Mannich bases of 3H-pyrrolo[3,2-f]quinoline having vasorelaxing activity

Article abstract of DOI:10.1016/S0223-5234(02)01355-7

Mannich bases obtained by aminoalkylation of 3H-pyrrolo[3,2-f]quinoline were designed and prepared as potential vasorelaxing agents. Compounds Ia-Va were characterised by IR, ¹H-NMR, mass spectral data and elemental analysis; IIb,c-Vb,c were also confirmed by ¹H-NMR spectra of reaction mixtures. To estimate their vascular activity, prototypes 1-(N,N-dimethylaminomethyl)- (Ia) and 1-(4-phenyl-piperazin-1-ylmethyl)- (IVa) 3H-pyrrolo[3,2-f]quinoline derivatives were studied in rat-tail arteries. In tissues precontracted with 0.5 μM 5-hydroxytryptamine (5-HT), 3 μM phenylephrine or 80 mM KCl, Ia and IVa showed endothelium-independent relaxing action. In a preliminary study on the cellular mechanisms of Ia, the influence of propranolol, a β-receptor antagonist, and ketanserin, a 5-HT_2A-receptor antagonist, was checked. In the presence of phenylephrine, the vasorelaxing effect of Ia was not affected by these inhibitors.

Full text of DOI:10.1016/S0223-5234(02)01355-7

Eur. J. Med. Chem. 37 (2002) 427–434
www.elsevier.com/locate/ejmech
Preliminary communication
Mannich bases of 3H-pyrrolo[3,2-f ]quinoline having vasorelaxing
activity
Maria Grazia Ferlin a, Gianfranco Chiarelotto a, Francesca Antonucci b,
Laura Caparrotta b, Guglielmina Froldi b,*
a Department of Pharmaceutical Sciences, Uni6ersity of Pado6a, Via Marzolo 5, I-35131 Padua, Italy
b Department of Pharmacology and Anesthesiology, Uni6ersity of Pado6a, Largo E. Meneghetti 2, I-35131 Padua, Italy
Received 18 July 2001; received in revised form 10 October 2001; accepted 14 February 2002
Abstract
Mannich bases obtained by aminoalkylation of 3H-pyrrolo[3,2-f ]quinoline were designed and prepared as potential vasorelax-
1
ing agents. Compounds Ia–Va were characterised by IR, H-NMR, mass spectral data and elemental analysis; IIb,c–Vb,c were
also confirmed by 1H-NMR spectra of reaction mixtures. To estimate their vascular activity, prototypes 1-(N,N-dimethy-
laminomethyl)- (Ia) and 1-(4-phenyl-piperazin-1-ylmethyl)- (IVa) 3H-pyrrolo[3,2-f ]quinoline derivatives were studied in rat-tail
arteries. In tissues precontracted with 0.5 mM 5-hydroxytryptamine (5-HT), 3 mM phenylephrine or 80 mM KCl, Ia and IVa
showed endothelium-independent relaxing action. In a preliminary study on the cellular mechanisms of Ia, the influence of
propranolol, a b-receptor antagonist, and ketanserin, a 5-HT2A-receptor antagonist, was checked. In the presence of
´
phenylephrine, the vasorelaxing effect of Ia was not affected by these inhibitors. © 2002 Published by Editions scientifiques et
me´dicales Elsevier SAS.
Keywords: 3H-pyrrolo[3,2-f ]quinoline derivatives; Mannich bases; vasorelaxation; 5-HT receptors
1. Introduction
nich bases of benzazoles have a relaxing effect on
smooth muscle [4]. Compounds with indole moiety,
present in 5-HT and related drugs, are well-known to
influence resistance in human vessels. Also, the indole
alkaloid gramine (N,N-dimethyl-3-aminomethylindole)
interacts with the 5-HT system. Recently, halogen-
gramine derivatives demonstrated a vasorelaxing action
in isolated rat aorta [5]. Heterocycles with aminoethyl
or piperazine side-chains in the C-3 substituent of the
indole were generally useful for interaction with 5-HT
receptors [6,7]. All these observations were the basis for
our project: Mannich synthesis was used to prepare a
new series of dialkyl-aminomethyl derivatives of 3H-
pyrrolo[3,2-f ]quinoline, a tricyclic nucleus containing
the indole moiety. The aminomethylation reaction of
pyrroloquinoline, obtained from its 9-chloro derivative
by reductive elimination, was accomplished, employing
formaldehyde and five secondary amines in a ratio of
1:1:1 and following a known procedure for the prepara-
tion of analogous indole derivatives [8]. The structure
of Mannich bases was established on the basis of
elemental analysis, IR, 1H-NMR and mass spectral
data.
