Synthesis, and cytotoxic activity of some novel indolo[2,3-b]quinoline derivatives: DNA topoisomerase II inhibitors

Article abstract of DOI:10.1016/S0968-0896(99)00200-X

A series of new 5H-indolo[2,3-b]quinoline derivatives bearing methoxy and methyl groups at C-2 and C-9 was synthesized (according to the modified Graebe-Ullmann reaction). These compounds were evaluated for their antimicrobial and cytotoxic activity and tested as inhibitors of DNA topoisomerase II. Lipophilic and calf thymus DNA binding properties of these compounds were also established. In the SAR studies we used quantum- mechanical methodology to analyze the molecular properties of the drugs. All of the 5H-indolo[2,3-b]quinolines tested were found to inhibit the growth of Gram-positive bacteria and pathogenic fungi at MIC ranging between 2.0 and 6.0 μM. They showed also cytotoxic activity in vitro against several human cancer cell lines of different origin (ID₅₀ varied from 0.6 to 1.4 μM), and stimulated the formation of topoisomerase-II-mediated pSP65 DNA cleavage at concentration between 0.2 and 0.5 μM. The most active indolo[2,3- b]quinolines which had the greatest contribution to the increase in the T(m) of DNA displayed also the highest DNA binding constants and the highest cytotoxic activity. The differences in DNA binding properties and cytotoxic activity seem to be more related to steric than electrostatic effects. (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00200-X

Bioorganic & Medicinal Chemistry 7 (1999) 2457±2464
Synthesis, and Cytotoxic Activity of Some Novel
Indolo[2,3-b]quinoline Derivatives:
DNA Topoisomerase II Inhibitors
—ukasz Kaczmarek, a Wanda Peczynska-Czoch, b,* Jaros•aw Osiadacz, c
Marian Mordarski, c W. Andrzej Sokalski, d Janusz Boratynski, e
Ewa Marcinkowska, e Halina Glazman-Kusnierczyk, e
and Czes•aw Radzikowski e
aPharmaceutical Research Institute, Rydygiera 8, 01-783 Warszawa, Poland
bInstitute of Organic Chemistry, Biochemistry & Biotechnology, Wroc•aw University of Technology,
Wybrzeze Wyspianskiego 27, 50-370 Wroc•aw, Poland
cDepartment of Microbiology, Institute of Immunology and Experimental Therapy, Polish Academy of Sciences,
Weigla 12, 53-114 Wroc•aw, Poland
dInstitute of Theoretical and Physical Chemistry, Wroc•aw University of Technology,
Wybrzeze Wyspianskiego 27, 50-370 Wroc•aw, Poland
eDepartment of Tumor Immunology, Institute of Immunology and Experimental Therapy,
Polish Academy of Sciences, Weigla 12, 53-114 Wroc•aw, Poland
Received 16 March 1999; accepted 26 May 1999
AbstractÐA series of new 5H-indolo[2,3-b]quinoline derivatives bearing methoxy and methyl groups at C-2 and C-9 was synthesized
(according to the modi®ed Graebe±Ullmann reaction). These compounds were evaluated for their antimicrobial and cytotoxic activity
and tested as inhibitors of DNA topoisomerase II. Lipophilic and calf thymus DNA binding properties of these compounds were also
established. In the SAR studies we used quantum-mechanical methodology to analyze the molecular properties of the drugs. All of the
5H-indolo[2,3-b]quinolines tested were found to inhibit the growth of Gram-positive bacteria and pathogenic fungi at MIC ranging
between 2.0 and 6.0 mM. They showed also cytotoxic activity in vitro against several human cancer cell lines of di€erent origin (ID50
varied from 0.6 to 1.4 mM), and stimulated the formation of topoisomerase-II-mediated pSP65 DNA cleavage at concentration between
0.2 and 0.5 mM. The most active indolo[2,3-b]quinolines which had the greatest contribution to the increase in the Tm of DNA displayed
also the highest DNA binding constants and the highest cytotoxic activity. The di€erences in DNA binding properties and cytotoxic
activity seem to be more related to steric than electrostatic e€ects. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
these compounds. We found that only derivatives
belonging to the 5H-series (i.e. those bearing a methyl
group on the pyridine nitrogen, which stabilizes the
positive charge of the molecule) showed signi®cant
activity against prokaryotic and eukaryotic organisms,
increased calf thymus DNA denaturation temperature
(Tm) and stimulated the formation of calf thymus
topoisomerase II mediated DNA cleavage.3
In our previous papers we reported the synthesis and
preliminary evaluation of indolo[2,3-b]quinoline deriva-
tives which appeared to be a new class of cytotoxic
DNA intercalators and topoisomerase II inhibitors.1,2
Owing to the novelty of the indolo[2,3-b]quinoline core,
a synthetic program was undertaken to explore the
structure±activity relationship with the objective to
increase the potency and speci®city of the action of
In our former SAR study we found that none of the
analogous indolo[2,3-b]quinolines belonging to the 6H-
series (i.e. those lacking in a methyl group on the pyri-
dine nitrogen) was active in appropriate tests.3 As part
of our structure±activity program, which aimed at
obtaining novel indolo[2,3-b]quinolines with improved
Key words: Indolo[2,3-b]quinolines; cytotoxic agents; topoisomerase
II inhibitors; quantum chemical calculations.
