Photochemical electrocyclisation of 3-vinylindoles to pyrido[2,3-a]-, pyrido[4,3-a]- and thieno[2,3-a]-carbazoles: Design, synthesis, DNA binding and antitumor cell cytotoxicity

Article abstract of DOI:10.1016/j.ejmech.2009.03.026

In the context of the design and synthesis of DNA ligands, some new hetarene annelated carbazoles were synthesized. As lead structure the intercalating tetracyclic systems pyrido[2,3-a]- and pyrido[4,3-a]-carbazoles and in one case a thieno[2,3-a]-carbazo

Full text of DOI:10.1016/j.ejmech.2009.03.026

European Journal of Medicinal Chemistry 44 (2009) 3235–3252
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Photochemical electrocyclisation of 3-vinylindoles to pyrido[2,3-a]-,
pyrido[4,3-a]- and thieno[2,3-a]-carbazoles: Design, synthesis,
DNA binding and antitumor cell cytotoxicity
,
a
b
,
,
,c
Thomas Lemster a , Ulf Pindur , Gaelle Lenglet c, Sabine Depauw b c, Christelle Dassi b
,
*
b,c,
**
´ `
Marie-Helene David-Cordonnier
a Institute of Pharmacy, Department of Pharmaceutical and Medicinal Chemistry, Johannes Gutenberg-University, Staudingerweg 5, D-55099 Mainz, Rheinland-Pfalz, Germany
b INSERM U837-JPARC (Jean-Pierre Aubert Research Center), Team ‘‘Molecular and Cellular Targetting for Cancer Treatment’’, Institut de Recherches sur le Cancer de Lille,
Place de Verdun, F 59045 Lille Cedex, France
c
´
´ ´
Universite de Lille 2, IMPRT-Institut Federatif de Recherche 114, Lille, France
a r t i c l e i n f o
a b s t r a c t
Article history:
In the context of the design and synthesis of DNA ligands, some new hetarene annelated carbazoles were
synthesized. As lead structure the intercalating tetracyclic systems pyrido[2,3-a]- and pyrido[4,3-a]-
carbazoles and in one case a thieno[2,3-a]-carbazole were taken into account. A dialkyl amino amidic
chain was introduced to the planar chromophoric system with the intent to generate minor groove
binding properties. The cytotoxicity of some compounds was examined by the NCI antitumor screening.
Furthermore, biophysical as well as biochemical studies were performed in order to get some infor-
mation about the DNA-binding properties and inhibition of DNA related functional enzymes of this new
series of molecules.
Received 17 October 2008
Received in revised form
13 March 2009
Accepted 19 March 2009
Available online 27 March 2009
Keywords:
DNA ligands
Pyridocarbazoles
DNA intercalation
DNA minor groove binding
Crown Copyright Ó 2009 Published by Elsevier Masson SAS. All rights reserved.
1. Introduction
were carried out with compounds bearing an angled pyr-
idocarbazole anellation [6,11]. Thus, in continuation of our
Since the discovery of the natural products ellipticine 1 and its
regioisomeric olivacine 2 annelated indole and carbazole deriva-
tives with pyrido[4,3-b]carbazole framework constitute an inter-
esting class of antitumor active drugs [1–4]. Whereas ellipticine 1
as well as 9-methoxyellipticine 3 shows activity against a variety of
human tumor cell lines, especially against leukaemia, the quarter-
nary pyridinium salt ellipticinium acetate (CelliptiumÒ) 4 was
deployed against metastatic breast cancer [3,5]. The mechanism of
action in vivo of ellipticine and variants via DNA-intercalation
process, bioactivation due to oxidation at C-9 and indirect topo-
isomerase II-inhibition is well described [4,6]. In this context many
derivatives based on the linear annelated carbazole ellipticine and
olivacine were synthesized for the development of new DNA-
intercalating anticancer drugs [2,3,7]. However, only a few studies
synthetic efforts in carbazole and carbazole alkaloid chemistry
[12,13], we studied new short and flexible routes to some hetarene-
[a]-annelated carbazoles of type 10. Moreover, derived from some
literature reports [3,7,14] and from our own results in the netropsin
series [15–19], the selective introduction of a dimethylaminoalkyl-
carboxamide group at the coplanaric chromophoric system should
be of significant importance for a selective DNA binding and anti-
tumor activity. For example DACA (N-[2-(dimethylamino)-ethyl]-
acridine-4-carboxamide) 5 is a dual topoisomerase I/II inhibitor and
DNA intercalator with potent cytotoxicity against leukaemia cell
lines. Its carboxamide group is essential to the biological activity
[14]. The olivacine derivatives 6 (n ¼ 2: S16020-2) and the
g-car-
boline 7 are further examples of drugs which bear a dimethylami-
noalkylamino functionality as an important DNA-binding structure
element which increases the cytotoxicity and antitumor activity
(Scheme 1) [3,7–10].
On the background of these findings we describe in the present
paper the preparation of new pyrido- and thieno-[a]-annelated
carbazoles 10 bearing a dimethylaminoalkyl-carboxamide moiety at
the carbazole ring C. In one case a morpholino moiety as cyclic
* Corresponding author.
** Corresponding author.
E-mail addresses: lemster@uni-mainz.de (T. Lemster), marie-helene.david@
inserm.fr (M-H. David-Cordonnier).
0223-5234/$ – see front matter Crown Copyright Ó 2009 Published by Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.03.026

3236
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
H3C
H3C
N
N
N
N
H
H
CH3
CH3
Ellipticine
Olivacine
1
2
CH3
H3C
H3C
H3CO
HO
N
N
H3CCOO
N
N
H
H
CH3
CH3
®
9-Methoxyellipticine
Celliptium
4
3
H
N
O
N(CH3)2
N
DACA 5
N(CH3)2
H
HO
N(CH3)2
N
(CH2)n
O
HN
HO
N
N
n = 2,3
N
N
CH3
CH3
CH3
CH3
7
6
Scheme 1. Lead compounds: examples of annelated N-heterocycles as intercalating antitumor active drugs.
structural element was introduced. These products can be consid-
ered formally as regioisomeric and/or bioisosteric ellipticine deriv-
atives combining an appropriate carboxamide structural element.
6p-electrocyclisation of 2-phenyl-3-vinylindoles has been estab-
lished [20]. This convenient method has now been transferred to
produce new pyrido- and thieno[a]carbazoles 8, 9 as chromophoric
systems starting from the corresponding 2-hetarene-3-methyl-
The synthetical key step comprises a photochemically induced 6p-
electrocyclisation, a convenient procedure derived from synthetic
developments in our group[13,20]. For the establishmentof primary
structure–activity relationships (DNA-binding cyctotoxicity) of
these new functionalised carbazoles some biophysical, biochemical
and cell biological studies were performed.
acrylate substituted indoles. Preorientating DTm-value measure-
mentsof basic compounds8 and 9 [18–20] exhibit no DNA bindingat
all. Thus, in relation to compounds 5–7 an amidic chain as potential
binding element into the minor groove of DNA has been introduced
(10) (Scheme 2). As already described above, a dimethylamino-
ethylamino- and the homologue dimethylamino-propylamino-
function approved to have an optimal distance between the conju-
2. Chemistry
gated
p-system and the tertiary aliphatic nitrogen as potential H-
In former studies in our group, the synthesis of new ben-
zo[a]-annelated carbazoles via photochemically induced
bond acceptor. Moreover the basic substituted carboxamide func-
tion comprises additional DNA-binding element (groove binding).

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3237
O
(CH2)n NR2´
R´=
R
COOCH3
COOCH3
N
H
N(CH3)2
N
H
N
H
N
H
O
N
S
N
n = 2, 3
R = H, 2-Cl, 4-Cl
8
9
= 2-, 4-pyridyl
2-thienyl
10
Scheme 2. Frameworks of the synthesized compounds (8, 9), general formula of the [a]-annelated carbazole amides (10).
The access to the new compounds 10 is displayed exemplarily in
Scheme 3. The synthetic sequence starts with the synthesis of the
corresponding hydrazones 13 readily available from the reaction of
phenylhydrazines 11 with 2- and 4-acetylpyridine 12 in very good
yields. Subsequent Fischer cyclization of 13 gave rise to indoles 14
in the presence of polyphosphoric acid in all cases. The obtained
indoles were formylated according to Vilsmeier–Haak reaction
using phosphoryl chloride in DMF. The resulting carboxaldehydes
15 were converted stereoselectively into the (E)-methylacrylates 16
via Wittig reaction in good yield (46–76%). In the following key step
reaction the 2-hetaryl-3-vinylindoles 16 were submitted to ultra-
program against 60 human tumor cell lines. In the main screening,
the sixty human tumor cell lines derived from the nine cancer
types, leukaemia, non-small cell lung cancer, colon cancer, CNS
cancer, melanoma, ovarian cancer, renal cancer, prostate cancer and
breast cancer were used. The compounds’ dose–response curves for
each cell line were measured at a minimum of five concentrations
at 10-fold dilutions in a protocol of 48 h continuous drug exposure,
and a sulfurhodamine B (SRB) protein assay was used to estimate
cell viability or growth. The concentration causing 50% cell growth
inhibition (GI50), total cell growth inhibition (TGI, 0% growth) and
50% cell death (LC50, ꢀ50% growth) compared with the control was
calculated. In general, log10 GI50 values (the molar concentration of
the drug resulting in inhibition of cell growth to 50% of control)
were used for comparative discussion.
violet irradiation (
directly the desirable [a]-annelated carbazole derivatives 17 via 6
l
¼ 200–600 nm) for a period of 4–8 h achieving
p-
electrocyclisation and subsequent dehydrogenation. Standard
hydrolysis in 20% aqueous sodium hydroxide/ethanol gave rise to
the carboxylic acid 18 completely. In the last step the introduction
of the dimethylaminoalkyl linker succeeded by reaction of the free
acid 18 with N,N-dimethylalkylene diamine in DMF with EDCI/
HOBT as catalysts (Scheme 3). The amides 19 were obtained in good
yields. Thus, for fine tuning developments and for establishing
structure–activity relationships (DNA binding, topoisomerases
inhibition, cytotoxicity) structural variations were performed. In
one case the tertiary amine function was implemented in mor-
pholine ring system (29).
The best mean graph midpoint value MG_MID (activity is
defined as a log10 GI50 of < 100 mM), which displays an averaged
activity parameter over all cell lines, is ꢀ5.6 for the chloro-
substituted compound 22. The unsubstituted pyridocarbazoles
display values of ꢀ5.2 for 20 and ꢀ4.9 for 21 (Table 2). From these
results, it can be concluded that a C-2 side chain gives preference to
the C-3 and furthermore the chloro-derivative is more active
probably as a consequence of its higher lipophilicity.
It should be mentioned that compound 20 shows high potency
against an ovarian (OVCAR-3; GI50 [M]: 2.54 ꢁ10ꢀ8, log10GI50
:
The chloro derivatives 22–27 were synthesized in order to
evaluate the influence of increased polar hydrophobicity and
decreased density of electrons in the chromophoric structure.
Moreover the exchange of the pyridine ring against a more lipo-
ꢀ7.60) and a renal (CAKI-1; GI50 [M]: 5.15 ꢁ10ꢀ7, log10GI50: ꢀ6.28)
cell line suggesting that some antitumor-type cell selectivities can
be adjudged to this compound.
In order to compare the whole series, the cytotoxic effect was
measured on HT-29 cell line treated with increasing concentrations
of the various compounds. The corresponding GI50 are presented in
Table 3 and evidenced that compounds 24, 25 (and to a lesser
extend compounds 22, and 27) are the most cytotoxic one with GI50
values in the micromolar range. All four compounds are chlorine-
substituted molecules. Particularly, the pyrido-[4,3a]carbazoles 24
and 25 are more cytotoxic than their respective pyr-
ido[2,3a]carbazole derivatives 22 and 23. Surprisingly, in the
dimethylamino-propyl series, the 10-chloro-substituted compound
philic and more p-electron rich thiophene ring (see compound 28
in Table 1) should expand the spectrum of structure–activity rela-
tionship. Furthermore in the planar system the position isomeric
pyrido[4,3-a]carbazoles 24, 25 with a localized more peripheral
nitrogen atom in the planar system were synthesized. These
compounds should expand the activity pattern in comparison to
the more lipophilic pyrido[2,3-a]carbazoles 20–23 relating to their
biological/biophysical activities. The studies with the thiophene
derivative 28 on DNA binding and topoisomerase inhibition
demonstrated no significant advantage in comparison to the pyr-
ido-derivatives, thus in the thiophene series no further synthetic
variations were pursued so far (Table 1).
27 is more cytotoxic (GI50 ¼ 4.5
derivative 23 (GI50 ¼ 9.2 M) whereas the opposite observation is
made when the 8- or 10-chloro-substitution are present in the
dimethylamino-ethyl-substituted series 26 (GI50 ¼ 14 M) and 22
(GI50 ¼ 3.2 M).
mM) than its 8-chloro-substituted
m
m
m
3. Biological and biophysical/biochemical methods
3.1. Cytotoxicity measurements
3.2. Cell cycle effect of compounds 20–29
Due to restrictive criteria only three pyridocarbazole carbox-
amides were selected so far by the US National Cancer Institute [21]
for evaluation in the in vitro preclinical antitumor screening
The effect of compounds 20–29 on cell cycle was then assessed
on HT-29 colon carcinoma cells. Those cells were chosen based on
the interesting result obtained on cytotoxicity measurements