Drugs are available for the treatment of diseases such
as hypertension, intermittent claudication and Ray-
naud’s disease, but the possibility of new therapeutic
tools is still a goal of medical research. Vascular tone
depends on the simultaneous influences of constricting
and dilating stimuli on vascular smooth muscle. 5-Hy-
droxytryptamine (5-HT) plays an important modula-
tory function. In human vessels, 5-HT interacts with
endothelial receptors to stimulate nitric oxide synthase
and to evoke vascular relaxation, or, with other 5-HT
receptors, to elicit contraction in smooth muscle [1].
The vascular effect of 5-HT depends on the extent of
endothelial damage and the presence of various types of
5-HT receptors [2,3].
The aim of this research was to synthesise com-
pounds with vasorelaxing activity interacting with the
5-HT system. Data from literature showed that Man-
* Correspondence and reprints.
E-mail address: g.froldi@unipd.it (G. Froldi).
´
0223-5234/02/$ - see front matter © 2002 Published by Editions scientifiques et me´dicales Elsevier SAS.
PII: S0223-5234(02)01355-7

428
M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
Preliminary studies on vascular activity focused on
1-(N,N-dimethyl-aminomethyl)- and 1-(4-phenyl-piper-
azin-1-ylmethyl)-3H-pyrrolo[3,2-f ]quinoline derivatives
Ia and IVa, used as prototypes. These compounds were
tested on rat artery tissues precontracted by 5-HT,
phenylephrine or KCl, and had an appreciable endothe-
lium-independent vasorelaxing action. Experiments
Fig. 2. Products (a, b, c) identified from aminoalkylation reaction of
3H-pyrrolo[3,2-f ]quinoline, carried out as described in Section 6.
were also carried out to investigate the role of b-recep-
tors and 5-HT2A receptors in the action of Ia.
2. Chemistry
Mannich bases were prepared as described in Fig. 1.
Pyrroloquinoline was obtained from the 9-chloro
derivative by reductive elimination [9], performed with
H2 in the presence of 5% Pd/C, in a 90% yield. The
dialkyl-aminomethylation reaction was carried out fol-
lowing the classical method, which employs formalde-
hyde, secondary amine, and the reactive compound in a
1:1:1 ratio. In all cases, except for the dimethyl-
aminomethyl derivative, the formation of three com-
Fig. 1. Synthetic route to 3H-pyrrolo[3,2-f ]quinoline Mannich bases
I–Va, R=H.
1
pounds was confirmed by TLC analysis and H-NMR
Table 1
spectral data (Table 1): the desired 1-Mannich base (a),
1,3-bis-Mannich base (b), and 1-Mannich base-3-hy-
droxymethyl derivative (c) (Fig. 2).
Chemical shifts of methylene groups of compounds I–V
Compound
a
b
c
ꢀCꢁCH2ꢁN
ꢁNꢁCH2ꢁN
–
ꢀCꢁCH2ꢁN
ꢁNꢁCH2ꢁOH
–
ꢀCꢁCH2ꢁN
3.86a
3. Pharmacology
I
II
3.91a
3.78b
3.81 (s, CH2)c
5.62 (d, CH2)
6.61 (t, OH)
3.82b
4.82
1-(N,N-Dimethyl-aminomethyl)- and 1-(4-phenyl-
piperazin-1-ylmethyl)-3H-pyrrolo[3,2-f ]quinoline deri-
vatives Ia and IVa were tested on rat-tail artery as
prototypes of the synthesised Mannich bases. More
precisely, the compounds were studied in varying condi-
tions of precontracture (5-HT, phenylephrine or KCl),
with or without endothelium. Further experiments were
carried out with Ia to evaluate the cellular mechanism
of the vasorelaxing action, using propranolol and ke-
tanserin, well-known antagonists selective respectively
for b and 5-HT2A receptors.
III
IV
V
3.87b
3.82b
4.92
5.75
3.93b
3.87c
3.81b
3.88c
5.18
3.87b
5.75
3.80b
4.92
3.79 (s, CH2)c
5.67 (d, CH2)
6.57 (t, OH)
a Deutero-methanol.
b Deutero-chloroform.
c Deutero-dimethylsulphoxide.

M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
429
4. Results
4.1. Chemical characteristics
In comparison with the behaviour of indole in the
Mannich reaction [10], pyrrolo[3,2-f ]quinoline showed
greater reactivity of the indole moiety, certainly due to
the fused pyridine ring: the 3-H atom becomes more
acid and higher stability of 3-substituted derivatives
also follows. Similar results were obtained when differ-
ent reagent ratios and reaction conditions were
adopted, in order to obtain better yields of the 1-
aminomethyl derivative. Obviously, reaction perfor-
mance was partly conditioned by which secondary
amine was used in the Mannich reaction, as indicated
by the different but constant amounts in which the
three compounds a, b and c formed in each reaction.
The mutual ratios were noted by means of integral
values in 1H-NMR spectra of raw reaction material
(data not reported). Among labelled b and c com-
pounds, only IVb was isolated.
Fig. 3. Concentration–effect curve of 5-HT on rat-tail artery. Values
are mean9S.E.M. of eight experiments. S.E.M.s not indicated fall
within respective symbols.
