New aromatase inhibitors. Synthesis and biological activity of aryl- substituted pyrrolizine and indolizine derivatives

Article abstract of DOI:10.1016/S0968-0896(00)00024-9

We report herein the design and the synthesis of some aryl-substituted pyrrolizine and indolizine derivatives, on the basis of a hypothetical pharmacophore structure designed to fit the catalytic site of the human cytochrome P450 aromatase. The in vitro b

Full text of DOI:10.1016/S0968-0896(00)00024-9

Bioorganic & Medicinal Chemistry 8 (2000) 945±955
New Aromatase Inhibitors. Synthesis and Biological Activity of
Aryl-Substituted Pyrrolizine and Indolizine Derivatives
Pascal Sonnet, a Patrick Dallemagne, a,* Jean Guillon, a Cecile Enguehard, a
Silvia Stiebing, a Julien Tanguy, a Ronan Bureau, a Sylvain Rault, a Pierrõck Auvray, b
Safa Moslemi, b Pascal Sourdaine b and Gilles-Eric Seralini b
a
Â
Centre d'Etudes et de Recherche sur le Medicament de Normandie, Laboratoire de Pharmacochimie,
Â
UFR des Sciences Pharmaceutiques, 1, rue Vaubenard, 14032 Caen Cedex, France
b
Â
IBBA, Laboratoire de Biochimie et Biologie Moleculaire Esplanade de la Paix, 14032 Caen Cedex, France
Received 15 January 1999; accepted 6 December 1999
AbstractÐWe report herein the design and the synthesis of some aryl-substituted pyrrolizine and indolizine derivatives, on the basis
of a hypothetical pharmacophore structure designed to ®t the catalytic site of the human cytochrome P450 aromatase. The in vitro
biological evaluation of these compounds allowed us to point out two new potent non-steroidal aromatase inhibitors, MR 20494
and MR 20492, with IC50 values in the range of 0.1 mM. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
towards equine aromatase (IC50>10 mM and Ki/
Km=100.0‹6.4), used as a useful comparative mole-
cular model to understand the mammalian active site of
this enzyme.13
Inhibitors of the cytochrome P450 aromatase are ther-
apeutic agents for the treatment of estrogen dependent
diseases such as breast cancer,1 and constitute today an
alternative to antihormone treatment.2 More recently,
antisense oligodeoxynucleotides were also developed to
inhibit human aromatase gene expression.3 Moreover,
several nonsteroidal inhibitors of aromatase have
hitherto been reported. Most of them belong to the
azole-type and are commercially available or currently
under clinical evaluation (fadrozole, vorozole, letrozole,
and anastrozole).4 6 However, and particularly in
relation to results obtained in vivo and with side e€ects
due to these compounds,7,8 the synthesis of more pow-
erful and more speci®c aromatase inhibitors still
remains a challenge.
This previous study also showed, on the basis of the
characteristics of the UV di€erence spectrum (type II),
an interaction between the pyridin nitrogen atom of 1
and the heme iron atom of the human enzyme. As
shown by comparative molecular modeling between
human and equine enzymes, this interaction led to dis-
pose of the amino group of 1 at the entry of an extra-
hydrophobic pocket, already described by Laughton et
al.14 In the case of equine aromatase, we proposed that
the primary structure and the conformation of the
enzyme cannot allow this position.13
In order to con®rm this hypothesis and to improve the
activity of some related compounds, we undertook the
pharmacomodulation of 1. In particular, we carried out
the substitution of its amino group by an hydrophobic
moiety likely to occupy, in an optimal way, the extra-
hydrophobic pocket within the active site of the enzyme.
This work allowed us to report, in a preliminary
paper,15 the synthesis and the biological activity of
compound 2 (MR 20496, Scheme 1), in which the amino
group of 1 was replaced by a phenyl ring, while its pyr-
idin moiety was frozen in a Z geometry. Derivative 2
inhibits strongly human aromatase in vitro (IC50=
0.472‹0.092 mM).
With this view, some benzocycloalkene-type inhibitors
have been designed and synthesized,9 12 and among
them, we have recently described the synthesis and the
biological evaluation of new 3-amino-2-arylmethyl
indenones.13 We showed that derivative 1 (MR 20814,
Scheme 1) inhibits eciently in vitro human aromatase
activity with an IC50 of 3.5‹1.2 mM and an apparent
Ki/Km of 6.7‹0.6, while it was considerably less active
*Corresponding author. Tel.: +33-02-31-56-59-10; fax: +33-02-31-
93-11-88; e-mail: dallemagne@pharmacie.unicaen.fr
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00024-9

946
P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
Scheme 1.
These results altogether led us to design a hypothetical
pharmacophore structure, responsible for the aromatase
inhibition, and we report herein the synthesis and the
biological evaluation of some new aryl-substituted pyr-
rolizine and indolizine derivatives 3, ®tting this model.
tribromide in dichloromethane.17 This ecient ring
closure method led us to reinvestigate the synthesis of
the pyrrolizinone series and particularly to develop it
starting from the b-amino acids 5a±e. We previously
described18,19 5a, 5b and 5e, that we obtained, starting
from benzaldehydes 4a and 4b or from thiophene-3-
carboxaldehyde 4e, according to the Rodionow±Johnson
reaction.20,21 Application of the same treatment, invol-
ving malonic acid and ammonium acetate, to 4-cyano-
4c or 4-tri¯uoromethoxybenzaldehyde 4d led to the new
aminophenylpropionic acids 5c and 5d (Scheme 2).
Esteri®cation of the carboxylic acid group of 5a±c,e was
carried out, without protection of the amino group,
through a treatment with thionyl chloride in ethanol,
followed by displacement of the intermediary salt with
ammonia in ether.22 The expected ethyl aminopropio-
nates 6a±c,e were then involved in a Clauson±Kaas
reaction,23,24 using 2,5-dimethoxytetrahydrofuran in
acetic acid, and yielded the ethyl arylpyrrolylpropionates
Chemistry
We previously reported the access to 3-phenyl-2,3-di-
hydro-1H-pyrrolizin-1-one 8a, starting from 3-amino-3-
phenylpropionic acid 5a.16 The synthetic pathway we
used involved successively a Clauson±Kaas reaction and
subsequently a Vilsmeier intramolecular cyclisation
leading to 8a after an alkaline hydrolysis of the Vilsmeier
salt.
Recent works showed that the pyrrole ring could be
acylated by an ester function under the action of boron
Scheme 2. (i) CH2(CO2H)2, AcONH4, EtOH or MeCN; (ii) SOCl2, EtOH; (iii) NH3 gas, EtO2; (iv) 2,5-diOMeTHF, AcOH; (v) BBr3, CH2Cl2.

P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
947
Scheme 3. (i) 2,5-diOMeTHF, AcOH; (ii) TEA, (CH3)2CO; (iii) ClCO2Et, (CH3)2CO; (iv) pyrrolidine, (CH3)2CO; (v) POCl3, toluene; (vi) HClO4,
H2O; (vii) NaOH, H2O.
7a±c,e. Cyclisation of the latter into the pyrrolizinones
8a±c,e ®nally took place using boron tribromide in
methylene chloride at room temperature.
carried out, either in an ethanolic solution of sodium
hydroxide or in a heterogeneous system (dichloro-
methane/water) using tetrabutylammonium hydrogen
sulfate as catalyst. In all cases, the reaction led selec-
tively to the Z forms of 12±29, as deduced from their 1H
NMR spectra (data not shown).
However, we could not apply this short sequence to 3-
amino-3-(4-tri¯uoromethoxyphenyl)-propionic acid 5d,
due to the sensitivity of the alkoxy groups to the action
of boron tribromide. For this reason, the tri¯uoro-
methoxypyrrolizinone 8d was prepared, in a similar
manner as for the initial synthesis of 8a, via a Vilsmeier
cyclisation, using phosphorus oxychloride as reagent,
and a subsequent hydrolysis (Scheme 3).