3238
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
R1
R1
R1
5
R2
CH3
O
(EtOH)
- H2O
PPA
R2
NH2
+
N
R2
2
N
H
N
H
N
H
7
- NH3
CH
3
11
12
13
14
11, 11, 13
R1
a
b
c
d
e
H
4-Cl
4-Cl
2-Cl
H
N
N
N
N
R2
S
POCl3
DMF
R1
O
R1
R1
COOCH3
COOCH3
(H5C6)3P=CHCOOCH3
(toluene)
H
.
h
R2
R2
HET
N
H
N
N
H
H
16
15
14, 15, 16
R1
a
b
c
d
e
H
5-Cl
5-Cl
7-Cl
H
N
N
N
N
R2
S
- 2H
R3
O
R1
R1
R1
R3
+ H2N
(CH2)2
COOH
COOCH3
N
H
8
NaOH
5
EDCI, HOBT
(DMF)
(H2O / C2H5OH)
HET
HET
HET
N
H
N
10
N
H
H
18
17
19
17, 18, 19
R1
a
b
c
d
e
H
8-Cl
8-Cl
10-Cl
H
HET
R3
N
S
N
N
N
O
N(CH3)2
H2C
N
N(CH3)2
Scheme 3. General synthesic route to hetarene [a]-annelated carbazole amides 19 (for the formula of the final products 20–29 see Table 1).

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3239
Table 1
Table 3
The final products.
HT-29 cells cytotoxicity of the various compounds measured using MTS assays.
R
R0
Compound
Compound
GI50 (mM)
–H
–N(CH3)2
20
20
21
22
23
24
25
26
27
28
29
11.5 ꢂ 3.4
10.1 ꢂ 1.4
3.20 ꢂ 1.25
9.23 ꢂ 3.12
1.08 ꢂ 0.28
2.20 ꢂ 1.07
14.0 ꢂ 2.3
4.49 ꢂ 1.48
10.8 ꢂ 0.5
ND
´
R
O
–H
–H2C–N(CH3)2 21
–N(CH3)2 22
–H2C–N(CH3)2 23
26
–10-Cl –H2C–N(CH3)2 27
–8-Cl
–8-Cl
N
H
8
R
–10-Cl –N(CH3)2
10
N
H
N
O
N
–10-Cl
29
24
–8-Cl
–8-Cl
–N(CH3)2
–H2C–N(CH3)2 25
´
R
O
N
8
H
phases. No differences were observed depending on the length of
the side chain (compounds 20 and 22 versus 21 and 23).
Using compound 24, cells accumulate in G2/M phases and then
in S phase. The cell cycle is less affected by treatment using
compounds 25–27 with a much weaker accumulation in G2/M
phase.
R
N
H
10
N
´
R
O
N
H
R
–H
–H2C–N(CH3)2 28
.
HCl
N
H
S
(Tables 2 and 3). Results in Fig. 1 evidenced a strong accumulation
of cells in G2/M phases using the unsubstituted pyridocarbazoles
20 and 21 with a very strong reduction of cells in G1 phase. By
contrast, cells treated with the chloro-pyridocarbazole compounds
22 and 23 first accumulate in G2/M phase and then in G1 phase. The
increase in G2/M phase also correlate with a large decrease of the
G1 phase and a smaller increase in the S phase portion from
20.4 ꢂ1.3 to 32.6 ꢂ 0.3 and 31.7 ꢂ 3.4 for 25
compounds 20 and 21, respectively, and to 37.0 ꢂ 1.5 and 26.1 ꢂ0.8
for 10 M of chloro-substituted pyridocarbazoles 22 and 23. At
higher concentrations (25 M), the G2/M phase arrest observed
mM of unsubstituted
m
m
using the chloro-pyridocarbazoles 22 and 23 is reduced with cells
accumulating in G1 phase whereas using the unsubstituted pyr-
idocarbazoles 20 and 21 cells stay in a large majority in G2/M
Table 2
Some results of the 60 panel NCI antitumor screening. GI50: molar concentration of
compounds, which induces inhibition of cell proliferation of 50%. MG_MID: mean
graph midpoint, activity is defined as log10G50 of <100 mM.
Compound
Tumor cell line
GI50 [M]
log10GI50
MG_MID
20
OVCAR-3 (ovarian)
CAKI-1 (renal)
IGROV-1 (ovarian)
UO-31 (renal)
2.54 ꢁ 10ꢀ8
5.15 ꢁ 10ꢀ7
1.14 ꢁ 10ꢀ6
2.13 ꢁ 10ꢀ6
5.27 ꢁ 10ꢀ6
ꢀ7.60
ꢀ6.28
ꢀ5.96
ꢀ5.68
ꢀ5.28
HT-29 (colon)
ꢀ5.2
ꢀ4.9
ꢀ5.6
21
22
CCRF-CEM (leukaemia)
COLO 205 (colon)
ACHN (renal)
4.31 ꢁ 10ꢀ6
4.43 ꢁ 10ꢀ6
5.24 ꢁ 10ꢀ6
1.03 ꢁ 10ꢀ5
ꢀ5.37
ꢀ5.35
ꢀ5.28
ꢀ5.99
HT-29 (colon)
UO-31 (renal)
1.10 ꢁ 10ꢀ6
1.34 ꢁ 10ꢀ6
1.47 ꢁ 10ꢀ6
2.33 ꢁ 10ꢀ6
ꢀ5.96
ꢀ5.87
ꢀ5.83
ꢀ5.63
SK-MEL-5 (melanoma)
SW-620 (colon)
HT-29 (colon)
Fig. 1. Cell cycle effect of compounds 20–23 on HT-29 cells. HT-29 cells in exponential
growth were exposed to the indicated concentrations of compounds 20–23 (mM) for
24 h prior to be fixed and labelled with PI for flow cytometric analysis. The various cell
cycle phases and sub-G1 portion are localized in the first panel.

3240
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
1
0.75
0.5
0.25
0
20
21
250
250
250
250
300
300
300
300
350
350
350
350
400
250
250
250
250
300
300
300
300
350
350
350
350
400
2
23
22
1.5
1
0.5
0
400
400
1
0.75
0.5
0.25
0
24
25
400
400
1
27
26
0.75
0.5
0.25
0
400
400
1
0.75
0.5
28
29
0.25
0
250
300 350
wavelength (nm)
400
250
300 350
wavelength (nm)
400
Fig. 2. UV/vis absorption spectra of pyrido[2,3-a]carbazoles in the absence (upper line) or presence of calf-thymus DNA (phosphate/drug ratio from 1 to 20). Arrows Y and / refer
to hypochromic and bathochromic effects, respectively.

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3241
Table 4
Quantification of hypochromic and bathochromic effects.
a good DNA binding of the ligand. Table 5 represents the
of this series of the new compounds.
D
Tm data
Compound
l
1a (nm)
l
2b (nm)
Dlc (nm)
Hd
%
Derived from these data, all compounds but compound 29
reflect a significant affinity to the poly(dAdT)2-DNA sequence ( Tm
values > 18 ꢃC). Comparison between
Tm values obtained from
D
20
21
22
23
24
287
287
288
288
254
281
305
254
281
305
286
250
286
276
284
293
292
298
297
þ6
þ5
þ10
þ9
62
56
48
51
56
62
37
56
65
25
46
D
unsubstituted (20, 21) and chloro-substituted pyridocarbazoles
(22, 23) clearly reveals the increase of DNA binding caused by the
halogen atom. Both DTm values of poly(dAdT)2 and CT-DNA confirm
this structure–activity relationship. It is mentioned that the chloro
derivatives possess a higher DNA-binding affinity as the leading
substance ellipticine (1), probably caused by their polar hydro-
phobicity. Concerning the length of the amidic chain, no obvious
preference could be noticed from comparison of the relative affinity
for DNA of compounds 20, 22 and 24 versus 21, 23 and 25,
respectively. However, comparison of the relative affinity for CT-
DNA or poly(dAdT)2 of compounds 26 and 27 evidenced an increase
in the DNA affinity in the presence of a dimethylamino-propyl
chain (27) instead of a dimethylamino-ethyl one (26). For this latter,
the reduced DNA affinity could also be attributed to the orientation
of the chlorine on the carbazole ring. Indeed, comparison of the
relative DNA affinity for compounds 22 and 26 enlightens
309
þ4
25
312
298
þ7
26
27
þ12
45e
54e
50
291e
284
297
þ5
28
29
þ8
þ13
48
a
l
1 ¼ lmax compound alone.
b
c
l
2 ¼ lmax compoundþCT-DNA.
Dl
¼
l
10ꢀ
l2. Positive values evidenced bathochromic effects.
d
alone
Hypochromic effect H ¼ [100 ꢁ (Abscompound
ꢀ AbscompoundþCT-DNA)/
Abscompound alone] at the specified wavelength (
l).
e
Hypochromic and bathochromic effects are measured at the minimal point
regarding the following increase in the y value due to complex binding mode.
a
decrease in DNA-binding potency when the chlorine is
substituted at position 10 (26) instead of position (22).
The integration of a -electron rich thiophene moiety (28) does not
8
p
Quantification of cells in each cell cycle steps was performed as
untreated cells (0) or cells treated with the increasing concentra-
tion of each compounds (arrow), as specified on the left of each
graph (mM), is performed as described in the Material and methods
section. The localisation of subG1, G1/G0, S, G2/M phases is speci-
fied on the first graph.
enhance the affinity for DNA compared to its pyrido analogues.
Evidently, the electron density of the chromophoric system is not as
determining for the DNA interaction as the introduction of a more
lipophilic group. Finally, the presence of dimethyl-amine at the end
of the lateral chain is, as attempted, crucial for the DNA binding as
evidenced from comparison of the results using compounds 26 and
29, in correlation with previous observation using compounds 8
3.3. DNA binding and inhibition of topoisomerases I and II
and 9 lacking any side chain and that failed to induce any DTm
values.
3.3.1. UV/vis spectral absorbance
In order to address the DNA-binding ability of this series of
pyridocarbazoles in the primary screening, we first looked at the
modification of the absorption spectra of the various compounds in
the absence and presence of increasing concentrations of calf-
thymus DNA (phosphate/drug ratio from 1 to 20) based on the
principle that the UV/Vis-spectra of a substance changes if it
interacts with the DNA. The spectra of pyridocarbazole derivatives
are presented in Fig. 2. All these compounds show hypochromic
and bathochromic effects (arrows Y and /, respectively) that are
quantified in Table 4. Comparison between compounds 20 and 21
to their corresponding chloro-containing derivatives 22 and 23
reveals similar hypochromic effects but different bathochromic
shifts (Fig. 2 and Table 4). This bathochromic effect implies
a stronger interaction between DNA and the chromophore using
the chloro-substituted compounds from comparison with the
unsubstituted pyridocarbazoles.
3.4. Fluorescence studies
Compounds 20–25 and 28 present intrinsic fluorescence prop-
erties which could be used to evaluate their respective DNA-binding
propensities (Fig. 3). Compounds chloro substituted in position 10
are not fluorescent. From the titration of compounds 24 and 25, the
emission maxima undergoes a clear red shift after binding to
increasing concentration of CT-DNA, reflecting
a more polar
environments. The additional decrease of the fluorescence intensi-
ties indicates a marked quenching upon binding to the DNA. The
quenching constant Ksv was deduced from Stern–Volmer plots to be
6.72 ꢁ 10þ3 and 9.95 ꢁ10þ3 Mꢀ1 for compounds 24 and 25,
Table 5
Variation of DNA-melting temperature
D
Tm [ꢃC] of calf-thymus DNA and poly
(dAdT)2 from complexation with various pyridocarbazoles (20–27, 29) and a thie-
nocarbazole (28). r ¼ molar ratio of ligand/DNA.
D
Tm (ꢃC)
3.3.2. DNA-melting temperature
Compound
The DNA ligand affinity can be determined by UV measurement
of the DNA-melting temperature, which is enhanced in case of
ligand binding (base-pair separation). In order to confirm the DNA-
binding ability of those compounds, we then looked at the stabi-
lization of the DNA helix by the various hetarene annelated
carbazoles using melting temperature studies. UV spectroscopic
determination of the melting curve of DNA compared with the
melting curve of a DNA ligand complex gives a basic information
about the binding potency of a ligand to DNA [22] and is used to
CT-DN
Poly(dAdT)2
r
r
0.1
0.25
0.5
2.9
4.4
7.3
1
0.1
2.9
2.9
6.1
0.25
0.5
1
20
21
22
23
24
25
26
27
28
29
0
0
3.0
4.4
7.4
7.4
7.9
6.6
1.5
5.0
1.5
0.7
7.3
11.8
8.9
14.7
9.2
14.5
11.6
18.9
20.4
24.7
22.7
13.6
23.6
16.4
5.3
18.9
17.5
24.9
27.8
29.6
28.3
8.8
29.8
20.9
5.3
7.3
11.7
13.2
15.1
13.9
7.9
13.8
5.8
0.6
1.5
3.0
5.3
3.6
1.1
2.6
0
8.8
6.1
11.2
10.7
5.1
10.4
4.4
11.6
10.3
2.8
9.7
5.9
2.1
18.9
17.3
6.2
17.6
11.9
3.9
compare the relative affinity of a series of derivatives. The
DTm
values (Tm[CT-DNAþdrug] ꢀ Tm[CT-DNAalone]) were determined using
calf-thymus DNA (CT-DNA) and poly(dAdT)2 at DNA/drug ratio
0
0.4
from 0.1 to 1. A DTm value of more than 10 is a convenient sign for

