Pyrrolo-quinoline derivatives as potential antineoplastic drugs

Article abstract of DOI:10.1016/S0968-0896(00)00060-2

Some novel pyrrolo-quinoline derivatives have been synthesized as potential antineoplastic agents. They contain an angular aromatic tricyclic or tetracyclic system, to which the methanesulfon-anisidide side chain typical of amsacrine as such, or lacking the m-methoxy substituent, is connected. A methyl group can be present at position 7 of the pyrrolo-quinoline ring. The novel compounds exhibit interesting cell growth inhibitory properties when tested against the NCI panel of cell lines, in particular those obtained from solid tumors like CNS-, melanoma- and prostate-derived cells. The mechanism of cytotoxic action does not seem to be related to topoisomerase II poisoning ability. Most active proved to be compound 4a, which lacks both methyl and methoxy substituents, followed by 5a, having the methoxy group only. Biological activity is less pronounced in the tetracyclic family of derivatives 6 and 7. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00060-2

Bioorganic & Medicinal Chemistry 8 (2000) 1415±1422
Pyrrolo-Quinoline Derivatives as Potential Antineoplastic Drugs
M. G. Ferlin, B. Gatto, G. Chiarelotto and M. Palumbo*
Department of Pharmaceutical Sciences, University of Padova, via Marzolo 5, 35131 Padova, Italy
Received 20 September 1999; accepted 16 February 2000
AbstractÐSome novel pyrrolo-quinoline derivatives have been synthesized as potential antineoplastic agents. They contain an
angular aromatic tricyclic or tetracyclic system, to which the methanesulfon-anisidide side chain typical of amsacrine as such, or
lacking the m-methoxy substituent, is connected. A methyl group can be present at position 7 of the pyrrolo-quinoline ring. The
novel compounds exhibit interesting cell growth inhibitory properties when tested against the NCI panel of cell lines, in particular
those obtained from solid tumors like CNS-, melanoma- and prostate-derived cells. The mechanism of cytotoxic action does not
seem to be related to topoisomerase II poisoning ability. Most active proved to be compound 4a, which lacks both methyl and
methoxy substituents, followed by 5a, having the methoxy group only. Biological activity is less pronounced in the tetracyclic
family of derivatives 6 and 7. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
the m-methoxy substituent. The novel structures are
presented in Figure 1.
Well known cancer chemotherapeutic agents, such as
anthracyclines, camptothecin and amsacrine, are char-
acterized by planar polycyclic systems.1 They are able to
interfere with DNA-processing enzymes (topoisomer-
ases I and II) by forming a ternary (cleavable) complex
involving the drug, the DNA and the enzyme.2,3
Although the molecular details of the ternary complex
are still not fully characterized, a number of models are
available for both topo I and topo II inhibition.4 6 It is
proposed that an active drug consists of two pharma-
cophoric domains, one forming appropriate contacts
with the nucleic acid and the other binding to the
enzyme.7 Examining the structural determinants for
optimal topoisomerase poisoning, we recently demon-
strated that the pharmacophores must occupy well
de®ned reciprocal positions within the drug. Indeed,
moving one domain with reference to the other gave
structural isomers completely di€ering in their ability of
recognizing the cleavable complex and producing dif-
ferent biological and pharmacological responses.8
Results and Discussion
Chemistry
The preparation of the desired compounds was accom-
plished according to Schemes 1 and 2. Intermediates 1a
and 1b (Scheme 1) were prepared adopting known
methods for the synthesis of the quinoline ring such as
the Conrad±Limpach9 and the Gould±Jacobs reac-
tions,10 respectively. For both, the starting material 5-
nitroindole was catalytically reduced to the correspond-
ing amino derivative. In order to obtain the 7-methyl-9-
hydroxy-pyrrolo[3,2-f]quinoline 1a, the acid-catalyzed
condensation with ethyl acetoacetate in re¯uxing etha-
nol yielded the enamine-derivative crotonate, which
then underwent thermal cyclization in boiling phenyl
ether. Likewise, condensation of equimolar amounts of
5-aminoindole with diethyl ethoxymethylene malonate
yielded the corresponding enamine acrylate which,
without requiring puri®cation, was cyclized in boiling
phenyl ether to produce exclusively 9-hydroxy-pyr-
rolo[3,2-f]quinoline ethyl ester as the result of a regio-
selective reaction.11 Alkaline hydrolysis, followed by
thermal decarboxylation a€orded intermediate 1b.
Treatment of 1a,b with phosphorus oxychloride allowed
us to replace the hydroxyl group at C-9 with chlorine to
yield 2a,b, as shown in Scheme 2. Compounds 4a,b and
5a,b were obtained as the hydrochlorides after nucleo-
philic substitution of the chlorine at C-9 in 2a,b with
Continuing an investigation aimed at de®ning the basic
rules that govern structure±activity relationships, we
synthesized and investigated a number of new deriva-
tives, characterized by an angular 9-anilino-3H-pyr-
rolo[3,2-f]quinoline planar nucleus connected to the
methanesulfon-anisidide residue characteristic of the
known anticancer drug m-AMSA, or its analogue lacking
*Corresponding author. Tel.: +39-049-827-5711; fax: +39-049-827-
5366; e-mail: mpalumbo@purple.dsfarm.unipd.it
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00060-2

1416
M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
p-methanesulfonamido-aniline or o-methoxy-p-methane-
sulfonamido-aniline in re¯uxing ethanol±methanol mix-
ture. Moreover, hydroxy derivatives 1a,b were subjected
to the Vilsmeier±Haach reaction with dimethylformamide
and phosphorus oxychloride to provide the corre-
sponding 9-chloro-1-formylate compounds 3a,b. The
®nal compounds 6a,b and 7a,b, in which the amino-
group of the aniline moiety is incorporated into an
indole-naphthyridine tetracyclic structure, were obtained
as the mono-hydrochlorides by a one-step reaction
consisting of a reductive alkylation of methanesulfon-
amido-anilines; at the same time due to a nucleophilic
substitution of chlorine at C-9 the cyclization reaction
occurred for probable favorable steric factors. Because
the required p-methanesulfonamido-anilines were not
available commercially, we prepared them by means of
known strategies,12 with just one signi®cant modi®ca-
tion. In fact, the carbamoyl ethyl ester protecting group
was used instead of the acetyl function, which allowed
milder reaction conditions and gave higher global
yields.
