New R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(phenylaminothiocarbonylamino)-2H- 1-benzopyrans structurally related to (±)-cromakalim as tissue-selective pancreatic β-cell KATP channel openers

Article abstract of DOI:10.1016/j.bmc.2008.03.065

The present work was aimed at exploring a series of R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(phenylaminothiocarbonylamino)-2H- 1-benzopyrans structurally related to (±)-cromakalim and differently substituted at the 4- and 6-positions. The biological effects of these putative activators of ATP-sensitive potassium channels (K_ATP) were characterized in vitro on the pancreatic endocrine tissue (inhibition of insulin release) and on the vascular smooth muscle tissue (relaxation of aorta rings). The biological activity of these new dimethylchroman derivatives was further compared to that of (±)-cromakalim, (±)-pinacidil, diazoxide and BPDZ 73. Structure-activity relationships indicated that an improved potency for the pancreatic tissue was obtained by introducing a meta- or a para-electron-withdrawing group such as a chlorine atom on the C-4 phenyl ring, independently of the nature of the halogen atom at the 6-position of the benzopyran nucleus. Most original dimethylchroman thioureas were more potent than their 'urea' homologues and even more potent than diazoxide at inhibiting insulin release. Moreover, and unlike (±)-cromakalim or (±)-pinacidil, such compounds appeared to be highly selective towards the pancreatic tissue. Radioisotopic and fluorimetric investigations indicated that the new drugs activated pancreatic K_ATP channels. Lastly, conformational studies suggested that the urea/thiourea dimethylchromans can be regarded as hybrid compounds between cromakalim and pinacidil.

Full text of DOI:10.1016/j.bmc.2008.03.065

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5704–5719
New R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(phenylaminothiocar-
bonylamino)-2H-1-benzopyrans structurally related to
( )-cromakalim as tissue-selective pancreatic
b-cell KATP channel openers
a
b
Sophie Sebille,a, Pascal de Tullio,a, Xavier Florence, Benedicte Becker,
´ ´
b
c
c
`
´
Marie-Helene Antoine, Catherine Michaux, Johan Wouters,
Bernard Pirottea,*,à and Philippe Lebrunb,à
a
` ˆ
Drug Research Center, Laboratoire de Chimie Pharmaceutique, Universite´ de Liege, 1, Avenue de l’Hopital,
`
tour 4 (+5) Sart-Tilman, B-4000 Liege, Belgium
bLaboratoire de Pharmacodynamie et de The´rapeutique, Universite´ Libre de Bruxelles, 808 route de Lennik, B-1070 Bruxelles, Belgium
cLaboratoire de Chimie biologique structurale, Faculte´s Universitaires Notre-Dame de la Paix, rue de Bruxelles,
B-5000 Namur, Belgium
Received 30 November 2007; accepted 25 March 2008
Available online 30 March 2008
Abstract—The present work was aimed at exploring a series of R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(phenylaminothiocarbonyla-
mino)-2H-1-benzopyrans structurally related to ( )-cromakalim and differently substituted at the 4- and 6-positions. The biological
effects of these putative activators of ATP-sensitive potassium channels (KATP) were characterized in vitro on the pancreatic endo-
crine tissue (inhibition of insulin release) and on the vascular smooth muscle tissue (relaxation of aorta rings). The biological activity
of these new dimethylchroman derivatives was further compared to that of ( )-cromakalim, ( )-pinacidil, diazoxide and BPDZ 73.
Structure–activity relationships indicated that an improved potency for the pancreatic tissue was obtained by introducing a meta- or
a para-electron-withdrawing group such as a chlorine atom on the C-4 phenyl ring, independently of the nature of the halogen atom
at the 6-position of the benzopyran nucleus. Most original dimethylchroman thioureas were more potent than their ‘urea’ homo-
logues and even more potent than diazoxide at inhibiting insulin release. Moreover, and unlike ( )-cromakalim or ( )-pinacidil,
such compounds appeared to be highly selective towards the pancreatic tissue. Radioisotopic and fluorimetric investigations indi-
cated that the new drugs activated pancreatic KATP channels. Lastly, conformational studies suggested that the urea/thiourea dim-
ethylchromans can be regarded as hybrid compounds between cromakalim and pinacidil.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
cellular ADP concentration increases, thus linking mem-
brane potential to the metabolic state of the cell.1
ATP-sensitive potassium channels (KATP channels) rep-
resent an important class of ionic channels whose func-
tion is mainly regulated by changes in the intracellular
levels of adenosine triphosphate. These channels are
closed when intracellular ATP levels are elevated and
opened when intracellular ATP declines whereas intra-
KATP channels have been identified in a wide range of
cell types. They have initially been described in cardiac
myocytes2 and later found in endocrine cells,3 skeletal
and smooth muscle cells4,5 and central neurons.6 Such
channels have been shown to be involved in major phys-
iological processes such as hormone secretion, smooth
muscle contractile activity and neurotransmitter
release.7
Keywords: Benzopyrans; Potassium channel openers; Cromakalim
analogues; KATP channels.
The KATP channel is an octameric complex of a sulfo-
nylurea receptor (SURx) and a pore-forming inwardly
rectifying potassium channel (Kir6.·) in a 4 + 4 stoi-
chiometry.8 According to their tissue localization,
*
Corresponding author. Tel.: +32 4 366 43 65; fax: +32 4 366 43 62;
e-mail: B.Pirotte@ulg.ac.be
These authors equally contributed.
à
These authors assumed an equal supervision.
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.03.065

