standard fragments and the resulting geometries
were optimized by molecular mechanics. Conformational
analysis of the compounds by Systematic Search protocol
around rotable bonds was performed using the MMFF94
force field. The most frequent conformer for each com-
pound was selected and geometry optimization was carried
out with semiempirical AM1 method. Due to the basic
properties of the tertiary amino side chain at 2-position, the
protonated form of the compounds with this substituent was
used in the docking studies and neutral form was also used.
DNA affinity and intercalation
DNA intercalation was determined from the displacement
of ethidium bromide from DNA (Loza-Mejía et al., 2008).
Sterile solutions of high molecular weight DNA from calf
thymus (Gibco, BRL, New York, USA) in a 0.1 M Tris-HCl
buffer at pH 7.4, 0.15M NaCl and 5 mM ethidium bromide
(ultrapure from Gibco, BRL, New York, USA) were mixed
with serial additions of the compounds to be tested dis-
solved in 100 % dimethylsulphoxide (DMSO); the fluores-
cence intensity of the solution was recorded at 584 nm with
an excitation light of 546 nm. The DMSO concentration
never exceeded 8 %. The effect of this amount of DMSO
was small and had no effect on the shape of emission or
excitation fluorescence spectra of a DNA-ethidium bromide
complex as compared with that determined in 100 % aqu-
eous buffers. The recorded fluorescence change was cor-
rected for the dilution caused by serial additions of this
solvent. The concentration of the compounds tested varied
in the range of 1–100 μM depending on their respective
solubility. The precipitation of the compound from the
solution was detected from an increase in 600 nm light
dispersal at 90° in a Shimadzu RF5000U fluorescence
spectrophotometer. The displacement curves were fitted by
non-linear regression analysis to a rectangular hyperbola.
Semiempirical LMO calculations of DNA–DNA intercalator
complexes
The representative geometry of selected DNA–DNA-inter-
calator complexes, obtained from the molecular docking
predictions with the best score belonging to the largest
cluster, was edited to add all missing hydrogen atoms with
the phosphate groups unprotonated. MOPAC 2012 (Stew-
art, 2012; Maia et al., 2012) was used to improve the
molecular geometry using the semiempirical PM7 method
and the Localized Molecular Orbital theory as implemented
in the MOPAC MOZYME code. Implicit solvation was set
using the conductor-like screening model (COSMO) with a
dielectrical constant of 78.4 in a non-periodical box and in
the absence of other explicit atoms, such as sodium ions.
Geometry optimization was run in two steps, first heavy-
atoms were fixed and hydrogen positions were optimized to
a gradient of 20.0 kcal mol−1 Å−1. When required, anom-
alous hydrogen positions were corrected and the step was
repeated. When a chemically consistent geometry for all
hydrogen atoms was obtained a free optimization of all
atoms was run to a gradient of 10.0 kcal mol−1 Å−1 (a
recommended target for large systems). Analysis of mole-
cular graphics was done using Jmol v. 13.0. (http://www.
jmol.org/).
Molecular docking
The crystal structure of an 9-acridine-peptide drug in
complex with a DNA dodecamer was downloaded from
Protein Data Bank (PDB ID: 1G3X) (Malinina et al., 2002)
and edited using PyMOL v.1.4.1 software (http://www.
pymol.org), after edition the structure was minimizing with
WHAT IF: A molecular modeling and drug design program
(http://swift.cmbi.ru.nl/whatif/) (Vriend, 1990) and ready
for docking studies. All docking studies were performed
with AutoDock 4.2. software (Morris et al., 1996, 1998)
employing the Lamarkian Genetic Algorithm, generating 20
independent docking poses for each compound. In all the
cases the population size was set to 150 and the maximal
number of evaluations was set to 5,000,000. The position of
the docking grid was centered at the position of the original
co-crystallized ligand which was removed. The dimension
of the grid was 100 Å × 100 Å × 100 Å points with spacing
of 0.375 between the grid points. The DNA was considered
as rigid molecule, while the ligands were considered as
flexible molecules. The best binding mode was selected
Acknowledgments We thank Maricela Gutiérrez, Rosa Isela del
Villar, Victor M. Arroyo, Georgina Duarte and Margarita Guzmán for
determination of all spectra. A.K.L.R. thanks DGAPA-UNAM for a
scholarship. We also thank DGAPA-UNAM for financing project
PAPIIT IN2128910 and PAPIIT IN221113, as well as Facultad de
Química for financial support (PAIP 6390-10, 6290-09), and UNAM-
DGTIC-SC16-1-IR-111.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interests.