Topoisomerase poisoning
The extent of DNA damage induced by the test drugs in
the presence of human topoisomerase II a was eval-
uated by gel electrophoresis techniques.13 While the well
known drug m-AMSA produced the expected cleavage
of the nucleic acid in a concentration dependent man-
ner, the test compounds failed to show appreciable
e€ects up to 100 mM, with the only exception of 4a. This
compound, as shown in Figure 2, exhibits a slight
increase of topo II-mediated DNA cleavage at 10 mM
Figure 1. Chemical structure of the test pyrrolo-quinoline derivatives
and m-AMSA.
Scheme 1.

M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
1417
Scheme 2.
concentration (compare with lane showing the back-
ground cleavage of topoisomerase II alone on pBR322
plasmid), while, at higher concentration, the drug, as
well as all tested derivatives, inhibited topo II activity.
This e€ect is clearly due to the DNA-binding properties
of the tested compounds. Moreover, to check whether
these amsacrine derivatives could exert their activity
through topoisomerase I, as other DNA-targeted agents
with dual activity do,7 we tested the drugs' ability to
inhibit the human enzyme. None of them could inhibit
topoisomerase I-mediated relaxation of the negatively
supercoiled plasmid, or, similarly to CPT, induce
nicking of DNA in the presence of the enzyme. Rather,
similarly to the intercalator m-AMSA, at doses higher
than 1 mM they could unwind DNA and in¯uence the
migration rate of the plasmid in ethidium bromide-
containing agarose gels. The results for representative
compounds and appropriate controls are shown in
Figure 3. From these data we can conclude that the
topoisomerase-dependent mechanism of DNA damage
does not appear to be appreciably operating in this
case.
Cell cytotoxicity
The novel pyrrolo-quinoline derivatives were tested as
cytotoxic agents in the in vitro primary NCI antitumor
screen.14

1418
M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
The results are reported in Table 1 as the average of the
data obtained with cell lines corresponding to speci®c
types of cancer and as the global average.
of the condensed tetracyclic system. Hence the double 7-
methyl-40-methoxy substitution is clearly detrimental to
drug activity. Compounds bearing the 7-methyl sub-
stituent but not the methoxy group (4b and 6b) are more
potent than the 5b and 7b congeners, but 2-fold less
potent than 7-unsubstituted derivatives 4a and 6a. The
latter are indeed the most active in the tricyclic and tet-
racyclic series. Recalling that the planar portion of
important anticancer compounds is often a DNA-inter-
calator, the above ®ndings are consistent with the fact
that methyl substitution renders drug intercalation into
DNA base-pairs less ecient. Methoxy substitution also
appears to reduce the cytotoxic response, in particular
when comparing 6a with 7a. Finally, the closure of
an extra ring in compounds 6 and 7 lowers potency.
The lower conformational freedom experienced by the
p-amino methanesulfonyl anilino moiety when the p-
amino group is linked to the pyrrole function possibly
accounts for these ®ndings.
The overall potency is in the range 10±100 mM for all
compounds except 5b, which proved to be almost inac-
tive. Also, compound 7b is the least active in the series
As far as the mechanism of action is concerned, topo-
isomerase II-mediated DNA damage does not seem to
be the major cause of cell death. In fact, even consider-
ing the relatively low potency exhibited by the novel
compounds, very little enzyme-mediated DNA cleavage
could be detected in the range of drug concentrations at
which remarkable cytotoxic e€ects are observed. Hence,
despite the structural similarity with m-AMSA shown
by compounds 4 and 5, the changes occurring in the
planar, DNA-interactive region cause a major loss of
topoisomerase poisoning ability. Given their remarkable
biological response, interference with other important
biological target(s) should be operating in the novel
compounds.
Figure 2. Topoisomerase II-stimulated DNA cleavage. The com-
pounds tested were examined for their ability to stimulate human
topoisomerase II-mediated DNA cleavage. A representative gel show-
ing the activity of the most cytotoxic compound 4a (see Table 1)
alongside 4b and the parent compound m-AMSA is reported in the
®gure.
A COMPARE15 analysis was performed with com-
pounds 4a and 6a to check whether they resemble pre-
viously identi®ed anticancer drugs, including m-AMSA.
The highest correlation coecients are of the order of
0.7, indicating a modest level of similarity with the test
drugs. The 20 compounds most highly correlated to 4a
and 6a in the COMPARE analysis belong to di€erent
structural families and do not exhibit a common phar-
macophore. In addition, m-AMSA was not found in the
above list, which con®rms a di€erent mechanism of
cytotoxic action.
Figure 3. Pyrrolo-quinoline derivatives do not inhibit human topo-
isomerase I. Representative compounds were tested for their ability to
inhibit the relaxation activity of human topoisomerase I toward nega-
tively supercoiled pBR322. 250 ng of plasmid were treated with 1 unit
of human topoisomerase I and the indicated concentrations of drugs
as described in Experimental, and run on ethidium bromide-containing
gels.