Fig. 3. Fluorescence spectra of the various pyridocarbazoles upon incubation with graded concentrations of CT-DNA. A fixed concentration of the various derivatives (20
mM) was
incubated with increasing concentrations of CT-DNA from 10 to 200 M for a phosphate/compound ratio from 0.4 to 8. Arrows indicate the orientations of the fluorescence
m
quenching (Y) or enhancement ([). The excitation wavelength peaks for the various compounds are indicated at the top of the related graphs. Stern–Volmer plots expressing F0/F as
a function of the concentration of CT-DNA is used to deduce the quenching and enhancement constants for the indicated compounds on the three bottom graphs.

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3243
Fig. 4. Circular dichroism spectra of CT-DNA with the various pyridocarbazoles. A fixed concentration of the various derivatives (50
mM) was incubated with increasing concen-
trations of CT-DNA from 10 to 200
concentrations, respectively.
mM (phosphate/compound ratio from 0.4 to 8). Open and full arrows indicate the orientations of the induced CD at lower and upper CT-DNA

3244
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
21
20
23
22
26
29
Fig. 5. Localisation of the DNA binding using DNase I footprinting analysis. The 30-end labelled EcoRI-PvuII fragment of pBS vector was incubated with increasing amount of the drug
as specified on the top of the appropriate lanes ( M) prior to be digested by DNase I. Control tracks labelled ‘‘DNA’’ contained no drug. Lanes labelled ‘‘G-track’’ were used as marker
for guanine positions. Lane ‘‘0’’ evidenced the untreated radio-labelled DNA fragment used as quality control.
m
respectively. The DNA-binding mode for compound 28 appears to be
more complex regarding the two slopes evidenced from Stern–-
Volmer plot. Such result suggests that two types of fluorophores are
present in solution, one being compound 28 alone and the second
being the complex formed between compound 28 and DNA base
pairs. The quenching constant for compound 28 was calculated to be
of 2.65 ꢁ10þ4 Mꢀ1 from the first points at CT-DNA concentrations up
a negative absorbance between
l
¼ 295–315 nm (Fig. 4) for higher
CT-DNA concentrations, suggesting the displacement of those
compounds from DNA groove binding to an intercalation binding
mode by increasing the concentration of base pairs (Fig. 2).
Surprisingly, the pyrido-[4,3a]carbazoles 24 and 25 present
different ICDs with a negative ICD at wavelength around 300 nm
using the lowest CT-DNA concentrations shifting to a positive ICD at
wavelength around 285 and 330 nm using the highest phosphate/
drug ratio. A unique high positive ICD at 300 nm is observed using
the 10-chloro-derivative 27 with a saturation of the effect for
a phosphate/drug ratio of more than 4.8. The other 10-chloro-
derivative 26 bearing a dimethyl-amino-ethyl arm instead of
a dimethyl-amino-propyl (27) presents a positive ICD at the same
wavelength for up to a phosphate/drug ratio of 1.2 followed by
a negative ICD for higher CT-DNA/compound ratio. The morpholino
derivative 29 which can be considered as a cyclic analogue of
dimethylamino end of compound 26 to give abolished the forma-
tion of ICD at the wavelength of the compound, in agreement with
the crucial role of the positively charged amino-termini for DNA-
binding stabilization.
to 80 mM. Contrasting with the fluorescence quenching through DNA
interaction of those compounds, an enhancement of the fluorescent
is observed using compounds 20 and 23, as well as to a lesser extend
using compound 21. The enhancement constants for 8-chloro (23) or
unsubstituted (21) derivatives were deduced from plotting F0/F
versus the concentration of CT-DNA to be 4.84 ꢁ10þ3 Mꢀ1 and
5.94 ꢁ10þ2 Mꢀ1, respectively.
3.4.1. Circular dichroism measurements
The binding of an achiral ligand to the chiral DNA causes an
induced optical activity of the ligand. These measurements with
CT-DNA were used to get an insight in the DNA-binding mode –
intercalation and/or groove binding – of the various annelated
carbazoles. The most common feature (compounds 20–23) is
a positive induced CD signal in the absorption area of the pyr-
idocarbazole ligand around l ¼ 310–315 nm (Fig. 4) suggesting
a groove binding of the compounds. This positive induced CD (ICD)
occurs using the lowest CT-DNA concentrations and shifted to
3.4.2. Localisation of DNA-binding site
The DNA minor groove is a good docking site for sequence
selective DNA ligand. The sequence selectivity for a compound can
be achieved using DNase I footprinting on a radio-labelled DNA

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3245
Topo I
21
20
23
22
Rel + Nck
Topoisomers
Sc
Topo I
28
25
24
Rel + Nck
Topoisomers
Sc
Topo I
27
26
29
Rel + Nck
Topoisomers
Sc
Fig. 6. Topoisomerase I-induced DNA relaxation. The effect of the various pyridocarbazoles and the thienocarbazoles on the relaxation of supercoiled plasmid DNA by top-
oisomerase I was established using native supercoiled pUC19 DNA (130 ng; lanes DNA) incubated with topoisomerase I in the absence (lanes Topo I) or presence of increasing
concentrations (mM) of the various compounds. The different forms of plasmid DNA were separated by electrophoresis on a 1% agarose gel which was then stained with ethidium
bromide. Rel, relaxed; Sc, supercoiled; Nck, nicked.
fragment. We evaluated the potential for sequence selective
binding of this series of compounds on a 117 bp 30-end labelled
DNA fragment (Fig. 5). Incubation of increasing concentrations of
each compounds failed to evidence any sequence selectivity but
none-specific DNA interaction as attempted for DNA-intercalating
compounds. The non-specific DNA binding can particularly well be
seen using compounds 20–27 whereas compound 28 only weakly
affect the DNase I activity and compound 29 totally failed to modify
the DNA cleavage profile. Those results are in perfect correlation
with previous observation from spectroscopic analysis.
I-mediated cleavage of supercoiled DNA. As presented in Fig. 6, all
compounds but compound 29 present a change in the topoisomer
profile from negatively supercoiled to relaxed DNA (Rel) and back
to (positively) supercoiled position (Sc). Those DNA relaxation
profiles suggest that the compounds are potent DNA base pair
intercalators thus acting on the twist of the DNA helix. The lack of
DNA relaxation profile for compound 29 is in agreement with
previous results evidencing no DNA double strand stabilization
(Table 5), which is usually strong for DNA intercalators, and no ICD
in Fig. 4. Therefore, since no groove binding could be evidenced in
CD measurements, the hypochromic effect seen in UV/vis absor-
bance measurements with compound 29 in the presence of CT-DNA
could presumably only be attributed to electrostatic binding of
compound 29 to the phosphate backbone. From the DNA relaxation
experiment, the intercalation efficiencies could be classified from
weaker to stronger: 26 < 20 < 21, 27, 28 < 22, 23, 24, 25
and correlate with their efficiency to stabilize the DNA helix (CT-
DNA and poly(dAdT)2) evidenced using melting temperature
studies: 26 < 20, 21 <27, 28, 22–25.
The 30-end labelled EcoRI-PvuII fragment of pBS vector was
incubated with increasing amount of the drug as specified on the
top of the appropriate lanes (mM) prior to be digested by DNase I.
Control track labelled ‘‘DNA’’ contained no drug. Lanes labelled
‘‘G-track’’ were used as marker for guanine positions. Lane ‘‘0’’
evidenced the untreated radio-labelled DNA fragment used as
quality control.
3.4.3. Inhibition of topoisomerases I and II
In order to get a further insight in the mode of binding of the
pyridocarbazole derivative on DNA, we used the DNA relaxation
properties of topoisomerase I. This method is a powerful approach
to evidence DNA-intercalation process as an enhancer of DNA
relaxation creating more relaxed topoisomers from topoisomerase
Regarding the differences for DNA binding and DNA-intercala-
tion potencies of the unsubstituted compounds 20 and 21 versus
the chloro-substituted equivalent 22 and 23, those four compounds
were evaluated for topoisomerases I and II poisoning effect. From
comparison with the respective reference compounds

3246
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
Topo I
A
21
20
23
22
28
Nck
Sc
Rel
Topo II
B
Etop
21
20
23
22
28
Lin
Sc
Rel
Fig. 7. Topoisomerase I (A) and II (B) poisoning effects of various pyridocarbazoles. Native supercoiled pUC19 plasmid DNA (130 ng, lanes DNA) was incubated with topoisomerase I
or II in the absence (lanes Topo I or Topo II) or presence of 20 or 50 M of the various derivatives. Camptothecin (CPT) and etoposide (Etop) were used as specific controls. DNA
m
samples were separated under electrophoresis on an ethidium bromide containing 1% agarose gel. The gels were photographed under UV light. Sc, supercoiled; Rel, relaxed; Nck,
nicked; Lin, linear.
camptothecin (CPT, Fig. 7A) and etoposide (Etop, Fig. 7B), none of
those compounds are topoisomerase I or II inhibitors as evidenced
by the lack of nicking form (topoisomerase I) and double strand
break form of the DNA (linear DNA, topoisomerase II). The gel shift
migration of the supercoiled form correlates with the binding
efficiency of the compounds: the chloro-derivatives 22 and 23 are
more potent DNA-binding compounds than the unsubstituted
compounds 20 and 21, in perfect correlation with previous
conclusions.
This could be due to an increase of the polar hydrophobicity of the
chloro derivatives relatively to the unsubstituted pyridocarbazoles.
According to the DNA-melting temperature measurements, the
DTm values of chloro-pyridocarbazoles 22–25 and 27 range
between about 25 and 30 ꢃC, highlighting their potent DNA
binding. Using this assay, the pyridocarbazoles 24 and 27 are the
strongest DNA binders. Concerning the length of the amidic chain,
no obvious preference could be noticed from comparison of the
relative affinity for DNA of the ethylamides 20, 22 and 24 versus the
corresponding propylamides 21, 23 and 25, respectively. However,
comparison of the relative DNA affinity for CT-DNA or poly(dAdT)2
of compounds 26 and 27 evidenced an increased DNA affinity in the
presence of a dimethylamino-propyl chain (27) instead of a dime-
thylamino-ethyl (26). For this latter, the reduced DNA affinity could
also be attributed to the orientation of the chlorine on the carbazole
ring. Indeed, comparison of the relative DNA affinity for compounds
22 and 26 enlightens a decrease in DNA-binding potency when the
chlorine is substituted at position 10 (26) instead of position 8 (22).
4. Discussion and conclusion
Only carbazoles 20, 21 and the chloro derivative 22 were chosen
for NCI antitumor screening. The chloro derivative 22 which
displays a Mean Graph Midpoint value (MG_MID) of - 5,6 proved to
be the most active compound in general, whereas compound 20
shows significant activity against single ovarian (GI50 [M]:
2.54 ꢁ10ꢀ8) and renal (GI50 [M]: 5.15 ꢁ10ꢀ7) tumor cell lines
(Table 2). Comparison of the cytotoxicity of the whole series of
those newly synthesized carbazoles was established in the lab on
the colon carcinoma cell line HT-29 and evidenced that compounds
22 but also compounds 24 and 25 (and to a lesser extend compound
The integration of a p-electron rich thiophene moiety (28) does not
enhance the affinity for DNA compared to its pyrido analogues.
Finally, the presence of dimethyl-amine at the end of the lateral
chain is, as attempted, crucial for the DNA binding as evidenced
from comparison of the
correlation with previous observation using compounds 8 and 9
lacking any side chain and that failed to induce any Tm values.
DTm results using compounds 26 and 29, in
27) are the most cytotoxic agents in this series with GI50 in the mM
range (Table 3). The cellular effects of those compounds are asso-
ciated with a major G2/M phase arrest (Fig. 1).
D
The fluorescence titration studies are throughout positively by
compounds 24, 25 and 28 with the exception of the morpholine
derivative 29. The induced CD (ICD) on drug/DNA-binding
evidences a common positive ICD signal compounds 20–23 sug-
gesting a groove binding. By contrast, the pyrido-[4,3a]carbazoles
On the basis of the UV absorption (hypochromic and bath-
ochromic shifts) and
DTm values the chloro derivatives bearing
a dimethylaminoalkyl-side chain reveal a significantly stronger
DNA-binding potency than the basic related compounds 20 and 21.