Biology
This study showed that compounds 12±20, 25 and 26
did not inhibit at 10 mM human aromatase in placental
microsomes. According to the results presented in Table
1, compounds 24 (MR 16088) and 27 (MR 20493) were
On the other hand, using the same pathway as for 8d,
we also reported the synthesis of 6-(4-chlorophenyl)-
5,6,7,8-tetrahydroindolizin-8-one 8f,25 starting from the
corresponding g-amino acid, baclofen 5f, a commer-
cially available antispastic drug (Scheme 4). The vinyl
homologue of the latter, vigabatrin 5g, lent itself to the
previous sequence using boron tribromide and led to 5-
vinyl-5,6,7,8-tetrahydroindolizin-8-one 8g, in a similar
manner as for 8a±c,e (Scheme 5).
All the synthesised pyrrolizinones 8a±e and indolizi-
nones 8f,g were ®nally involved in an aldolisation reac-
tion, using various arylaldehydes such as benzaldehyde,
2,4-dimethoxybenzaldehyde, pyridine-2-, -3- or -4-car-
boxaldehydes (Schemes 6 and 7). The reactions were
Scheme 4.

948
P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
Scheme 5. (i) SOCl2, EtOH; (ii) NH3 gas, EtO2; (iii) 2,5-diOMeTHF, AcOH; (iv) BBr3, CH2Cl2.
Scheme 6. (i) AR-CHO, NaOH, EtOH; (ii) AR-CHO, nBu4NH+ HSO4, CH2Cl2, H2O.

P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
949
Scheme 7. (i) AR-CHO, nBu4NH+ HSO4, CH2Cl2, H2O.
Table 1. (A) Inhibition of human aromatase in placental microsomes
by pyrrolizine derivatives. (B) Inhibition of human aromatase in
placental microsomes by indolizine derivativesa
Discussion
For the pyrrolizinone series, the compounds bearing a
meta or para pyridinylmethylene substituent are actives.
The main interaction of these ligands with the receptor
is the coordination bond between the nitrogen of pyr-
idine group and the heme iron.13 This bond imposes the
orientation of the ligand and therefore the position of
the hydrophobic group inside the active site. These vari-
ations of positions of the ligands inside the active site
could explain the variation of anity. The comparison
of the three dimensional structures of these compounds
and steroid aromatase inhibitors, like the compound
3026 with hydrophobic group, showed a close position
of the hydrophobic groups between the two structures
when the pyridin group for these series and C19 methyl
group of steroid inhibitor have the same orientation
(Fig. 1). These results con®rm the presence of the
extrahydrophobic surface, explaining the activities of
these series. By comparing 21 to 22 and 23 to 24, two
opposite e€ects are observed, in terms of activity, with
the 4-tri¯uoromethoxyphenyl group. This group is very
hydrophobic with a theoretical logP27 of 3.2 (compared
to a value of 2.8 for para chloro phenyl or 2.1 for 4-
methoxyphenyl group). The improvement of activity of
22 compared to 21 could be explained by this factor.
For 24 compared to 23, a steric hindrance could explain
this result. Indeed, the better anity of 23 compared to
22 must be due to a closer hydrophobic contact between
the ligand and residues of the receptor, and this rein-
forces the idea of the steric hindrance with a 4-tri-
¯uoromethoxyphenyl group.
Compounds
IC50 (mM)
Ki/Km
A
B
4-OHA
0.45‹0.06
3.0‹0.6
1.6‹0.4
0.65‹0.07
9.4
1.9‹0.8
12.7‹0.1
2.1‹0.2
Ð
MR 20445 (21)
MR 20491 (23)
MR 16089 (22)
MR 16088 (24)
Ð
Fadrozole
0.06‹0.03
7.4‹1.6
0.11‹0.06
0.15‹0.01
0.1‹0.0
7.3‹0.9
0.5‹0.3
1.0‹0.3
MR 20493 (27)
MR 20494 (28)
MR 20492 (29)
IC50 of compounds 12±20, 25±26 were >10 mM.
Km (Á4)=10.7‹3.3 nM
aAromatase activity was evaluated by measuring 3H2O released from
200 nM [1b,2b-3H]androstenedione at 37 ꢀC for 15 min, in the presence
of inhibitors. 4-Hydroxyandrostenedione (4-OHA) was used as con-
trol. The inhibition of aromatase in human placental (195 mg) micro-
somes was performed in the presence of 60 mM NADPH, H+. The
blank was realised without adding NADPH, H+. The results are
expressed as % to a standard control which was incubated with
NADPH, H+ and without inhibitor, and are the mean of triplicate
experiments‹SD. (Ð): not determined. For legend, see above.
Fadrozole was used as control.
weak inhibitors in these conditions, while 21 (MR
20445), 23 (MR 20491) and 22 (MR 16089) were as
potent as 4-OHA. Finally, compounds 28 (MR 20494)
and 29 (MR 20492) were very potent inhibitors of
human aromatase, slightly less than Fadrozole.

950
P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
Figure 2. One equatorial conformation of compound 30.
For the tetrahydroindolizinone series, these compounds
have a better ¯exibility with an envelope conformation
allowing us to place the hydrophobic group in two dif-
ferent equatorial positions (see Fig. 2 for one con-
formation). This ¯exibility is surely the main factor
explaining the improvement of the activities. It is inter-
esting to see the same activities for the meta or para
pyridinylmethylene group. From these data, by molecular
modeling, we can design new structures which will be a
combination of the three dimensional characteristics of
these two ligands. This work is actually in progress.
0.1 mol) and ammonium acetate (15.4 g, 0.2 mol). The
reaction mixture was re¯uxed for 48 h. The precipitate
was ®ltered, washed with water and dried to give 5c as a
colorless solid (26%) which was used without further
puri®cation; mp>260 ꢀC; IR (KBr) 3100±2450 (+NH3
1
CO2 ), 2230 (CN), 1630 (CO); H NMR (DMSO-d6) d:
0
0
0
0
0
7.82 (d, 2H, JH-3 H-2 =JH-5 H-6 =7.5 Hz, H-3 and H-
50), 7.60 (d, 2H, JH-2 H-3 =JH-6 H-5 =7.5 Hz, H-2 and
H-60), 4.33 (m, 1H, H-3), 3.5 (b s, 3H, +NH3), 2.49
(m, 1H, H-2a), 2.37 (m, 1H, H-2b); anal. C10H10N2O2
(C, H, N).
0
0
0
0
0
Experimental Protocols
Chemistry
3-Amino-3-(4-tri¯uoromethoxyphenyl)propionic acid 5d.
To a solution of 4-tri¯uoromethoxybenzaldehyde 4d
(19.0 g, 0.1 mol) in ethanol (200 mL) were added malo-
nic acid (10.7 g, 0.1 mol) and ammonium acetate (15.4 g,
0.2 mol). The reaction mixture was re¯uxed for 6 h. The
precipitate was ®ltered, washed with hot ethanol and
dried to give 5d as a colorless solid (58%) which was
used without further puri®cation; mp >260ꢀC; IR (KBr)
Melting points were determinated on a Ko¯er block and
are uncorrected. IR spectra were recorded on a Mattson
1000 FTIR spectrophotometer and only noteworthy
absorptions (reciprocal centimeters) are listed. 1H NMR
spectra were recorded at 400 MHz with tetramethyl-
silane as an internal standard using a JEOL JNM-LA
400 spectrometer. Splitting patterns have been desig-
nated as follows: s=singlet; b s=broad singlet; d=
doublet; t=triplet; q=quartet; qt=quintuplet; dd=
double doublet; m=multiplet. Analytical TLC was car-
ried out on 0.25 precoated silica gel plates (POLY-
GRAM SIL G/UV254) with visualisation by irradiation
with a UV lamp. Silica gel 60 (70±230 mesh) was used
for column chromatography. Analyses indicated by
the symbols of the elements were within ‹0.4% of the
theoretical values.
3100±2450 (+NH3 CO2 ), 1630 (CO), 1260, 1210, 1165
1
0
0
(CF3); H NMR (DMSO-d6) d0: 7.51 (d, 2H, JH-3
=
=
H-2
0
0
0
0
JH-5 H-6 =8.5 Hz, H-3 and H-5 ), 7.27 (d, 2H, JH-2 H-3
0
0
0
0
0
JH-6 H-5 =8.5 Hz, H-2 and H-6 ), 4.26 (t, 1H, J=8 Hz,
H-3), 3.6 (b s, 3H, +NH3), 2.5 (m, 2H, H-2a and H-2b);
anal. C10H10NO3F2 (C, H, N).