4.2. Pharmacological characteristics
4.2.1. Acti6ity of Ia and IVa
1 - (N,N - Dimethylaminomethyl) - 3H - pyrrolo[3,2-f ]-
quinoline (Ia) and 1-(4-phenyl-piperazin-1-ylmethyl)-
3H-pyrrolo[3,2-f ]quinoline (IVa) were tested in a range
of concentrations from 10 nM to 0.5 mM, and main-
tained in an organ bath solution for 15 min at each
concentration. In this condition, the compounds were
unable to modify the resting tension of rat-tail artery
(data not reported).
4.2.2. Effect of Ia in the presence of 5-HT, phenyl-
ephrine or KCl
In order to obtain the EC50 value, we tested 5-HT
from 50 nM to 100 mM on rat-tail arterial rings (Fig.
3). The pD2 was 6.590.2 (CE50ꢀ0.3 mM). With pre-
contracture by 0.5 mM 5-HT, a concentration slightly
higher than that of the EC50, Ia evoked concentration-
related relaxation (Fig. 4). The pD2 was 5.090.1.
a1-Adrenoceptor agonist phenylephrine, tested from 0.1
to 100 mM, evoked concentration-related contraction in
arterial rings (Fig. 5). The pD2 value was 5.490.1
(CE50ꢀ3.6 mM). With precontracture by
3 mM
phenylephrine (EC50), Ia caused concentration-related
relaxation (Fig. 4). The pD2 was 4.590.1. Increased
isometric tension in artery rings can be obtained by
direct depolarisation of tissue with KCl (80 mM). In
this condition too, Ia evoked concentration-related re-
laxation (Fig. 4). The pD2 was 4.390.1. The potency
of Ia was higher in tissues precontracted with 5-HT
than with phenylephrine or KCl.
Fig. 4. Example of original tracing (A) obtained by pyrrolo[3,2-
f ]quinoline Ia on rat-tail artery precontracted with 0.5 mM 5-HT, and
concentration–response curves (B) of Ia in rat-tail artery precon-
tracted with 0.5 mM 5-HT (ꢀ), 3 mM PE (ꢁ) and 80 mM KCl (ꢂ).
Values are mean9S.E.M. of 6–8 experiments. * PB0.05 vs. Ia in
tissues precontracted with 5-HT.

430
M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
they were first contracted with 10 mM phenylephrine
and then, at the plateau of contraction, 30 mM acetyl-
choline was added. In this condition, acetylcholine was
able to evoke vasorelaxation only if the endothelium
was not damaged [11,12]. The vasorelaxing action of Ia
and IVa was studied in rings with uninjured endothe-
lium as well as in damaged endothelium; the curve-con-
traction effects did not change in the presence or
absence of endothelium (data not shown).
Preliminary studies on cellular mechanisms using Ia
were also carried out, since this compound showed
selectivity in the 5-HT system. Derivative Ia in arterial
rings precontracted by 3 mM phenylephrine evoked
vasodilatation. In order to evaluate the role of b-adre-
nergic receptors, Ia was tested in the presence of 1 mM
propranolol, a known blocker of b-adrenoceptor, which
also has 5-HT1A/5-HT1B receptor antagonist activity
[13,14]. Propranolol in itself did not modify the effect
of phenylephrine used to precontract arterial rings. In
this condition, the concentration-dependent relaxation
curve of Ia was not modified (Fig. 7).
Fig. 5. Concentration–effect curve of phenylephrine (PE) in rat-tail
artery. Values are mean9S.E.M. of seven experiments. SEMs not
indicated fall within respective symbols.
The role of 5-HT2A receptor was also examined,
using ketanserin, a well-known 5-HT2A antagonist
[15,16]. Alone ketanserin (50 nM) did not influence
precontracture by phenylephrine, and did not change
the relaxation induced by derivative Ia (Fig. 7).
5. Discussion
Compounds Ia and IVa evoked concentration-related
relaxation in rat-tail artery precontracted with various
active agents such as 5-HT, phenylephrine and potas-
sium chloride. Concentration–response curves were ob-
tained in a range from 0.1 to 500 mM. The vasorelaxing
effects were endothelium-independent. Comparing the
Fig. 6. Comparison among concentration–response curves of
pyrrolo[3,2-f ]quinoline IVa in rat-tail artery precontracted with 0.5
mM 5-HT (ꢃ), 3 mM PE (ꢄ) and 80 mM KCl (ꢅ). Values are
mean9S.E.M. of 6–8 experiments. SEMs not indicated fall within
respective symbols.
4.2.3. Effect of IVa in the presence of 5-HT,
phenylephrine or KCl
In the presence of 0.5 mM 5-HT, 3 mM phenylephrine
or 80 mM KCl, IVa evoked concentration-related re-
laxation. The pD2 were 5.990.3, 5.990.5 and 5.69
0.2,
respectively.