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3247
24 and 25 evidence negative ICD present from the lowest CT-DNA
concentrations and followed by a positive ICD at the highest CT-
DNA concentrations. This suggests a preliminary groove binding
(amidic side chain) shifting then to an intercalation on the core of
the compound between adjacent base pairs at saturating concen-
trations. Interestingly, the 10-chloro-derivative 27 only evidences
a high positive ICD in the wavelength of the compound suggesting
unique groove binding contrasting with the 10-chloro-derivative
26 presenting groove binding followed by good intercalative
process. This suggests that the presence of both the chlorine and
the shorter arm alters the correct positioning of the chromophore
between adjacent base pairs with the chlorine protruding on one
side of the DNA helix (major groove) and the dimethylaminoalkyl-
side chain on the other side (minor groove), thus leading to the
unique groove binding of this compound. Positioning of the chlo-
rine at position 8 does not affect the intercalation of the chromo-
phore within the DNA since it will be orientated properly relatively
to the phosphate backbone of the DNA helix.
DNase I footprinting studies for analysing sequence selectivity of
binding reveals for compounds 20–27 throughout a non-specific
binding whereas carbazole 28 only weakly affects the DNase I
activity and compound 29 totally failed to modify the DNA damage
profile. These results are in perfect correlation with previous
observations from spectroscopic analysis.
Finally, the topoisomerase I and II poisoning effects are in good
correlation with other results. Indeed, the gel shift migration of the
supercoiled form correlates with the binding efficiency of the
compounds: the chloro derivatives 22 and 23 are more potent
binding compounds than the unsubstituted compounds 20 and 21
in perfect correlation with previous conclusions.
>0.3%. Despite using different solvent mixtures in TLC only one spot
could be detected, therefore the testing compounds were consid-
ered analytically pure.
5.2. General procedure for the synthesis of hydrazones
The acetyl hetarenes [11.2 ml (100 mmol) 2- or 4-acetylpyridine
resp. 10.8 ml (100 mmol) 2-acetylthiophene] were dissolved in
300 ml ethanol and the hydrazine [9.8 ml (100 mmol) phenyl-
hydrazine resp. 17.0 g (100 mmol) 2- or 4-chlorophenylhy-
drazine$HCl] were added. The resulting yellow suspension was
refluxed for 5 h. After cooling, the precipitated hydrazone war
filtered off, washed with ethanol and petrol ether and dried.
5.2.1. 2-[(1-(2-Phenylhydrazono)ethyl)pyridine] (13a) [23]
Yellow solid, (19.4 g) (90 mmol) (90%), m.p. 153 ꢃC; 1H NMR
(CDCl3):
d
¼ 2.38 (s, 3H, CH3), 6.89 (m, 1H, phenyl-H-4), 7.16–7.31
(m, 4H, phenyl-H), 7.71 (m, 2H, pyridyl-H), 8.14 (pd, 1H, J ¼ 8.1 Hz,
pyridyl-H-3), 8.56 (pd, 1H, J ¼ 5.0 Hz, pyridyl-H-6).
5.2.2. 2-{1-[2-(4-Chlorophenyl)hydrazono]ethyl}pyridine (13b)
Yellow solid, (21.3 g) (86 mmol) (86%), m.p. 238 ꢃC; 1H NMR
(DMSO-d6):
d
¼ 2.40 (s, 3H, CH3), 7.32 (pd, 2H, J ¼ 8.8 Hz, phenyl-H-
2, H-6) 7.63 (pd, 2H, J ¼ 8.8 Hz, phenyl-H-3, H-5), 7.77 (pt, 1H,
pyridyl-H-5), 8.24 (pd, 1H, J ¼ 8.1 Hz, pyridyl-H-3), 8.43 (pt, 1H,
pyridyl-H-4), 8.73 (pd, 1H, J ¼ 5.7 Hz, pyridyl-H-6), 10.57 (s, NH).
5.2.3. 4-{1-[2-(4-Chlorophenyl)hydrazono]ethyl}pyridine (13c) [24]
Yellow solid, (21.7 g) (87 mmol) (87%), m.p. 307 ꢃC (decompo-
It accomplished to synthesize a series of pyrido- and thieno-
sition); 1H NMR (DMSO-d6):
J ¼ 7.6, phenyl-H), 7.45 (pd, 2H, J ¼ 7.6, phenyl-H), 8.25 (pd, 2H,
J ¼ 5.6, pyridyl-H-3 þ 5), 8.73 (pd, 2H, J ¼ 5.6, pyridyl-H-2 þ 6),
10.55 (s, NH).
d
¼ 2.33 (s, 3H, CH3), 7.34 (pd, 2H,
carbazole amides via photochemically induced
6
p-electro-
cyclisation. The biological and biophysical/biochemical results
reveal high DNA-binding properties especially of chlorine-
substituted pyridocarbazoles which exceed considerably the
DNA-binding affinity of the leading substance ellipticine. The
replacement of pyridine with an electron rich thiophene does not
seem to bring a benefit on DNA binding. The introduction of
a cyclic residue in the amidic group as side chain (morpholine)
decreases DNA binding compared to acyclic series. Concerning the
binding mode of active compounds 10, we suggest groove binding
and intercalation. The synthesis of regio-isomers and particularly
the introduction of a further halogen atom in the chloropyr-
idocarbazole framework give rise to develop promising candidates
as potent DNA ligands.
5.2.4. 2-{1-[2-(2-Chlorophenyl)hydrazono]ethyl}pyridine (13d)
Yellow solid, (21.6 g) (88 mmol) (88%), m.p. 246 ꢃC (decompo-
sition); 1H NMR (DMSO-d6):
d
¼ 2.44 (s, 3H, CH3), 6.97 (m, 1H,
phenyl-H), 7.34 (m, 1H, phenyl-H), 7.43 (m, 1H, phenyl-H), 7.73 (m,
1H, phenyl-H), 7.96 (pd, 1H, J ¼ 8.1 Hz, pyridyl-H), 8.30 (m, 2H,
pyridyl-H), 8.72 (pd, 1H, J ¼ 5.5 Hz, pyridyl-H), 8.92 (s, NH).
5.2.5. 2-[1-(2-Phenylhydrazono)ethlyl]thiophene (13e) [25]
Yellow solid, (17.5 g) (81 mmol) (81%), m.p. 94 ꢃC; 1H NMR
(DMSO-d6):
d
¼ 2.25 (s, 3H, CH3), 6.34 (m,1H, thienyl-H), 7.01 (m,1H,
thienyl-H), 7.12–7.24 (m, 5H, phenyl-H), 7.39 (pd, 1H, thienyl-H-5),
8.92 (s, NH).
5. Experimental
5.1. Chemistry
5.2.6. 2-Pyridin-2-yl-indole (14a) [26]
40 ml polyphosphoric acid was added into a beaker and
heated under continuous stirring at 100–120 ꢃC. 4.2 g (20 mmol)
2-[1-(2-Phenylhydrazono)ethyl]pyridine (13a) was added in small
amounts over a period of 10 min and the brown liquid was kept
at m.p. 120 ꢃC for 30 min. After cooling the mixture was hydro-
lyzed carefully with ice and neutralized with aqueous sodium
hydroxide. The resulting yellow solid was filtered and refluxed in
300 ml acetone for about 10 min. The hot suspension was fil-
trated and the solution was evaporated to dryness. The residue
was chromatographed on silica gel (petrol ether/ethyl acetate
(3:1)). White solid (2.0 g) (10.3 mmol) (52%), 152 ꢃC; 1H NMR
Melting points were measured with a Bu¨chi 510 instrument and
are uncorrected. IR spectra were recorded on a Thermoelectron
Avatar 330 FT-IR using ZnSe crystal (AMTIR) (n
in cmꢀ1). 1H NMR,
13C NMR spectra including spin echo were recorded on a Bruker AC-
300 apparatus (300 MHz). The samples were dissolved in CDCl3 or
DMSO-d6. pd: pseudodublet; pt: pseudotriplet. The chemical shift
values are reported in parts per million (ppm,
d units) and spin–
spin coupling J were listed in hertz. 70 eV EI-mass spectra were
obtained with a Mascom 311-A apparatus and FD mass spectra with
a Finnigan MAT 7 instrument. Photochemical apparatus: Normag
company, mercury vapour medium pressure lamp with 150 W and
(CDCl3):
d
¼ 7.05 (m, 1H, indole-H-3), 7.11–7.26 (m, 3H, indole-H-
an area of wavelength
l
¼ 200–600 nm. Column chromatography
5-7), 7.37 (pd, 1H, J ¼ 8.1 Hz, indole-H-4), 7.67–7.75 (m, 2H, pyr-
idyl-H-4 þ 5), 7.83 (pd, 1H, J ¼ 8.1 Hz, pyridyl-H-3), 8.59 (pd, 1H,
J ¼ 4.3 Hz, pyridyl-H-6), 10.23 (s, N-H); EI-MS: m/z 194 [Mþ],
C13H10N2.
was performed on silica gel (Merck, silica gel 60). Petrol ether of
boiling point 40–60 ꢃC was used. Non-stoichiometric inclusion of
solvent molecules gave always C, H, N analysis with divergence