* Corresponding author. Tel.: +4871-320-24-59; e-mail: peczyns-
ka@kchf.ch.pwr.wroc.pl
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00200-X

2458
—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
therapeutic potential, we synthesized the methoxy and
methyl-methoxy derivatives of our lead compound: 5-
11-dimethyl-5H-indolo[2,3-b]quinoline (DiMIQ, com-
pound 8) and evaluated them as cytotoxic agents in
vitro and in vivo. The same derivatives were also inves-
tigated as topoisomerase II inhibitors. With the aim to
recognize the contribution of the methoxy and methyl
groups at C-2 and/or C-9 to the activity of 5H-
indolo[2,3-b]quinolines, we selected three sets of com-
pounds for preparation: (i) substituted with a methoxy
group at C-2; (ii) bearing this substituent at C-9; (iii)
disubstituted at C-2 and C-9 with methoxy, methyl and
methoxy or methoxy and methyl groups, respectively.
The compounds obtained were evaluated for their anti-
microbial properties and cytotoxicity against several
human cell lines in vitro. Investigated was also the abil-
ity of these compounds to induce in vitro the formation
of calf thymus DNA topoisomerase II mediated clea-
vable complexes. Additionally, we determined the lipo-
philicity, pKa values, DNA binding constant (Kapp)4 and
thermal stability of the indolo[2,3-b]quinoline-DNA
complex (ÁTm). In the SAR study we used the quan-
tum-mechanical methodology to analyze the geometry
and the electrostatic properties of the drugs. The geo-
metry of the investigated compounds was optimized,
using both ab initio LCAO MO SCF (6-31G and 6-
31GÃ basis sets), and semi-empirical (AM1 and PM3)
methods. We also calculated the distribution of the
molecular electrostatic potentials (MEPs) on the Van
der Waals surfaces in order to correlate it with some
measurable properties of the compounds.
The values of the hydrophobic parameter (logP) corro-
borate this ®nding. Thus, the logP values for the com-
pounds examined, which were positive at pH 5 and 9,
became negative at pH 2.2 (Fig. 1). As we expected, it
was the dimethoxy derivative (compound 7d) that
showed the highest hydrophilicity.
Like the previously tested methyl-5H-indolo[2,3-b]qui-
nolines, compounds 7a±e were found to display toxicity
against prokaryotic and eukaryotic cells. Table 2 sum-
marizes the antimicrobial activity of the compounds,
expressed as minimum inhibitory concentration values
(MICs) against the representatives of Gram-negative and
Gram-positive bacteria and pathogenic fungi. All deriva-
tives showed signi®cant activity against Gram-positive
bacteria and fungi in the MIC range 0.015 to 0.12 mM. No
antimicrobial activity against Gram-negative bacteria
was observed for these compounds.
The cytotoxic properties of compounds 7a±e were deter-
mined in vitro against several human malignant lines: HL-
60 promyelotic leukemia, T-470 breast carcinoma, SW-
707 colon adenocarcinoma, A-599 lung carcinoma, A-431
epidermal carcinoma, and BM-melanoma. All the com-
pounds were found to be cytotoxic, ED50 ranged from 0.2
to 2.48 mM. The most sensitive lines were promyelocytic
leukemia HL-60 and colon adenocarcinoma SW-707. The
least active compounds were our lead compound DiMIQ
8, and ellipticine 10 (Table 3), whereas the most cytotoxic
were compounds 7c, 7d and 9 (compound 9, which had
been selected as the most active against the KB cell line in
our previous study,3 was examined for comparison).
The di€erences in the DNA binding properties between
the compounds referred to as 7a±e were estimated by
examining the increase of calf thymus DNA denatura-
tion temperature (ÁTm) in their presence and by deter-
mining the binding constants (Kapp). Measurements of
DNA denaturation temperature were carried out at a
drug: DNA ratio of 1:10 in 5 mM Tris±HCl bu€er. The
ÁTm values ranged from 12 to 19ꢀC. For DiMIQ 8, 2,9-
dimethyl-DiMIQ 9, and ellipticine 10 (used as control)
the ÁTm values were found to be 11, 19, and 9.8ꢀC,
respectively (Table 4).
Results and Discussion
The 5H-indolo[2,3-b]quinoline synthetic pathways are
outlined in Scheme 1. 6H-Indolo[2,3-b]quinolines 6 were
synthesized by decomposition of corresponding triazoles
5 in polyphosphoric acid (PPA) at 110±160ꢀC and then
transformed into 5-methyl-5H-indolo[2,3-b]quinolines 7
by quarternization with dimethyl sulfate followed by
alkalization of quarternary salts with 20% NaOH. The
starting triazoles 5 were prepared either by heating with
corresponding chloroquinoline 1 and with benzotriazole
2 at 110±120ꢀC (5d) or by condensation (5a±c) sub-
stituted in benzotriazole ring) of 1 with o-nitroaniline
derivatives 3 and by subsequent reduction of the nitro
group in intermediate 4 followed by diazotization of the
amines obtained. Compounds 8 and 9 have been descri-
bed in literature a number of times (Schemes 1 and 2).
We found that the binding of indolo[2,3-b]quinolines to
calf thymus DNA was accompanied by variations in the
UV spectra according to the pattern reported pre-
viously.3 And this ®nding allowed for the determination
of the r values (mol of drug bound per mol of nucleo-
tide) and of the c values (free drug concentration). The
Scatchard plots were analyzed according to the McGhee
and Von Hippel4 model by nonlinear ®tting of the data
according to the equation r=c ˆ Kapp à ꢁ1 n à r† à ‰ꢁ1
Our earlier investigations into the 5H-indolo[2,3-
b]quinoline group indicated that all the compounds
belonging to this series were partially protonated under
physiological conditions.3,5 Therefore the values of the
dissociation constants for the novel compounds 7a±e
were deduced from their UV spectra at ®xed wavelength
taken in bu€ers at pH 2.2 to 8.3. The pKa values ranged
from 6.02 (for the C-9 mono methoxy-derivative) to
8.07 (for the dimethoxy-derivative) with an isosbestic
point at 276±293 nm (Table 1). This means that, at low
pH, 5H-indolo[2,3-b]quinolines occur in the form of
salts and become easy soluble in water.