The
concentration-dependent
relaxation curves obtained with IVa, in the various
conditions of precontraction, are compared in Fig. 6.
Derivative IVa showed comparable potency in vasore-
laxing arterial tissues precontracted with 5-HT,
phenylephrine, and KCl.
Fig. 7. Vasorelaxing effects induced by pyrrolo[3,2-f ]quinoline Ia in
rings precontracted with 3 mM PE, in the absence (control, ꢁ) and
presence of 1 mM propranolol (") or 50 nM ketanserin (ꢀ). Values
are mean9S.E.M. of 6–8 experiments. SEMs not indicated fall
within respective symbols.
4.2.4. Studies on 6asorelaxing mechanisms
All vascular preparations were subjected to a starting
procedure to determine the viability of endothelium:

M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
431
relaxing effect induced by Ia with that of IVa, it was
evident that IVa had a higher potency but less selecti-
vity on the 5-HT system. These data indicate that the
1-(N,N-dimethylaminomethyl) group in 3H-pyrrolo-
[3,2-f ]quinoline Ia is important for selectivity on 5-HT
receptors, whereas the 1-(4-phenyl-piperazin-1-yl-
methyl) group eventually is useful for increasing po-
tency. The other synthesised compounds, IIa, IIIa and
Va, will be investigated pharmacologically in the near
future.
are endothelium-independent. Data also show a slight
selectivity of Ia on the 5-HT system and exclude the
involvement of b-receptors. These derivatives may rep-
resent a source of endothelium-independent vasorela-
xing agents.
6. Experimental protocols
6.1. Chemistry
Drugs with relaxing activity on vascular vessels inde-
pendent of endothelium are of great value, because the
endothelial layer is often damaged in cardiovascular
diseases [17]. In these pathological conditions, physio-
logical modulators such as acetylcholine and ATP lose
their relaxing action on arterial tone, so that endothe-
lium-independent relaxing agents are useful.
Melting points were determined on a Gallenkamp
MFB-595-010M Melting Point Apparatus, and are
uncorrected.
1H-NMR spectra were obtained on a Varian Gemini
200MHz spectrometer using indicated solvents and
TMS as internal reference. 1H-NMR signals are re-
ported in parts per million (ppm), and are characterised
as singlets (s), doublets (d), triplets (t), quartets (q), or
multiplets (m). In the case of multiplets, the quoted
chemical shift corresponds to the multiplet centre. Inte-
grals corresponded satisfactorily to those expected on
the basis of compound structure. Coupling constants
are expressed in Hertz (Hz).
As regards the mechanism adopted by derivatives
inducing relaxation, some preliminary experiments on
Ia were carried out, which was the more selective on the
5-HT system. b-Adrenoceptors are expressed in vascu-
lar tissues regulating resistance [18–20]. Pretreatment
with 1 mM propranolol, a well-known b-antagonist, did
not change the vasorelaxing action of Ia; this experi-
mental observation excludes the direct activation of Ia
on b-adrenoceptors. It was also shown that propranolol
is an antagonist of 5-HT1A/5-HT1B receptors [13,14], so
that the involvement of 5-HT1A/5-HT1B receptors in
vasorelaxing action by Ia. is not probable, partly be-
cause this subtype of receptor is poorly represented in
rat-tail artery [21]. The effect of ketanserin, a well-
known antagonist of 5-HT2A receptors [15,16], with
slight a-antagonist activity was also studied here
[22,23]. Ketanserin did not change the vasorelaxing
action of Ia, indicating that the compound does not act
through 5-HT2A receptors. This is an intricate point: as
it was not possible to carry out experiments with ke-
tanserin with 5-HT precontracture since 5-HT is largely
inhibited by ketanserin, we worked with vessels precon-
tracted by phenylephrine. However, therefore we prob-
ably lost the component of Ia on 5-HT2A: compound Ia
cannot inhibit 5-HT receptors if they are not preacti-
vated! Obviously, more research is necessary to clarify
the cellular mechanisms by which Ia induces relaxation
of precontracted rat-tail artery. The only consideration
that, at this moment, may be made on dimethy-
laminomethylpyrrolo[3,2-f ]quinoline Ia is based on its
different potency in the various conditions of precon-
tracture studied here. Its potency in reducing vascular
tone is greater in rat-tail rings precontracted by 5-HT,
less with phenylephrine, and the least by KCl. Future
binding studies on the 5-HT receptors potentially in-
volved may clarify this important point.
Mass spectra were obtained on a Varian Mat 112
mass spectrometer.
Elemental analyses were performed using Perkin–
Elmer elemental analysis apparatus, model 240B; re-
sults for C, H, and N were within 90.4% of theoretical
values.
IR spectra were recorded in cm−1 on a Perkin–
Elmer 1760FTIR spectrophotometer as potassium bro-
mide pressed disks.
Thin-layer chromatography (TLC) was performed
using Merck silica gel on F254 polystyrene-backed
plates.