3248
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
5.2.7. 5-Chloro-2-(pyridin-2-yl)-indole (14b)
5.3.1. 2-(Pyridin-2-yl)-indole-3-carbaldehyde (15a) [28]
See Section 5.3.1. 4.9 g (20 mmol) 2-{1-[2-(4-chlorophenyl)hy-
drazono]ethyl}pyridine (13b). The resulting yellow solid was
filtered and refluxed in 200 ml ethanol instead of acetone for about
10 min. White solid (2.6 g) (11.4 mmol) (57%), m.p. 172 ꢃC; 1H NMR
See Section 5.4. The chromatographic isolation was performed
with ethyl acetate/petrol ether (2:1). Bright yellow solid (0.9 g)
(41%) (4.1 mmol), m.p. 213 ꢃC; 1H NMR (DMSO-d6):
d
¼ 7.22–7.33
(m, 2H, indole-5 þ 6), 7.54 (m, 2H, indole-H-7, pyridyl-H-3), 8.01–
8.11 (m, 2H, pyridyl-H-4 þ 5), 8.24 (pd, 1H, J ¼ 7.4 Hz, indole-H-4),
8.79 (pd, 1H J ¼ 4.8 Hz, pyridyl-H-6), 10.59 (s, 1H, formyl-H), 12.22
(s, 1H, N-H); EI-MS: m/z 222.1 [Mþ], C14H10N2O.
(CDCl3):
d
¼ 6.95 (m, 1H, indole-H-3), 7.13–7.28 (m, 3H, indole-H-4,
6, 7), 7.60 (m, 1H, pyridyl-H-3), 7.71–7.81 (m, 2H, pyridyl-H-4 þ 5),
8.57 (m, 1H, pyridyl-H-6), 10.06 (s, N-H); EI-MS: m/z 228.1 [Mþ],
C13H9ClN2.
5.3.2. 5-Chloro-2-(pyridin-2-yl)-indole-3-carbaldehyde (15b)
See Section 5.4. The chromatographic isolation was performed
with ethyl acetate/petrol ether (2:1). Bright yellow solid (1.2 g)
(47%) (4.7 mmol), m.p. 290 ꢃC (decomposition); 1H NMR (DMSO-
5.2.8. 5-Chloro-2-(pyridin-4-yl)-indole (14c)
See Section 5.3.1. 4.9 g (20 mmol) 4-{1-[2-(4-Chlorophenyl)hy-
drazono]ethyl}pyridine (13c). The temperature was kept at m.p.
140 ꢃC for 30–45 min. The resulting yellow solid was filtered and
dissolved in 100 ml dimethyl formamide. The solution was evapo-
rated under vacuum to dryness and chromatographed on silica gel
(ethyl acetate/ethanol (2:1)). White solid (2.4 g) (10.5 mmol) (53%),
d6):
d
¼ 7.32 (pd, 1H, J ¼ 8.8 Hz, indole-H-6), 7.53–7.58 (m, 2H), 8.06
(pd, 1H, J ¼ 1.8 Hz, pyridyl-H-4), 8.11 (m, 1H), 8.21 (pd, 1H, 1H,
J ¼ 1.8 Hz, pyridyl-H-3), 8.80 (pd, 1H, J ¼ 4.8 Hz, pyridyl-H-6), 10.55
(s, 1H, formyl-H), 12.77 (s, 1H, N-H); FD-MS: m/z 256.0 [Mþ],
C14H9ClN2O.
260 ꢃC (decomposition); 1H NMR (DMSO-d6):
d
¼ 7.14 (m, 2H,
indole-H-3 þ 6), 7.45 (pd, 1H, J ¼ 8.6 Hz, indole-H-7), 7.63 (pd, 1H,
J ¼ 1.9 Hz, indole-H-4), 7.91 (pd, 2H, J ¼ 6.0, pyridyl-H-3 þ 5), 8.61
(pd, 2H, J ¼ 6.0, pyridyl-H-2 þ 6), 12.03 (s, N-H); 13C NMR (DMSO-
5.3.3. 5-Chloro-2-(pyridin-4-yl)-indole-3-carbaldehyde (15c)
See Section 5.4. The chromatographic isolation was performed
with ethyl acetate/ethanol (15:1). Bright yellow solid (1.0 g) (39%)
(3.9 mmol), m.p. 275 ꢃC (decomposition); 1H NMR (DMSO-d6):
d6):
d
¼ 99.9 (t), 111.2 (t), 119.6 (2 t), 120.2 (t), 123.5 (t), 124.6 (q),
129.7 (q),136.3 (q),136.6 (q),137.2 (q),150.8 (2 t); FD-MS: m/z 228.3
[Mþ], C13H9ClN2.
d
¼ 7.33 (pd, 1H, J ¼ 8.8 Hz, indole-H-6), 7.55 (pd, 1H, J ¼ 8.8 Hz,
indole-H-7), 7.79 (pd, 2H, J ¼ 5.9 Hz, pyridyl-H-3 þ 5), 8.19 (pd, 1H,
J ¼ 1.9 Hz, indole-H-4), 8.78 (pd, 2H, J ¼ 5.9 Hz, pyridyl-H-2 þ 6),
10.01 (s, 1H, formyl-H), 12.10 (s, 1H, N-H); 13C NMR (DMSO-d6):
5.2.9. 7-Chloro-2-(pyridin-2-yl)-indole (14d)
See Section 5.3.1. 4.4 g (20 mmol) 2-{1-[-(2-chlorophenyl)hy-
drazono]ethyl}pyridine (13d). The reaction time of cyclization was
limited to 15 min. After neutralization the crude product is
extracted three times with ethyl acetate. The combined organic
solutions were dried with magnesium sulfate and after removing of
the solvent the residue was recrystallized with methanol. White
solid (2.8 g) (12.2 mmol) (61%), m.p. 121 ꢃC; 1H NMR (CDCl3):
d
¼ 114.2 (t), 114.3 (q), 120.6 (q), 124.3 (q), 124.7 (q), 127.1 (2t), 127.7
(t), 135.0 (t), 137.0 (t), 146.6 (2t), 150.6 (q), 185.7 (q); FD-MS: m/z
256.7 [Mþ], C14H9ClN2O.
5.3.4. 7-Chloro-2-(pyridin-2-yl)-indole-3-carbaldehyde (15d)
See Section 5.4.1. The chromatographic isolation was performed
with petrol ether/ethyl acetate (1:1). Bright yellow solid (1.1 g)
(43%) (4.3 mmol), 256 ꢃC (decomposition); 1H NMR (DMSO-d6):
d
¼ 7.02–7.07 (m, 2H, indole-H-3 þ 5), 7.17–7.25 (m, 2H, indole-H-
4 þ 6), 7.54 (pd, 1H, J ¼ 7.9 Hz, pyridyl-H-3), 7.69–7.80 (m, 2H, pyr-
idyl-4 þ 5), 8.59 (pd, 1H, J ¼ 5.0, pyridyl-H-6), 9.76 (s, N-H); FD-MS:
m/z 199.1 [Mþ], C12H9NS.
d
¼ 7.22 (m, 1H, indole-H), 7.41 (m, 1H), 7.59 (m, 1H), 8.08 (m, 1H),
8.20 (m, 2H, pyridyl-H), 8.79 (m, 1H, pyridyl-H-6), 10.49 (s, 1H, N-
H), 12.67 (s, 1H, formyl-H); FD-MS: m/z 256.1 [Mþ], C14H9ClN2O.
5.2.10. 2-Thiophen-2-yl-indole (14e) [27]
See
Section
5.3.1.
4.0 g
(20 mmol)
2-[1-(2-phenyl-
5.3.5. 2-(Thiophen-2-yl)-indole-3-carbaldehyde (15e) [27]
hydrazono)ethyl]thiophene (13e). The reaction time of cyclization
2.5 ml (26.5 mmol) POCl3 were added dropwise to 1 ml
(13 mmol) DMF at 0 ꢃC. Subsequently a solution of 2.0 g (10 mmol)
14e in 10 ml dry diethyl ether was added slowly at 0 ꢃC.
The mixture was stirred 30 min at 0 ꢃC and after warming up at
room temperature it was heated for 3 h at 90–100 ꢃC. The reaction
was stopped by adding 30 ml ice water and neutralized with
aqueous sodium hydroxide solution (20%). The suspension was
extracted three times with CH2Cl2 and after drying of the combined
organic phases. The solvent of the suspension was evaporated and
the precipitate recrystallized in ethanol. Bright yellow solid (1.4 g)
was limited to 15 min. White solid (1.9 g) (9.6 mmol) (48%), m.p.
165 ꢃC; 1H NMR (DMSO-d6):
1H, indole-H), 7.06–7.14 (m, 2H, indole-H), 7.35 (pd, 1H, J ¼ 7.9,
indole-H-7), 7.50 (m, 3H, thophene-H), 11.55 (s, N-H); FD-MS: m/z
228.1 [Mþ], C13H9ClN2.
d
¼ 6.65 (s, 1H, indole-H-3), 6.98 (m,
5.3. General procedure for the synthesis of 2-(pyridinyl)-1H-indole-
3-carbaldehydes
(62%) (6.2 mmol), m.p. 228 ꢃC; 1H NMR (DMSO-d6):
d
¼ 7.20–7.32
(m, 3H), 7.47 (m, 1H), 7.74 (m, 1H), 7.89 (m, 1H), 8.16 (m, 1H, indole-
H-4), 10.23 (s, 1H, formyl-H), 12.44 (s, 1H, N-H); 13C NMR (DMSO-
1.94 g of the indole 14a resp. 2.28 g of the chloro-indoles 14b–
d (10 mmol) were dissolved in 10 ml of dry DMF under nitrogen
atmosphere and cooled down to ꢀ5 ꢃC. 1.12 ml (12 mmol) POCl3
were added dropwise over a period of 10 min where the temper-
ature had to be kept under 5 ꢃC. The suspension was stirred for
30 min at 0 ꢃC, 30 min at room temperature and further 4 h at
90 ꢃC. Subsequent the mixture was hydrolyzed carefully with water
and neutralized with aqueous sodium hydroxide solution (20%).
The suspension was filtered, the filtrate extracted with ethyl acetate
and the organic phase combined with the solid. The solvent of the
suspension was evaporated and the precipitate was chromato-
graphed on silica gel.
d6):
d
¼ 112.2 (q), 113.8 (t), 121.2 (q), 124.4 (q), 126.2 (t), 128.8 (q),
130.2 (q), 130.8 (t), 136.1 (t), 136.3 (t), 141.5 (t), 141.7 (t), 186.1 (q);
FD-MS: m/z 227.3 [Mþ], C13H9NOS.
5.4. General procedure for the synthesis of (E)-methyl 3-(2-
hetarylindol-3-yl)acrylate
10 mmol of the carbaldehyde (2.2 g of 15a, 2.6 g of 15b–d, 2.3 g
of 15e) and 4.3 g (13 mmol) of (methoxycarbonylmethylen)-tri-
phenylphosphorane are refluxed in 300 ml dry toluene for 5–6 h.
The progress of the reaction was followed by TLC (silicagel, petrol