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5705
KATP channels exist in different isoforms resulting
from the assembly, in multiple combinations, of the
Kir6.x (Kir6.1 and Kir6.2) and the SUR· (SUR1,
SUR2A and SUR2B) subunits. For example, SUR1
combines with Kir6.2 to form the pancreatic (insu-
lin-secreting)-b cell KATP channels.9 The cardiac and
skeletal muscle types consist of SUR2A and Kir6.2
subunits whereas the smooth muscle KATP channels
are composed of SUR2B and Kir6.1 or Kir6.2
subunits.10
O
N
O
H
N
H
N
N
C
OH
N
N
C
N
(-)-Cromakalim (1)
( )-Pinacidil (2)
H
N
H
N
H
N
CH3
Given their numerous physiological functions, KATP
channels represent promising drug targets. Therefore,
one main challenge in the development of new KATP
channel modulators as therapeutic agents is the discov-
ery of compounds exhibiting the highest selectivity for
a single KATP channel subtype.
N
N
Cl
S
Cl
S
O
O
O
O
Diazoxide (3)
BPDZ 73 (4)
R
H
A variety of compounds, named ‘potassium channel
openers’ (or PCOs), have been reported to activate KATP
channels. The opening of these K+ channels leads to
plasma membrane hyperpolarization and subsequent
reduction in cell excitability. As a result, the activation
of KATP channels has been shown to inhibit endocrine
and/or neurotransmitter release, to relax vascular and
non-vascular smooth muscle and to shorten cardiac ac-
tion potentials.11 Thus, and according to their tissue
selectivity, PCOs may be expected to become new ther-
apeutic agents for diseases such as type 1/type 2 diabe-
tes, obesity, hyperinsulinism, arterial hypertension,
angina pectoris, bronchial asthma or urinary inconti-
nence.11–16
N
H
N
O
5
4
6
X
3
7
2
O
1
8
Benzopyran phenylurea
derivatives (5)
Figure 1. Chemical structure of (ꢀ)-cromakalim (1), ( )-pinacidil (2),
diazoxide (3), BPDZ 73 (4) and benzopyran phenylurea derivatives (5).
(R, Fig. 1, compound 5) was also shown to enhance
the pancreatic activity and selectivity of these original
R/S-4,6-disubstituted 2,2-dimethylchromans.22
Potassium channel openers consist of a group of com-
pounds with a wide range of chemical structures such
as (ꢀ)-cromakalim (1), ( )-pinacidil (2) and diazoxide
(3) (Fig. 1). (ꢀ)-Cromakalim (1) is the prototype of
the benzopyran potassium channel openers. This drug
has been found to exert a marked myorelaxant activ-
ity17 but cromakalim is also known to be only slightly
active as an inhibitor of insulin secretion, in contrast
to other PCOs like diazoxide (3) or BPDZ 73 (4)
(Fig. 1).18–20
According to these structure–activity relationships de-
duced from our previous investigations, the present
work aimed at developing more active and more insu-
lin-secreting cell selective dimethylchroman derivatives
by replacing the C-4 phenylurea moiety by a bioisosteric
phenylthiourea group.
The new compounds were examined as putative potas-
sium channel openers on rat pancreatic islets (inhibition
of glucose-induced insulin release) and on rat aorta rings
(myorelaxant effect on KCl-precontracted aorta rings) in
order to evaluate their potency and tissue selectivity.
The biological effects of the dimethylchroman thioureas
were also compared with those of their urea analogues
and the effects of compounds 40 and 42 were further
characterized on pancreatic transmembrane cationic
movements.
In the search for new pancreatic b-cell selective PCOs,
we have recently prepared a series of R/S-4,6-disubsti-
tuted 2,2-dimethylchromans structurally related to cro-
makalim.21,22 Some of these original drugs were found
to be less effective as myorelaxants and much more
active as inhibitors of insulin release than the refer-
ence molecule cromakalim. These cromakalim ana-
logues can be considered as the first series of
benzopyrans that markedly activate the pancreatic b-
cell KATP channels.21,22 Structure–activity relationships
further indicated that the nature of the substituent at
the 4-position of the benzopyran nucleus played a cru-
cial role for the development of an inhibitory effect on
insulin release. Indeed, a C-4 phenylurea moiety
(Fig. 1, compound 5) was preferred to a C-4 alkylu-
rea, aralkylurea, alkylthiourea, aralkylthiourea, aryl-
sulfonylurea, alkylcarbamate or alkylcarboxamide
group.21,22 Moreover, the introduction of a meta- or
a para-electron-withdrawing group on the phenyl ring
It should be noted that, due to the existence of a possible
isosterism between arylurea/arylthiourea and arylcyano-
guanidine moieties, such thiourea analogues as well as
their C-4 phenylurea counterparts22 might be regarded
as hybrid compounds between cromakalim and pinacidil
(Fig. 2). Thus, a conformational study was conducted to
judge putative similarities between these dimethylchro-
man ureas/thioureas and the prototype compounds cro-
makalim and pinacidil.

5706
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
R
H
N
N
O
N
O
N
N
H
N
C
OH
H
N
+
X
H
CN
X
N
O
cromakalim
pinacidil
hybrid compounds
X = O, S, NCN
Figure 2. Benzopyran phenylurea and phenylthiourea derivatives as hybrid compounds between cromakalim and pinacidil.
2. Chemistry
X
X
O
i
Access to R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyrans (13–42)
is described on Scheme 1.21,23 Such compounds were
obtained from the appropriate para-halogenophenols
in seven steps.
OH
6 a-c
O
CH3
7 a-c
ii
O
O
O
First, para-halogenophenols 6a–c were acetylated to
provide 4-halogenophenyl acetates 7a–c. Fries rear-
rangement of the esters was carried out using aluminium
chloride at 160 ꢁC to give hydroxyacetophenones 8a–c.
The acetophenone intermediates were then involved in
a ring closure reaction in the presence of acetone and
pyrrolidine and the resulting ketones 9a–c were reduced
to alcohols 10a–c with sodium borohydride. The 4-ace-
tylaminobenzopyrans 11a–c were prepared by the Ritter
reaction from chromanols 10a–c. This reaction occurred
in acetonitrile supplemented with concentrated sulfuric
acid. The subsequent hydrolysis of 11a–c with concen-
trated hydrochloric acid led to the aminochromans
12a–c. Finally, these intermediates were converted to
the final R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyrans (13–42)
by reaction with the appropriate isothiocyanate (R–
N@C@S). All these derivatives (13–42) were crystallised
X
X
iii
CH3
OH
9 a-c
iv
8 a-c
O
H
OH
N
CH3
X
v
X
O
O
11 a-c
10 a-c
vi
1
H
R
S
from appropriate solvents and characterized by IR, H
N
C
NMR and elemental analyses to obtain the final materi-
als with the chemical purity required before pharmaco-
logical evaluations.
H
N
NH2
O
X
X
vii
O
3. Results and discussion
13-42
12 a-c
The ability of the newly synthesized dimethylchromans
(compounds 13–42) to inhibit the glucose-induced insu-
lin secretion was evaluated on isolated rat pancreatic is-
lets and the myorelaxant activity of the compounds was
determined on K+-depolarized rat aorta rings. ( )-Cro-
makalim, ( )-pinacidil, diazoxide and BPDZ 73 were
used as reference PCOs (Table 1).
a : X = F
b : X = Cl
c : X = Br
Scheme 1. Synthesis of R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyrans. Reagents: (i) (CH3CO)2O,
H2SO4; (ii) AlCl3; (iii) acetone, pyrrolidine; (iv) NaBH4, CH3OH; (v)
CH3CN, H2SO4; (vi) HCl 37%; (vii) RNCS, CH2Cl2 (13–42).
( )-Cromakalim and ( )-pinacidil have previously been
shown to be quite inactive at inhibiting insulin release
whilst diazoxide and BPDZ 73 reduced the glucose-in-
duced insulin output (Table 1). BPDZ 73 was found to
be much more active than diazoxide (% of residual insu-
lin secretion at 10 lM = 73.9 4.4% for diazoxide and
4.9 0.4% for BPDZ 73).