Ethyl 3-amino-3-arylpropionates and ethyl 4-amino-4-
vinylbutyrate 6a±c,e or g. General method: to a stirred
suspension of 3-amino-3-arylpropionic acid 5a±c or e or
4-amino-4-vinylbutyric acid (0.050 mol) 5g in ethanol
(100 mL) was added dropwise at room temperature
thionyl chloride (0.055 mol). Stirring was maintained for
20 min and the solution was then evaporated to dryness
under reduced pressure. The solid residue was triturated
in ether (50 mL), ®ltered and dissolved in a mixture of
water (30 mL) and ether (300 mL). The reaction mixture
was then bubbled for 2 min at room temperature with
an ammonia gas ¯ow. The organic layer was separated,
dried over sodium sulfate, ®ltered and the solvent was
removed under reduced pressure to give 6a±c,e or g as
an oil which was used without further puri®cation.
3-Amino-3-(4-cyanophenyl)propionic acid 5c. To a solu-
tion of 4-cyanobenzaldehyde 4c (13.1 g, 0.1 mol) in ace-
tonitrile (200 mL) were added malonic acid (10.7 g,
Ethyl 3-amino-3-phenylpropionate 6a. Yellow oil (53%);
IR (KBr) 3200 (NH2), 1740 (CO); 1H NMR (DMSO-d6)
d: 7.2 (m, 5H, Ph), 4.15 (m, 1H, H-3), 3.95 (q, 2H,
J=7.2 Hz, CH2), 2.54 (m, 2H, H-2); 1.8 (b s, 2H, NH2),
Figure 1. Superimposition of compounds 21 and 30.

P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
951
1.05 (t, 3H, J=7.2 Hz, CH3); anal. C11H15NO2 (C, H,
N).
CH2), 3.38 (dd, 1H, JH-2a H-2b=16 Hz, JH-2a H-3=9 Hz,
H-2a), 3.28 (dd, 1H, JH-2b H-2a=16 Hz, JH-2b
6.3 Hz, H-2b), 1.07 (t, 3H, J=7.1 Hz, CH3); anal.
=
H-3
Ethyl 3-amino-3-(2-chlorophenyl)propionate 6b. Yellow
1
C15H16NO2Cl (C, H, N).
oil (71%); IR (KBr) 3200 (NH2), 1750 (CO); H NMR
0
0
0
(DMSO-d6) d: 7.62 (d, 1H, JH-3 H-4 =7.7 Hz, H-3 ), 7.3
=
Ethyl 3-(4-cyanophenyl)-3-pyrrol-1-ylpropionate 7c.
1
(m, 3H, H-60, H-50 and H-40), 4.62 (dd, 1H, JH-3 H-2b
Orange oil (80%); IR (KBr) 2230 (CN), 1725 (CO); H
0
0
0
0
8.3 Hz, JH-3 H-2a=4.8 Hz, H-3), 4.03 (q, 2H, J=7 Hz,
CH2), 2.56 (dd, 1H, JH-2a H2b=15 Hz, JH-2a H-3=4.8 Hz,
NMR (CDCl3) d: 7.61 (d, 2H, JH-3 H-2 =JH-5
=
=
H-6
8.5 Hz, H-30 and H-50), 7.22 (d, 2H, JH-2 H-3 =JH-6 H-5
0
0
0
0
H-2a), 2.46 (dd, 1H, JH-2b H-2a=15 Hz, JH-2b
8.3 Hz, H-2b), 1.7 (b s, 2H, NH2), 1.02 (t, 3H, J=7 Hz,
CH3); anal. C11H14NO2Cl (C, H, N).
=
8.5 Hz, H-20 and H-60), 6.72 (m, 2H, H-a pyrrol), 6.19
(m, 2H, H-b pyrrol), 5.71 (dd, 1H, JH-3 H-2a=9 Hz, JH-3
H-2b=6.3 Hz, H-3), 4.10 (q, 2H, J=7 Hz, CH2), 3.23
(dd, 1H, JH-2a H-2b=16 Hz, JH-2a H-3=9 Hz, H-2a), 3.13
(dd, 1H, JH-2b H-2a=16 Hz, JH-2b H-3=6.3 Hz, H-2a),
1.17 (t, 3H, J=7 Hz, CH3); anal. C16H16N2O2 (C, H,
N).
H-3
Ethyl 3-amino-3-(4-cyanophenyl)propionate 6c. Yellow
oil (80%); IR (KBr) 3380, 3310 (NH2), 2230 (CN), 1730
(CO); 1H NMR (CDCl3) d: 7.56 (d, 2H, JH-3 H-2 =JH-5
0
0
0
0
0
0
0
0
0
H-6 =8.5 Hz, H-30 and H-5 0), 7.43 (d, 2H, JH-2 H-3 =JH-6
0
H-5 =8.5 Hz, H-2 and H-6 ), 4.41 (t, 1H, J=7 Hz, H-3),
4.06 (q, 2H, J=7 Hz, CH2), 2.56 (d, 2H, J=7 Hz, H-2a
and H-2b), 1.7 (b s, 2H, NH2), 1.16 (t, 3H, J=7 Hz,
CH3); anal. C12H14N2O2 (C, H, N).
Ethyl 3-(pyrrol-1-yl)-3-(thien-3-yl)propionate 7e. Orange
oil (59%); IR (KBr) 1740 (CO); 1H NMR (CDCl3)
0
0
0
d: 7.22 (d, 1H, JH-5 H4 =4.8 Hz, H-5 ), 7.02 (s, 1H,
H-20), 6.85 (d, 1H, JH-4 H5 =4.8 Hz, H-4 ), 6.69 (m,
2H, H-a pyrrol), 6.12 (m, 2H, H-b pyrrol), 5.67 (t, 1H,
JH-3 H-2=7 Hz, 1H, H-3), 4.08 (q, 2H, J=7 Hz, CH2),
3.12 (d, 2H, JH-2 H-3=7.7 Hz, H-2a and H-2b), 1.14 (t,
3H, J=7 Hz, CH3); anal. C13H15NO2S (C, H, N).
0
0
0
Ethyl 3-amino-3-thien-3-ylpropionate 6e. Orange oil
(44%); IR (KBr) 3200 (NH2), 1750 (CO); 10H NMR
0
0
(CDCl3) d: 7.19 (d, 1H, JH-5 H-4 =4.8 Hz, H-5 ), 7.07 (s,
1H, H-20), 6.97 (d, 1H, JH-4 H-5 =4.8 Hz, H-4 ), 4.40 (dd,
1H, JH-3 H-2b=8.8Hz, JH3 H-2a=4.8Hz, H-3), 4.04 (q,
2H, J=72 Hz, CH2), 2.60 (dd, 1H, JH-2a H-2b= 15.7Hz,
0
0
0
Ethyl 4-(pyrrol-1-yl)-4-vinylbutyrate 7g. Orange oil
1
(78%); IR (KBr): 1740 (CO); H NMR (DMSO-d6) d:
JH-2a H-3=4.8Hz, H-2a), 2.53 (dd, 1H, JH-2b
15.7 Hz, JH-2b H-3=8.8 Hz, H-2b), 1.8 (b s, 2H, NH2), 1.14
(t, 3H, J=7.2Hz, CH3); anal. C9H13NO2S (C, H, N).
=
6.74 (t, 2H, H-a pyrrol), 6.19 (m, 3H, H-b pyrrol and
H-vinyl), 5.10 (d, 1H, J=10 Hz, H-vinyl), 5.05 (d, 1H,
J=17 Hz, H-vinyl), 4.55 (m, 1H, H-4), 4.03 (q, 2H,
J=7 Hz, CH2), 2.1 (m, 4H, H-2a, H-2b, H-3a and H-
3b), 1.15 (t, 3H, J=7 Hz, CH3); anal. C12H17NO2 (C, H,
N).
H-2a
Ethyl 4-amino-4-vinylbutyrate 6g. Colorless oil (93%);
IR (KBr) 3200 (NH2), 1750 (CO); 1H NMR (DMSO-d6)
d: 5.72 (m, 1H, H-vinyl), 5.04 (d, 1H, J=10 Hz, H-
vinyl), 5.04 (d, 1H, J=17 Hz, H-vinyl), 4.03 (q, 2H, J=
7 Hz, CH2), 3.15 (m, 1H, H-4), 2.28 (t, 2H, J=7.5 Hz,
H-2a and H-2b), 1.6 (m, 4H, H-3a, H-3b and NH2),
1.16 (t, 3H, J=7 Hz, CH3); anal. C8H15NO2 (C, H, N).