Column flash chromatography was performed with
Merck silica gel (250–400 mesh ASTM).
40% Dimethylamine was purchased from Fluka, all
other secondary amines from Aldrich, and 37% formal-
dehyde from Acros.
6.1.1. 3H-Pyrrolo[3,2-f ]quinoline
A solution of 9-chloro-pyrroloquinoline [9] (2.39
mmol) in 200 ml methanol was slowly added to 5%
palladium in a charcoal–methanol suspension saturated
with hydrogen, and the mixture was hydrogenated at
atmospheric pressure at 40 °C until the starting material
disappeared, according to TLC analysis (eluent ethyl
acetate). After 2 h, the mixture was filtered and the
filtrate evaporated to dryness. The residue was used in
the Mannich reaction without purification. Yield 90%,
pinkish-white solid; m.p. 181–82 °C (aceton) [24]; r.f.
0.41 (eluent ethyl acetate–n-hexane 7:3); 1H-NMR
In conclusion, the most interesting observation is
that pyrrolo[3,2-f ]quinoline derivatives Ia and IVa
have concentration-dependent relaxing effects, which
(DMSO-d6), l: 7.43 (m, 1H, HC-1), 7.78 (t, 1H, J2,3,1
=
2.74 Hz, HC-2), 8.10–8.03 (m, 2H, HC-5, HC-8), 8.29

432
M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
(d, 1H, J4,5=9.52 Hz, HC-4), 9.09 (dd, 1H, J7,8=5.38
Hz, J7,9=1.40 Hz, HC-7); 9.52 (d, 1H, J9,8=7.70 Hz,
HC-9), 12.43 (bs, 1H, indolic NH).
J=9.1 Hz, HC-4), 7.83 (d, 1H, J=9.1 Hz, HC–5),
8.67(dd, 1H, J=4.5 and 1.8 Hz, HC–9), 9.18 (dd, 1H,
J=8.4 and 1.8 Hz, HC–7), 11.57 (bs, 1H, NH DMSO-
d6). Anal. calc. for C16H17N3O: C, 71.89; H, 6.41; N,
15.72. Found: C, 71.60; H, 6.58; N, 15.51%.
6.1.2. General procedure for preparation of
1-alkyl-aminomethyl-pyrroloquinolines (Mannich bases)
I–V
6.1.2.3. 1-(4-Methyl-piperazin-1-ylmethyl)-3H-pyrrolo-
[3,2-f ]quinoline (IIIa). This compound was prepared
according to the general procedure but, as it did not
solidify, it was extracted with chloroform and extracts
evaporated to dryness. As the residue was composed of
three compounds, IIIa–c, IIIa was separated by frac-
tionated recrystallisation with absolute ethanol. Yield
35%; m.p. 235–236 °C (pale yellowish); r.f. 0.17 (eluent
methanol); IR (KBr) 3136 (NH) cm−1; ms (70 eV) m/z
(relative abundance) 281 (M+, 30%), 182 (281–99,
85%), 112 (281–113, 9%), 99 (281–182, 43%), 56
An equimolar mixture of 37% formaldehyde and a
secondary amine, made acid with glacial CH3COOH,
was previously prepared in a vessel at room tempera-
ture. After about 30 min, powdered pyrroloquinoline
(1–3 mmol) was added, and the suspension was vigor-
ously stirred for some time. It was then heated to 40 °C
and maintained at that temperature until the end of the
reaction, as indicated by TLC analysis (about 24 h).
Next, the dark reaction mixture was poured into a cold
10% KOH water solution (6 ml) and kept at 0 °C under
stirring for several hours. A semisolid precipitate
formed and, if it became crystalline by standing at 0 °C
for a further 24 h, it was filtered out, washed with water
and dried; otherwise, it was extracted using a suitable
solvent.
1
(100%); H-NMR (DMSO-d6) l: 2.12 (s, 3H, ꢁCH3),
2.29 (bm, 4H, 3%- and 5%-CH2), 2.50 (bm, 4H, 2% and
6%-CH2, overlapped with DMSO signal), 3.78 (s, 2H,
CH2ꢁ), 7.39 (d, 1H, J=2.2 Hz, HC-2), 7.47 (dd, 1H,
J=8.5 and 4.2 Hz, HC-8), 7.66 (d, 1H, J=9.0 Hz,
HC-4), 7.81 (d, 1H, J=9.0 Hz, HC-5), 8.72 (dd, 1H,
J=4.2 and 1.5 Hz, HC-9), 8.96 (dd, 1H, J=8.5 and
1.5 Hz, HC-7), 11.54 (bs, 1H, NH). Anal. Calc. for
C17H20N4: C, 72.3; H, 7.19; N, 19.98. Found: C, 72.55;
H, 7.40; N, 19.70%.
Purification by recrystallisation or flash chromato-
graphy was needed in order to obtain pure products.