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5707
Table 1. Residual insulin secretion and myorelaxant activity of original dimethylchroman thioureas compared to ( )-cromakalim, diazoxide, ( )-
pinacidil and BPDZ 73
R2
R1
R3
H
N
N
C
H
S
X
O
Compounds
X
R1
R2
R3
% Residual insulin secretiona (10 lM)
Myorelaxant activity EC50 (lM)b
13
14
F
H
H
H
H
H
H
H
H
H
H
H
63.2 4.4 (22)
46.6 2.3 (31)
32.8 2.0 (23)
82.9 3.5 (23)
76.8 5.4 (20)
65.1 3.1 (20)
52.0 2.3 (24)
34.3 2.3 (21)
36.6 3.1 (19)
73.8 3.2 (24)
50.4 2.2 (24)
53.1 2.7 (23)
69.8 4.2 (16)
69.0 3.0 (24)
57.0 2.5 (29)
53.1 3.6 (24)
43.0 3.2 (16)
37.2 2.3 (22)
60.2 3.0 (32)
32.8 1.9 (16)
52.2 3.1 (15)
81.3 2.2 (24)
68.2 3.2 (19)
57.5 3.3 (26)
18.0 1.3 (22)
16.2 1.6 (22)
23.0 2.4 (31)
10.5 1.1 (15)
8.9 0.7 (15)
12.2 1.2 (20)
94.4 4.1 (32)c
92.1 3.9 (13)d
73.9 4.4 (16)d
4.9 0.4 (32)d
7.6 0.6 (4)
11.1 2.2 (4)
51.5 19.4 (8)
>10 (4)
Cl
Br
F
H
H
15
16
H
H
OCH3
OCH3
OCH3
H
H
17
18
Cl
Br
F
H
5.5 1.2 (4)
>10 (6)
H
19
20
OCH3
OCH3
OCH3
H
6.2 0.6 (4)
2.9 0.5 (4)
>10 (5)
Cl
Br
F
H
21
22
H
H
OCH3
OCH3
OCH3
H
21.6 3.4 (4)
12.7 4.3 (5)
>10 (4)
23
24
Cl
Br
F
H
H
H
H
25
26
CH3
CH3
CH3
H
H
11.1 0.8 (4)
>30 (4)
>30 (4)
Cl
Br
F
H
H
27
28
H
H
CH3
CH3
CH3
H
H
7.3 0.8 (4)
137.3 10.3 (4)
>30 (4)
29
30
Cl
Br
F
H
H
H
H
31
32
H
CH3
CH3
CH3
H
>30 (4)
>30 (4)
>30 (4)
Cl
Br
F
H
H
33
34
H
Cl
H
H
24.4 1.3 (4)
3.2 0.6 (4)
6.5 0.4 (5)
>10 (4)
35
36
Cl
Br
F
Cl
Cl
H
H
H
Cl
H
37
38
H
H
Cl
Br
F
H
Cl
Cl
H
>10 (4)
>10 (5)
39
40
H
H
Cl
H
H
>10 (4)
>10 (5)
41
42
Cl
Br
—
—
—
—
H
H
Cl
Cl
H
—
H
—
>10 (4)
( )-Cromakalim
( )-Pinacidil
Diazoxide
BPDZ 73
—
—
0.13 0.01 (7)c
0.35 0.02 (11)d
22.4 2.1 (11)d
36.3 2.2 (6)d
—
—
—
—
—
—
—
—
a Percentage of residual insulin release from rat pancreatic islets incubated in the presence of 16.7 mM glucose (means SEM (n)).
b EC50: drug concentration giving 50% relaxation of the 30 mM KCl-induced contraction of rat aorta rings (means SEM (n)). n refers to the
number of samples.
c Published results: Ref. 22.
d Published results: Ref. 24.
The present biological data revealed that the new R/
S-3,4-dihydro-2,2-dimethyl-6-halo-4-(phenylaminothio-
carbonylamino)-2H-1-benzopyrans (13–42), tested at a
10 lM concentration, were more active on pancreatic
b-cells than the reference compounds ( )-cromakalim
and ( )-pinacidil. Moreover, most compounds were
even found to be more potent than diazoxide. A
few drugs provoked a 90% inhibition of glucose-in-
duced insulin release, which can be considered as a
close to maximal effect relative to the glucose-insensi-
tive basal insulin release (5–10% residual insulin
secretion).
Compounds 13, 14 and 15, devoid of substituent on the
C-4 phenyl ring, exhibited a pancreatic effect between
that of diazoxide and that of BPDZ 73. The nature of
the 6-substituent affected the activity and the rank order
of potency was 6-Br (15) > 6-Cl (14) > 6-F (13)
(p < 0.05).
The introduction of an electron-donating group on the
C-4 phenyl ring [methoxy (compounds 16–24) or methyl
(compounds 25–33)] did not improve the activity on
pancreatic b-cells. Most of these compounds were,
again, more potent than diazoxide and less potent than