3-Aryl-2,3-dihydro-1H-pyrrolizin-1-ones and 5-vinyl-
5,6,7,8-tetrahydroindolizin-8-one 8a±c,e or g. General
method: to a stirred solution of ethyl 3-aryl-3-(pyrrol-1-
yl)propionates 7a±c or e or ethyl 4-(pyrrol-1-yl)-4-
vinylbutyrate 7g (0.019 mol) in methylene chloride
(100 mL) was added dropwise at 0 ꢀC a solution of
boron tribromide in methylene chloride (1 M, 110 mL).
The reaction mixture was stirred for 30 min at room
temperature and hydrolysed with sodium hydrogen
carbonate at 0 ꢀC. The organic layer was washed with
water (2Â100 mL), dried over calcium chloride and
evaporated to dryness. The solid residue was puri®ed by
a silica-gel column with methylene chloride as eluent to
give 8a±c,e or g as crystals.
Ethyl 3-aryl-3-(pyrrol-1-yl)propionates and ethyl 4-(pyr-
rol-1-yl)-4-vinylbutyrate 7a±c,e or g. General method:
to a solution of ethyl 3-amino-3-arylpropionates 6a±c or
e or ethyl 4-amino-4-vinyl butyrate 6g (0.045 mol) in
acetic acid (100 mL) was added 2,5-dimethoxytetra-
hydrofuran (5.9 mL, 0.045 mol). The reaction mixture
was re¯uxed for 1 h and then evaporated to dryness.
The residue was taken up in ether (100 mL) and ®ltered.
The ®ltrate was washed with water, dried over sodium
sulfate and evaporated under reduced pressure to give
7a±c,e or g as an oil.
3-Phenyl-2,3-dihydro-1H-pyrrolizin-1-one 8a. Colorless
crystals (50%); mp 98 ꢀC (literature).
Ethyl 3-phenyl-3-(pyrrol-1-yl)propionate 7a. Orange oil
1
(86%); IR (KBr) 1750 (CO); H NMR (DMSO-d6) d:
3-(2-Chlorophenyl)-2,3-dihydro-1H-pyrrolizin-1-one 8b.
Colorless crystals (59%); mp 112 ꢀC; IR (KBr) 1685
7.3 (m, 5H, Ph), 6.94 (m 2H, H-a pyrrol), 6.03 (m, 2H,
H-b pyrrol), 5.66 (t, 1H, JH-3 H-2=6.8 Hz, H-3), 4.01 (q,
2H, J=7.1 Hz, CH2), 2.32 (m, 2H, H-2), 1.07 (t, 3H,
J=7.1 Hz, CH3); anal. C15H17NO2 (C, H, N).
1
0
0
(CO); H NMR (CDCl3) d: 7.50 (dd, 1H, JH-3
=
=
H-4
0
0
7.7 Hz, JH-3 H-5 =1.4 Hz, H-3 ), 7.35 (dt, 1H, JH-4 H-3
0
0
0
0
0
7.7 Hz, JH-4 H-5 =7.7 Hz, JH-4 H-6 =1.7 Hz, H-4 ), 7.28
0
0
0
0
0
0
0
0
0
(dt, 1H, JH-5 H-4 =7.7 Hz, JH-5 H-6 =7.7 Hz, JH-5 H-3
=
=
=
=
=
Ethyl 3-(2-chlorophenyl)-3-(pyrrol-1-yl)propionate 7b.
Orange oil (78%); IR (KBr) 1750 (CO); 1H NMR
(DMSO-d6) d: 7.3 (m, 4H, Ph), 6.90 (m, 2H, H-a pyrrol),
6.02 (m, 3H, H-b pyrrol and H-3), 4.03 (q, 2H, J=7 Hz,
1.7 Hz, H-50), 7.16 (dd, 1H, JH-5 H-6=2.2 Hz, JH-5 H-7
1.1 Hz, H-5), 6.72 (dd, 1H, JH-7 H-6=3.9 Hz, JH-7 H-5
1.1 Hz, H-7), 6.68 (dd, 1H, JH-6 H-5 =7.7 Hz, JH-6 H-4
1.4 Hz, H-60), 6.56 (dd, 1H, JH-6 H-7=3.9 Hz, JH-6 H-5
0
0
0
0

952
P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
2.2 Hz, H-6), 6.04 (dd, 1H, JH-3 H-2a=8 Hz, JH-3 H-2b
3.1 Hz, H-3), 3.65 (dd, 1H, JH-2a H-2b=18 Hz, JH-2a H-3
8 Hz, H-2a), 2.73 (dd, 1H, JH-2b H-2a=18 Hz, JH-2b H-3
3.1 Hz, H-2b); anal. C13H10NOCl (C, H, N).
=
=
=
phenyl)propionic acid 5d (10 g, 0.04 mol) in acetic acid
(150 mL) was added 2,5-dimethoxytetrahydrofuran
(7.8 mL, 0.04 mol). The reaction mixture was re¯uxed
for 1 h and then evaporated to dryness. The residue was
taken up in ether (200 mL) and the suspension was ®l-
tered. The ®ltrate was washed with water (2Â100 mL),
then dried over sodium sulfate and the solvent was
removed under reduced pressure to give 9 as a colorless
oil (95%); IR (KBr) 3300±2500 (OH), 1715 (CO), 1260,
3-(4-Cyanophenyl)-2,3-dihydro-1H-pyrrolizin-1-one 8c.
Yellow crystals (80%); mp 104 ꢀC (ether); IR (KBr) 2230
1
(CN), 1695 (CO); H NMR (CDCl3) d: 7.61 (d, 2H,
0
0
0
0
0
0
JH-3 H-2 =JH-5 H-6 =8.5 Hz, H-3 and H-5 ), 7.12 (d,
0
0
2H, JH-2 H-3 =JH-6 H-5 =8.5 Hz, H-2 and H-6 ), 6.78
(m, 2H, H-5 and H-7), 6.53 (dd, 1H, JH-6 H-7=3.8 Hz,
JH-6 H-5=2.5 Hz, H-6), 5.54 (dd, 1H, JH-3 H-2a=9 Hz,
JH-3 H-2b=3 Hz, H-3), 3.53 (dd, 1H, JH-2a H-2b=18 Hz,
JH-2a H-3=9 Hz, H-2a), 2.81 (dd, 1H, JH-2b H-2a=18 Hz,
JH-2b H-3=3 Hz, H-2b); anal. C14H10N2O (C, H, N).
1215, 1165 (CF3); 1H NMR (DMSO-d6) d: 10.5 (b s, 1H,
0
0
0
0
0
0
0
0
0
OH), 7.42 (d, 2H, JH-3 H-2 =JH-5 H-6 =8.5 Hz, H-3 and
H-50), 7.29 (d, 2H, JH-2 H-3 =JH-6 H-5 =8.5 Hz, H-2
and H-60), 6.95 (m, 2H, H-a pyrrol), 5.99 (m, 2H, H-b
pyrrol), 5.66 (t, 1H, J=7 Hz, H-3), 3.25 (m, 2H, H-2a
and H-2b); anal. C14H12NO3F3 (C, H, N).