6.1.2.1. 1-(N,N-Dimethylaminomethyl)-3H-pyrrolo[3,2-
f ]quinoline (Ia). Yield 50%, pale pink crystals; m.p.
192 °C dec. (methanol); r.f. 0.15 (eluent ethyl acetate–
methanol 8:2); IR (KBr) 3186 (NH) cm−1; ms (70 eV),
m/z (relative abundance): 226 (M+, 15%), 182 (226–44,
100%), 58 (226–168, 14%); 1H-NMR (deutero-
methanol), l: 2.33 (s, 6H, 2 CH3); 3.85 (s, 2H,
ꢁCH2ꢁN); 7.34 (s, 1H, CH-2); 7.54 (dd, 1H, J=8.5 and
4.4 Hz, HC-8); 7.70 (d, 1H, J=9.0 Hz, HC-4); 7.83 (d,
1H, J=9.0 Hz, HC-5); 8.67 (dd, 1H, J=4.4 and 1.6
Hz, HC-9); 8.00 (dd, 1H, J=8.5 and 1.6 Hz, HC-7),
11.54 (bs, 1H, NH DMSO-d6). Anal. calc. for
C14H15N3: C, 74.64; H, 6.71; N, 18.65. Found: C, 74.31;
H, 6.93; N, 18.33%.
6.1.2.4. 1-(4-Phenyl-piperazin-1-ylmethyl)-3H-pyrrolo-
[3,2-f ]quinoline (IVa–c). Prepared according to the
general procedure, a white precipitate formed from the
reaction mixture after 12–24 h at 0 °C. It was filtered
out and dried. As it was composed of three compounds
(TLC, 1H-NMR), it was chromatographed by flash
chromatography (eluent ethyl acetate–n-hexane 7:3),
affording pure IVa and IVb.
IVa: yield 42%; m.p. 246–249 °C dec.; r.f. 0.25 (elu-
ent ethyl acetate–n-hexane 7:3); IR (KBr) 3146 (NH)
cm−1; ms (70 eV) m/z (relative abundance) 343 (M+,
45%), 182 (343–162, 66%), 162 (343–182, 52%), 121
(100%), 77 (343–266, 31%), 56 (96%); 1H-NMR
(CDCl3) l: 2.73 (bt, 4H, 3% and 5% ꢁCH2ꢁ), 3.20 (bt, 4H,
2% and 6% ꢁCH2ꢁ), 3.93 (s, 2H, ꢁCH2ꢁN), 6.84 (t, 1H,
J=7.2 Hz, phenyl), 6.92 (d, 2H, J=7.8 Hz, phenyl),
7.26 (t, 2H, J=7.2 Hz, phenyl), 7.44 (d, 1H, J=2.2,
HC-2), 7.5 (dd, 1H, J=8.3 and 4.2 Hz, HC-8), 7.75 (d,
1H, J=9.1 Hz, HC-4), 7.87 (d, 1H, J=9.0 Hz, HC-5),
8.81 (dd, 1H, J=2.7 and 4.3 Hz, HC-9), 9.03 (bs, 1H,
NH), 9.15 (dd, 1H, J=8.3 and 2.7 Hz, HC-7), 11.6 (s,
1H, NH DMSO). Anal. Calc. for C22H22N4: C, 77.16;
H, 6.48; N, 16.36. Found: C, 76.91; H, 6.30; N, 16.52%.
IVb: Yield 30%; m.p. 220–23 °C dec.; r.f. 0.18 (eluent
ethyl acetate–n-hexane 7:3); IR (KBr) 1594 and 1502
6.1.2.2.
1-Morpholin-4-ylmethyl-3H-pyrrolo[3,2-f ]-
quinoline (IIa). This compound was prepared according
to the general procedure but, as it did not solidify, it
was extracted with ethyl acetate and extracts were
evaporated to dryness. As the resulting crude product
contained three compounds, IIa–c (1H-NMR), having
similar r.f. at TLC analysis, IIa was isolated by frac-
tionated recrystallisation with absolute ethanol. Yield
20%; m.p. 160–63 °C dec.; r.f. 0.71 (eluent methanol);
IR (KBr) 3125 (NH) cm−1; ms (70 eV) m/z (relative
abundance) 267 (M+, 4%), 182 (267–86, 100%), 100
(267–167, 13%), 86 (267–182, 13%); 1H-NMR
(deutero-methanol) l: 2.57 (t, 4H, J=4.7 Hz,
CH2ꢁOꢁCH2), 3.67 (t, 4H, J=4,7 Hz, morpholinic
CH2ꢁNꢁCH2) 3.91 (s, 2H, CH2), 7.35 (s, 1H, HC-2),
7.54 (dd, 1H, J=8.4 and 4.5 Hz, HC-8), 7.70 (d, 1H,
1
(phenyl) cm−1; H-NMR (DMSO-d6) l: 2.66 (bm, 8H,
bis-piperazinic 3% and 5% ꢁCH2ꢁ), 3.11 (bm, 8H, bis-
piperazinic 2% and 6% ꢁCH2ꢁ), 3.88 (s, 2H, CH2), 5.18 (s,