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3249
ether/ethyl acetate (2:1)). The solvent was evaporated and the
5.5. General procedure for the synthesis of hetarene annelated [a]-
residue chromatographed on silica gel.
carbazoles via photochemical electrocyclisation
5.4.1. (E)-Methyl-3-[2-(pyridin-2-yl)-indol-3-yl]acrylate (16a) [29]
See Section 5.5. The isolation was performed with petrol ether/
ethyl acetate (2:1). Bright yellow solid (2.1 g) (76%) (7.6 mmol), m.p.
2 mmol of the methyl-3-(2-hetaryl-indol-3-yl)acrylate 16 and
some crystals of iodine were dissolved in 450 ml of the specified
solvent and irradiated with a wavelength band
l
¼ 200–600 nm at
158 ꢃC; 1H NMR (CDCl3):
d
¼ 3.85 (s, 3H, CH3), 6.66 (d, 1H,
17–20 ꢃC. The reaction progress was controlled by TLC. After
complete conversion the solvent was let off and evaporated to
dryness. The resulting residue was purified by column chroma-
tography on silica gel.
3J ¼ 16.0 Hz, ]CH), 7.27–7.34, (m, 3H), 7.46 (m, 1H), 7.82–7.88 (m,
2H), 7.96 (pd, 1H, J ¼ 7.6, pyridyl-H), 8.27 (d, 1H, 3J ¼ 16.0 Hz, ]CH),
8.70 (m, 1H), 9.97 (s, 1H, N-H); 13C (DMSO-d6):
d
¼ 51.1 (p), 110.9 (q),
112.4 (t), 119.9 (t), 120.2 (t), 120.6 (t), 122.7 (t), 123.0 (t), 123.3 (t),
126.7 (q), 135.2 (q), 136.3 (q), 137.5 (2t), 149.8 (t), 150.9 (q), 166.9
(q); FT-IR (cmꢀ1): 2900–3200, 1698, 1609, 1422, 1182; FD-MS: m/z
278.6 [Mþ], C17H14N2O2.
5.5.1. Methyl-11H-pyrido[2,3-a]carbazole-5-
carboxylate (17a)
557 mg of 16a was dissolved in 450 ml CH2Cl2 and irradiated
for 6 h. The progress of reaction via TLC and the isolation of the
pyridocarbazole by column chromatography on silica gel were
performed with petrol ether/ethyl acetate (3:1). White solid
(400 mg) (72%) (1.4 mmol), m.p. 228 ꢃC; 1H NMR (DMSO-d6):
5.4.2. (E)-Methyl-3-[5-chloro-2-(pyridin-2-yl)-indol-3-
yl]acrylate (16b)
See Section 5.5. The isolation was performed with ethyl acetate/
petrol ether (1:1). Bright yellow solid (2.0 g) (64%) (6.4 mmol), m.p.
d
¼ 3.97 (s, 3H, CH3), 7.32 (m, 1H, H-8), 7.50 (m, 1H, H-9), 7.72 (m,
182 ꢃC; 1H NMR (DMSO-d6):
d
¼ 3.71 (s, 3H, CH3), 6.50 (d, 1H,
2H, H-4 þ10), 8.34 (m, 1H, H-7), 9.01 (m, 1H, H-3), 9.13 (m, 1H, H-
3J ¼ 16.2 Hz, ]CH), 7.29 (m, 1H, indole-H-6), 7.46–7.54 (m, 2H,
indole-H þ pyridyl-H), 7.83 (pd, 1H, J ¼ 7.8, pyridyl-H-3), 7.94, (m,
1H, indole-H-4), 8.03 (m, 1H, pyridyl-H), 8.46 (d, 1H, 3J ¼ 16.2 Hz,
]CH), 8.79 (pd, 1H, J ¼ 4.0), (pyridyl-H-6), 12.38 (s, 1H, N-H); 13C
6), 9.47 (m, 1H, H-2), 11.96 (s, 1H, N-H); 13C NMR (DMSO-d6):
d
¼ 52.3, 112.7, 116.3, 118.7, 120.8 (2), 122.4, 123.5, 125.4, 126.3,
126.4, 134.8, 137.1, 138.7, 139.9, 149.1, 167.4; FTIR (cmꢀ1): 3180,
2946, 1708, 1518, 1429, 1346, 1235, 1220, 1131, 1059, 727; FD-MS:
m/z 276.6 [Mþ], C17H12N2O2.
(DMSO-d6):
d
¼ 51.5 (p), 109.7 (q), 114.5 (t), 114.5 (t), 120.2 (t), 123.8
(t), 124.0 (t), 126.4 (q), 127.2. (q), 135.6 (q), 137.8 (t), 139.0 (2t), 140.9
(q), 149.9 (q), 150.3 (t), 167.9 (q); EI-MS: m/z 312.1 [Mþ],
C17H13ClN2O2.
5.5.2. Methyl-8-chloro-11H-pyrido[2,3-a]carbazole-5-carboxylate
(17b)
625 mg of 16b was dissolved in 450 ml CH2Cl2 and irradiated for
6 h. The progress of the reaction via TLC and the isolation of the
chloropyridocarbazole by column chromatography on silica gel
were performed with petrol ether/ethyl acetate (2:1). White solid
5.4.3. (E)-Methyl-3-[5-chloro-2-(pyridin-4-yl)-indol-3-
yl]acrylate (16c)
See Section 5.5. The isolation was performed with ethyl
acetate/ethanol (6:1). Bright yellow solid (2.2 g) (70%) (7.0 mmol),
(410 mg) (66%) (1.3 mmol), m.p. 256 ꢃC; 1H NMR (CDCl3):
d
¼ 4.06
m.p. 251 ꢃC (decomposition); 1H NMR (DMSO-d6):
d
¼ 4.0 (s, 3H,
(s, 3H, CH3), 7.45–7.55 (m, 2H), 7.66 (m, 1H), 8.15 (m, 1H), 8.92 (m,
CH3), 6.52 (d, 1H, 3J ¼ 16.0 Hz, ¼ CH), 7.31 (m, 1H, indole-H-6),
7.52 (m, 1H, indole-H-7), 7.59 (pd, 2H, J ¼ 6.0, pyridyl-H-3 þ 5),
7.77 (d, 1H, 3J ¼ 16.0 Hz, ¼ CH), 8.0 (m, 1H, indole-H-4), 8.79 (pd,
2H, J ¼ 6.0, pyridyl-H-2 þ 6), 12.31 (s, 1H, N-H); 13C (DMSO-d6):
1H), 9.06 (m, 1H), 9.75 (m, 1H), 10.69 (s, 1H, N-H); 13C NMR (DMSO-
d6):
d
¼ 52.3 (p), 114.1 (t), 116.9 (q), 118.0 (q), 120.5 (t), 122.7 (t),
124.8 (q), 125.1 (q), 125.7 (q), 126.2 (t), 126.8 (t), 134,9 (t), 137.1 (q),
138.4 (q), 139.4 (q), 149.2 (t), 167.2 (q); FD-MS: m/z 310.2 [Mþ],
C17H11ClN2O2.
d
¼ 51.6 (p), 109.5 (q), 114.3 (t), 144.8 (t), 120.2 (t), 122.3 (t), 123.9
(2t), 124.1 (t), 126.6 (q), 126.9 (q), 135.8 (q), 138.2 (q), 140.9
(q), 150.6 (2t), 167.7 (q); FT-IR (cmꢀ1): 2746–3120, 1704, 1622,
1597, 1299, 1283, 998, 826; FD-MS: m/z 312.3 [Mþ],
C17H13ClN2O2.
5.5.3. Methyl-8-chloro-11H-pyrido[4,3-a]carbazole-5-
carboxylate (17c)
625 mg of 16c was dissolved in 450 ml ethanol and irradiated
for 6 h. The progress of reaction via TLC and the isolation of the
chloropyridocarbazole by column chromatography on silica gel
were performed with ethyl acetate/ethanol (3:1). White solid
(390 mg) (64%) (1.3 mmol), m.p. 285–290 ꢃC (decomposition); 1H
5.4.4. (E)-Methyl-3-[7-chloro-2-(pyridin-2-yl)-indol-3-
yl]acrylate (16d)
See Section 5.5. The isolation was performed with petrol ether/
ethyl acetate (2:1). Bright yellow solid (2.1 g) (67%) (6.7 mmol), m.p.
NMR (DMSO-d6):
d
¼ 4.00 (s, 3H, CH3), 7.54 (m, 1H, H-9), 7.74 (m,
152–155 ꢃC; 1H NMR (CDCl3):
d
¼ 3.84 (s, 3H, CH3), 6.63 (d, 1H,
1H, H-10), 8.44 (m, 1H, H-1), 8.54 (m, 1H, H-7), 8.75 (m, 1H, H-2),
3J ¼ 16.0 Hz, ]CH), 7.17 (m,1H, indole-H), 7.27 (m,1H), 7.34 (m,1H),
7.83 (m, 3H), 8.23 (d, 1H, 3J ¼ 16.0 Hz, ]CH), 8.71 (pd, 1H, J ¼ 4.5,
9.18 (m, 1H, H-6), 10.30 (m, 1H, H-4), 11.76 (s, 1H, N-H); 13C NMR
(DMSO-d6):
d
¼ 52.5 (p), 114.0 (t), 115.6 (t), 117.8 (q), 118.2 (q), 120.8
pyridyl-H-6), 9.91 (s, 1H, N-H); 13C NMR (CDCl3):
d
¼ 51.7 (p), 112.0
(t), 124.5 (q), 124.7 (q), 125.5 (q), 126.6 (t), 127.4 (t), 137.3 (q), 138.4
(q), 143.4 (t), 143.5 (q), 150.5 (t), 167.2 (q); FD-MS: m/z 310.8 [Mþ],
C17H11ClN2O2.
(2q), 117.5 (q), 117.7 (t), 119.6 (t), 122.4 (t), 123,4 (t), 123.7 (t), 124.0
(t), 128.3 (q), 133.7 (q), 137.3 (t), 37.7 (t), 148.5 (q), 149.4 (t), 168.2
(q); FTIR (cmꢀ1): 3391, 1710, 1615, 1429, 1292, 1163; FD-MS: m/z
312.4 [Mþ], C17H13ClN2O2.
5.5.4. Methyl-10-chloro-11H-pyrido[2,3-a]carbazole-5-
carboxylate (17d)
5.4.5. (E)-Methyl 3-[2-(thiophen-2-yl)-indol-3-yl]acrylate
(16e) [29]
625 mg of 16d was dissolved in 450 ml CH2Cl2 and irradiated for
6 h. The progress of reaction via TLC and the isolation of the
chloropyridocarbazole by column chromatography on silica gel
were performed with petrol ether/ethyl acetate (2:1). White solid
See Section 5.5. The isolation was performed with chloroform/
ethyl acetate (4:1). Bright yellow solid (1.3 g) (46%) (4.6 mmol), m.p.
172 ꢃC; 1H NMR (CDCl3):
d
¼ 3.82 (s, 3H, CH3), 6.61 (d, 1H,
(425 mg) (68%) (1.4 mmol), m.p. 180 ꢃC; 1H NMR (CDCl3):
d
¼ 4.06
3J ¼ 16.0 Hz, ]CH), 7.18 (m, 1H), 7.26–7.34 (m, 3H), 7.41 (m, 1H),
7.48 (m,1H), 7.95 (m,1H), 8.15 (d,1H, 3J ¼ 16.0 Hz, ]CH), 8.47 (s,1H,
N-H); FD-MS: m/z 283.5 [Mþ], C16H13NO2S.
(s, 3H, CH3), 7.25 (m, 1H), 7.32 (m, 1H), 7.53 (m, 1H), 7.68 (m, 1H),
8.07 (m,1H), 8. 94 (m,1H), 9.06 (m,1H, H-6), 9.74 (s,1H); FD-MS: m/
z 310.0 [Mþ], C17H11ClN2O2.