Med Chem Res
novel thiazolo[5,4-b]quinoline derivatives. Bioorg Med Chem
16:1142–1149
References
Loza-Mejía MA, Olvera-Vázquez S, Maldonado-Hernández K,
Guadarrama-Salgado K, González-Sánchez I, Rodríguez-
Hernández F, Solano JD, Rodríguez-Sotres R, Lira-Rocha A
(2009) Synthesis, cytotoxic activity, DNA topoisomerase-II
inhibition, molecular modeling and structure–activity relation-
ship of 9-anilinothiazolo[5,4-b]quinoline derivatives. Bioorg Med
Chem 17:3266–3277
Maia JDC, Carvalho GAU, Mangueira CP, Santana SR, Cabral LAF,
Rocha GB (2012) GPU linear algebra libraries and GPGPU
programming for accelerating MOPAC semiempirical quantum
chemistry calculations. J Chem Theory Comput 8:3072–3081
Malinina L, Soler-López M, Aymamí J, Subirana JA (2002) Inter-
calation of an acridine-peptide drug in an AA/TT base step in the
crystal structure of [d(CGCGAATTCGCG)](2) with six duplexes
and seven Mg(2+) ions in the asymmetric unit. Biochemistry
41:9341–9348
Medhi C, Mitchell JBO, Price SL, Tabor AB (1999) Electrostatic
factors in DNA intercalation. Biopolymers 52:84–93
McConnaughie WA, Jenkins TC (1995) Novel acridine-triazenes as
prototype combilexins: synthesis, DNA binding, and biological
activity. J Med Chem 38:3488–3501
Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK,
Olson AJ (1998) Automated docking using a Lamarckian genetic
algorithm and an empirical binding free energy function. J
Comput Chem 19:1639–1662
Morris GM, Goodsell DS, Huey R, Olson AJ (1996) Distributed
automated docking of flexible ligands to proteins: parallel
applications of AutoDock 2.4. J Comput Aided Mol Des
10:293–304
Quintero A, Pelcastre A, Dolores J, Guzmán E, Díaz E (1999) Anti-
tumoral activity of new pyrimidine derivatives of sesquiterpene
lactones. J Pharm Pharm Sci 2:108–112
Rodríguez-Loaiza P, Quintero A, Rodríguez-Sotres R, Solano JD,
Lira-Rocha A (2004) Synthesis and evaluation of 9-anilinothia-
zolo[5,4-b]quinoline derivatives as potential antitumorals. Eur J
Med Chem 39:5–10
Sato K, Hyodo M, Aoki M, Zheng X-Q, Noyori R (2001) Oxidation of
sulfides to sulfoxides and sulfones with 30 % hydrogen peroxide
under organic solvent- and halogen-free conditions. Tetrahedron
57:2469–2476
Baldwin JJ, Engelhardt EL, Hirschmann R, Ponticello GS, Atkinson
JG, Wasson BK, Sweet CS, Scriabine A (1980) Heterocyclic
analogues of the antihypertensive beta-adrenergic blocking agent
(S)-2-[3-(ter-butylamino)-2-hydroxypropoxy]-3-cyanopyridine. J
Med Chem 23:65–70
Bondarev DA, Skawinski WJ, Venanzi CA (2000) Nature of inter-
calator amiloride-nucelobase stacking. an empirical potential
and ab initio electron correlation study. J Phys Chem B
104:815–822
Chen Y-L, Chen I-L, Lu C-M, Tzeng C-C, Tsao L-T, Wang J-P (2004)
Synthesis and anti-inflammatory evaluation of 4-anilinofuro[2,3-
b]quinoline and 4-phenoxyfuro[2,3-b]quinoline derivatives.