Table 1. Cell growth inhibition properties of the test pyrrolo-quinolines
IG50 (mM)b
Type of cancer cell linea
4a
4b
5a
5b
6a
6b
7a
7b
Leukemia (5)
Non-small cell lung (9)
Colon (7)
Central nervous system (5)
Melanoma (7)
Ovary (6)
Kidney (7)
Prostate (2)
Breast (7)
7.41
8.91
19.9
14.2
17.8
19.9
15.8
16.6
15.9
17.8
14.1
19.9
>100
>100
>100
>100
>100
>100
>100
>100
>100
10.7
16.6
20.9
17.8
13.5
15.8
26.9
17.4
17.8
24.5
22.4
61.6
10.5
89.1
46.8
26.3
70.8
66.1
23.4
37.2
37.2
31.6
32.3
38.0
52.5
28.2
38.0
45.7
47.9
93.3
26.9
79.4
43.6
56.2
10.7
10.9
8.13
10.0
15.8
18.2
10.0
16.6
23.4
14.8
17.8
20.9
31.6
19.5
17.4
100
46.8
Average
12.0
18.2
16.0
>100
17.4
36.3
35.5
53.7
aNumber of tested lines in brackets.
bDrug concentration inhibiting tumor cell growth by 50%.

M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
1419
Conclusion
as the catalyst, the mixture was re¯uxed for 24±36 h to
drive the reaction to completion. CaSO4 was ®ltered o€
and the ®ltrate was evaporated to dryness a€ording a
yellow±brown solid condensation product. A pure sam-
ple of the enamine compound was obtained by crystal-
lization from ethanol:ethyl acetate 1:1 (v/v) mixture.
Yield 94%; mp 114 ꢀC; Rf 0.8 (eluent a); 1H NMR
(acetone-d6): d 1.23 (t, 3H, J=7.1 Hz, ethyl CH3), 1.91
(s, 3H, CH3), 4.09 (q, 2H, J=7.1 Hz, CH2), 4.62 (s, 1H,
ˆCH-), 6.48 (m, 1H, HC-3), 6.95 (dd, J6,4=2.1 Hz and
J6,7=8.6 Hz, HC-6), 7.39 (m, 2H, HC-2 and HC-4),
7.44 (d, 1H, J7,6=8.6 Hz, HC-7), 10.38 (bs, 1H, NH).
In conclusion, among all tested compounds, 4a proved
to be the most valuable. Besides being e€ective against
leukemia cell lines, it appears to be remarkably active
against other usually poorly sensitive lines from solid
tumors like CNS-, melanoma- and prostate-derived
cells. To further assess its potential as a therapeutic
agent, compound 4a is presently undergoing in vivo
testing at the National Cancer Institute.
Experimental
Chemistry
The enamine crotonate derivative was suspended in
boiling phenyl ether (50 mL), heated to re¯ux for 30
min to produce the pyrrolo-quinoline, cooled to room
temperature and diluted with petroleum ether. The
resulting precipitate was collected and washed several
times with ethyl acetate and desiccated. Recrystallization
from 70% ethanol a€orded pure 1a. Yield 93%; mp
Melting points were determined with a Gallenkamp
MFB-595±010M Melting Point Apparatus, and are
uncorrected.
1H NMR spectra were obtained with a Varian Gemini
200 MHz spectrometer using the indicated solvents and
300 ꢀC dec.; Rf 0.54 (eluent b); H NMR (DMSO-d6):
1
2.35 (s, 3H, CH3), 5.95 (s, 1H, HC-8), 7.25 (d, 1H, J5,4
=
1
TMS as internal reference. The H NMR signals are
8.7 Hz, HC-5), 7.41 (m, 1H, HC-3), 7.50 (m, 1H, HC-2),
7.69 (d, 1H, J4,5=8.7 Hz, HC-4), 11.43 (bs, 1H, OH),
11.50 (bs, 1H, NH). Anal. calcd for C12H10N2O: 72.71
C, 5.09 H, 14.14 N; found: 71.97 C, 5.14 H, 14.52 N.
reported in parts per million (d ppm), and are char-
acterized as singlet (s), doublet (d), triplet (t), quartet
(q), or multiplet (m). In the case of multiplets, the quo-
ted chemical shift corresponds to the multiplet center.
Integrals corresponded satisfactorily to those expected
on the basis of compound structure. Coupling constants
are expressed in Hertz (Hz). Mass spectra were obtained
with a mass spectrometer Mat 112 Varian Mat Bremen.
3H-9-Hydroxy-pyrrolo[3,2-f]quinoline (1b). According
to a previously published procedure,9 5-aminoindole,
obtained as described above, and diethylethoxy methyl-
ene malonate were mixed together and heated slowly to
130 ꢀC. After 2±3 h, the reaction product ethanol was
evaporated under reduced pressure.
Elemental analyses were performed at the microanalysis
laboratory, Department of Pharmaceutical Sciences,
University of Padova, Italy, using a Perkin±Elmer ele-
mental analysis apparatus model 240B; results for C, H,
N, Cl and S were within ‹0.4% of the theoretical values.
Cooling to room temperature converted the semisolid
crude malonate into a compact solid. This was used
without further puri®cation in the following condensa-
tion reaction, accomplished adding portions of enamine
malonate derivative to boiling phenyl ether, heating to
re¯ux for 30 min to produce the pyrrolo-quinoline ethyl
ester, cooling to room temperature and ®ltering. The
collected precipitate was washed with ethyl acetate and
recrystallized from 70% ethanol. Yield 77%; mp 288±
1
IR spectra were recorded in cm with a Perkin±Elmer
1760FTIR spectrophotometer as potassium bromide
pressed disk.
Thin layer chromatography (TLC) was performed with
Merck silica gel F254 polystyrene-backed plates. Elu-
ents used in TLC analysis: (a) (ethyl acetate:n-hexane 7:3
v/v), (b) (methanol:n-hexane 9:1 v/v). Column ¯ash
chromatography was performed with Merck silica gel
(250±400 mesh ASTM).
290 ꢀC; Rf 0.68 (eluent b); H NMR (DMSO-d6): d 1.30
1
(t, 3H, J=7.1 Hz, CH3), 4.23 (q, 2H, J=7.1 Hz, CH2),
7.33 (d, 1H, J5,4=8.8 Hz, HC-5), 7.52 (m, 1H, HC-2),
7.58 (m, 1H, HC-1), 7.78 (d, 1H, J4,5=8.8 Hz, HC-4), 8.45
(s, 1H, HC-7), 11.58 (bs, 1H, OH), 12.2 (bs, 1H, NH).