3250
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
5.5.5. Methyl-10H-thieno[2,3-a]carbazole-4-carboxylate (17e)
567 mg of 16e was dissolved in 450 ml CH2Cl2 and irradiated for
3 h. The progress of reaction via TLC and the isolation of the thie-
nocarbazole by column chromatography on silica gel were per-
formed with petrol ether/ethyl acetate (3:1). Bright yellow solid
(253 mg) (47%) (0.9 mmol), m.p. 213 ꢃC; 1H NMR (DMSO-d6):
d
¼ 2.03 (m, 2H, C–CH2–C), 2.78 (d, 6H, N(CH3)2), 3.18 (m, 2H, ꢀCH2
next to amine-N), 3.44 (m, 2H, next to amidic-N), 7.32 (m, 1H), 7.50
(m, 1H), 7.73 (m, 2H), 8.30 (m, 1H), 8.71 (m, 1H), 8.88 (m, 1H), 9.06
(m, 2H), 10.50 (s,1H, N-H),12.08 (s, 1H, indole-N-H); 13C NMR of the
hydrochloride (methanol-d4): 26.2 (s), 38.0 (s), 43.7 (p), 56.9 (s),
113.4 (t), 121.6 (t), 122.1 (t), 122.4(t), 123.1 (q), 123.8 (q), 123.9 (t),
125.2 (q), 126.2 (q), 128.7 (t), 132.1 (q), 132.7 (q), 141.2 (q), 142.1 (t),
145.9 (t), 171.4 (q); FD-MS: m/z 346.1 [Mþ], 347.1 [27%]
C21H23ClN4O.
d
¼ 3.95 (s, 3H, CH3), 7.27 (m, 1H), 7.45 (m, 1H), 7.58 (m, 1H, H-9),
7.94 (pd, 1H, 3J ¼ 5.5, H-3), 8.27 (m, 1H, H-6), 8.34 (pd, 1H, 3J ¼ 5.5,
H-2), 8.89 (m, 1H, H-5), 12.39 (s, 1H); 13C NMR (DMSO-d6):
d
¼ 52.1
(p), 111.9 (t), 115.8. (q), 117.7 (q), 120.5 (t), 120.7 (t), 122.3 (t), 123.5
(q), 125.8 (t), 126.1 (t), 127.9 (t), 137.4 (q), 137.9 (q), 139.9 (q), 140.1
(q), 167.2 (q); FD-MS: m/z 281.5 [Mþ], C16H11NO2S.
5.7.3. 8-Chloro-N-[2-(dimethylamino)ethyl]-11H-pyrido[2,3-
a]carbazole-5-carboxamide (22)
310 mg of 18b and 0.2 ml N,N-dimethylethylenediamine were
5.6. General procedure for the preparation of the hetarene
added. White solid (288 mg) (75%) (0.8 mmol), m.p. 255 ꢃC; 1H
annelated [a]-carbazole carboxylic acids via hydrolysis
NMR (DMSO-d6):
d
¼ 2.30 (s, 6H, N(CH3)2), 2.59 (t, 2H, –CH2 next to
amine-N), 3.49 (t, 2H, –CH2 next to amidic-N), 7.46 (m, 1H), 7.66 (m,
2H), 8.36 (m, 1H), 8.56 (m, 2H), 8.91 (m, 1H), 8.99 (m, 1H), 12.04
2 mmol of the hetarene annelated [a]-carbazole 17 (553 mg of
17a, 622 mg of 17b–d, 563 mg of 17e) was suspended in 20 ml
ethanol and 1 ml of an aqueous sodium hydroxide solution and
refluxed for 2 h. The mixture was cooled down at room tempera-
ture, neutralized first with concentrated hydrochloric acid and
subsequent with acetic acid. After 1 h the precipitate was filtered
off, dried for 24 h by remaining at the air and then under vacuum.
The purity of the free carboxylic acids was sufficient, so that they
could be applied for further reactions.
(s, 1H, N-H); 13C NMR (DMSO-d6):
d
¼ 37.6 (s), 45.4 (p), 58.4 (s),
114.0 (t), 118.0 (q), 120.1 (t), 120.9 (t),121.8 (t), 124.5 (q), 124.6 (q),
124.8 (q), 125.6 (q), 125.7 (t), 135.1 (t), 137.2 (q), 137.4 (q), 138.2 (q),
149.2 (t), 138.5 (q); FD-MS: m/z 366.2 [Mþ], 368.3 [34%]
C20H19ClN4O.
5.7.4. 8-Chloro-N-[(2-(dimethylamino)propyl)-11H-pyrido[2,3-
a]carbazole-5-carboxamide] (23)
310 mg of 18b and 0.2 ml N,N-dimethylpropylenediamine were
5.7. General procedure for the preparation of the hetarene
annelated [a]-carbazole amides by use of EDI/HOBt as
coupling reagents
added. White solid (290 mg) (73%) (0.8 mmol), m.p. 256 ꢃC; 1H
NMR (DMSO-d6):
d
¼ 1.75 (m, 2H, C–CH2–C), 2.17 (s, 6H, N(CH3)2),
2.34 (t, 2H, –CH2 next to amine-N), 3.38 (t, 2H, –CH2 next to amidic-
N), 7.46 (m, 1H), 7.66 (m, 2H), 8.37 (m, 1H), 8.57 (m, 1H), 8.62 (m,
1H), 8.87, (m, 1H), 8.99 (m, 1H), 12.04 (s, 1H, indole-N-H); 13C NMR
2.6 mmol (404 mg) 1-ethyl-3-[3-(dimethylamino)propyl]carbo-
diimide (EDCI), 1.045 mmol (141 mg) 1-hydroxybenzotriazole
(HOBt), 1.5 mmol of the amine and 1.045 mmol of the hetarene
annelated [a]-carbazole carboxylic acids 18 were dissolved in 10 ml
DMF and stirred for 24 h under nitrogen atmosphere. The urea was
filtered off, the filtrate was evaporated to dryness and the residue
was chromatographed on silica gel (methanol/NH3 (25%), 97:3).
(DMSO-d6)
d
¼ 27.5 (s), 38.1 (s), 45.5 (p), 57.2 (s), 114.0 (t), 118.0 (q),
120.1 (t), 120.9 (t), 121.8 (t), 124.5 (q), 124.6 (q), 124.8 (q), 125.7 (t),
125.9 (q), 135.0 (t), 137.2 (q), 137.3 (q), 138.2 (q), 149.2 (t), 168.5 (q);
FD-MS: m/z 380.2 [Mþ], 382.2 [34%] C21H21ClN4O.
5.7.5. 8-Chloro-N-[2-(dimethylamino)ethyl]-11H-pyrido[4,3-
a]carbazole-5-carboxamide (24)
5.7.1. N-[2-(Dimethylamino)ethyl]-11H-pyrido[2,3-a]carbazole-5-
carboxamide (20)
310 mg of 18c and 0.2 ml N,N-dimethylethylenediamine were
added. White solid (260 mg) (68%) (0.7 mmol), m.p. 289 ꢃC; 1H
274 mg of 18a and 0.2 ml N,N-dimethylethylenediamine were
NMR (methanol-d6):
d
¼ 2.40 (s, 6H, N(CH3)2), 2.71 (t, 2H, –CH2 next
added. White solid (253 mg) (73%) (0.8 mmol), m.p. 217 ꢃC; 1H NMR
to amine-N), 3.68 (t, 2H, –CH2 next to amidic-N), 7.48 (m, 1H), 7.62
(DMSO-d6):
d
¼ 2.29 (s, 6H, N(CH3)2), 2.58 (t, 2H, –CH2 next to
(m, 1H), 8.19 (m, 1H), 8.25 (m, 1H), 8.57 (m, 2H), 9.77 (m, 1H); 13C
amine-N), 3.48 (t, 2H, –CH2 next to amidic-N), 7.28 (m, 1H), 7.46 (m,
1H), 7.65 (m, 2H), 8.24 (m, 1H), 8.56 (m, 2H), 8.94 (m, 1H), 8.98 (m,
NMR (DMSO-d6)
d
¼ 7.9 (s), 45.6 (p), 58.6 (s),113.9 (t),115.4 (t),118.2
(2q), 120.4 (t), 121.4 (t), 124.3 (2q), 124.8 (q), 126.2 (t), 126.7 (q),
135.2 (q), 138.2 (q), 143.5 (t), 150.9 (t), 168.3 (q); FD-MS: m/z 366.6
[Mþ], 368.6 [35%] C20H19ClN4O.
1H),12.21 (s,1H, indole-N-H); 13C NMR (DMSO-d6):
d
¼ 37.6 (s), 45.4
(p), 58.4 (s), 112.5 (t), 118.7 (q), 120.1 (t),120.5 (t), 120.8 (t), 121.4 (t),
123.5 (q), 124.5 (q), 125.1 (q), 125.8 (t), 135.1 (t), 136.6 (q), 137.2 (q),
139.8 (q), 148.9 (t), 168.7 (q); FD-MS: m/z 332.2 [Mþ], C20H20N4O.
5.7.6. 8-Chloro-N-[2-(dimethylamino)propyl]-11H-pyrido[4,3-
a]carbazole-5-carboxamide (25)
5.7.2. N-[2-(Dimethylamino)propyl]-11H-pyrido[2,3-a]carbazole-
5-carboxamide hydrochloride (21)
310 mg of 18c and 0.2 ml N,N-dimethylpropylenediamine were
added. White solid (271 mg) (71%) (0.8 mmol), m.p. 283 ꢃC; 1H
274 mg of 18a and 0.2 ml N,N-dimethylpropylenediamine were
added. White solid (280 mg) (77%) (0.8 mmol). Because of unclear
correlations concerning the NMR signals of aliphatic protons in the
high-field the free base had do convert into her corresponding
hydrochloride. Therefore the free base was dissolved in ether and
after adding some drops of ethereal HCl the salt precipitated
immediately. Free base: bright yellow solid m.p. 227 ꢃC; hydro-
chloride: white solid, m.p. 295 ꢃC; 1H NMR (DMSO-d6) of the free
NMR (DMSO-d6):
d
¼ 1.76 (m, 2H, C–CH2–C), 2.20 (s, 6H, N(CH3)2),
2.38 (t, 2H, –CH2 next to amine-N), 3.42 (t, 2H, –CH2 next to amidic-
N), 7.51 (m, 1H), 7.72 (m, 1H), 8.39 (m, 1H), 8.42 (m, 1H), 8.59 (m,
1H), 8.69 (m, 2H), 9.70 (m, 1H), 11.98 (s, 1H, indole-N-H); 13C NMR
(DMSO-d6)
d
¼ 27.3 (s), 28.0 (s), 45.4 (p), 57.1 (s), 113.9 (t), 115.4 (t),
118.2 (2q), 120.3 (t), 121.5 (t), 124.3 (2q), 124.8 (q), 126.2 (t), 126.6
(q), 135.3 (q), 138.2 (q), 143.5 (t), 150.8 (t), 168.3 (q); FD-MS: m/z
381.1 [Mþ], 383.1 [36%] C21H21ClN4O.
base:
d
¼ 1.75 (m, 2H, C–CH2–C), 2.16 (s, 6H, N(CH3)2), 2.33 (t, 2H,
–CH2 next to amine-N), 3.39 (t, 2H, next to amidic-N), 7.28 (m, 1H),
7.46 (m, 1H), 7.66 (m, 2H), 8.25 (m, 1H), 8.53 (s, 1H), 8.65 (m, 1H),
8.90 (m, 1H), 8.98 (m, 1H), 12.21 (s, 1H, N-H); FD-MS: m/z 346.1
[Mþ], C21H22N4O; 1H NMR (DMSO-d6) of the hydrochloride:
5.7.7. 10-Chloro-N-[2-(dimethylamino)ethyl]-11H-pyrido[2,3-
a]carbazole-5-carboxamide (26)
310 mg of 18d and 0.2 ml N,N-dimethylethylenediamine were
added. Bright yellow solid (215 mg) (56%) (0.6 mmol), m.p. 199 ꢃC;

T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
3251
1H NMR (methanol-d6):
d
¼ 2.40 (s, 6H, N(CH3)2), 2.72 (t, 2H, –CH2
a Becton Dickinson FACScan flow cytometer using the LYSYS II
software. PI was excited at 488 nm and the emitted fluorescence
was quantified at 620 nm on channel Fl-3. WinMdi software was
used to determine the percentage of cells in G1, S or G2/M phases.
Each experiment was performed in duplicate.
next to amine-N), 3.69 (t, 2H, –CH2 next to amidic-N), 7.18 (m, 1H),
7.41 (m, 2H), 7.95 (m, 1H), 8.38 (m, 1H), 8.77 (m, 2H); 13C NMR
(DMSO-d6):
d
¼ 37.8 (s), 45.6 (p), 58.5 (s), 116.9 (q), 119.3 (q), 119.4
(t), 120.6 (t), 121.2 (t), 121.8 (t), 124.9 (q), 125.5 (q), 125.6 (t), 126.6
(q), 135.1 (t), 136.8 (q), 137.1 (q), 137.4 (q), 149.1 (t), 168.4 (q); FD-
MS: m/z 365.8 [Mþ], 367.8 [31%] C20H19ClN4O.
5.9. UV/vis absorption spectroscopy and DNA-melting
temperature studies
5.7.8. 10-Chloro-N-[2-(dimethylamino)propyl]-11H-pyrido[2,3-
a]carbazole-5-carboxamide (27)
310 mg of 18d and 0.2 ml N,N-dimethylpropylenediamine were
added. Bright yellow solid (225 mg) (59%) (0.6 mmol), m.p. 195 ꢃC;
CT-DNA and double stranded poly(dAdT)2 oligonucleotide were
purchased from Sigma. CT-DNA was prepared as previously
described [12]. The synthesized compounds were prepared as
10 mM solutions in DMSO and further freshly diluted in the
appropriate aqueous buffer.
UV/vis spectral absorption measurement was conducted using
20 mM of the various tested drugs incubated in 1 ml of BPE buffer
(6 mM Na2HPO4, 2 mM NaH2PO4, 1 mM EDTA, pH 7.1) in the pres-
ence or absence of increasing concentrations of CT-DNA (10, 20, 30,
40, 50, 60, 70, 80, 90,100,120,140,160,180, 200 mM of base pairs) in
1H NMR (methanol-d6):
d
¼ 2.0 (m, 2H, C–CH2–C), 2.42 (s, 6H,
N(CH3)2), 2.65 (t, 2H, –CH2 next to amine-N), 3.55 (t, 2H, –CH2 next
to amidic-N), 7.24 (m, 1H, H-8), 7.44 (m, 1H, H-9), 7.51 (m, 1H, H-3),
8.05 (m, 1H, H-7), 8.41 (s, 1H, H-6), 8.84 (m, 2H, H-2 þ 4); 13C NMR
(DMSO-d6):
d
¼ 27.3 (s), 38.0 (s), 45.4 (p), 57.1 (s), 117.0 (q), 119.3 (q),
119.4 (t), 120.6 (t), 121.2 (t), 121.8 (t), 124.9 (q), 125.5 (q), 125.6 (t),
126.6 (q), 135.1 (t), 136.8 (q), 137.1 (q), 137.5 (q), 149.1 (t), 168.4 (q);
FD-MS: m/z 380.4 [Mþ], 382.4 [33%]; C21H21ClN4O.
a quartz cuvette of 10 mm path length. The UV/vis spectra were
recorded from 230 nm to 430 nm using an Uvikon 943 spectro-
photometer and are referenced against a cuvette containing DNA at
the same concentration than in the sample cuvette to directly
substrate intrinsic DNA absorption.
5.7.9. N-[3-(Dimethylamino)propyl]-10H-thieno[2,3-a]carbazole-
4-carboxamide hydrochloride (28)
286 mg of 18e and 0.2 ml N,N-dimethylpropylenediamine were
added. The free base 28 was dissolved in ether and precipitated
with ethereal HCl to generate the hydrochloride. Bright yellow solid
(156 mg) (39%) (0.4 mmol), m.p. 243 ꢃC (decomposition); 1H NMR
The variation of melting temperature was calculated from the
melting temperature measurement of 20 mM of CT-DNA or pol-
(DMSO-d6):
d
¼ 1.99 (m, 2H, C–CH2–C), 2.78 (s, 6H, N(CH3)2), 3.16 (t,
y(dAdT)2 incubated alone (control Tm) or with increasing
concentration of the various compounds (drug/base pair ratio of
0.1, 0.25, 0.5 or 1) in 1 ml BPE buffer. The absorbency of DNA at
260 nm was measured in quartz cells using the Uvikon 943
spectrophotometer thermostated with a Neslab RTE111 cryostat.
Absorption value was measured every minute over a range of
20–100 ꢃC with an increment of 1 ꢃC/min. The Tm values were
obtained from the midpoint of the hyperchromic transition.
2H, –CH2), 3.42 (t, 2H, –CH2), 7.26 (m,1H), 7.43 (m,1H), 7.58 (m,1H),
7.82 (m, 1H), 8.15 (m, 2H), 8.56 (s, 1H), 8.71 (m, 1H), 10.15 (s, 1H,
indole-N-H); 13C NMR (DMSO-d6):
d
¼ 24.8 (s), 36.7 (s), 42.5 (p),
55.0 (s), 111.9 (t), 117.4 (q), 118.0 (t), 120.1 (t), 120.3 (t), 122.4 (q),
134.4 (q), 123.6 (q), 125.7 (t), 125.8 (t), 126.4 (t), 135.8 (q), 137.2 (q),
140.0 (q), 168.8 (q); FD-MS: m/z 351.0 [Mþ]; C20H21N3OS.
5.8. Cytotoxicity and cell cycle analysis on HT-29 cell line
D
Tm values ¼ Tm[DrugþDNA] ꢀ Tm[DNA alone]
.
HT-29 colon carcinoma cells were grown in DMEM-glutaMAX
medium (Gibco) supplemented with 10% fetal calf serum (FCS),
5.10. Fluorescence spectroscopy
Fluorescence spectral measurement was recorded using 20
penicillin (100 IU/ml) and streptomycin (100
atmosphere at 37 ꢃC under 5% CO2.
mg/ml) in a humidified
mM
The cytotoxicity of the various compounds was assessed using
the MTS cell proliferation assay developed by Promega (CellTiter
96Ò AQueous one solution cell proliferation assay). Briefly, 3 ꢁ103
exponentially growing HT-29 cells were seeded in 96-well micro-
culture plates for 24 h prior to be subjected to treatment using
graded concentrations of the tested compounds from 1 nM to
of the fluorescent drugs incubated in 1 ml of BPE buffer in the
presence or absence of increasing concentrations of CT-DNA (10, 20,
30, 40, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200 mM of base
pairs) in a quartz cuvette of 10 mm path length. The excitation
wavelengths were 312 (compound 20), 319 (compounds 21, 22),
318 (compound 23), 276 (compounds 24, 25) and 300 nm
(compound 28). The quenching constant Ksv was deduced from
Stern–Volmer method where the ratio of fluorescence of the
compound alone (F0) over the fluorescence of the compound in the
presence of CT-DNA (F) is presented as a function of CT-DNA
concentration. In this configuration, F0/F ¼ 1 þ Ksv[CT-DNA]. The
slope Ksv is considered as an equilibrium constant for the static
quenching process.
100
m
M in 3 different series of duplicate or triplicate points. After
72 h of incubation at 37 ꢃC, 20
ml of the tetrazolium dye were added
to each well and the samples were incubated for a further 2 h at
37 ꢃC. Plates were analyzed on a Labsystems Multiskan MS (type
352) reader at 492 nm. The GC50 was determined as the concen-
tration for which the number of cell is one half of that obtained in
the non-treated control wells (6 individual points per plate) using
graphical analyses using SoftMax Pro 4.7.1 software.
For flow cytometric analysis of DNA content, 3 ꢁ 105 cells in
exponential growth phase were treated with graded concentra-
tions of the test drug for 24 h as indicated in figure legends. The
cells were then washed twice with phosphate buffered saline (PBS),
5.11. Circular dichroism
The different drugs (50 mM) were incubated in 1 ml of sodium
cacodylate (1 mM, pH 7.0) with or without (control) increasing
concentration of CT-DNA (10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 120,
incubated with 100
again with 500 L of culture medium to inhibit trypsin and har-
vested. Cells were then fixed using 70% ethanol. Finally, 500 L of
PBS containing propidium iodide (PI, FluoProbes) at 50 g/ml and
RNase (Sigma) at 100 g/ml were added for 30 min at room
temperature in the darkness. Cell samples were analyzed on
m
L of trypsin (Gibco) for 5 min at 37 ꢃC, washed
m
m
140, 160, 180, 200 mM). The CD spectra were collected from 430 to
m
230 nm with a resolution of 0.1 nm, in a quartz cell of 10 mm path
length, using a J-810 Jasco spectropolarimeter at 20 ꢃC controlled
by a PTC-424S/L peltier type cell changer (Jasco).
m