0
0
0
0
0
3-(4-Tri¯uoromethoxyphenyl)-2,3-dihydro-1H-pyrrolizin-
1-one 8d. A suspension of N(3-(4-tri¯uoromethoxy-
phenyl)-2,3-dihydro-1H-pyrrolizin-1-yl)pyrrolidinium
perchlorate 11 (10.8 g, 0.025 mol) in an aqueous solution
of sodium hydroxide (1 N, 50 mL) was heated at 60 ꢀC
for 30 min. The cooled reaction mixture was then
extracted with chloroform (2Â100 mL). The organic
layers were collected, washed with water (100 mL), dried
over magnesium chloride and evaporated to dryness
under reduced pressure to give a green oil which was
dissolved in hot ethanol (20 mL). The solution was
charcoaled, ®ltered and cooled to give, by dilution with
water (10 mL), 8d as a colorless solid (37%); mp 84 ꢀC
(propan-2-ol); IR (KBr) 1695 (CO), 1265, 1220, 1165
N-(3-(4-Tri¯uoromethoxyphenyl)-3-pyrrol-1-ylpropionyl)-
pyrrolidine 10. To a stirred solution of 3-(4-tri¯uoro-
methoxyphenyl)-3-pyrrol-1-ylpropionic acid 9 (18.3 g,
0.06 mol) in acetone (150 mL) at 0 ꢀC was added drop-
wise triethylamine (9.3 mL, 0.066 mol). After 30 min,
ethyl chloroformate (5.9 mL, 0.06 mol) was added
dropwise at 0 ꢀC. After 30 min, pyrrolidine (5.1 mL,
0.06 mol) was added in the same conditions. The reac-
tion mixture was re¯uxed for 1 h and then evaporated to
dryness. The oily residue was taken up in ether (250 mL)
and ®ltered. The ®ltrate was washed with water
(2Â100 mL), dried over calcium chloride and the solvent
was removed under reduced pressure to give 10 as a
colorless solid (40%); mp 85 ꢀC (ether/petroleum ether);
1
1
(CF3); H NMR (CDCl3) d: 7.26 (s, 1H, H-5), 7.23 (d,
IR (KBr) 1624 (CO), 1280, 1195, 1160 (CF3); H NMR
0
0
(CDCl3) d: 7.2 (m, 4H, H-20, H-30, H-50 and H-60), 6.72
(m, 2H, H-a pyrrol), 6.16 (m, 2H, H-b pyrrol), 4.10 (m,
1H, H-3), 3.20 (m, 5H, H-a pyrrolidine and H-2a),
2.14 (m, 1H, H-b), 1.10 (m, 4H, H-b pyrrolidine); anal.
C18H19N2O2F3 (C, H, N).
0
0
0
0
2H, JH-3 H-2 =JH-5 H-6 =8.5 Hz, H-3 and H-5 ), 7.13
0
0
0
0
(s, 1H, H-7), 6.84 (d, 2H, JH-2 H-3 =JH-6 H-5 =8.5 Hz,
H-20 and H-60), 6.57 (m, 1H, H-6), 5.57 (dd, 1H,
JH-3 H-2a=9 Hz, JH-3 H-2b=3 Hz, H-3), 3.56 (dd, 1H,
JH-2a H-2b=18 Hz, JH-2a H-3=9 Hz, H-2a), 2.92 (dd, 1H,
JH-2b H-2a=18 Hz, JH-2b H-3=3 Hz, H-2b); anal. C14H10
NO2F3 (C, H, N).
N-(3-(4-Tri¯uoromethoxyphenyl)-2,3-dihydro-1H-pyrrolizin-
1-yl)pyrrolidinium perchlorate 11. Phosphoryl chloride
(4.4 mL, 0.048 mol) was added to a solution of N-(3-(4-
tri¯uoromethoxyphenyl)-3-pyrrol-1-ylpropionyl)pyrro-
lidine 10 (8 g, 0.024 mol) in toluene (100 mL). The reac-
tion mixture was re¯uxed for 1 h and then evaporated to
dryness. The oily residue was washed with petroleum
ether (2Â100 mL) and poured into water (100 mL). The
mixture was re¯uxed again for another 1 h and ®ltered.
The ®ltrate was cooled and adjusted to pH=9 with
sodium hydrogen carbonate. Perchloric acid was ®nally
added to acidify the solution to pH=2. The precipitate
was ®ltered, washed with water (50 mL) and dried to
give 11 as a green solid (95%); mp 100 ꢀC (dec) (etha-
3-(Thien-3-yl)-2,3-dihydro-1H-pyrrolizin-1-one 8e. Color-
less crystals (45%); mp 110 ꢀC; IR (KBr) 01680 (CO);
1H NMR (DMSO-d6) d: 7.50 (m, 2H, H-2 and H-50),
0
0
0
7.13 (d, 1H, JH-4 H-5 =4.8 Hz, H-4 ), 6.99 (d, 1H, JH-5
H-6=2.2 Hz, H-5), 6.64 (d,1H, JH-7 H-6=3.8 Hz, H-7),
6.49 (dd, 1H, JH-6 H-7=3.8 Hz, JH-6 H-5=2.2 Hz, H-6),
5.82 (dd, 1H, JH-3 H-2a=7.8 Hz, JH-3 H-2b=3.6 Hz, H-3),
3.50 (dd, 1H, JH-2a H-2b=18 Hz, JH-2a H-3=7.8 Hz, H-
2a), 2.95 (dd, 1H, JH-2b H-2a=18 Hz, JH-2b H-3=3.6 Hz,
H-2b); anal. C11H9NOS (C, H, N).
5-Vinyl-5,6,7,8-tetrahydroindolizin-8-one 8g. Colorless
crystals (49%); mp 66ꢀC; IR (KBr) 1680 (CO); 1H NMR
nol); IR (KBr) 1670 (CN), 1260, 1220, 1170 (CF3), 1090
1
0
0
(CDCl3) d: 6.91 (dd, 1H, JH-1 H-2=3.9Hz, JH-1
1.5 Hz, H-1), 6.76 (dd, 1H, JH-3 H-2=2.5 Hz, JH-3 H-1
1.5 Hz, H-3), 6.16 (dd, 1H, JH-2 H-1=3.9 Hz, JH-2 H-3
=
=
=
(ClO4); H NMR (CDCl3) d: 07.30 (d, 2H, JH-3
=
=
H-3
H-2
0
0
0
0
0
JH-5 H-6 =8.5 Hz, H-3 and H-5 ), 07.22 (d, 2H, JH-2 H-3
0
0
0
JH-6 H-5 =8.5 Hz, H-2 and H-6 ), 7.11 (m, 1H, H-5),
6.98 (m, 1H, H-7), 6.74 (m, 1H, H-6), 5.75 (m, 1H, H-3),
4.34 (dd, 1H, JH-2a H-2b=18 Hz, JH-2a H-3=9 Hz, H-2a),
3.90 (m, 4H, H-a pyrrolidinium), 3.52 (dd, 1H, JH-2b
H-2a=18 Hz, JH-2b H-3=3 Hz, H-2b), 2.22 (m, 4H, H-b
pyrrolidinium); anal. C18H18N2O5ClF3 (C, H, N).
2.5 Hz, H-2), 5.83 (m, 1H, H-vinyl), 5.22 (dd, 1H,
J=10.2 Hz, J=1 Hz, H-vinyl), 4.90 (dd, 1H, J=17 Hz,
J=1 Hz, H-vinyl), 4.62 (m, 1H, H-5), 2.50 (m, 1H, H-
7a), 2.38 (m, 1H, H-7b), 2.26 (m, 1H, H-6a), 2.06 (m,
1H, H-6b); anal. C10H11NO (C, H, N).
3-(4-Tri¯uoromethoxyphenyl)-3-pyrrol-1-ylpropionic acid
9. To a solution of 3-amino-3-(4-tri¯uoromethoxy-
(Z) 2-Alkylidene-3-aryl-2,3-dihydro-1H-pyrrolizin-1-ones
and (Z) 7-alkylidene-5-vinyl-5,6,7,8-tetra-hydroindolizin-

P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
953
8-ones 12±27. General method A: an aqueous sodium
hydroxide solution (6 N, 2 mL) was added to a solution
of 3-aryl-2,3-dihydro-1H-pyrrolizin-1-ones 8a±e or 5-
vinyl-5,6,7,8-tetrahydroindolizin-8-one 8g (0.002 mol) in
ethanol (5 mL). Arylcarboxaldehyde (0.0024 mol) was
then added to the reaction mixture wich was stirred at
room temperature for 12 h. The precipitate was ®ltered,
washed with water (2Â20 mL) and ether (2Â20 mL). A
silica-gel column was used, with cyclohexane:ethyl ace-
tate (70:30) as eluent, to give the Z-form of 12±18, 21,
25 or 26.