3H-7-Methyl-9-hydroxy-pyrrolo[3,2-f]quinoline (1a). A
solution of 5-nitroindole (Aldrich) in absolute ethanol
was slowly added to a 5% palladium on charcoal etha-
nol suspension saturated with hydrogen and the mixture
was hydrogenated at atmospheric pressure at the tem-
perature of 40 ꢀC until the starting material disappeared
according to TLC analysis (eluent a). After 3±6 h the
mixture was ®ltered and the ®ltrate evaporated to dry-
ness. The residue (yield 98%) was used in the next con-
densation step without puri®cation.
A mixture of crude pyrroloquinoline ester was re¯uxed
in 2 N NaOH with stirring. After 2 h, the reaction mix-
ture was acidi®ed with acetic acid (AcOH), and the
resulting precipitate was ®ltered o€, washed with water,
and dried. The solid was recrystallized from dimethyl-
formamide (DMF) to yield 9-hydroxy-pyrrolo[3,2-f]-
quinoline-8-carboxylic acid. Yield 88%; mp 282±284 ꢀC
dec. (lit.7 mp 307±308 ꢀC); Rf 0.55 (eluent b); H NMR
1
(DMSO-d6): d 7.54 (m, 2H, HC-1 and HC-4), 7.66 (m,
1H, HC-2), 7.99 (d, 1H, J5,4=8.8 Hz, HC-5), 8.81 (s,
1H, HC-7), 11.91 (bs, 1H, OH), 12.3 (bs, 1H, NH).
Equimolar amounts of 5-aminoindole (11±12 mM) and
ethyl acetacetate were dissolved in 70 mL absolute eth-
anol. After adding 3±4 g of CaSO4 (drierite) and AcOH
Crude pyrrolo-quinoline acid was suspended in boiling
phenyl ether and allowed to re¯ux for 20 min. After

1420
M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
.
12.47 (s, 1H, indole NH). Anal. calcd for C18H16N4O2S
HCl: 55.66 C, 4.41 H, 14.43 N, 9.01 Cl, 8.24 S; found:
55.25 C, 4.32 H, 14.10 N, 9.29 Cl, 7.93 S.
cooling to room temperature, the crystalline product
was separated, ®ltered, washed with ethyl ether, and
recrystallized twice from ethanol:methanol mixture (1:1)
to produce 1b. Yield 87%; mp 320 ꢀC, dec.; Rf 0.83
(eluent b); 1H NMR (DMSO-d6): d 6.8 (d, 1H, J8,7=7.4
Hz, HC-8), 7.27 (d, 1H, J5,4=8.8 Hz, HC-5), 7.45 (t,
1H, J=2.7 Hz, HC-2), 7.55 (bs, 1H, HC-1), 7.73 (d, 1H,
J4,5=8.8 Hz, HC-4), 7.85 (d, 1H, J7,8=7.4 Hz, HC-7),
11.5 (bs, 1H, OH), 11.7 (bs, 1H, NH). Anal. calcd for
C11H8N2O: 71.73 C, 4.38 H, 15.21 N; found: 71.58 C,
4.30 H, 15.30 N.
3H-7-Methyl-9-(p-methanesulfonamido-aniline)-pyrrolo-
[3,2-f]quinoline hydrochloride (4b). The crude hydro-
chloride was dissolved in water and the solution was
made alkaline with 5% (w/v) NH4OH in water. The
organic product was extracted with ethyl acetate,
washed with water and dried over anhydrous sodium
sulfate. After the solution was evaporated in vacuo, the
resulting yellowish solid was puri®ed by ¯ash-chroma-
tography using silica gel as the solid support and eluted
with methanol:n-hexane mixture (9:1 v/v). A crystalline
yellow product was obtained, which, after a single crys-
tallization from methanol, gave analytically pure 3H-7-
methyl-9-(p-methanesulfonamido-aniline)-pyrrolo[3,2-f]-
quinoline. Mp 215 ꢀC; Rf 0.27 (eluent b); 1H NMR
(DMSO-d6): d 2.48 (s, 3H, CH3), 3.38 (s, 3H, CH3), 6.97
(s, 1H, HC-8), 7.15 (s, 1H, HC-1), 7.19 (d, 2H, J=9.1
Hz HC-30 and HC-50), 7.27 (d, 2H, HC-20 and HC-60),
7.46 (t, 1H, J=2.6 Hz, HC-2), 7.45 (d, 1H, J5,4=8.9 Hz,
HC-5), 7.77 (d, 1H, J4,5=8.9 Hz, HC-4), 8.03 (bs, 1H,
sulfonamide NH), 11.66 (bs, 1H, indole NH). Anal.
calcd for C19H18N4O2S: 62.28 C, 4.95 H, 15.30 N, 8.73
S; found: 62,26 C, 4.86 H, 15.67 N, 8.61 S.
9-Chloro-pyrrolo-quinolines (2a,b). 9-Hydroxy-pyrrolo-
quinolines 1a,b (6±7 mmol) were added to phosphorus
oxychloride (5 mL) and heated to 140 ꢀC for 1 h. Upon
completion of the reaction, as monitored by silica gel
thin layer chromatography, excess of phosphorus oxy-
chloride was evaporated and the mixture cooled. Then,
a cold 25% (w/v) NH4OH water solution was added
(pH 8) and the solid precipitate was ®ltered, washed
with water and dried.