3252
T. Lemster et al. / European Journal of Medicinal Chemistry 44 (2009) 3235–3252
5.12. DNase I footprinting
were run without ethidium bromide and then stained using a bath
containing ethidium bromide. For topoisomerase I DNA cleavage
assays, the samples were treated as for the relaxation assays but
loaded on a 1% agarose gel that was prepared with ethidium
DNase I footprinting experiments were performed essentially as
described in Ref. [30]. A 117 bp DNA fragment was obtained from
double digestion of the pBS plasmid (Stratagene, La Jolla, CA) at
EcoRI and PvuII restriction sites for 1 h with the corresponding
enzymes in their respective buffers. The generated DNA fragment
bromide. Camptothecin (CPT, 20
topoisomerase poisoning effect. For topoisomerase II DNA
cleavage assays, supercoiled pUC19 plasmid (130 ng) was incubated
with 20 or 50 M of compounds 20–24, or etoposide as a control,
mM) was used as a control for
I
was 30-end labelled by incubation with
a
-[32P]-dATP (3000 Ci/
m
mmol, GE Healthcare, Buckinghamshire, England) and 10 units of
Klenow enzyme (BioLabs) for 30 min at 37 ꢃC. The 117 bp radio-
labelled DNA fragment was then isomated on a 6% polyacrylamide
gel under native conditions in TBE buffer (89 mM Tris base, 89 mM
boric acid, 2.5 mM Na2 EDTA, pH 8.3), cut off from the gel, crushed
and dialyzed overnight against 400 ml of elution buffer (10 mM
Tris–HCl pH 8.0, 1 mM EDTA, 100 mM NaCl). This suspension was
separated from polyacrylamide gel by filtration through a Millipore
prior to the addition of 4 units of human topoisomerase II (Topogen,
USA) at 37 ꢃC for 45 min in the appropriate cleavage buffer. SDS
(0.25%) and proteinase K (250 mg/ml) were then added to stop the
reaction during 30 min at 50 ꢃC. DNA samples were loaded on 1%
agarose gels containing ethidium bromide for 2 h at 120 ꢃC in TBE
buffer. After migration, both gels were washed and photographed
under UV light.
0.22 mm membrane. The DNA was precipitated with cold ethanol
Acknowledgments
followed by centrifugation. The radio-labelled DNA fragment was
then dried and the pellet dissolved in water. Appropriate concen-
trations of the various ligands (as indicated in the figure legends)
were incubated with the 117-bp radio-labelled DNA fragment
15 min incubation at 37 ꢃC to ensure equilibrium prior to digestion
of the DNA by the addition of DNase I (final concentration 0.001
unit/ml) in 20 mM NaCl, 2 mM MgCl2, 2 mM MnCl2, pH 7.3. After
3 min of digestion, the reaction was stopped by freeze-drying.
This work was supported by grants from the Ligue Nationale
contre le Cancer (Comite du Nord) and the IRCL to M.-H.D.-C. G.L.
thanks the Universite´ de Lille 2 and the Conseil Re´gional Nord/Pas-
de-Calais for a Ph.D. fellowship.
´
References
Samples were lyophilized and subsequently dissolved in 4 mL of
[1] H. Mastalarz, R. Jasztold-Howorko, F. Rulko, A. Croisy, D. Carrez, Arch. Pharm.
Pharm. Med. Chem. 337 (2004) 434–439.
[2] M.M. Harding, A.R. Grummitt, Mini Rev. Med. Chem. 3 (2003) 67–76.
denaturing loading buffer (80% formamide solution containing
tracking dyes). The DNA samples were heated at 90 ꢃC for 4 min
and chilled on ice for 4 min prior to electrophoresis. DNA cleavage
products were resolved under polyacrylamide electrophoresis in
denaturing conditions (0.35 mm thick, 8% polyacrylamide con-
taining 8 M urea). After a 90 min electrophoresis at 65 W in TBE
buffer, gels were soaked in 10% acetic acid for 10 min, transferred to
Whatman 3 MM paper and dried under vacuum at 80 ꢃC. A
Molecular Dynamics STORM 860 was used to collect data from
storage screens exposed to dried gels overnight at room tempera-
ture. The identity of the bases was established from comparison of
the relative position of the bands to the guanines sequencing
standard (G-track) classically obtained using dimethyl-sulfate and
piperidine treatment of the same DNA fragment.
´
[3] A. Pierre, G. Atassi, M. Devissaguet, E. Bissagli, Drugs Future 22 (1997) 53–59.
[4] G.W. Gribble, The Alkaloids, , In: Synthesis and Antitumor Activity of the
Ellipticine Alkaloids and Related Compounds, vol. 39 (1990).
´
´
[5] J. Rouesse, M. Spielman, F. Turpin, T. Le Chevalier, M. Azab, J.M. Mondesir, Eur.
J. Cancer 29A (1993) 856–859.
[6] U. Pindur, Pharm. Unserer Zeit. 16 (1987) 47–52.
[7] M.F. Brana, M. Cacho, A. Gradillas, B. de Pascual-Teresa, A. Ramos, Curr. Pharm.
Des. 7 (2001) 1745–1780.
[8] C. Guillonneau, A. Nault, E. Raimbaud, S. Le´once, L. Kraus- Berthier, A. Pierre´,
S. Goldstein, Bioorg. Med. Chem. 13 (2005) 175–184.
´
[9] R. Jasztold-Howorko, C. Landras, A. Pierre, G. Atassi, N. Guilbaud, L. Kraus-
Berthier, S. Le´once, Y. Rolland, J.F. Prost, E. Bisagni, J. Med. Chem. 37 (1994)
2445–2452.
[10] C. Asche, M. Demeunynck, Anti-cancer agents, Med. Chem. 7 (2007) 247–267.
[11] D. Pelaprat, R. Oberlin, I. Le Guen, B.P. Roques, J.B. Pecq, J. Med. Chem. 23
(1980) 1330–1335.
[12] U. Pindur, T. Lemster, Recent Res. Dev. Org. Bioorg. Chem. 1 (1997) 33–53.
[13] T. Lemster, U. Pindur, Recent Res. Dev. Org. Bioorg. Chem. 5 (2002) 99–115.
[14] R. Martinez, L. Chacon-Garcia, Curr. Med. Chem. 12 (2005) 127–151.
[15] M.H. David-Cordonnier, M.P. Hildebrand, B. Baldeyrou, A. Lansiaux, C. Keuser,
K. Benzschawel, T. Lemster, U. Pindur, Eur. J. Med. Chem. 42 (2007) 752–771.
[16] C. Keuser, U. Pindur, Pharmazie 4 (2006) 261–268.
[17] U. Pindur, M. Jansen, T. Lemster, Curr. Med. Chem. 12 (2005) 2805–2847.
[18] C. Hotzel, A. Marotto, U. Pindur, Eur. J. Med. Chem. 36 (2003) 189–197.
[19] U. Pindur, T. Lemster, Recent Res. Dev. Org. Bioorg. Chem. 5 (2002) 99–115.
[20] C. Perlick. Dissertation: 2,3-Divinyl- und 2-Aryl-3-vinylindole als Bausteine zu
Carbazolen und Versuche zur photochemischen und Ultraschall-induzierten
1,6-Elektrocyclisierung, 2003.
[21] National Cancer Institute (NCI). Bethesda. http://dtp.nci.nih.gov.
[22] C.H. Spink, S.E. Wellman, Methods Enzymol. 340 (2001) 193–211.
[23] Muth, et al., J. Heterocycl. Chem. 9 (1972) 1299.
[24] Zhaiwei Lin, Pankaja Kadaba, Heterocycl. Commun. 8 (2002) 565–568.
[25] A.Y. Hassan, M.M. Ghorab, O.M. Nassar, Phosphorus, Sulfur Silicon Relat. Elem.
134 (1998) 77–86.
5.13. Topoisomerase I-mediated DNA relaxation, topoisomerases I
or II cleavage assays
Topoisomerase I-mediated DNA relaxation experiments were
performed as previously described in Ref. [15] with the following
modifications. Graded concentrations of the tested compounds
were incubated in the presence of supercoiled pUC19 plasmid DNA
(130 ng) prior to the addition of 4 units of human topoisomerase I
(Topogen, USA) at 37 ꢃC for 45 min in relaxation buffer [50 mM
tris(hydroxymethyl)aminomethane (pH 7.8), 50 mM KCl, 10 mM
MgCl2, 1 mM dithiothreitol, 1 mM EDTA and 1 mM ATP]. The reac-
tions were stopped upon addition of SDS to 0.25% and proteinase K
to 250 m ml
g/ml and incubation of the samples for 30 min at 50 ꢃC. 3
[26] K.B. Prasad, G.A. Swan, J. Chem. Soc. (1958) 2024–2038.
[27] H.B. Shivarama, A. Sarvottam, J. Ind. Chem. Soc. 51 (1974) 965–966.
[28] M. Somei, S. Sayama, K. Naka, F. Yamada, Heterocycles 27 (1988) 1585–1587.
[29] I. Fernandez, C. Galvez, L. Urpi, An. Quim. 87 (1991) 936–939.
[30] C. Bailly, M.J. Waring, J. Biomol. Struct. Dyn. 12 (1995) 869–898.
of the electrophoresis dye mixture were added to the DNA samples
which were then loaded on a 1% agarose gel. The various DNA forms
were separated by electrophoresis at room temperature for 2 h at
120 V in TBE buffer (89 mM Tris–borate pH 8.3, 1 mM EDTA). Gels