5708
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
BPDZ 73 at inhibiting the glucose-induced insulin re-
lease. Biological results obtained with methoxy drugs
(16–24) indicated that the nature of the halogen atom
at the 6-position of the benzopyran nucleus affected
the pancreatic activity. Indeed, 6-Cl and 6-Br derivatives
appeared to be significantly more potent than their 6-F
homologues (compare the activity of 16 vs 18
(p < 0.05), 19 vs 20 and 21 (p < 0.05), 22 vs 23 and 24
(p < 0.05)). Furthermore, meta-methoxy and para-meth-
oxy drugs were shown to be more active than their
ortho-methoxy counterparts (compare the activity of
16 vs 19 (p < 0.05), 17 vs 20 and 23 (p < 0.05), 18 vs
21 and 24 (p < 0.05)). Such structure–activity relation-
ships were less obvious for methyl derivatives 25–33.
The unsubstituted compounds 13, 14 and 15 also in-
duced vasorelaxant effects although they were less po-
tent than ( )-cromakalim and ( )-pinacidil (p < 0.05).
Unfortunately, most methoxy-, methyl- and chlorophe-
nylaminothiocarbonylamino drugs (compounds 16–24,
25–33 and 34–42) precipitated in the medium before
reaching their maximal activity, making difficult to
determine their respective EC50 values. Nevertheless,
all drugs were found to be less active on the vascular tis-
sue than ( )-cromakalim and ( )-pinacidil. Some com-
pounds were even less potent than diazoxide and
BPDZ 73 (Table 1).
According to these in vitro data, R/S-3,4-dihydro-2,2-
dimethyl-6-halo-4-(3- or 4-chlorophenylaminothiocar-
bonylamino)-2H-1-benzopyrans (37–42) have particu-
larly drawn our attention. Indeed, these compounds
were potent inhibitors of the insulin-releasing process
(all drugs, tested at a 10 lM concentration, were
much more active than diazoxide) while displaying
a weak myorelaxant activity (all drugs were less
potent than ( )-cromakalim and ( )-pinacidil). Thus,
IC50 pancreatic value and IC50/EC50 ratio were eval-
uated in order to assess insulin-secreting cell potency
and tissue selectivity (pancreatic vs vascular tissue)
(Table 2).
By contrast, and compared to the unsubstituted deriva-
tives 13–15, the introduction of a weak electron-with-
drawing group (a chlorine atom) on the C-4 phenyl
ring (compounds 34–42) generally enhanced the activity
on the pancreatic tissue. Thus, except for ortho-chloro
compounds 34, 35 and 36, meta- (37–39) and para-
(40–42) chloro products provoked a notable inhibitory
effect on the glucose-induced insulin secretion
(p < 0.05). At a 10 lM concentration, these compounds
(37–42) induced an 80–90% inhibition of the insulin
secretory rate. The nature of the halogen atom at the
6-position of the benzopyran nucleus of these com-
pounds did not affect the activity on the pancreatic
tissue (p > 0.05, NS). It is noteworthy that the m-chloro-
and p-chloro-substituted phenylureas counterparts were
also the most potent at inhibiting insulin release.22
The IC50 values of the original dimethylchroman thio-
ureas 37, 38, 39, 40, 41 and 42, although being higher
than that of BPDZ 73, were much lower than those of
( )-cromakalim, ( )-pinacidil and even diazoxide (Ta-
ble 2). Moreover, and compared to their parent com-
pound ( )-cromakalim, the IC50/EC50 ratios indicated
that these new drugs exhibited a clear-cut inversion of
tissue selectivity (Table 2). Incidentally, and because
such new derivatives present a stereogenic centre at
the C-4 position of the benzopyran nucleus, the tissue
selectivity of the individual optical isomers remains to
be determined.
On the vascular model, and as previously reported,21,24,25
diazoxide and BPDZ 73 provoked a moderate myore-
laxant activity (EC50 = 22.4 2.1 lM and 36.3
2.2 lM, respectively), while ( )-cromakalim and ( )-
pinacidil exhibited marked vasorelaxant properties
(EC50 = 0.13 0.01 lM and 0.35 0.02 lM, respec-
tively) (Table 1).
Table 2. Effects of R/S-3,4-dihydro-2,2-dimethyl-6-halo-4-(3- or 4-chlorophenylaminothiocarbonylamino)-2H-1-benzopyrans, ( )-cromakalim,
diazoxide, ( )-pinacidil and BPDZ 73 on insulin secretion from rat pancreatic islets and on the KCl-induced contractions of rat aorta rings
Compounds
Rat pancreatic b-cells
% Residual insulin secretiona
1 lM
Rat aorta rings
IC50 (lM)b
EC50 (lM)c
IC50/EC50
d
10 lM
37
38
18.0 1.3 (22)
16.2 1.6 (22)
23.0 2.4 (31)
10.5 1.1 (15)
8.9 0.7 (15)
12.2 1.2 (20)
94.4 4.1 (32)e
92.1 3.9 (13)f
73.9 4.4 (16)f
4.9 0.4 (32)f
90.2 4.6 (38)
89.5 3.7 (23)
74.6 4.0 (21)
77.5 4.0 (15)
87.9 3.0 (16)
75.7 3.2 (24)
95.3 3.8 (31)e
97.7 6.7 (19)f
87.5 5.0 (15)f
36.2 2.4 (31)f
3.25
3.12
>10 (4)
>10 (4)
<0.32
<0.31
<0.25
<0.23
<0.27
<0.22
>770
>285
1.01
39
40
2.59
2.30
>10 (5)
>10 (4)
>10 (5)
41
42
2.75
2.26
>10 (4)
( )-Cromakalim
( )-Pinacidil
Diazoxide
BPDZ 73
>100e
>100f
22.6f
0.73f
0.13 0.01 (7)e
0.35 0.02 (11)f
22.4 2.1 (11)f
36.3 2.2 (6)f
0.02
a Percentage of residual insulin release from rat pancreatic islets incubated in presence of 16.7 mM glucose (means SEM (n)).
b IC50: drug concentration giving 50% inhibition of insulin release (estimated value).
c EC50: drug concentration giving 50% relaxation of the 30 mM KCl-induced contraction of rat aortic rings (means SEM (n)).
d Estimated selectivity ratio: pancreatic versus vascular tissue.
e Published results: Ref. 22.
f Published results: Ref. 24.

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5709
Among these new drugs, compounds 40 and 42 were
shown to be quite selective for the pancreatic tissue
and the most potent at inhibiting the glucose-induced
insulin secretion from incubated rat pancreatic islets.
Therefore, the next objective of the present study was
to determine, by using radioisotopic and fluorimetric ap-
proaches, whether the inhibitory effects of such com-
pounds were related to the activation of KATP channels.
lease from islets exposed throughout to insulinotropic
glucose concentrations (16.7 mM). In the presence of
extracellular Ca2+ in the perifusing medium, the addi-
tion of compound 40 (10 lM) provoked an immediate,
pronounced and reversible inhibition of both 45Ca
FOR (Fig. 4, upper panel, •) and insulin release
(Fig. 4, lower panel, •). Under such experimental condi-
tions, a decrease in 45Ca FOR is known to result from a
reduction in 40Ca2+ entry through voltage-sensitive Ca2+
channels 31. Thus, the inhibitory effect of 40 on 45Ca out-
flow can be viewed as the result of a reduction in 40Ca2+
entry with subsequent reduction in the insulin secretory
rate. Such a proposal is further substantiated by the lack
of effect of compound 40 (10 lM) on 45Ca outflow from
islets perifused throughout in the absence of extracellu-
lar Ca2+ (Fig. 4, upper panel, ꢁ).
Figure
3 illustrates the effects of compound 40
(10 lM) on 86Rb fractional outflow rate (FOR) from
prelabeled and perifused rat pancreatic islets exposed
throughout to 5.6 mM glucose and extracellular
Ca2+. In the absence of glibenclamide in the perifusing
medium (•), 40 provoked a rapid, sustained and rap-
idly reversible increase in the rate of 86Rb outflow.
The presence in the perifusate of the hypoglycemic
sulfonylurea glibenclamide (10 lM, ꢁ), a pharmacolog-
ical tool known to block the KATP channels,26,27 com-
pletely abolished the stimulatory effect of compound
40 on 86Rb outflow. Identical results were obtained
when testing, under the same experimental conditions,
the effects of compound 42 (data not shown). Such
data support the view that compounds 40 and 42 in-
creased the pancreatic b-cell membrane K+ permeabil-
ity through the activation of ATP-sensitive potassium
channels.20,28–30
40
2.0
1.0
Because the activation of KATP channels should hyper-
polarize the insulin-secreting cells and restrict Ca2+ in-
flow, we further examined the effects of compounds 40
(10 lM) and 42 (10 lM) on 45Ca FOR and insulin re-
40
0.0
2.0
3.0
2.0
1.0
1.0
0.0
30
40
50
60
70
80
90
30
40
50
60
70
80
90
TIME (min)
TIME (min)
Figure 3. Effect of 40 (10 lM) on 86Rb outflow from rat pancreatic
islets perifused throughout in the absence (•) or presence (ꢁ) of
glibenclamide (10 lM). Basal media contained 5.6 mmol/L glucose and
extracellular Ca2+. Mean values ( SEM) refer to six individual
experiments.
Figure 4. Effect of 40 (10 lM) on 45Ca outflow (upper panel) and
insulin release (lower panel) from rat pancreatic islets perifused
throughout in the presence of 16.7 mM glucose. Basal media contained
extracellular Ca2+ (•) or were deprived of Ca2+ and enriched with
EGTA (ꢁ). Mean values ( SEM) refer to four individual experiments.