n
1
n à r†=ꢁ1 ꢁn 1† à r†Š
in order to obtain the binding
constant to an isolated site and apparent site size. The
DNA binding parameters were found to range from
2.32Â106 M (for the binding constant Kapp) and 2.28
1
(for the number of nucleotides/binding site (n) for C-2
M
methoxy substituted compound 7c to 63.2Â106
1
(Kapp) and 2.42 for the n values for dimethoxy sub-
stituted compound 7d. In our experiments with reference
indolo[2,3-b]quinolines: unsubstituted DiMIQ 8 and

—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
2459
Scheme 1. General scheme of synthesis of 5H-indolo[2,3-b]quinolines. Conditions: (a) 110±150ꢀC, (b) 150±160ꢀC; (c) SnCl2/HCl, then NaNO2/HCl;
(d) polyphosphoric acid, 110±160ꢀC; (e) dimethyl sulfate, than NaOH.
dimethyl substituted compound 9, Kapp was 0.23Â106
method, and some of the investigated derivatives were
found to be bent (Table 5). The bending of the structure
occurred between ring A and ring B (de®ned in Figure
2). The values of the torsional (valence) angles (also
de®ned in Figure 2) j1 (C4-C4a-N5-C5a) and j2 (C1-
C11a-C11-C10b) are listed in Table 5. As shown by
these data, the most distinct bending is the one in com-
pound 7d (selected as the most active in the present
study) and the one in compound 9 (selected previously
as the most active3). Only DiMIQ 8 and ellipticine 10
are planar.
1
1
M
tively. Analogous values found for ellipticine were: Kapp
at n=4.96 and 63.2Â106 M at n=2.42, respec-
=
4.87Â106 M 1 at n equal to 4.1 (Table 4).
Our previous studies regarding the DiMIQ-induced
double stranded DNA breaks formation mediated by
topoisomerase II implied the furtherance of our enzy-
matic assay in vitro for compounds 7a±e under condi-
tions established earlier.3 We found that all of the 5H-
indolo[2,3-b]quinolines stimulated DNA cleavage, pro-
ducing a maximum of the cleavable complexes in the
concentration range of 0.6 to 1.4 mM, when puri®ed calf
thymus topoisomerase II and circular pSP65 DNA as
substrate were used (Table 4).
To resolve the question of planarity, the geometry of
these compounds was optimized by other methods, i.e.
the semi-empirical PM3 hamiltonian and the ab initio
procedure involving the 6-31G and 6-31GÃ basis sets.
This time no signi®cant bends were found between the
A-ring and the B-ring (Table 5). Thus, the ab initio
Initially the geometry of compounds 7a±e, DiMIQ and
ellipticine was optimized by the semi-empirical AM1

2460
—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
procedure using the 6-31GÃ basis set (with polarization
functions) was much more reliable and con®rmed the
planarity of the investigated compounds. These results
were also con®rmed by the alternative semi-empirical
PM3 approach. In this case, the bending of the
indolo[2,3-b]quinolines observed in AM1 calculations,
as well as correlation between calculated bending and
other (physicochemical and biological) properties seems
to be fortuitous.
Electrostatic properties play a key role in the interaction
between strongly charged DNA polymers and inter-
calating ligands.6 These interactions are responsible for
the short-range e€ects resulting from intercalation in a
de®ned geometry (hydrogen bonds, stacking), and for
the long-range stacking e€ects accounting for the
sequence speci®c docking of the intercalator in the
DNA helix. Long-range electrostatic interactions are
responsible for the relative angular orientation of the
molecular (DNA/intercalator) complexes. This holds
Figure 1. Octanol:water partition coecient (logP) of compounds
7aBe. Water±octanol partition coecients were determined on pKb-
100 column (4.6Â250 mm, 42ꢀC, 1,0 mL/min) using 0.01% H2SO4 in
methanol as mobile phase. All compounds were eluted in the well
de®ned peaks. LogP de®ned as logP=log (Cf(octanol)/Cf (water)). Pre-
sented values are mean values of ®ve experiments.
Table 2. Antimicrobial activity MIC (mM)a
Scheme 2. Structures of reference compounds. (8) 5,11-dimethyl-5H-
indolo[2,3-b]quinoline; (9) 2,5,9,11-tetramethyl-5H-indolo[2,3-b]quin-
oline; (10) ellipticine.
Strains
Compd
1
2
3
4
5
6
7a
7b
7c
7d
7e
8b
n.a.c
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
0.12
0.03
0.03
0.03
0.06
0.12
0.06
0.06
0.015
0.03
0.03
0.06
0.06
0.03
0.015
0.12
0.03
0.03
0.12
0.03
0.015
0.12
0.12
0.06
Table 1. Isosbestic points and pKa values of compounds 7a±ea
Compd.
Isosbestic point [nm]
pKa
l [nm]
7a
7b
7c
7d
7e
293
287
282
290
276
6.02‹0.09
7.87‹0.10
7.47‹0.10
8.07‹0.11
7.64‹0.11
273
304
278
285
281
a
Strains used: 1. Escherichia coli PCM 271; 2. Pseudomonas aerugi-
nosa PCM 499; 3. Staphylococcus aureus PCM 458; 4. Micrococcus
luteus PCM 525; 5. Candida albicans (clinical isolate); 6. Trichophyton
mentagrophytes (clinical isolate); MICs were determined at the con-
centration range of 0.01 to 5 mM.
a
Determination of pKa value was performed at indicated wavelength
using Henderson±Hasselbach equation (pKa=pH log ([A ]/[HA]))
under conditions elaborated earlier.3 Presented values are mean values
of ®ve experiments; values of standard deviation are given.
b
Compound 8 was used as a standard.
n.a., no activity (MIC > 5 mM).
c
Table 3. Cytotoxic activity ED50 (mM)a
Tumor cell lines
Compd
HL-60
T-47D
SW-707
A-549
A-431
BM
7a
7b
7c
7d
7e
8b
n.d.c
n.d.