(Z) 3-(2-Chlorophenyl)-2-(2,4-dimethoxyphenyl)methylene-
2,3-dihydro-1H-pyrrolizin-1ꢀ-one 16. Yellow crystals
1
(method A, 94%); mp 183 C; IR (KBr) 1680 (CO); H
NMR (DMSO-d6) d: 7.80 (s, 1H, H-vinyl), 7.50 (d, 1H,
00
2H, H-7 and H-3), 6.71 (d, 1H, JH-6
00
00
JH-3
=7.7 Hz, H-3 ), 7.3 (m, 4H, Ph and 5), 6.7 (m,
00
H-4
00
00
=7.7Hz, H-6 ),
0
=
H-5
6.45 (m, 2H, H-6 and H-30), 6.35 (d, 1H, JH-5
0
H-6
8.8 Hz, 1H, H-50), 3.82 (s, 3H, CH3), 3.72 (s, 3H, CH3);
anal. C22H18NO3Cl (C, H, N).
(Z) 2-(2,4-Dimethoxyphenyl)methylene-3-(thien-3-yl)-2,3-
dihydro-1H-pyrrolizin-1-one 17. Yellow crystals (method
ꢀ
1
General method B: to a solution of 3-aryl-2,3-dihydro-
1H-pyrrolizin-1-ones 8a±e or 5-vinyl-5,6,7,8-tetra-
hydroindolizin-8-one 8g (0.002 mol) and arylcarbox-
aldehyde (0.0022 mol) in methylene chloride (20 mL)
were added a solution of sodium hydroxide (6 N,
0.006 mol) and tetrabutylammonium hydrogen sulfate
(catalytic amount). The reaction mixture was stirred at
room temperature for 2 h. After addition of methylene
chloride (40 mL), the organic layer was separated,
washed with water, dried over calcium chloride and
evaporated to dryness. A silica-gel column was used,
with cyclohexane:ethyl acetate (70:30) as eluent, to give
the Z-form of 19, 20, 22±24 or 27.
A, 44%); mp 112 C; IR (KBr) 1670 (CO); H NMR
0
=
0
(DMSO-d6) d: 7.77 (s, 1H, H-vinyl), 7.56 (d, 1H, JH-5 H-4
4.8 Hz, H-50), 7.4 (m, 2H, H-600 and H-20), 7.18 (d, 1H,
0
0
JH-5 H-6=2.3 Hz, H-5), 6.84 (d, 1H, JH-4 H-5 =4.8 Hz,
H-40), 6.80 (s, 1H, H-3), 6.72 (d, 1H, JH-7 H-6=3.4 Hz,
H-7), 6.53 (s, 1H, H-300), 6.46 (dd, 1H, JH-6 H-7=3.4 Hz,
00
00
=8.5 Hz,
JH-6 H-5=2.3 Hz, H-6), 6.33 (d, 1H, JH-5
H-6
H-50), 3.85 (s, 3H, CH3), 3.78 (s, 3H, CH3); anal.
C20H17NO3S (C, H, N).
(Z) 3-Phenyl-2-(pyridin-2-ylmethylene)-2,3-dihydro-1H-
pyrrolizin-1-one 18. Yellow crystals (method A, 68%);
mp 196 ꢀC; IR (KBr) 1690 (CO); H NMR (DMSO-d6)
1
0
0
0
0
d: 8.55 (d, 1H, JH-6 H-5 =5 Hz, H-6 ), 7.7 (m, 2H, H-4
(Z) 2-Benzylidene-3-phenyl-2,3-dihydro-1H-pyrrolizin-1-
one 12. Yellow crystals (method A, 90%); mp 182 ꢀC;
and H-30), 7.46 (s, 1H, H-vinyl), 7.2 (m, 7H, Ph, H-50
and H-5), 6.97 (d, 1H, JH-7 H-6=3.9 Hz, H-7), 6.84 (dd,
1H, JH-6 H-7=3.9 Hz, JH-6 H-5=2.2 Hz, H-6), 6.50 (s,
1H, H-3); anal. C19H14N2O (C, H, N).
1
IR (KBr) 1650 (CO); H NMR (DMSO-d6) d: 7.53 (s,
1H, H-vinyl), 7.49 (m, 2H, Ph), 7.3 (m, 8H, Ph), 7.14 (d,
1H, JH-5 H-6=2.1 Hz, H-5), 6.85 (s, 1H, H-3), 6.81 (d,
1H, JH-7 H-6=3.7 Hz, H-7), 6.49 (dd, 1H, JH-6
3.7 Hz, JH-6 H-5=2.1 Hz, H-6); anal. C20H15NO (C, H,
N).
=
(Z) 3-(4-Cyanophenyl)-2-(pyridin-2-ylmethylene)-2,3-di-
hydro-1H-pyrrolizin-1-one 19. Yellow crystals (method
B, 85%); mp 102 ꢀC (ethanol/water); IR (KBr) 2230
H-7
1
(CN), 1695 (CO); H NMR (CDCl3) d0: 8.51 (d, 1H,
0
H-vinyl), 7.52 (d, 2H, JH-3
0
JH-6 H-5 =5 Hz, H-6 ), 7.65 (m, 1H, H-5 ), 7.54 (s, 1H,
0
(Z) 2-Benzylidene-3-(2-chlorophenyl)-2,3-dihydro-1H-pyr-
rolizin-1-one 13. Yellow crystals (method A, 85%); mp
00
00
00
00
=8.5 Hz, H-
H-6
=JH-5
00
H-2
128 ꢀC; IR (KBr) 1690 (CO); H NMR (CDCl3) d: 7.62
300 and H-500), 7.43 (d, 2H, JH-2
=JH-6
=8.5 Hz,
1
00
00
00
H-3
H-5
0
H-200 and H-600), 7.41 (d, 1H, JH-3 H-4 =5 Hz, H-3 ),
7.16 (m, 1H, H-40), 6.96 (d, 1H, JH-5 H-6=3.5 Hz, H-5),
6.93 (m, 1H, H-7), 6.86 (m, 1H, H-6), 6.53 (s, 1H, H-3);
anal. C19H13N3O (C, H, N).
0
0
(s, 1H, H-vinyl), 7.2 (m, 10H, Ph and H-5), 6.84 (s, 1H,
H-3), 6.81 (d, 1H, JH-7 H-6=3.9 Hz, H-7), 6.38 (dd, 1H,
JH-6 H-7=3.9 Hz, JH-6 H-5=2.2 Hz, H-6); anal. C20H14
NOCl (C, H, N).
(Z) 2-Benzylidene-3-(thien-3-yl)-2,3-dihydro-1H-pyrroli-
zin-1-one 14. Yellow crystals (method A, 77%); mp
148 C; IR (KBr) 1670 (CO); H NMR (DMSO-d6) d:
(Z) 3-(4-Tri¯uoromethoxyphenyl)-2-(pyridin-2-ylmethyl-
ene)-2,3-dihydro-1H-pyrrolizin-1-one 20. Yellow crystals
(method B 70%); mp 152 ꢀC (ether); IR (KBr) 1685
ꢀ
1
7.65 (s, 1H, H-vinyl),0 7.50 (m, 3H, Ph), 7.40 (d, 1H,
(d, 1H, JH-5 H-6=2.3 Hz, H-5), 6.95 (s, 1H, H-3), 6.90
(CO), 1265, 1215, 1165 (CF3); 1H NMR (CDCl3)0d: 8.54
0
0
(s, 1H, H-vinyl), 7.39 (d, 1H, JH-3 H-4 =5 Hz, H-3 ),
0
JH-5 H-4 =4.8 Hz, H-5 ), 7.31 (m, 3H, Ph et H-2 ), 7.25
0
0
(d, 1H, JH-6 H-5 =5 Hz, H-6 ), 7.63 (m, 1H, H-5 ), 7.52
0
0
0
0
0
00
00
0
0
00
00
00
00
=8.5 Hz, H-3 and H-5 ),
H-6
(d, 1H, JH-4 H-5 =4.8 Hz, H-4 ), 6.80 (d, 1H, JH-7 H-6
3.4 Hz, H-7), 6.50 (dd, 1H, JH-6 H-7=3.4 Hz, JH-6 H-5
2.3 Hz, H-6); anal. C18H13NOS (C, H, N).
=
=
7.32 (d, 2H, JH-3
=JH-5
H-2
7.14 (m, 1H, H-40), 7.04 (d, 2H, JH-2
=JH-6
00
00
00
00
=
H-5
H-3
8.5 Hz, H-200 and H-600), 6.95 (d, 1H, JH-5 H-6=3.5 Hz,
H-5), 6.92 (m, 1H, H-7), 6.89 (m, 1H, H-6), 6.53 (s, 1H,
H-3); anal. C20H13N2O2F3 (C, H, N).