3H-9-Chloro-pyrrolo[3,2-f[quinoline (2a). Recrystalliza-
tion from ethanol:ethyl acetate (1:1 v/v) a€orded pure
compound. Yield 83%; mp 235 ꢀC; Rf 0.86 (eluent b);
1H NMR (DMSO-d6): d 7.75 (m, 2H, HC-1 abd HC-2),
7.70 (d, 1H, J7,8=4.7 Hz, HC-7), 7.76 (d, 1H, J5,4=9
Hz, HC-5), 7.96 (d, 1H, J4,5=9 Hz, HC-4), 8.67 (d, 1H,
J8,7=4.7 Hz, HC-8), 12 (bs, 1H, NH). Anal. calcd for
C13H9ON2Cl: 63.82 C, 3.71 H, 11.45 N, 14.49 Cl;
found: 63.94 C, 3.68 H, 11.57 N, 14.54 Cl.
By treatment of an absolute ethanol solution of the
above compound with dry HCl, the desired hydrochloride
4b was obtained, which was recrystallized from metha-
nol. Yield 86%; mp 304 ꢀC. Anal. calcd for C19H18
.
N4O2ClS 0.5H2O: 55.54 C, 4.66 H, 13.64 N, 8.43 Cl,
7.80 S; found: 55.63 C, 4.71 H, 13.41 N, 8.83 Cl, 8.03 S.
3H-7-Methyl-9-chloro-pyrrolo[3,2-f]quinoline (2b). Dou-
ble recrystallization from ethanol a€orded the pure
compound. Yield 80%; mp 225 ꢀC dec.; Rf 0.48 (eluent
3H-9-(o-Methoxy-p-methanesulfonamido-aniline)-pyrrolo-
[3,2-f]quinoline (5a). Pure 5a was obtained as the
hydrochloride following the same procedure described
for 4b. Yield 75%; mp 280 ꢀC dec. Rf 0.29 (eluent b); 1H
NMR (DMSO-d6): d 3.12 (s, 3H, CH3), 3.76 (s, 3H,
1
a); H NMR (deuterio DMSO): d 2.38 (s, 3H, CH3),
7.36 (s, 1H, HC-1), 7.42 (sa, 1H, HC-2), 7.47 (s, 1H,
HC-8), 7.53 (1H, d, J5,4=8.4 Hz, HC-5), 7.78 (d, 1H,
J4,5=8.4 Hz, HC-5), 11.76 (bs, 1H, NH). Anal. calcd
for C12H7ON2Cl: 62.49 C, 3.06 H, 12.15 N, 15.37 Cl;
found 61.35 C, 3.25 H, 12.02 N, 15.24 Cl.
OCH3), 6.55 (d, 1H, J8,7=6.8 Hz, HC-8), 7.00 (dd, 1H,
0
0
0
0
0
J5 ,6 =8.3 Hz and J05 ,3 =2.1 Hz, HC-5 ), 7.10 (d, 1H,
0
0
0
J3 ,5 =2.1 Hz, HC-3 ), 7.45 (d, 1H, J6 ,5=8.3 Hz, HC-
60), 7.72 (m, 1H, HC-1), 7.77 (d, 1H, J5,4=8.9 Hz, HC-
5), 7.85 (t, 1H, J2,1=2.7 Hz, HC-2), 8.20 (d, 1H,
J4,5=8.9 Hz, HC-4), 8.40 (d, 1H, J7,8=6.9 Hz, HC-7),
8.96 (d, 1H, sulfonamide NH), 10.04 (d, 1H, anilino
NH), 12.45 (bs, 1H, indole NH).). Anal. calcd for
C19H19N4O3ClS: 53.33 C, 4.71 H, 13.09 N, 8.29 Cl, 7.49
S; found: 53.63 C, 4.41 H, 12.96 N, 8.64 Cl, 7.77 S.
9-Anilino-pyrrolo-quinoline hydrochlorides (4a,b and 5a,b).
A solution of 9-chloro-pyrrolo-quinolines 2a,b (1.5±
2 mmol) and an equimolar amount of p-methanesulfon-
amido-aniline or o-methoxy-p-methanesulfonamido-
aniline in methanol:water mixture (9:1 v/v) and in the
presence of 36% HCl as the catalyst, was re¯uxed until
completion as monitored by TLC (24±48 h). After con-
centration of methanolic solution and cooling to 4 ꢀC, a
yellow crystalline precipitate was formed, which was ®l-
tered, washed with water and cold ethanol and dried.
3H-7-Methyl-9-(o-methoxy-p-methanesulfonamido-ani-
line)-pyrrolo[3,2-f]quinoline hydrochloride (5b). Pure 5b
was obtained following the same procedure described
for 4b. Yield 78%; mp 290 ꢀC dec.; Rf 0.25 (eluent b);
1H NMR (DMSO-d6): d 2.63 (s, 3H, CH3), 3.12 (s, 3H,
3H-9-(p-Methanesulfonamido-aniline)-pyrrolo[3,2-f]quino-
line hydrochloride (4a). Double recrystallization from
methanol yielded the pure compound. Yield 78%; mp
CH3), 3.78 (s, 3H, OCH3), 6.44 (s, 1H, HC-8), 7.00 (dd,
300 ꢀC; Rf 0.46 (eluent b); H NMR (DMSO-d6): d 3.08
1H, J5 ,6 =8.3 Hz 0and J5 ,3 =2 Hz, HC-5 ), 7.10 (d, 1H,
1
0
0
0
0
0
0
0
0
0
0
(s, 1H, CH3), 6.85 (d, 1H, J8,7=6.8 Hz, HC-8), 7.39 (d,
2H, J=8.9 Hz, HC-20 and HC-60), 7.50 (d, 2H, J=8.5
Hz, HC-30and HC-50), 7.66 (bs, 1H, HC-1), 7.81 (d, 1H,
J5,4=9 Hz, HC-5), 7.84 (m, 1H, HC-2), 8.20 (d, 1H,
J4,5=9 Hz, HC-4), 8.43 (d, 1H, J7,8=7 Hz, HC-7), 9.39
(s, 1H, sulfonamide NH), 10.03 (s, 1H, amino NH),
J3 ,5 =2 Hz, HC-3 ), 7.45 (d, 1H, J6 ,5 =8.3 Hz, HC-6 ),
7.67 (bs, 1H, HC-1), 7.82 (s, 1H, HC-2), 7.83 (d, 1H,
J5,4=8.7 Hz, HC-5), 8.16 (d, 1H, J4,5=8.7 Hz, HC-4),
8.79 (s, 1H, sulfonamide NH), 10.08 (s, 1H, aniline
NH), 12.45 (s, 1H, indole NH). Anal. calcd for
.