M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
433
2H, CH2), 6.74 (m, 2H, bis-phenylic), 6.89 (m, 4H,
bis-phenylic) 7.17 (m, 4H, bis-phenylic), 7.53 (m, 2H,
HC-8 and HC-2), 7.74 (d, 1H, J4,5=9.1 Hz, HC-4),
8.15 (d, 1H, J5,4=9.1 Hz, HC-5), 8.74 (dd, 1H, J=1.4
and 4.4 Hz, HC-7), 9.04 (dd, 1H, J=1.0 and 8.5 Hz,
HC-9). Anal. Calc. for C33H36N6: C, 76.71; H, 7.02; N,
16.27. Found: C, 76.48; H, 7.13; N, 16.44%.
mM acetylcholine; this procedure are also useful to
check the condition of endothelium [22].
In experiments carried out on depolarised tissue by
80 mM KCl, the Krebs–Ringer solution was modified
in order to counterbalance the increased concentration
of KCl by an equal reduction of NaCl. Pharmacologi-
cal experiments with untreated and treated tissues were
performed on the same preparation.
6.1.2.5. 1-Piperidin-1-ylmethyl-3H-pyrrolo[3,2-f ]quino-
line (Va). Following the general procedure, a solid beige
precipitate separated from the reaction mixture, was
filtered out, washed with water, and dried. As TLC
analysis indicated the presence of several compounds
having similar r.f., its isolation by means of fraction-
ated crystallisation with various solvents was necessary.
Yield 15%; m.p. 189.193 °C.; r.f. 0.45 (eluent
methanol); IR (KBr) 3146 (NH) cm−1; ms (70 eV) m/z
(relative abundance) 266 (M+, 9%), 182 (266–84,
100%), 98 (266–169, 30%), 84 (266–182, 39%), 56
(18%). 1H-NMR (CDCl3) l: 1.0 (bm, 6H, piperidinic 3%,
4% and 5% ꢁCH2ꢁ), 2.51 (bm, 4H, piperidinic 2% and 6%
ꢁCH2ꢁ), 3.81 (s, 2H, CH2ꢁN), 7.23 (d, 1H, J=2.1 Hz,
HC-2), 7.46 (dd, 1H, J=8.3 and 4.3 Hz, HC-8), 7.74
(d, 1H, J=9.1 Hz, HC-4), 7.87 (d, 1H, J=9.1 Hz,
HC-5), 8.82 (dd, 1H, J=4.3 and 1.7 Hz, HC-9), 9.14
(dd, 1H, J=8.3 and 1.7 Hz, HC-7), 11.81 (s, 1H, NH,
DMSO). Anal. Calc. for C17H19N3: C, 76.95; H, 7.22;
N, 15.84. Found: C, 7716; H, 7.35; N, 15.59%.
6.2.1. Drugs
5-HT,
L-phenylephrine hydrochloride and acetyl-
choline chloride were dissolved in saline. Ketanserin
tartrate and propranolol hydrochloride were prepared
in dimethylsulphoxide and ethanol, respectively. Ia and
IVa were dissolved in dimethylsulphoxide. The concen-
tration of organic solvents added never exceeded 0.2%
V/V of the organ bath; in this condition, the solvents
did not modify the contractility of vascular tissue.
6.2.2. Data analysis and statistics
All values presented are means9S.E.M. for 7–8
experiments. Changes in tension are expressed in mN or
as percentages of the maximum contraction obtained.
For statistical comparisons between treatment and con-
trol data, student’s two-tail paired t-test was used; the
difference between mean group data was significant at
PB0.05.
6.2. Pharmacological methods
Acknowledgements
Male rats (Wistar, 250–350 g) were anaesthetised
with inhalation of methoxyfluoran and then sacrificed
by cervical dislocation. The proximal 2 cm of the tail
artery were carefully removed and cleaned of adhering
tissue. Arterial rings (3 mm long) were suspended be-
tween two tungsten wires to measure isometric tension
and placed in a tissue bath containing 15 ml of Krebs–
Ringer solution at 37 °C, with the following composi-
tion (mM): NaCl 118.3, KCl 4.7, CaCl2·2H2O 2.5,
The authors are indebted to Dr. Barbara Silvestrin
for performing some experiments while preparing her
degree thesis. The authors are grateful to Mr. Enrico
Secchi for excellent technical assistance.
References
[1] T. Nilsson, J. Longmore, D. Shaw, E. Pantev, J.A. Bard, T.
Branchek, L. Edvinsson, Eur. J. Pharmacol. 372 (1999) 49–56.
[2] G.R. Martin, Pharmacol. Ther. 62 (1994) 283–324.
[3] P. De Vries, P.A. De Visser, J.P.C. Heiligers, C.M. Villalo´n, P.R.