(Z) 2-(2,4-Dimethoxyphenyl)methylene-3-phenyl-2,3-di-
hydro-1H-pyrrolizin-1-one 15. Yellow crystals (method
A, 87%); mp 131 ꢀC; IR (KBr) 1650 (CO); H NMR
(Z) 3-Phenyl-2-(pyridin-3-ylmethylene)-2,3-dihydro-1H-
pyrrolizin-1-one 21. Yellow crystals (method A, 45%);
1
0
0
(DMSO-d6) d: 7.80 (s, 1H, H-vinyl), 7.35 (d, 1H, JH-6 H-5
=
=
8.8 Hz, H-60), 7.2 (m, 5H, Ph), 7.08 (d, 1H, JH-5 H-6
mp 166 ꢀC; IR (KBr) 1700 (CO); H NMR (CDCl3) 0d:
1
0
8.6 (s, 1H, H-20), 8.44 (d, 1H, JH-6 H-5 =4 Hz, H-6 ),
0
0
2.4 Hz, H-5), 6.7 (m, 2H, H-3 and H-7), 6.50 (d, 1H,
0
JH-6 H-5=2.4 Hz, H-6), 6.38 (dd, 1H, JH-5 H-6 =8.8 Hz,
0
JH-3 H-5 =2 Hz, H-3 ), 6.45 (dd, 1H, JH-6 H-7=3.6 Hz,
0
0
7.64 (s, 1H, H-vinyl), 7.54 (d, 1H, JH-4 H-5=7.7 Hz,
0
H-40), 7.4 (m, 5H, Ph), 7.14 (dd, 1H, JH-5 H-4 =7.7 Hz,
0
0
0
0
CH3); anal. C22H19NO3 (C, H, N).
0
0
JH-5 H-3 =2 Hz, H-5 ), 3.82 (s, 3H, CH3), 3.74 (s, 3H,
0
0
0
JH-5 H-6 =4 Hz, H-5 ), 6.93 (d, 1H, JH-7 H-6=3.9 Hz,
H-7), 6.90 (d, 1H, JH-5 H-6=2.2 Hz, H-5), 6.50 (dd, 1H,

954
P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
JH-6 H-7=3.9 Hz, JH-6 H-5=2.2 Hz, H-6), 6.32 (s, 1H, H-
3); anal. C19H14N2O (C, H, N).
(Z) 7-(Pyridin-4-ylmethylene)-5-vinyl-5,6,7,8-tetrahydro-
indoliziꢀn-8-one 27. Yellow crystals (method B, 57%);
1
mp 70 C; IR (KBr) 1660 (CO); H NMR (C0DCl3) d:
0
0
0
0
(Z) 3-(4-Tri¯uoromethoxyphenyl)-2-(pyridin-3-ylmethyl-
ene)-2,3-dihydro-1H-pyrrolizin-1-one 22. Yellow crystals
(method B, 65%); mp 136 ꢀC (ether); IR (KBr)
8.60 (d, 2H, JH-2 H-3 =JH-6 H-5 =5.1 Hz, H-2 and H-
60), 7.70 (s, 1H, H-vinyl), 7.2 (m, 3H, H-30, H-50 and
H-1), 6.85 (d, 1H, JH-3 H-2=2.5 Hz, H-3), 6.29 (dd, 1H,
JH-2 H-1=4 Hz, JH-2 H-3=2.5 Hz, H-2), 5.80 (m, 1H, H-
vinyl), 5.22 (d, 1H, J=10 Hz, H-vinyl), 4.97 (d, 1H,
J=17 Hz, H-vinyl), 4.68 (m, 1H, H-5), 3.20 (dd, 1H,
JH-6a H-6b=15 Hz, JH-6a H-5=4.4 Hz, H-6a), 3.07 (dd,
1H, JH-6b H-6a=15 Hz, JH-6b H-5=6.4 Hz, H-6b); anal.
C16H14N2O (C, H, N).
1
1690 (CO), 1265, 1215, 1165 (CF3); H NMR (CDCl3)
0
d: 8.61 (s, 1H, H-20), 8.47 (d, 1H, JH-6 H-5 =4 Hz, H-60),
0
0
0
7.66 (s, 1H, H-vinyl), 7.51 (d, 1H, JH-4 H-5 =8Hz, H-4 ),
0
00
00
00
00
00
7.27 (d, 2H, JH-3
=JH-5
=8.5 Hz, H-3 and H-
H-2
H-6
500), 7.18 (m, 1H, H-50), 7.11 (d, 2H, JH-2 H-3 =JH-6
00
00
00
00
=
H-5
8.5 Hz, H-200 and H-600), 6.96 (d, 1H, JH-5 H-6=2.2 Hz,
H-5), 6.91 (m, 1H, H-7), 6.54 (m, 1H, H-6), 6.37 (s, 1H,
H-3); anal. C20H13N2O2F3 (C, H, N).
Pharmacology
(Z) 3-Phenyl-2-(pyridin-4-ylmethylene)-2,3-dihydro-1H-
pyrrolizin-1-one 23. Yellow crystals (method B, 33%);
Preparation of microsomes. Human placental and
equine testicular microsomes were prepared as pre-
viously described.13,28 Brie¯y, we have taken in each
instance the available tissue presenting the highest spe-
ci®c aromatase activity. Fresh tissue washed with 0.50
M KCl was ®rst homogenised in 50 mM phosphate
bu€er pH 7.5 containing 0.25 M sucrose, 1 mM DTT,
and 4 mM androstenedione in order to preserve the
enzyme active site, and then centrifuged at 20,000 g. The
supernatant was further ultracentrifuged at 100,000 g
and the ®nal pellet was dissolved in the same bu€er
containing 20% glycerol, 1 mM DTT, 0.2 mM EDTA-
4 Na, 4 mM androstenedione, and stored at 80 ꢀC until
use. Protein concentration was evaluated according to
Bradford29 using bovine serum albumin as standard,
and Coomassie brilliant blue as dye-reagent.
mp 148 ꢀC; IR (KBr) 1685 (CO); H NMR (CDCl3) d:
1
0
0
0
0
0
8.48 (d, 2H, JH-2 H-3 =JH-6 H-5 =4.9 Hz, H-2 and H-
60), 7.6 (m, 3H, H-30, H-50 and H-vinyl), 7.2 (m, 5H, Ph),
6.95 (d, 1H, JH-7 H-6=3.9Hz, H-7), 6.90 (d, 1H, JH-5 H-6
2.2 Hz, H-5), 6.31 (dd, 1H, JH-6 H-7=3.9 Hz, JH-6 H-5
=
=
2.2 Hz, H-6), 6.16 (s, 1H, H-3); anal. C19H14N2O (C,
H, N).
(Z) 3-(4-Tri¯uoromethoxyphenyl)-2-(pyridin-4-ylmethyl-
ene)-2,3-dihydro-1H-pyrrolizin-1-one 24. Yellow crystals
(method B, 55%); mp 92 ꢀC (ether/petroleum ether);
1
IR (KBr) 1685 (CO), 1270, 1215, 1165 (CF3); H NMR
0
=
0
0
0
0
(CDCl3) d: 8.52 (d, 2H, JH-2 H-3 =JH-6 H-5 =5 Hz, H-2
and H-60), 7.62 (s, 1H, H-vinyl), 7.26 (d, 2H, JH-3
00
00
H-2
00
00
0
00
00
JH-5
=8.5 Hz, H-3 and H-5 ), 7.15 (m, 4H, H-3 ,
H-6
H-50, H-200 and H-6), 6.98 (d, 1H, JH-5 H-6=Hz, H-5),
6.91 (m, 1H, H-7), 6.56 (m, 1H, H-6), 6.33 (s, 1H, H-3);
anal. C20H13N2O2F3 (C, H, N).