C20H21N4O3ClS 0.5H2O: C 53.27, H 5.14, N 12.42, Cl

M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
1421
7.86, S 7.11; found: C 53.09, H 4.85, N 12.36, Cl 8.13, S
7.32.
3H-5,6-Dihydro-6-(40 -methanesulfonamido-benzen)-8-
methyl-indolo[3,3a,4,5-c,d,e][1,6]naphthyridine chloride
(6b). This compound was recrystallized several times
from an ethanol:methanol mixture (1:1 v/v). Yield 30%;
1-Formyl-pyrrolo-quinolines (3a,b). A solution of 9-
hydroxy-pyrrolo-quinolines 1a,b (3±4 mmol) in DMF
was dropped into a mixture of 2 mL POCl3 and 3 mL
DMF cooled at 10 ꢀC. Then heating was raised to 35 ꢀC
and continued with stirring until the starting material
was no longer detectable by TLC. The volatile products
were removed in vacuo and the residue diluted with
water/ice before alkalinization with 25% NH4OH (v/v).
The precipitate was ®ltered, washed with water and
dried.
Rf 0.21 (eluent b); IR (KBr): 3377, 2875, 1323±1152
1
cm
;
1H NMR (DMSO-d6): d 2.53 (s, 3H, CH3-8),
3.13 (s, 3H, CH3SO2), 5.45 (s, 2H, HC-5a and HC-5b),
6.07 (s, 1H, HC-7), 7.36 (s, 0 1H, HC-4), 7.43 (d, 2H,
0
0
0
0
0
J2 , 3 and 6 , 5 =8.9 Hz, HC-2 , HC-6 ), 7.44 (d, 1H,
0
0
0
0
J2,1=8.9 Hz, HC-2), 7.58 (d, 2H, J3 , 2 and 5 , 6 =8.9 Hz,
HC-30 and HC-50), 7.90 (d, 1H, J1,2=8.9 Hz, HC-1),
10.14 (bs, 1H, sulfonamide NH), 11.76 (bs, 1H, indole
NH); MS: 382 (M+), 301, 208, 180, 108, 92, 65. Anal.
.
calcd for C20 H18N4O2S 0.5H2O: 56.67 C, 4.76 H, 13.22
N, 8.36 Cl, 7.56 S; found: 57.00 C, 4.82 H, 13.20 N, 8.36
Cl, 7.73 S.
3H-9-Chloro-1-formyl-pyrrolo[3,2-f]quinoline (3a). This
was obtained in almost quantitative yield; mp 255 ꢀC
(from ethanol); Rf 0.35 (eluent a); IR (KBr): 3480, 3126,
2885, 1655 cm 1; H NMR (DMSO-d6): d 7.75 (d, 1H,
3H-5,6-Dihydro-6-(20 -methoxy-40 -methanesulfonamido-
1
J7,8=4.7 Hz, HC-7), 7.88 (d, 1H, J5,4=9.0 Hz, HC-5),
8.00 (d, 1H, J4,5=9.0 Hz, HC-4), 8.21 (s, 1H, HC-2),
8.75 (d, 1H, J8,7=4.7 Hz, HC-8), 10.50 (s, 1H, CHO).
benzen)-indolo[3,3a,4,5-c,d,e][1,6]naphthyridine
hydro-
chloride (7a). The compound was recrystallized from
ethanol; yield 40%; Rf 0.22 (eluent b); IR (KBr): 3300,
3227, 1639, 1333, 1278, 1142, 1031, 806 cm 1; 1H NMR
(DMSO-d6): d 3.15 (s, 3H, CH3), 3.78 (s, 3H, OCH3),
5.31 (d, 1H, J5b,5a=5.6 Hz, HC-5b), 5.38 (d, 1H,
J5a,5b=5.6 Hz, HC-5a), 6.08 (d, 1H, J7,8=6.9 Hz, HC-
3H-9-Chloro-7-methyl-1-formyl-pyrrolo[3,2-f]quinoline
(3b). Yield 76%; mp 165±168 ꢀC; Rf 0.33 (eluent b); IR
(KBr): 3420, 2915, 2875, 1645 cm 1; 1H NMR (DMSO-
d6): d 2.69 (s, 1H, CH3), 7.59 (s, 1H, HC-8), 7.83 (d, 1H,
J5,4=9 Hz, HC-5), 8.00 (d, 1H, J4,5=9 Hz, HC-4), 8.11
(d, 1H, J2,NH=3.1 Hz, HC-2), 10.63 (s, 1H, CHO).
0
0
0
0
7), 7.02 (dd, 1H, J5 ,3 =2.2 Hz and J5 ,6 =8.4 Hz, HC-
0
50), 7.1 (d, 1H, J3 ,5 =2.2 Hz, HC-3 ), 7.38 (d, 1H,
J4,3=1.6 Hz, HC-4), 7.48 (d, 1H, J1,2=8.9 Hz, HC-1),
0
0
0
0
0
7.55 (d, 1H, J6 ,5 =8.4 Hz, HC-6 ), 7.95 (d, 1H, J2,1=8.9
Hz, HC-2), 8.26 (d, 1H, J8,7=6.9 Hz, HC-8), 10.18 (bs,
1H, sulfonamide NH), 11.80 (bs, 1H, indole NH); MS:
394 (M+), 364, 315, 285, 194, 186, 79.64. Anal. calcd
for C20H19N4O3ClS H2O: 53.51 C, 4.71 H, 12.48 N,
7.90 Cl, 7.14 S; found: 53.36 C, 4.91 H, 12.67 N, 7.60 Cl,
7.32 S.
Indolo-naphthyridine chlorides 6a,b and 7a,b. A metha-
nolic solution of equimolar amounts of 1-formyl-pyrrolo-
quinolines 3a,b (1±1.5 mmol) and of p-methane-sulfon-
amido-aniline or o-methoxy-p-methanesulfonamido-
aniline was acidi®ed to pH=6 with glacial acetic acid.