0.90‹0.04
0.47‹0.03
0.47‹0.03
0.52‹0.04
2.48‹0.05
2.70‹0.04
0.73‹0.02
6.40‹0.03
0.70‹0.03
1.40‹0.04
0.60‹0.03
1.60‹0.04
1.18‹0.03
2.07‹0.03
1.70‹0.02
3.49‹0.04
n.d.
n.d.
0.22‹0.03
0.20‹0.02
0.22‹0.05
n.d.
1.60‹0.05
0.40‹0.02
1.20‹0.04
1.25‹0.05
0.91‹0.05
1.25‹0.03
n.d.
1.60‹0.03
2.00‹0.05
12.5‹0.1
0.50‹0.02
0.34‹0.02
0.35‹0.02
n.d.
1.43‹0.04
0.57‹0.03
2.00‹0.04
1.35‹0.04
0.90‹0.03
0.88‹0.03
n.d.
2.55‹0.05
1.62‹0.04
4.25‹0.05
9b
10b
a
Cell lines: HL-60, promyelocytic leukemia; T-470, breast carcinoma; SW-707, colon adenocarcinoma; A-549, lung carcinoma; A-431, epidermal
carcinoma; BM, melanoma. Presented values are mean values of nine experiments; standard deviations are given.
Compounds 8±10 were used as a standard.
n.d., not determined.
b
c

—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
2461
Table 4. Interaction of compounds 7a±e with calf thymus DNA and DNA topoisomerase II
l (nm)b
Kapp (M 1) (Â106)
n
CCmax (mM)c
% CC
a
Compd.
ÁTm
7a
7b
7c
7d
7e
8d
12.0‹0.3
14.6‹0.1
15.3‹0.2
19.0‹0.3
12.4‹0.4
11.0‹0.2
19.0‹0.4
9.8‹0.3
344.0
345.0
347.0
350.0
350.0
340.0
345.0
301.0
2.35‹0.43
17.4‹4.92
2.32‹0.32
63.2‹0.00
6.20‹1.22
0.23‹0.77
39.48‹0.76
1.44‹0.70
2.31‹0.09
2.57‹0.07
2.28‹0.08
2.42‹0.00
2.71‹0.12
4.96‹0.1
1.4‹0.1
1.0‹0.1
0.8‹0.1
0.6‹0.1
1.2‹0.1
5.0‹0.1
0.4‹0.1
n.d.e
18‹1
30‹1
32‹2
35‹2
20‹1
22‹1
24‹1
n.d.
9d
2.83‹0.14
5.53‹0.28
10d
a
Increase (ÁTm) of calf-thymus DNA denaturation temperature. Measurements were taken in 5.0 mM pH 7.0 Tris±HCl bu€er containing 50 (M
calf thymus DNA (Boeringer-Mannheim, average molecular weight 1.2Â106 D) 50 mM EDTA, 5.0 (M drug, 5% DMSO.
b
Binding constant values (Kapp) and n values (nucleotide/binding site) were calculated from Schatchard plot using McGhee and von Hippel4 model
of DNA-ligand interaction. Measurements were taken at indicated wavelength, by titration of the 10 mM solution of drug with 1.33 mM solution of
DNA.
c
Topoisomerase II-induced DNA cleavage (CCmax) by indolo[2,3-b]quinolines. Measurements were taken according to a previously elaborated
procedure.3
Compounds 8±10 were used as a standards.
n.d., not determined; standard deviations are given.
d
e
Table 5. Torsional angles of optimized indolo[2,3-b]quinolines
AM1
PM3
6-31G
6-31GÃ
Compd
j1
j2
j1
j2
j1
j2
j1
j2
7a
7b
7c
7d
7e
8a
180.0
179.9
179.8
167.2
179.7
179.6
162.6
179.6
179.8
179.9
179.7
168.2
179.8
179.7
167.9
179.9
179.7
176.8
177.1
178.9
177.1
179.7
179.7
179.1
179.7
177.1
176.4
178.8
176.4
179.7
179.6
176.6
179.9
180.0
180.0
180.0
179.9
180.0
179.9
180.0
179.9
179.0
180.0
180.0
179.9
179.9
179.9
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
180.0
9a
10a
a
Compounds 8±10 were used as a standard.
Figure 2. Geometry of 5H-indolo[2,3-b]quinolines. Torsional angles
j1 C1-C11a-C-11-C10b, and j2 C4-C4a-N5-C5a.
not only for hydrogen-bonded systems7 but also for
sandwich complexes of aromatic molecules.8 Such spe-
ci®c interactions could be reasonably described by mul-
tipole expansions7±9 representing the anisotropy of the
molecular charge distribution at the atomic level. The
speci®c electronic properties of intercalating molecules
may be visualized by the molecular electrostatic poten-
tial (MEP) plotted at Van der Waals surfaces.10 Such
simple model of electrostatic properties describes the
anisotropy of the molecular charge distribution. Higher
multipole moments introduced to the MEP model may
bring about signi®cant changes in the surface MEP dis-
tribution10 and could be related to some properties of the
molecules (lipophilicity, long-range interactions, ligand ±
receptor binding constant, etc.). The MEP surfaces
obtained for DiMIQ and for the most active 2,9-dime-
thoxy-derivative 7d form quite similar patterns (Fig. 3).