5710
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
Experiments conducted with compound 42 gave essen-
tially the same results (data not shown).
responses to high K+ concentrations are known to be
sensitive to Ca2+ channel blockers but resistant to
PCOs.17,20,28,31
Calcium fluorimetry experiments were performed to
confirm the ability of compounds 40 and 42 to reduce
the pancreatic b-cell cytosolic Ca2+ concentration, as a
result of their capacity to activate the plasma membrane
KATP channels.
These radioisotopic and fluorimetric data suggest that
R/S-4-(4-chlorophenylaminothiocarbonylamino)-3,4-dihy-
dro-2,2-dimethyl-6-fluoro-2H-1-benzopyran 40 and R/
S-6-bromo-4-(4-chlorophenylaminothiocarbonylamino)-
3,4-dihydro-2,2-dimethyl-2H-1-benzopyran 42 activate
the pancreatic b-cell ATP-sensitive potassium channels.
Such an event, in turn, decreases Ca2+ inflow, reduces
the cytosolic Ca2+ concentration and, ultimately, inhib-
its the insulin secretory rate.
Figure 5 clearly shows that a rise in the extracellular
concentration of glucose from 2.8 to 20 mM (upper pa-
nel) or in the extracellular concentration of K+ from 5 to
50 mM (lower panel) provoked a marked increase in
[Ca2+]i. The subsequent addition of drug 40 (10 lM)
or 42 (10 lM, data not shown) dramatically reduced
the glucose-induced rise in cytosolic free Ca2+ concen-
tration (Fig. 5, upper panel), indicating that these drugs
were able to counteract the increase in [Ca2+]i provoked
by an insulinotropic glucose concentration. On the other
hand, compounds 40 (10 lM) and 42 (10 lM; data not
From a structural point of view, the present dimethylch-
roman thioureas, as well as their previously described
‘urea’ homologues,22 might be regarded as hybrid com-
pounds between cromakalim and pinacidil (Fig. 2). In-
deed, such dimethylchroman derivatives bear the
benzopyran nucleus of cromakalim onto which an
arylurea/arylthiourea moiety was introduced at the 4-
position, the latter moieties being recognized as isosteric
substitutes of arylcyanoguanidines (cfr pinacidil; Fig. 2).
In order to substantiate this assumption, conforma-
tional studies were conducted with a double objective.
shown) barely affected the KCl-induced rise in [Ca2+
]
i
(Fig. 5, lower panel), demonstrating that the main ac-
tion of the molecules was accounted for by their capac-
ity to raise K+ permeability. Indeed, physiological
Glucose 20 mM
The first objective was to investigate the flexibility of the
phenyl moiety located on the ‘pinacidil-like’ part of ur-
eas/thioureas and to compare it with the flexibility of the
pyridine ring of pinacidil. Such an analysis should indi-
cate if arylureas/thioureas are able to adopt an energet-
ically accessible conformation close to the preferred
conformation of pinacidil.
40
120
80
40
0
The second objective was to appreciate the fit of the ur-
eas/thioureas with cromakalim when their benzopyran
nucleuses were superimposed. The critical position taken
by the exocyclic carbonyl group of cromakalim was then
compared to the position adopted by the carbonyl/thio-
carbonyl group of the ureas/thioureas.
0
300
600
900
1200
1500
1800
Geometry optimization first revealed that the preferred
conformation adopted by arylureas, arylthioureas and
pinacidil confers the same parallel orientation to the
two N–H groups of the urea/thiourea/cyanoguanidine
function.
TIME (sec)
KCl 50 mM
40
120
80
40
0
A conformational scan was then performed by varying
the so-called T1 dihedral angle (1–2–3–4; see Fig. 6)
from 0ꢁ to 360ꢁ by steps of 15ꢁ (calculations from simpli-
fied structures in the R configuration).
The conformation, stabilized by a CHarom. . .O bond,
into which the urea and the phenyl groups were planar
(T1 = 0ꢁ) correspond to the most stable conformation
for urea derivatives (Fig. 6A). However, as the energetic
barriers to the internal rotation T1 were quite low
(around 6 kcal/mol), all conformations could be
observed.
0
300
600
900
1200
1500
1800
TIME (sec)
Figure 5. Effect of 10 lM 40 on glucose (20 mM; upper panel)- and
KCl (50 mM; lower panel)-induced increase in [Ca2+]i. Basal media
contained 2.8 mM glucose and extracellular Ca2+. Each graph is a
representative experiment conducted on a single rat pancreatic b-cell.
The conformational space of arylthiourea was quite
different (Fig. 6B). The most stable conformation
(T1 = 45ꢁ) did not favour neither CHarom. . .S bond

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5711
7
6
5
4
3
2
1
0
A
3
2
NH
1
4
O
NH
O
0
0
0
50
50
50
100
100
100
150
150
150
200
T1 (°)
250
300
350
20
18
16
14
12
10
8
B
3
2
NH
1
4
S
NH
6
O
4
2
0
200
T1 (°)
250
300
350
40
35
30
25
20
15
10
5
C
N
3
2
NH
1
4
N
N
NH
0
200
250
300
350
T1 (°)
Figure 6. Conformational scan around T1 angle of urea derivative (A), thiourea derivative (B) and pinacidil (C).
nor electronic delocalization between the phenyl and the
thiourea groups, though the energetic barrier to reach
planarity was very low. Conformations with T1’s rang-
ing between 150ꢁ and 200ꢁ were quite unstable
(DE > 10 kcal/mol).
ever, the CN moiety is not located in the vicinity of the
hydrogen atom at the 3-position of the pyridine ring.
Conformations with the CN moiety in close contact
with such a hydrogen atom are not energetically
favourable.
Finally, the conformational space of pinacidil was found
to be relatively restrained (Fig. 6C). Conformations with
T1 = 0ꢁ and 180ꢁ, where the pyridine and NH moiety
are coplanar, were the most stable. In such a case, how-
The conformations with T1 close to 0ꢁ, energetically
accessible for the three families of compounds (arylure-
as, arylthioureas and arylcyanoguanidines), are there-
fore expected to be the bioactive conformations.