Stirring was continued at room temperature until the
starting compounds were no longer detectable by TLC,
while the iminium salt (Rf 0.18, eluent b) appeared as a
new ¯uorescent band. Then, NaBH3CN (slight molar
excess) dissolved in 3 mL methanol was added to the
solution, and the reduction reaction was carried out at
room temperature with stirring until completion (30±48
h, TLC monitoring). On partial evaporation of the sol-
vent and cooling at 4 ꢀC, a crystalline product pre-
cipitated from the reaction mixture. The precipitate was
®ltered, washed with methanol and dried yielding a pure
product.
.
3H-5,6-Dihydro-6-(20 -methoxy-40 -methanesulfonamido-
benzen)-8-methyl-indolo[3,3a,4,5-c,d,e][1,6]naphthyridine
chloride (7b). This compound was recrystallized from
absolute ethanol; yield 45%; Rf 0.2 (eluent b); IR (KBr):
3400, 3267, 2925, 2855, 1639, 1323, 1147, 1027, 800
1
cm 1; H NMR (DMSO-d6): d 2.51 (s, 3H, CH3), 3.15
(s, 3H, CH3SO2), 3.78 (s, 3H, OCH3), 5.29 (d, 1H,
J5b,5a=3.96 Hz, HC-5b), 5.32 (d, 1H, J5a,5b=3.96 Hz,
0
0
HC-5a), 5.95 (s, 1H, HC-8), 0 6.99 (dd, 1H, J5 ,6 =8.68
0
0
0
0
Hz and J5 ,3 =2.12 Hz, HC-5 ), 7.10 (d, 1H, J3 ,5 =2.12
Hz, HC-30), 7.33 (ss, 1H, J4,3=1.54 Hz, HC-4), 7.41 (d,
1H, J1,2=8.67 Hz, HC-1), 7.50 (d, 1H, J2,1=8.67 Hz,
3H-5,6-Dihydro-6-(40-methanesulfonamido-benzen)-indolo-
[3,3a,4,5-c,d,e][1,6]naphthyridine hydrochloride (6a).
Double recrystallization from ethanol:methanol mix-
ture (1:1 v/v) a€orded the pure compound. Yield 50%;
Rf 0.18 (eluent b); IR (KBr): 3350, 3026, 2875, 1639,
0
0
0
HC-2), 7.88 (1H, d, J6 ,5 =8.68 Hz, HC-6 ), 10.2 (bs,
1H, sulfonamide NH), 11.7 (bs, 1H, indole NH); MS:
408 (M+), 329, 208, 200, 179, 121, 79, 64. Anal. calcd
.
for C21H21N4O3ClS H2O: 54.48 C, 5.01 H, 12.10 N,
7.66 Cl, S 6.93; found: 54.76 C, 4.91 H, 11.97 N, 7.50 Cl,
7.32 S.
1
1328±1152 cm 1; H NMR (DMSO-d6): d 3.11 (s, 3H,
CH3), 5.44 (s, 2H, CH-5a and CH-5b), 6.16 (d, 1H,
J7,8=6.5 Hz, HC-7), 7.31 (s, 1H, HC-4), 7.42 (d, 2H,
0
1H, J2,1=8.9 Hz, HC-2), 7.46 (d, 1H, J1,2=8.9 Hz, HC-
0
0
J2 ,3 =8.8 Hz and J6 ,5 =8.8 Hz, HC-2 , HC-6 ), 7.86 (d,
00
0
0
N-(2-Methoxy-4-nitro-benzen)-ethylcarbamate. To
a
solution of commercial 2-methoxy-4-nitro-aniline (2 g,
12 mM) in 50 mL dry THF, 1.4 mL (12 mmol) of
ethylchloroformate was added. The mixture was
re¯uxed until the starting compound disappeared
according to TLC analysis (eluent a). The yellow solu-
tion was evaporated to dryness giving a crystalline yel-
low product. Yield 95%; mp 154±156 ꢀC. Rf 0.3 (eluent
0
0
00
0
0
1), 7.58 (d, 2H, J3 ,2 =8.8 Hz and J5 , 6 =8.8 Hz, HC-3 ,
HC-50), 8.25 (d, 1H, J8,7=6.5 Hz, HC-8), 10.1 (bs, 1H,
sulfonamide NH), 11.65 (bs, 1H, indole NH); MS: 364
(M+), 285, 166,79,76. Anal. calcd for C19H16N4
.
O2S HCl: 56.93 C, 4.27 H, 13.98 N, 8.84 Cl, 8.00 S;
found: 56.57 C, 4.36 H, 13.96 N, 8.60 Cl, 7.79 S.

1422
M. G. Ferlin et al. / Bioorg. Med. Chem. 8 (2000) 1415±1422
a). 1H NMR (acetone-d6): d 1.29 (t, 3H, J=7.1 Hz,
CH3), 4.06 (s, 3H, OCH3), 4.23 (q, 2H, J=7.1 Hz,
CH2), 7.82 (d, 1H, J3,5=2.6 Hz, HC-3), 7.92 (dd, 1H,
J5,3=2.6 and J5,6=9.2 Hz, HC-5), 8.21 (sa, 1H, NH),
8.31 (d, 1H, J6,5=9.2 Hz, HC-6).
of human topoisomerase I (TopoGEN, Inc., Columbus,
OH) in the presence of the indicated amount of drugs.