In both cases, strongly negative potentials (free electron
Figure 3. Molecular Electrostatic Potential (MEP) on the Van der Waals
surfaces for selected indolo[2,3-b]quinolines. (7d) 2,9-dimethoxy-5,11-
dimethyl-5H-indolo[2,3-b]quinoline; (8) 5,11-dimethyl-5H-indolo[2,3-
b]quinoline; (9) 2,5,9,11-tetramethyl-5H-indolo[2,3-b]quinoline.

2462
—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
pairs occurring in base-forms of the compounds) are
located near the N-6 atom and shifted outside towards
the D-ring (yellow spots), whereas weaker potentials are
located near N-5. The MEP surface near the external
hydrogen atoms in the A-ring (H-1, H-2, H-3, H-4) is
more positive (green spots) than the MEP surfaces near
the D-ring protons (H-7, H-8, H-9, H-10). The negative
potential above the molecular plan is shifted toward the
C-ring and the D-ring, which parallels the charge shift.
Elemental microanalyses were performed by the Analy-
tical Laboratory of the Wroc•aw University of Technol-
ogy (symbols of the analyzed elements are given only if the
element was within 0.4% of the theoretical value). Col-
umn chromatography was carried out using Merck Kie-
selgel 60 (230±400 mesh). The purity and identity of the
compounds were checked out by TLC, using Merck DC-
Alufolien Kieselgel F254. Abbreviations: DMF=di-
methylformamide; PPA=polyphosphoric acid.
The MEP maps obtained for other investigated com-
pounds do not di€er very much from one another (data
not shown). This suggests, that in our case, it is not so
much the electrostatics as the geometry of the molecule
that can be regarded as responsible for the di€erences in
the physicochemical and biological properties.
Synthesis of 4-methyl-2-(2-nitroanilino)quinolines 4a-d:
general procedure. The equimolar mixture of chlor-
oquinoline 111,12 and nitroaniline 3 was stirred at 150±
160ꢀC for 10 h, then cooled and crystallized from ethanol
containing 10% of ammonia.
2-(4-Methoxy-2-nitroanilino)-4-methylquinoline (4a). Yield
53%; mp 148±150ꢀC; IR n 3400, 1505 cm 1. Anal.
(C17H15N3O3): C, H, N.
Conclusion
The results of the study allow the following general-
izations to be made: (i) introduction of two methoxy
group into the indolo[2,3-b]quinoline system increases
the water-solubility of the compound, (ii) the presence
of substituents at C-2 and/or C-9 accounts for an
increase of the denaturation temperature of calf thymus
DNA, and (iii) the contribution of the position char-
acter and number of substituents in the aromatic area of
indolo[2,3-b]quinolines to the calf thymus DNA anity
becomes distinct. The highest Kapp values are found for
di-substituted (C-2 and C-9) dimethoxy (and dimethyl)
derivatives. This is correlated with their cytotoxic activ-
ities in vitro; (iv) the presence of substituents at C-2 and/
or C-9 also brings about an increase of cytotoxic activity
as compared to the lead compound (DiMIQ); (v) each
compound produces in vitro calf thymus topoisomerase
II-induced cleavable complexes of pSP65 DNA. The
most cytotoxic compounds stimulate the most eciently
formation of topoisomerase II-mediated DNA breaks;
(vi) methoxy (and methyl) substituents at C-2 and C-9
presumably account for a bending of the indolo[2,3-
b]quinoline system (as we found using the semi-empirical
AM1 method for geometry optimization). This bending
is likely to be responsible for the manner of DNA helix
unwinding, as well as for the manner by which topo-
isomerase recognizes the changes induced in the DNA
helix; (vii) there is very likely to be an interplay between
the character of the substituent, its in¯uence on the shape
of the molecule and the quality of the intercalator.
2-(4-Methoxy-2-nitroanilino)-4,6-dimethylquinoline (4b).
Yield 81%; mp 169±174ꢀC; IR n 3300, 1590, 1500, 1450,
1270, 1215, 1200 cm 1. Anal. (C18H17N3O3): C, H, N.
6-Methoxy-4-methyl-2-(4-methyl-2-nitroanilino)quinoline
(4c). Yield 81%; mp 146±154ꢀC; IR n 3315, 1620, 1605,
1570, 1525, 1360, 1215, 1205, 1175, 1150 cm 1. Anal. (C18
H17N3O3): C, H, N.
6-Methoxy-2-(4-methoxy-2-nitroanilino)-4-methylquino-
line (4d). Yield 96%; mp 150±156ꢀC; IR n 3330, 1520,
1510, 1280, 1220, 1210, 1180, 1140 cm 1. Anal.(C18H17
N3O4): C, H, N.
Synthesis of 2-(5-substituted-benzotriazol-1-yl)quinolines
5a±d: general procedure. Compound 4 (1 mol) was
added portionwise to the cooled solution of SnCl2 (5
mol) in 36% HCl (1.2 L) and ethanol (200 mL). The
mixture was then heated in a steam bath for 30 min, and
refrigerated overnight. The precipitate was collected,
stirred in 20% KOH (2 L) for 30 min, and ®ltered o€.
After drying it was dissolved in a mixture of 15%
HCl:ethanol 1:1 (3 L) and diazotized at 0±5ꢀC with a
solution of NaNO2 (76 g, 1.1 mol) in water (0.5 L). The
solid product was collected and crystallized from DMF.
2-(5-Methoxybenzotriazol-1-yl)-4-methylquinoline (5a).
Yield 58%; mp 161±165ꢀC; IR n 1500, 1310 cm 1. Anal.
(C17H14N4O): C, H, N.
2-(5-Methoxybenzotriazol-1-yl)-4,6-dimethylquinoline
(5b). Yield 45%; mp 234±237ꢀC; IR n 1605, 1505, 1310,
1105, 1050, 820 cm 1. Anal. (C18H16N4O): C, H, N.