5712
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
The conformation of the urea/thiourea dimethylchro-
mans was also compared with that of cromakalim,
by analysing the position and orientation of the car-
bonyl (or thiocarbonyl) group considered as being
potentially significant for biological activity. Another
conformational scan was performed by varying the
so-called dihedral T2 angle (5–6–7–8; see Fig. 7A)
The conformational space of thiourea, urea and ‘des-cy-
ano’-cromakalim was found to be identical (Fig. 7A, B
and C) with two energy minima corresponding to T2
values around 60ꢁ and 250ꢁ. Therefore, we can conclude
that the bioactive conformation of all these compounds
is quite similar.
(calculations from simplified structures in the
configuration).
R
Taken as a whole, the present conformational study re-
veals that the new urea and thiourea dimethylchromans
250
A
200
HN
150
8
HN
O
7
6
100
50
0
T2
5
O
0
50
100
150
200
T2 (°)
250
300
350
250
200
150
100
50
B
HN
HN
S
T2
O
0
0
50
100
150
200
T2 (°)
250
300
350
200
180
160
140
120
100
80
C
O
N
O
T2
OH
60
40
20
0
0
50
100
150
200
T2 (°)
250
300
350
Figure 7. Conformational scan around T2 angle of urea derivative (A), thiourea derivative (B) and ‘des-cyano’-cromakalim (C).

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5713
display structural similarities with cromakalim as well as
with pinacidil and may thus be considered as hybrid
compounds.
5. Experimental
5.1. Chemistry
Melting points were determined on a Buchi 530 capillary
¨
4. Conclusion
apparatus and were uncorrected. IR spectra were re-
corded as KBr pellets on a Perkin-Elmer 1000 FTIR
spectrophotometer. The 1H NMR spectra were recorded
on a Bruker Avance (500 MHz) instrument using
DMSO-d6 as solvent with TMS as an internal standard;
chemical shifts are reported in d values (ppm) relative to
internal TMS. The abbreviation s = singlet, d = doublet,
t = triplet, q = quadruplet, m = multiplet and b = broad
are used throughout. Elemental analyses (C, H, N, S)
were realized on a Carlo-Erba EA 1108-elemental ana-
lyser and were within 0.4% of the theoretical values.
All reactions were routinely checked by TLC on silica
In the search for new pancreatic selective PCOs, we
have synthesized and examined in vitro, on two dif-
ferent models (rat insulin-secreting cells and rat
aorta rings),
a
series of R/S-3,4-dihydro-2,2-di-
methyl-6-halo-4-(phenylaminothiocarbonylamino)-2H-1-
benzopyrans structurally related to ( )-cromakalim.
These 6-halo compounds (6-F, 6-Cl or 6-Br) were
devoid of a substituent, substituted by an electron-
donating group (a methoxy or a methyl moiety)
or by a weak electron-withdrawing group (a chlo-
rine atom) on the C-4 phenyl ring. Their biological
activity was compared to that of ( )-cromakalim,
( )-pinacidil, diazoxide and BPDZ 73, used as refer-
ence PCOs.
gel Merck 60 F254
.
5.1.1. Starting materials (12a–c). Starting materials for
the synthesis of the original R/S-3,4-dihydro-2,2-di-
methyl-6-halo-4-(phenylaminothiocarbonylamino)-2H-1-
benzopyrans (13–42) are R/S-4-amino-3,4-dihydro-2,
2-dimethyl-6-halo-2H-1-benzopyrans (12a–c). These
compounds were prepared from the appropriate para-
halogenophenols 6a–c according to previously described
synthetic procedures.21,23
The new drugs were found to be less active than
( )-pinacidil and ( )-cromakalim as vasorelaxant
agents.
On the other hand, data obtained on rat insulin-secret-
ing cells indicated that the new compounds, tested at a
10 lM concentration, were more active than ( )-pinaci-
dil or their parent molecule ( )-cromakalim at inhibiting
insulin release. Most drugs were even more effective than
diazoxide and compounds 40–42 exhibited an IC50 value
as low as 2–3 lM. Except for compounds 34–36, these
new dimethylchroman thioureas were more potent than
their ‘urea’ homologues at reducing the insulin secretory
rate.22
5.1.2. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyran (13). Phenyl
isothiocyanate (0.29 mL, 2.4 mmol) was added to a solu-
tion of 12a23 (0.4 g, 2 mmol) in methylene chloride
(5 mL). After 30 min, the solvent was removed under
vacuum and the crude product was triturated with ethyl
acetate. The insoluble was collected by filtration and
petroleum ether was added to the filtrate. The resulting
precipitate was collected by filtration, washed with
petroleum ether and dried (0.32 g, 47%): mp 164–
165 ꢁC; IR (KBr) t 3357, 3190 (N–H), 3036 (C–H aro-
The highest pancreatic activity was obtained with
derivatives bearing
a
weak electron-withdrawing
group on the C-4 phenyl ring, such as a chlorine
atom in the meta- or para-position, independently
of the nature of the halogen atom in the 6-position
of the benzopyran nucleus. Furthermore, the dim-
ethylchroman thioureas displayed an inverted tissue
selectivity (pancreatic tissue vs vascular smooth mus-
cle) compared to that observed with the reference
molecule ( )-cromakalim (IC50/EC50 6 0.32 and
>770, respectively).
matic), 2976, 2929 (C–H aliphatic), 1195 (C@S) cmꢀ1
;
1H NMR (DMSO-d6, 500 MHz) d 1.26 (s, 3H, CH3),
1.39 (s, 3H, CH3), 1.81 (m, 1H, HA of CH2), 2.19 (m,
1H, HB of CH2), 5.81 (m, 1H, CH), 6.76 (dd, 1H, 8-
H), 6.99 (m, 1H, 7-H), 7.05 (d, 1H, 5-H), 7.12 (t, 1H,
40-H), 7.33 (m, 2H, 30-H, 50-H), 7.46 (d, 2H, 20-H, 60-
H), 8.07 (d, 1H, NH(CH)), 9.61 (s, 1H, NH(C6H5)).
Anal. (C18H19FN2OS) theoretical: 65.43% C, 5.80% H,
8.48% N, 9.70% S; found: 65.09% C, 5.76% H, 8.56%
N, 9.55% S.
Combined radioisotopic and fluorimetric data indicated
that the drug-induced inhibition of insulin release was
mediated by the activation of pancreatic b-cell ATP-sen-
sitive K+ channels leading to a decrease in Ca2+ influx
and a subsequent reduction in cytosolic free Ca2+
concentration.
5.1.3. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyran (14). The ti-
tle compound was obtained as described for 13 starting
from 12b21 (0.4 g, 1.9 mmol) and phenyl isothiocyanate
(0.27 mL, 2.3 mmol) (0.36 g, 55%): mp 172.5–173 ꢁC;
IR (KBr) t 3368, 3192 (N–H), 2971 (C–H aliphatic),
1
Lastly, conformational studies revealed that the new
dimethylchroman thioureas and their previously re-
ported urea counterparts may be considered as ‘cro-
makalim’ as well as ‘pinacidil’ analogues. Such
hybrid compounds, however, exhibit a pharmacolog-
ical profile different from that of the two reference
PCOs.
1194 (C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d
1.27 (s, 3H, CH3), 1.40 (s, 3H, CH3), 1.82 (m, 1H, HA
of CH2), 2.19 (m, 1H, HB of CH2), 5.82 (m, 1H, CH),
6.78 (d, 1H, 8-H), 7.13 (t, 1H, 40-H), 7.18 (d, 1H, 7-
H), 7.27 (s, 1H, 5-H), 7.33 (m, 2H, 30-H, 50-H), 7.47
(d, 2H, 20-H, 60-H), 8.10 (bd, 1H, NH(CH)), 9.62 (s,
1H, NH(C6H5)). Anal. (C18H19ClN2OS) theoretical:

5714
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
62.33% C, 5.52% H, 8.08% N, 9.24% S; found: 62.57%
C, 5.61% H, 8.25% N, 9.05% S.
(s, 3H, CH3), 1.38 (s, 3H, CH3), 1.78 (m, 1H, HA of
CH2), 2.12 (m, 1H, HB of CH2), 3.84 (s, 3H, CH3O),
5.85 (m, 1H, CH), 6.71 (d, 1H, 8-H), 6.92 (m, 1H, 50-
H), 7.05 (d, 1H, 30-H), 7.17 (m, 1H, 40-H), 7.29 (d,
1H, 7-H), 7.36 (s, 1H, 5-H), 7.75 (d, 1H, 60-H), 8.11
(d, 1H, NH(CH)), 9.10 (s, 1H, NH(C6H4)). Anal.
(C19H21BrN2O2S) theoretical: 54.16% C, 5.02% H,
6.65% N, 7.61% S; found: 54.25% C, 5.32% H, 6.77%
N, 7.34% S.
5.1.4. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(pheny-
laminothiocarbonylamino)-2H-1-benzopyran (15). The ti-
tle compound was obtained as described for 13 starting
from 12c21 (0.4 g, 1.6 mmol) and phenyl isothiocyanate
(0.23 mL, 1.9 mmol) (0.40 g, 65%): mp 172–174 ꢁC; IR
(KBr) t 3363, 3189 (N–H), 3017 (C–H aromatic), 2974
1
(C–H aliphatic), 1194 (C@S) cmꢀ1; H NMR (DMSO-
d6, 500 MHz) d 1.27 (s, 3H, CH3), 1.39 (s, 3H, CH3),
1.81 (m, 1H, HA of CH2), 2.19 (m, 1H, HB of CH2),
5.82 (m, 1H, CH), 6.72 (d, 1H, 8-H), 7.13 (t, 1H, 40-
H), 7.29–7.35 (m, 3H, 7-H, 30-H, 50-H), 7.39 (s, 1H, 5-
H), 7.46 (d, 2H, 20-H, 60-H), 8.10 (d, 1H, NH(CH)),
9.62 (s, 1H, NH(C6H5)). Anal. (C18H19BrN2OS) theo-
retical: 55.25% C, 4.89% H, 7.16% N, 8.19% S; found:
55.07% C, 4.89% H, 7.24% N, 8.04% S.
5.1.8. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(3-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(19). The title compound was obtained as described for
13 starting from 12a23 (0.4 g, 2 mmol) and 3-methoxy-
phenyl isothiocyanate (0.34 mL, 2.4 mmol) (0.36 g,
48%): mp 147–148.5 ꢁC; IR (KBr) t 3181 (N–H), 3028
(C–H aromatic), 2978, 2926 (C–H aliphatic), 1260
1
(CH3–O–R), 1194 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.26 (s, 3H, CH3), 1.38 (s, 3H, CH3), 1.80
(m, 1H, HA of CH2), 2.19 (m, 1H, HB of CH2), 3.74
(s, 3H, CH3O), 5.81 (m, 1H, CH), 6.69 (d, 1H, 40-H),
6.76 (dd, 1H, 8-H), 6.95–7.01 (m, 2H, 7-H, 60-H), 7.06
(d, 1H, 5-H), 7.16 (s, 1H, 20-H), 7.22 (m, 1H, 50-H),
8.09 (d, 1H, NH(CH)), 9.63 (s, 1H, NH(C6H4)). Anal.
(C19H21FN2O2S) theoretical: 63.31% C, 5.87% H,
7.77% N, 8.90% S; found: 63.43% C, 6.17% H, 7.92%
N, 8.68% S.
5.1.5. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(2-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(16). The title compound was obtained as described for
13 starting from 12a23 (0.4 g, 2 mmol) and 2-methoxy-
phenyl isothiocyanate (0.32 mL, 2.4 mmol) (0.29 g,
40%): mp 132–134 ꢁC; IR (KBr) t 3246 (N–H), 3029
(C–H aromatic), 2978, 2924 (C–H aliphatic), 1254
1
(CH3–O–R), 1196 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.26 (s, 3H, CH3), 1.37 (s, 3H, CH3), 1.78
(m, 1H, HA of CH2), 2.12 (m, 1H, HB of CH2), 3.82
(s, 3H, CH3O), 5.83 (m, 1H, CH), 6.75 (dd, 1H, 8-H),
6.92 (m, 1H, 50-H), 6.98 (m, 1H, 7-H), 7.00–7.05 (m,
2H, 5-H, 30-H), 7.17 (m, 1H, 40-H), 7.77 (d, 1H, 60-H),
8.11 (d, 1H, NH(CH)), 9.08 (s, 1H, NH(C6H4)). Anal.
(C19H21FN2O2S) theoretical: 63.31% C, 5.87% H,
7.77% N, 8.90% S; found: 63.04% C, 6.25% H, 7.89%
N, 8.50% S.
5.1.9. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(3-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(20). The title compound was obtained as described for
13 starting from 12b21 (0.4 g, 1.9 mmol) and 3-methoxy-
phenyl isothiocyanate (0.32 mL, 2.3 mmol) (0.40 g,
56%): mp 152–154 ꢁC; IR (KBr) t 3308, 3167 (N–H),
2976, 2930 (C–H aliphatic), 1262 (CH3–O–R), 1201
1
(C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.27
(s, 3H, CH3), 1.39 (s, 3H, CH3), 1.81 (m, 1H, HA of
CH2), 2.19 (m, 1H, HB of CH2), 3.74 (s, 3H, CH3O),
5.82 (m, 1H, CH), 6.70 (d, 1H, 40-H), 6.77 (d, 1H, 8-
H), 6.96 (d, 1H, 60-H), 7.15–7.24 (m, 3H, 5-H, 7-H, 50-
H), 7.27 (s, 1H, 20-H), 8.11 (d, 1H, NH(CH)), 9.64 (s,
1H, NH(C6H4)). Anal. (C19H21ClN2O2S) theoretical:
60.55% C, 5.62% H, 7.43% N, 8.51% S; found: 60.77%
C, 6.00% H, 7.56% N, 8.27% S.
5.1.6. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(2-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(17). The title compound was obtained as described for
13 starting from 12b21 (0.4 g, 1.9 mmol) and 2-methoxy-
phenyl isothiocyanate (0.31 mL, 2.3 mmol) (0.39 g,
54%): mp 140–141 ꢁC; IR (KBr) t 3369, 3181 (N–H),
2976, 2951 (C–H aliphatic), 1260 (CH3–O–R), 1195
1
(C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.26
(s, 3H, CH3), 1.38 (s, 3H, CH3), 1.78 (m, 1H, HA of
CH2), 2.12 (m, 1H, HB of CH2), 3.83 (s, 3H, CH3O),
5.84 (m, 1H, CH), 6.76 (d, 1H, 8-H), 6.92 (m, 1H, 50-
H), 7.05 (d, 1H, 30-H), 7.16–7.19 (m, 2H, 7-H, 40-H),
7.24 (s, 1H, 5-H), 7.75 (d, J = 7.05 Hz, 1H, 60-H), 8.10
(d, 1H, NH(CH)), 9.10 (s, 1H, NH(C6H4)). Anal.
(C19H21ClN2O2S) theoretical: 60.55% C, 5.62% H,
7.43% N, 8.51% S; found: 60.59% C, 6.00% H, 7.58%
N, 8.16% S.
5.1.10. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(3-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(21). The title compound was obtained as described for
13 starting from 12c21 (0.4 g, 1.6 mmol) and 3-methoxy-
phenyl isothiocyanate (0.27 mL, 1.9 mmol) (0.40 g,
61%): mp 157–158 ꢁC; IR (KBr) t 3309, 3166 (N–H),
2974, 2930 (C–H aliphatic), 1262 (CH3–O–R), 1201
1
(C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.27
(s, 3H, CH3), 1.39 (s, 3H, CH3), 1.81 (m, 1H, HA of
CH2), 2.18 (m, 1H, HB of CH2), 3.74 (s, 3H, CH3O),
5.83 (m, 1H, CH), 6.69–6.73 (m, 2H, 8-H, 40-H), 6.95
(d, 1H, 60-H), 7.15 (s, 1H, 20-H), 7.22 (m, 1H, 50-H),
7.29 (d, 1H, 7-H), 7.39 (s, 1H, 5-H), 8.11 (d, 1H,
5.1.7. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(2-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(18). The title compound was obtained as described for
13 starting from 12c21 (0.4 g, 1.6 mmol) and 2-methoxy-
phenyl isothiocyanate (0.26 mL, 1.9 mmol) (0.41 g,
63%): mp 154–155 ꢁC; IR (KBr) t 3364, 3185 (N–H),
2976, 2950 (C–H aliphatic), 1258 (CH3–O–R), 1195
NH(CH)),
9.64
(s,
1H,
NH(C6H4)).
Anal.
(C19H21BrN2O2S) theoretical: 54.16% C, 5.02% H,
6.65% N, 7.61% S; found: 54.35% C, 5.33% H, 6.79%
N, 7.43% S.
1
(C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.26