After 30 min the reaction (®nal volume 20 mL) was
stopped by 1% SDS, and directly loaded on a 0.8%
agarose gel in Tris±borate 50 mM, EDTA 1 mM
and ethidium bromide 0.5 mg/mL, and run at 5 V/cm
for 4 h. The e€ects of drugs were examined up to 100
mM concentration.
N-(2-Methoxy-4-amino-benzen)-ethylcarbamate. Easily
air-oxidable intermediate. Yield 90%. Rf 0.6 (n-hexane:
ethyl acetate 1:1 v/v)
Antitumor screen
N-(2-Methoxy-4-methanesulfonamido-benzen)-ethylcarb-
amate. Yield 84%; mp 139±141 ꢀC; Rf 0.75 (eluent a).
1H NMR (acetone-d6): d 1.26 (t, 3H, J=7 Hz, CH3),
2.96 (s, 3H, SO2CH3), 3.88 (s, 3H, OCH3), 4.16 (q, 2H,
J=7 Hz, CH2), 6.9 (dd, 1H, J5,3=2.5 Hz and J5,6=8.7
Hz, HC-5), 7.04 (d, 1H, J3,5=2.5 Hz, HC-3), 7.58 (sa,
1H, urethane NH), 7.94 (d, 1H, J6,5=8.6 Hz, HC-6), 8.4
(sa, 1H, amide NH).
The test derivatives were submitted to an in vitro anti-
cancer testing at NCI.14 The cell panel consists of 60
lines against which compounds are tested at a minimum
of ®ve concentrations at 10-fold dilutions. A 48 h con-
tinuous drug exposure protocol is used, and a sulforho-
damine protein assay is used to estimate cell viability or
growth. Plot studies provide the basis for the detection
of patterns of interest.
2-Methoxy-4-methanesulfonamido-aniline.12 A solution
of the urethane derivative in aqueous 20% NaOH was
re¯uxed as appropriate (TLC monitoring). After cool-
ing at room temperature, the reaction mixture was
extracted with ethyl ether and the organic layer was
separated, dried over Na2SO4, and evaporated in vacuo
to dryness to give the desired aniline.
References
1. Hande, K. R. Biochim. Biophys. Acta 1998, 1400, 173.
2. Liu, L. F. Annu. Rev. Biochem. 1989, 58, 351.
3. Pommier, Y. Cancer Chemother. Pharmacol. 1993, 32, 103.
4. Berger, J. M.; Gamblin, S. J.; Harrison, S. C.; Wang, J. C.
Nature 1996, 379, 225.
5. Redinbo, M. R.; Stewart, L.; Kuhn, P.; Champoux, J. J.;
Hol, W. G. Science 1998, 279, 1504.
Topoisomerase II cleavage assay
The plasmid pBR322 was cut at the EcoR1 site, and
labeled by ®ll-in with the Large (Klenow) Fragment of
DNA Polymerase I and [a-32P]dATP. 0.1 mg of DNA
were reacted with the indicated concentrations of drugs
and 2 units of human topoisomerase II-a (TopoGEN,
Inc., Columbus, OH) at 37 ꢀC in 50 mM Tris±HCl (pH
8.0), 10 mM MgCl2, 120 mM KCl, 0.5 mM DTT, 30 mg/
mL BSA and 1 mM ATP. Following 30 min of incu-
bation at 37 ꢀC, samples were stopped by incubation
with 0.5 mg/mL proteinase K and 1% SDS at 60 ꢀC for
3 h, and then loaded on 1% agarose gel (0.09 M Tris±
borate, pH 8.3, 2.5 mM EDTA) and run for 16 h at 40
Volts. The gel was then dried and autoradiographed.
The e€ects of drugs were examined up to 100 mM
concentration.
6. Fan, Y.; Weinstein, J. N.; Kohn, K. W.; Shi, L. M.; Pom-
mier, Y. J. Med. Chem. 1998, 41, 2216.
7. Gatto, B.; Capranico, G.; Palumbo, M. Curr. Pharm. Des.
1999, 5, 195.
8. Capranico, G.; Guano, F.; Moro, S.; Zagotto, G.; Sissi, C.;
Gatto, B.; Zunino, F.; Menta, E.; Palumbo, M. J. Biol. Chem.
1998, 273, 12732.
9. Conrad, M.; Limpach, L. Ber. 1887, 20, 944 and 1891, 24,
2990.
10. Gould, R. G.; Jacobs, W. A. J. Am. Chem. Soc. 1939, 61,
2890.
11. Albrecht, R. Liebigs Ann. Chem. 1972, 762, 55.
12. Brennan, S. R.; Colbry, M. L.; Leeds, R. L.; Leja, B.;
Priele, S. R.; Reily, M. D.; Showalter, H. D. H.; Uhlendorf, S.
E.; Atwell, G. J.; Denny, W. A. J. Heter. Chem. 1989, 26,
1469.
13. Cornarotti, M.; Tinelli, S.; Willmore, E.; Zunino, F.;
Fisher, L. M.; Austin, C. A.; Capranico, G. Mol. Pharmacol.
1996, 50, 1463.
Topoisomerase I relaxation assay
14. Christian, M. C.; Pluda, J. M.; Ho, P. T.; Arbuck, S. G.;
Murgo, A. J.; Sausville, E. A. Semin. Oncol. 1997, 24, 219.
15. Boyd, M. R. In Cancer: Principles and Practice of Oncol-
ogy; DeVita, V. T., Jr.; Hellman, S.; Rosenberg, S. A., Eds.;
Lippincott: Philadelphia, PA, 1989; Vol. 3, pp 1±12.
Topoisomerase I catalytic activity assay was performed
by incubating 0.25 mg of negative supercoiled pBR322
at 37 ꢀC in 10 mM Tris±HCl, pH 7.5, NaCl 100 mM, b-
mercaptoethanol 1 mM and 1 mM PMSF, with 1 unit