Experimental
Chemistry
Melting points were taken in open capillary tubes on a
1
6-Methoxy-4-methyl-2-(5-methylbenzotriazol-1-yl)quino-
line (5c). Yield 60%; mp 196±198ꢀC; IR n 1520, 1270,
1235, 1090, 1045, 900, 820 cm 1. Anal. (C18H16N4O): C,
H, N.
Ko¯er-type apparatus and were uncorrected. H NMR
spectra were recorded on a Varian Gemini 200 (200
MHz) spectrometer. Chemical shifts are expressed in
ppm, using TMS as an internal standard. The IR spec-
tra were taken in KBr pellets, using a Beckman 4240
spectrometer. UV spectra were recorded in dioxane with
a Beckman Acta M-VI spectrometer. MS spectra were
obtained with an Intectra AMD-604 apparatus ( 70 eV).
6-Methoxy-2-(5-methoxybenzotriazol-1-yl)-4-methylqui-
noline (5d). Yield 41%; mp 197±199ꢀC; IR n 1525, 1505,
1425, 1310, 1240, 1120, 1055, 1035, 920, 830 cm 1. Anal.
(C18H16N4O2): C, H, N.

—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
2463
2-(Benzotriazol-1-yl)-6-methoxy-4-methylquinoline (5e).13
The mixture of 2-chloro-6-methoxy-4-methylquinoline
(1c) (97.0 g, 0.5 mol) was heated until an exothermic
reaction occurred (110ꢀC). Then it was stirred without
heating. The reaction temperature rose up to 140ꢀC.
When the reaction terminated, the solid was crystallized
from DMF±water to give compound 5e; yield 105.0 g
(72%); mp 160±162ꢀC (lit.13 163ꢀC).
150±160ꢀC for 12 h. After cooling the precipitate was
collected, washed with acetone and added to 20%
NaOH (50 mL). The mixture was then extracted with
CHCl3, the extract was dried over MgSO4 and chroma-
tographed on a silica gel column. The orange product 7
was eluted with CHCl3, the solvent was evaporated and
the residue was recrystallized.
9-Methoxy-5,11-dimethyl-5H-indolo[2,3-b]quinoline (7a).
Yield 30%, mp 206±207ꢀC (from toluene), NMR
(CDCl3-d1) d 8.17 (d-d, 1H, J=9±1 Hz), 7.77±7.64 (m,
4H), 7.44 (d-d-d, 1H, J=8±6±2 Hz), 7.16 (d-d, 1H,
J=9-2.5 Hz), 4.25 (s, 3H), 3.92 (s, 3H), 3.08 (s, 3H); IR
n 1635, 1495, 1475, 1465, 1215 cm 1; UV (lmax, log e)
291 nm (4.66); MS (m/e, rel intensity) 276 (56, M+), 261
(100). Anal. (C18H16N2O): C, H, N.
Synthesis of 6H-indolo[2,3-b]quinolines 6a±d: general
procedure. The corresponding triazole 5 (0.1 mol) was
suspended in PPA (150 mL) and the mixture was heated
at 110±160ꢀC, until gas evolution ceased. The mixture
was then cooled to 90ꢀC, poured into iced water (1.5 L)
and the precipitate was further cooled. After washing
with water the precipitate was heated in a steam bath in
25% NH3 (500 mL), ®ltered o€, washed with water,
crystallized from pyridine and recrystallized from DMF.
9-Methoxy-2,5,11-trimethyl-5H-indolo[2,3-b]quinoline
(7b). Yield 53%, mp 160±165ꢀC (from toluene:hexane),
NMR (CDCl3-d1) d 7.90 (bs, 1H), 7.66 (d, 1H, J=2.5
Hz), 7.64 (d, 1H, J=8.5 Hz), 7.55 (m, 2H), 7.15 (d-d, 1H,
9-Methoxy-11-methyl-6H-indolo[2,3-b]quinoline (6a).
Yield 14%, mp 283±285ꢀC, NMR (DMSO-d6) d 11.25
(bs, 1H), 8.34 (d-d, 1H, J=8.5-1 Hz), 7.94 (d-d, 1H,
J=8.5-1 Hz), 7.83 (d, 1H, J=2.5 Hz), 7.70 (d-d-d, 1H,
J=9-7-1.5 Hz), 7.50 (d-d, 1H, J=7-1.5 Hz), 7.42 (d, 1H,
J=8.5 Hz), 7.18 (d-d, 1H, J=9-2.5 Hz), 3.90 (s, 3H), 3.18
(s, 3H); IR n 3250±2700, 1670, 1490 cm 1; MS (m/e, rel
intensity) 262 (77, M+), 247 (100). Anal. (C17H14NO): C,
H, N.
J=8.5±2.5 Hz), 4.22 (s, 3H), 3.91 (s, 3H), 3.03 (s, 3H),
1
2.54 (s, 3H); IR n 1645, 1505, 1480, 1220, 1150, 830 cm
;
UV (lmax, log e) 292 nm (4.70); MS (m/e, rel intensity)
290 (47, M+), 275 (100). Anal. (C19H18N2O): C, H, N.
2-Methoxy-5,9,11-trimethyl-5H-indolo[2,3-b]quinoline
(7c). Yield 47%, mp 168±174ꢀC (from toluene:hexane),
NMR (CDCl3-d1) d 7.93 (bs, 1H), 7.64 (d, 1H, J=9 Hz),
7.61 (d, 1H, J=1.5 Hz), 7.50 (d, 1H, J=2.5 Hz), 7.40±7.32
9-Methoxy-2,11-dimethyl-6H-indolo[2,3-b]quinoline (6b).