Saxena, Naunyn-Schmiedeberg’s Arch. Pharmacol. 359 (1999)
331–338.
[4] F. Collino, S. Volpe, Boll. Chim. Farm. 121 (1982) 167–173,
221–229 and 328–333.
[5] S. Iwata, S. Saito, K. Kon-ya, Y. Shizuri, Y. Ohizumi, Eur. J.
Pharmacol. 432 (2001) 63–70.
MgSO4·7H2O 1.2, KH2PO4 1.2, NaHCO3 25, D-glucose
11.1, bubbled with 95% O2 and 5% CO2 (PO =34598
2
mmHg), pH 7.2. Tissues were stretched to a basal
tension of 20 mN, this amount of resting tension having
been shown to maximise the contractile response to
phenylephrine, a standard vasoconstricting agent. The
tail artery was preserved for a maximum period of 5
days in Krebs–Ringer solution at 4 °C [12]. The con-
tractility of vascular tissue was recorded by means of a
[6] R.A. Glennon, M. Bondarev, B. Roth, Med. Chem. Res. 9
(1999) 108–117.
high-sensitivity transducer connected to
a
chart
[7] M.B. van Niel, I. Collins, M.S. Beer, H.B. Broughton, S.K.F.
Cheng, S.C. Goodacre, A. Heald, K.L. Locker, A.M. MacLeod,
D. Morrison, C.R. Moyes, D. O’Connor, A. Pike, M. Rowley,
M.G.N. Russell, B. Sohal, J.A. Stanton, S. Thomas, H. Verrier,
A.P. Watt, J.L. Castro, J. Med. Chem. 42 (1999) 2087–2104.
[8] M. Tramontini, C. Angiolini, Tetrahedron 46 (1990) 1791–1837.
[9] M.G. Ferlin, B. Gatto, G. Chiarelotto, M. Palumbo, Bioorg.
Med. Chem. 8 (2000) 1415–1422.
recorder (Ugo Basile, type DYO and Unirecord Sys-
tem, Model 7050, Comerio (VA), Italy). Vessel seg-
ments were allowed to equilibrate for 60 min before
viability was assessed with a standard start procedure,
consisting in constriction by 10 mM phenylephrine and,
when contraction reached a plateau, relaxation by 50

434
M.G. Ferlin et al. / European Journal of Medicinal Chemistry 37 (2002) 427–434
[10] S. Swaminatan, K. Narasimhan, Chem. Ber. 3 (1966) 889–894.
[11] E. Vila, N.M. Vivas, A. Tabernero, J. Giraldo, S.M. Arribas, Br.
J. Pharmacol. 121 (1997) 1017–1023.
Cardiovasc. Pharmacol. 24 (1994) 216–228.
[18] S.R. O’Donnell, J.C. Wanstall, J. Pharmacol. Exp. Ther. 228
(1984) 733–738.
[12] C.A. Mcintyre, B.C. Williams, J.A. Mcknight, P.W.F. Hadoke,
Br. J. Pharmacol. 123 (1998) 1555–1560.
[19] M. Berlan, J. Galitzky, J.L. Montastruc, Fundam. Clin. Pharma-
col. 9 (1995) 234–239.
[13] P. Schoeffter, D. Hoyer, Naunyn-Schmiedeberg’s Arch. Pharma-
col. 340 (1989) 285–292.
[20] Y.B. Shen, P. Cervoni, T. Claus, S.F. Vatner, J. Pharmacol. Exp.
Ther. 278 (1996) 1435–1443.
[14] S.J. Peroutka, A.W. Schmidt, A.J. Sleight, M.A. Harrington, Ann.
NY Acad. Sci. 600 (1990) 104–112.
[21] C. Schoning, M. Flieger, H.H. Pertz, J. Anim. Sci. 79 (2001)
2202–2209.
[15] J.M. Van Nueten, P.A.J. Janssen, J. Van Beek, R. Xhonneux, T.J.
Verbeuren, P.M. Vanhoutte, J. Pharmacol. Exp. Ther. 218 (1981)
217–230.
[16] J.F. Marwood, G.S. Stokes, Clin. Exp. Pharmacol. Physiol. 11
(1984) 439–456.
[22] E.G. Spokas, G.C. Folco, Eur. J. Pharmacol. 100 (1984) 211–
217.
[23] R. Yoshio, T. Taniguchi, H. Itoh, I. Maramatsu, Jpn. J. Pharma-
col. 86 (2001) 189–195.
[24] A.P. Gryovznov, R.N. Akhvlediani, T.A. Volodina, A.M.
Vasil’ev, T.A. Babushkina, N.N. Suvorov, Khim. Geterotsikl.
Soedin. 3 (1977) 369–371.
[17] A. Chinellato, E. Ragazzi, L. Pandolfo, A. Palermo Alvaro, G.
Froldi, M. De Biasi, L. Caparrotta, G. Aliev, G. Fassina, J.