Bioorg Med Chem 12:387–392
Chen Y-L, Chen I-L, Wang T-C, Han C-H, Tzeng C-C (2005)
Synthesis and anticancer evaluation of certain 4-anilinofuro[2,3-
b]quinoline and 4-anilinofuro[3,2-c]quinoline derivatives. Eur J
Med Chem 40:928–934
Demecunynck H, Charmantray F, Martelli A (2001) Interest of acri-
dine derivatives in the anticancer chemotherapy. Curr Pharm Des
7:1703–1724
Denny WA (1995) The role of intercalation and DNA binding in the
activity of acridines and related tricyclic DNA-intercalating
agents. In: Foye WO (ed) Cancer chemotherapeutic agents.
ACS, Washington, pp 218–239
Denny WA (2002) Acridine derivatives as chemotherapeutic agents.
Curr Med Chem 9:1655–1665
González-Sánchez I, Solano JD, Loza-Mejía MA, Olvera-Vázquez S,
Rodríguez-Sotres R, Morán J, Lira-Rocha A, Cerbon M (2011)
Antineoplastic activity of the thiazolo[5,4-b]quinoline derivative
D3CLP in K-562 cells is mediated through effector caspases
activation. Eur J Med Chem 46:2102–2108
Gunawardana GP, Kohmoto S, Gunasekera SP, McConnell OJ, Koehn
FE (1988) Dercitin, a new biologically active acridine alkaloid
from a deep water marine sponge, Dercitus sp. J Am Chem Soc
110:4856–4860
He Y, Ma X, Ji HF, Zha XB, Jiang H, Lu M (2012) Selective oxidation
of sulfides to sulfoxides/sulfones by 30 % hydrogen peroxide.
Phosphorus Sulfur Silicon Relat Elem 187:822–830
Herrera A, Martínez-Alvarez R, Ramiro P, Molero D, Almy J (2006)
New easy approach to the synthesis of 2,5-disubstituted and
2,4,5-trisubstituted 1,3-oxazoles. The reaction of 1-(methylthio)
acetone with nitriles . Org Chem 71:3026–3032
Hoelder S, Clarke PA, Workman P (2012) Discovery of small mole-
cule cancer drugs: successes, challenges and opportunities. Mol
Oncol 6:155–176
Kerdesky FA, Seif LS (1995) A novel and efficient synthesis
of 5-(hydroxymethyl) thiazole: an important synthon for
preparing biologically active compounds. Synth Commun 25:
2639–2645
Lindsey RH, Pendleton M, Ashley RE, Mercer SL, Deweese JE,
Osheroff N (2014) Catalytic core of human topoisomerase IIα:
insights into enzyme-DNA interactions and drug mechanism.
Biochemistry 53:6595–6602
Stewart JJP MOPAC (2012) Stewart Computational Chemistry, Ver-
sion 15.140L web: http://OpenMopac.net
Siu F-M, Pommier Y (2013) Sequence selectivity of the cleavage sites
induced by topoisomerase I inhibitors: a molecular dynamics
study. Nucleic Acid Res 41:10010–10019
Vriend G (1990) WHAT IF: a molecular modeling and drug design
program. J Mol Graph 8:52–56
Xiao X, Antony S, Pommier Y, Cushman M (2005) On the binding of
indeno[1,2-c]isoquinolines in the DNA-topoisomerase I cleavage
complex. J Med Chem 48:3231–3238
Yin G, Wang Z, Chen A, Gao M, Wu A, Pan Y (2008) A new facile
approach to the synthesis of 3-methylthio-substituted furans,
pyrroles, thiophenes, and related derivatives. J Org Chem
73:3377–3383
Zumbrunn A (1998) The first versatile synthesis of 1-alkyl-3-fluoro-
1H-[1,2,4]triazoles. Synthesis 9:1357–1361
Loza-Mejía MA, Maldonado-Hernández K, Rodríguez-Hernández F,
Rodríguez-Sotres R, González-Sánchez I, Solano JD, Lira-Rocha
A (2008) Synthesis, cytotoxic evaluation, and DNA binding of