Yield 14%, mp above 300ꢀC, NMR (DMSO-d6) d 11.71
(bs, 1H), 8.15 (bs, 1H), 7.88 (d, 1H, J=8.6 Hz), 7.82 (d,
1H, J=2.5 Hz), 7.61(d-d, 1H, J=8.6-1.5 Hz), 7.45 (d, 1H,
J=8.6 Hz), 7.18 (d-d, 1H, J=8.8-2.5 Hz), 3.90 (s, 3H),
3.18 (s, 3H), 2.56 (s, 3H); IR n=3160±2840, 1600, 1490,
1265, 1220, 815 cm 1; MS (m/e, rel intensity) 276
(80,M+), 261 (100). Anal. (C18H16N2O): C, H, N.
(m, 2H), 4.28 (s, 3H), 3.95 (s, 3H), 3.02 (s, 3H), 2.55 (s,
1
3H); IR n 1645, 1500, 1490, 1480, 1475, 1240, 1225 cm
;
UV (lmax, log e) 291 nm, 4.73; MS (m/e, rel intensity) 290
(100,M+), 275 (42). Anal. (C19H18N2O): C, H, N.
2,9-Dimethoxy-5,11-dimethyl-5H-indolo[2,3-b]quinoline
(7d). Yield 39%, mp 165±171ꢀC (from toluene:hexane);
NMR (CDCl3-d1) d 7.66 (m, 2H), 7.61 (d, 1H, J=2.5
Hz), 7.51 (d, 1H, J=2.5 Hz), 7.38 (d-d, 1H, J=9±2.5
Hz), 7.16 (d-d, 1H, J=9±2.5 Hz), 4.28 (s, 3H), 3.97 (s,
3H), 3.92 (s, 3H), 3.00 (s, 3H); IR n 1640, 1630, 1500,
1480, 1460, 1230, 1220, 1040 cm 1; UV (lmax, log e) 294
nm (4.66); MS (m/e, rel intensity) 306 (50, M+), 291
(100). Anal. (C19H18N2O2): C, H, N.
2-Methoxy-9,11-dimethyl-6H-indolo[2,3-b]quinoline (6c).
Yield 30%, mp above 300ꢀC, NMR (DMSO-d6) d 11.67
(bs, 1H), 8.15 (s, 1H), 7.90 (d, 1H, J=9 Hz), 7.61 (d, 1H,
J=2.6 Hz), 7.46±7.39 (m, 2H), 7.34 (d-d, 1H, J=8.3±1.1
Hz), 3.96 (s, 3H), 3.17 (s, 3H), 2.52 (s, 3H); IR n 3180±
2860, 1640, 1250, 1230, 1195, 825, 730 cm 1; MS (m/e, rel
intensity) 276 (100, M+), 261 (67), 233 (66). Anal.
(C18H16N2O): C, H, N.
2-Methoxy-5,11-dimethyl-5H-indolo[2,3-b]quinoline (7e).13
Yield 55%, mp 149±150ꢀC.
2,9-Dimethoxy-11-methyl-6H-indolo[2,3-b]quinoline (6d).
Yield 21%, mp above 252ꢀC (decomp.), NMR (DMSO-d6)
d 11.33 (s, 1H), 7.85 (d, 1H, J=9.2 Hz), 7.81 (d, 1H, J=2.4
Hz), 7.58 (d, 1H, J=2.8 Hz), 7.40 (d, 1H, J=8.7 Hz), 7.38
(d-d, 1H, J=9.2±2.8 Hz), 7.16 (d-d, 1H, J=8.7±2.5 Hz),
3.96 (s, 3H), 3.90 (s, 3H), 3.14 (s, 3H); IR n 3180±2810,
1650, 1500, 1280, 1230, 830 cm 1; MS (m/e, rel. intensity)
292 (90, M+), 277 (100). Anal. (C18H16N2O2): C, H, N.
Determination of pKa
The pKa values of 5H-indolo[2,3-b]quinolines were
deduced from their UV absorption spectra. Measure-
ments were performed on a Pharmacia LKB Biochrom
Bio-4060 spectrophotometer, by a method elaborated
earlier.3 The spectra of the drugs (dissolved in DMSO)
were recorded in 0.02 M phosphate bu€ers pH 2.2, 5.35,
7.10, 8.30, in a 1 cm pathlength cuvette at 25ꢀC. The
®nal concentration of DMSO was 5%.
2-Methoxy-11-methyl-6H-indolo[2,3-b]quinoline (6e).13
Yield 25%, mp 305±307ꢀC.
Synthesis of 5,11-dimethyl-6H-indolo[2,3-b]quinoline 7a±
d: general procedure. The corresponding 6H-indolo
[2,3-b]quinoline 6 (0.01 mol) and dimethyl sulfate (1.5
mL) in toluene (20 mL) were heated in a sealed tube at
Determination of lipophilicity
The logP values for individual compounds were esti-
mated by the microscale-¯ask method combined with

2464
—. Kaczmarek et al. / Bioorg. Med. Chem. 7 (1999) 2457±2464
HPLC analysis of both water (bu€er) and octanol
phase. A detailed description of the procedure was
published elsewhere.3
MO SCF method (6-31G, and 6-31GÃ basis sets).16 Geo-
metry was optimized in vacuum, using steepest-descent
optimization algorithm for the free base form of the drug.