Inhibition studies with microsomes. Aromatase activity
was evaluated by measuring 3H2O released from
200 nM [1b,2b-3H]androstenedione (speci®c activity:
1554 GBq/mmol) at 37 ꢀC for 15 min according to
Auvray et al.,13 in the presence of various inhibitors
from 0 to 15 mM. Reactions with microsomes (195 and
20 mg of human and equine microsomal proteins
respectively for determination of IC50 values) were
initiated by adding 60 mM NADPH, H+ to a ®nal
volume of 0.5 mL and stopped by adding 1 mL of
chloroform. Steroids were then extracted by incubation
with a charcoal/dextran solution (7%/1.5%) and the
radioactivity of the aqueous phase was measured as
previously described.28 Control incubations were rea-
lised by incubating microsomes, substrate and inhibitors
without NADPH, H+ in the same conditions. Results
are the mean of triplicate experiments‹SD and are
(Z) 7-Benzylidene-5-vinyl-5,6,7,8-tetrahydroindolizin-8-
one 25. Yellow crystals (method A, 35%); mp 86 ꢀC; IR
1
(KBr) 1680 (CO); H NMR (CDCl3) d: 7.88 (s, 1H, H-
vinyl), 7.3 (m, 5H, Ph), 7.18 (d, 1H, JH-1 H-2=3.6 Hz,
H-1), 6.88 (d, 1H, JH-3 H-2=2.4 Hz, H-3), 6.32 (dd, 1H,
JH-2 H-1=3.6 Hz, JH-2 H-3=2.4 Hz, H-2), 5.88 (m, 1H,
H-vinyl), 5.25 (d, 1H, J=10 Hz, H-vinyl), 5.03 (d, 1H,
J=17 Hz, H-vinyl), 4.70 (m, 1H, H-5), 3.32 (dd, 1H,
JH-6a H-6b=14.9 Hz, JH-6a H-5=4.4 Hz, H-6a), 3.17 (dd,
1H, JH-6b H-6a=14.9 Hz, JH-6b H-5=6.8 Hz, H-6b); anal.
C17H15NO (C, H, N).
(Z) 7-(Pyridin-3-ylmethylene)-5-vinyl-5,6,7,8-tetrahydro-
indolizin-8-one 26. Yellow oil (method A, 28%); IR
.
expressed as pmol estrogen formed/min mg microsomal
proteins.
1
(KBr) 1670 (CO); H NMR (CDCl3) d: 8.53 (s, 1H, H-
0
20), 8.45 (d, 1H, JH-6 H-5 =4.8 Hz, H-6 ), 7.69 (s, 1H, H-
0
0
0
0
0
vinyl), 7.58 (d, 1H, JH-4 H-5 =7.9 Hz, H-4 ), 7.25 (dd,
Kinetic studies. Concentration ranges were 0±600 nM
for inhibitors and 6±100 nM for substrate. Aromatase
activity was evaluated at the linear portion of the
Michaelis±Menten plot by incubating 5 mg (human) or
4 mg (equine) of microsomal proteins with substrate and
various inhibitors at 37 ꢀC for 12 min with human aro-
matase or 8 min with the equine enzyme. Km and Ki
values were determined graphically by using respectively
Lineweaver±Burk and Dixon representations.
0
0
0
0
0
1H, JH-5 H-4 =7.9 Hz, JH-5 H-6 =4.8 Hz, H-5 ), 7.06 (d,
1H, JH-1 H-2=4 Hz, H-1), 6.81 (d, 1H, JH-3 H-2=2.5 Hz,
H-3), 6.22 (dd, 1H, JH-2 H-1=4 Hz, JH-2 H-3=2.5 Hz, H-
2), 5.78 (m, 1H, H-vinyl), 5.15 (d, 1H, J=10 Hz, H-
vinyl), 4.92 (d, 1H, J=17 Hz, H-vinyl), 4.65 (m, 1H, H-
5), 3.18 (dd, 1H, JH-6a H-6b=14.9 Hz, JH-6a H-5=4.7 Hz,
H-6a), 3.05 (dd, 1H, JH-6b H-6a=14.9 Hz, JH-6b
6.3 Hz, H-6b); anal. C16H14N2O (C, H, N).
=
H-5

P. Sonnet et al. / Bioorg. Med. Chem. 8 (2000) 945±955
955
Acknowledgements
Seralini, G. E.; Enguehard, C.; Dallemagne, P.; Bureau, R.;
Sonnet, P.; Rault, S. Eur. J. Med. Chem. 1998, 33, 451.
14. Laughton, C. A.; Zvelebil, M. J.; Neidle, S. J. Steroid
Biochem. Mol. Biol. 1993, 44, 399.
15. Sonnet, P.; Guillon, J.; Enguehard, C.; Dallemagne, P.;
Bureau, R.; Rault, S.; Auvray, P.; Moslemi, S.; Sourdaine, P.;
Galopin, S.; Seralini, G. E. Bioorg. Med. Chem. Lett. 1998, 8,
1041.
This work was supported by the Ligue Nationale contre
le Cancer (Comite du Calvados), for the chemical
syntheses, and by the Ligue Nationale contre le Cancer
(Comite de la Manche) for the biological evaluation,
with a studentship to P.A.
16. Tembo, O. N.; Dallemagne, P.; Rault, S.; Robba, M.
Heterocycles 1993, 36, 2129.
References
17. Je€ord, C. W.; Sienkiewicz, K.; Thornton, S. R. Helv.
Chim. Acta 1995, 78, 1511.
1. Cole, P. A.; Robinson, C. H. J. Med. Chem. 1990, 33, 2933.
2. Brodie, H. B.; Callard, G.; Robinson, C.; Roselli, C.; San-
ten, R. J. Steroid Biochem. Mol. Biol. 1993, 44, 321.
3. Auvray, P.; Sourdaine, P.; Seralini, G. E. Biochem. Biophys.
Res. Commun. 1998, 253, 1.
4. Trunet, P. F.; Bhatnagar, A. S.; Chaudri, H. A.; Hornber-
ger, U. Acta Oncol. 1996, 35, 15.
5. Wouters, W.; Snoeck, E.; De Coster, R. Breast Cancer Res.
Treat. 1994, 30, 89.
6. Dukes, M.; Edwards, P. N.; Large, M.; Smith, I. K.; Boyle,
T. J. J. Steroid Biochem. Mol. Biol. 1996, 58, 439.
7. Howell, A.; Downey, S.; Anderson, E. Eur. J. Cancer 1996,
32A, 576.
18. Rault, S.; Dallemagne, P.; Robba, M. Bull. Soc. Chim. Fr.
1987, 6, 1079.
19. Dallemagne, P.; Rault, S.; Cugnon de Sevricourt, M.;
Hassan, K. M.; Robba, M. Tetrahedron Lett. 1986, 27, 2607.
20. Rodionow, W. M.; Malewinskaya, E. T. Ber. 1926, 59,
2952.
21. Johnson, T. B.; Livak, J. E. J. Am. Chem. Soc. 1936, 58,
299.
22. Guillon, J.; Alsadi, A.; Dallemagne, P.; Rault, S. Pharm.
Pharmacol. Commun. 1998, 4, 213.
23. Clauson-Kaas, N.; Tyle, Z. Acta Chem. Scand. 1952, 6, 667.
24. Elming, N.; Clauson-Kaas, N. Acta Chem. Scand. 1952, 6,
867.
8. Miller, W. R. British J. Cancer 1996, 73, 415.
9. Bayer, H.; Batzl, C.; Hartmann, R. W.; Mannschreck, A. J.
Med. Chem. 1991, 34, 2685.
10. Bayer, H.; Hartmann, R. W. Arch. Pharm. 1991, 324, 815.
11. Bayer, H.; Hartmann, R. W. Arch. Pharm. 1991, 324, 833.
12. Hartmann, R. W.; Bayer, H.; Grun, G. J. Med. Chem.
1994, 37, 1275.
25. Dallemagne, P.; Sonnet, P.; Enguehard, C.; Rault, S. J.
Heterocyclic Chem. 1996, 33, 1689.
26. Oprea, T. I.; Garcia, A. E. J. Comp.-Aided Molecular
Design 1996, 10, 186.
27. www.daylight.com, CLOGP program.
28. Dintinger, T.; Gaillard, J. L.; Zwain, I.; Bouhamidi, R.;
Silberzahn, P. J. Steroid Biochem. 1989, 32, 537.
29. Bradford, M. Anal. Biochem. 1976, 72, 248.
13. Auvray, P.; Moslemi, S.; Sourdaine, P.; Galopin, S.;