Electronic molecular potentials were calculated from
Cumulative Atomic Multipole Moments (CAMM),
derived from corresponding wavefunctions and visualized
by the mol17 program.17
Antimicrobial activity study
MIC values were determined by serial dilution (according
to a routine) method.15
Acknowledgements
Cell cytotoxicity assay
The HL60 cell line was grown in an RPMI 1640 med-
ium supplemented with 10% FCS (fetal calf serum). The
A431 cell line was maintained in an MEM-a medium
enriched with 10% FCS. The remaining cell lines were
cultured in an opti-MEM medium supplemented with
5% FCS. All culture media also contained penicillin (50
mg/mL), streptomycin (50 mg/mL) and fresh l-glutamine
(2 mM). All cell lines were incubated at 37ꢀC in a
humidi®ed atmosphere of 95% air and 5% CO2. The
concentration of the drugs giving 50% growth inhibi-
tion (ED50) were determined according to the MTT
procedure.
The research was partially supported by the Polish State
Committee for Scienti®c Research. We wish to thank
the Wroc•aw Centre of Networking & Supercomputing
(WCSS Grant No. 10/96) as well as the Molecular
Modelling Laboratory (KBN Grant No. 6 P04A 060 09)
for providing CPU time and software.
References
1. Kaczmarek, —; Balicki, R.; Nantka-Namirski, P.; Peczy-
ska-Czoch, W.; Mordarski, M. Arch. Pharm. (Weinheim)
1988, 321, 463.
2. Pognan, F.; Saucier, J. M.; Paoletti, C.; Kaczmarek, —;
Nantka-Namirski, P.; Mordarski, M.; Peczynska-Czoch, W.
Biochem. Pharmacol. 1992, 44, 2149.
3. Peczynska-Czoch, W.; Pognan, F.; Kaczmarek, —; Bor-
atynski, J. J. Med. Chem. 1994, 37, 3503.
Determination of DTm
Experiments were carried out, using a Cary 3 Varian
spectrometer equipped with a DNA Thermal Application
Program according to the Cory method.14
4. McGhee, J. D.; Von Hippel, P. H. J. Mol. Biol. 1974, 86,
469.
Determination of binding constants (Kapp
)
5. Osiadacz, J.; Peczynska-Czoch, W.; Kaczmarek, —.;
Opolski, A. 6th International Symposium on Molecular
Aspects of Chemotherapy 9±12 July 1997, Gdansk, Poland.
6. Naray-Szabo, G.; Ferenczy, G. G. Chem. Rev. 1995, 95,
829.
7. Price, S. L.; LoCelso, F.; Treichel, J. A.; Goodfellow, J. M.;
Umrania, Y. J. Chem. Soc. Faraday Trans. 1993, 89, 3407.
8. Price, S. L.; Stone, A. J. J. Chem. Phys. 1987, 86, 2859.
9. Sokalski, W. A.; Poirier, R. Chem. Phys. Lett. 1983, 98, 86.
10. Sokalski, W. A.; Sawaryn, A. J. Chem. Phys. 1987, 87, 526.
11. Campbell, K. N.; Tipson, S. R.; Elder®eld, R. C.; Camp-
bell, B. K.; Clapp, M. A.; Gensler, W. J.; Morisson, D.;
Moran, W. J. J. Org. Chem. 1946, 11, 803.
Measurements were carried out using a Pharmacia LKB
Biochrom Bio-4060 spectrophotometer in a 5 cm path-
length cuvette, at indicated wavelength, by subsequent
adding of 5 mL of 1.33 mM DNA (in 0.025 M acetate
bu€er pH 5.0) to 10.0 mL of 10 mM solution of drug (in
0.025 M acetate bu€er pH 5.0, and 5% of DMSO)
Binding constant values (Kapp) and n-values (nucleotide/
binding site) were calculated from the Scatchard plot,
using the McGhee and von Hippel4 model of the DNA-
ligand interaction. Binding constants were evaluated, in
terms of the Marquardt procedure of nonlinear regres-
12. Kaslow, C. E.; Laver, W. M. Org. Synt. Coll. 3, 194.
13. Holt, S. J.; Petrov, V. J. Chem. Soc. 1948, 922.
14. Cory, M.; McKee, D. D.; Kagan, J.; Henry, D. W.; Miller,
A. J. J. Am. Chem. Soc. 1985, 107, 2528.
sion, using Kapp=1.0Â106 [M 1] and n=2.0 as starting
8
values, and Áss=1.0Â10
as a ®nal parameter of
optimization.
15. Approved standard M7-A2, In National Committee for
Clinical Laboratory Standards, 2nd ed.; Villanova, PA, 1990.
16. Frish; M. J.; Trucks; G. W.; Schlegel; H. B.; Gill; P. M.
W.; Johnson; B. G.; Robb; M. A.; Cheesman; J. R.; Keith; T.
A.; Petersson; G. A.; Montgomery; J. A.; Raghavachari; K.;
Al. -Laham; M. A.; Zakrzewski; V. G.; Ortiz; J. V.; Foresman;
J. B.; Cioslowski; J.; Stefanov; B. B.; Nanayakkara; A.; Chal-
lacombe; M.; Peng; C. Y.; Ayala; P. Y.; Chen; W.; Wong; M.
W.; Andres; J. L.; Replogle; E. S.; Gomperts; R.; Martin; R.
L.; Fox; D. J.; Binkley; J. S.; Defrees; D. J.; Baker; J.; Stewart;
J. P.; Head-Gordon; M.; Gonzales; C.; Pople; J. A. Gaussian
94 (Revision 3. 1), Gaussian, Inc., Pittsburgh, PA, 1995.
17. Sokalski, W. A.; Sneddon, S. J. Mol. Graphics 1991, 9, 86.
Determination of topoisomerase II inhibitory activity
Measurements were performed using a previously ela-
borated procedure3 (%CC, percent of generated pSP65
DNA cleavable complexes versus total pSP65 DNA
quantity).
Quantum chemistry calculations
All structures were ®rst optimized with semiempirical
AM1, and PM3 approaches and then ab initio the LCAO