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5715
5.1.11. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(4-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(22). The title compound was obtained as described for
13 starting from 12a23 (0.4 g, 2 mmol) and 4-methoxy-
phenyl isothiocyanate (0.32 mL, 2.4 mmol) (0.37 g,
51%): mp 158–159.5 ꢁC; IR (KBr) t 3182 (N–H), 3031
(C–H aromatic), 2972, 2930 (C–H aliphatic), 1253
6.95–7.01 (m, 2H, 5-H, 7-H), 7.15–7.25 (m, 4H, 30-H,
40-H, 50-H, 60-H), 7.81 (bd, 1H, NH(CH)), 9.22 (s, 1H,
NH(C6H4)). Anal. (C19H21FN2OS) theoretical: 66.25%
C, 6.14% H, 8.13% N, 9.31% S; found: 66.39% C,
6.36% H, 8.27% N, 9.00% S.
5.1.15. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(2-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (26).
The title compound was obtained as described for 13
starting from 12b21 (0.4 g, 1.9 mmol) and 2-methyl-
phenyl isothiocyanate (0.31 mL, 2.3 mmol) (0.35 g,
51%): mp 156.5–157.5 ꢁC; IR (KBr) t 3345, 3130 (N–
1
(CH3–O–R), 1196 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.25 (s, 3H, CH3), 1.37 (s, 3H, CH3), 1.81
(m, 1H, HA of CH2), 2.13 (m, 1H, HB of CH2), 3.74
(s, 3H, CH3O), 5.81 (m, 1H, CH), 6.75 (dd, 1H, 8-H),
6.90 (d, 2H, 30-H, 50-H), 6.97 (m, 1H, 7-H), 7.02 (d,
1H, 5-H), 7.28 (d, 2H, 20-H, 60-H), 7.85 (bd, 1H,
1
H), 2974 (C–H aliphatic), 1199 (C@S) cmꢀ1; H NMR
NH(CH)),
9.43
(s,
1H,
NH(C6H4)).
Anal.
(DMSO-d6, 500 MHz) d 1.25 (s, 3H, CH3), 1.37 (s,
3H, CH3), 1.84 (m, 1H, HA of CH2), 2.08 (m, 1H, HB
ofCH2), 2.23 (s, 3H, 20-CH3), 5.86 (m, 1H, CH), 6.75
(d, 1H, 8-H), 7.15–7.25 (m, 6H, 5-H, 7-H, 30-H, 40-H,
50-H, 60-H), 7.83 (bd, 1H, NH(CH)), 9.25 (s, 1H,
NH(C6H4)). Anal. (C19H21ClN2OS) theoretical:
63.23% C, 5.86% H, 7.76% N, 8.88% S; found: 63.42%
C, 6.12% H, 7.93% N, 8.65% S.
(C19H21FN2O2S) theoretical: 63.31% C, 5.87% H,
7.77% N, 8.90% S; found: 63.46% C, 6.26% H, 7.91%
N, 8.68% S.
5.1.12. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(4-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(23). The title compound was obtained as described for
13 starting from 12b21 (0.4 g, 1.9 mmol) and 4-methoxy-
phenyl isothiocyanate (0.31 mL, 2.3 mmol) (0.35 g,
48%): mp 158–160 ꢁC; IR (KBr) t 3192 (N–H), 3036
(C–H aromatic), 2974, 2929 (C–H aliphatic), 1249
5.1.16. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(2-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (27).
The title compound was obtained as described for 13
starting from 12c21 (0.4 g, 1.6 mmol) and 2-methyl-
phenyl isothiocyanate (0.26 mL, 1.9 mmol) (0.35 g,
55%): mp 162.5–163 ꢁC; IR (KBr) t 3347, 3168 (N–H),
1
(CH3–O–R), 1199 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.26 (s, 3H, CH3), 1.38 (s, 3H, CH3), 1.82
(m, 1H, HA of CH2), 2.14 (m, 1H, HB of CH2), 3.74
(s, 3H, CH3O), 5.82 (m, 1H, CH), 6.76 (d, 1H, 8-H),
6.90 (d, 2H, 30-H, 50-H), 7.16 (d, 1H, 7-H), 7.24 (s,
1H, 5-H), 7.28 (d, 2H, 20-H, 60-H), 7.87 (d, 1H,
2972 (C–H aliphatic), 1199 (C@S) cmꢀ1 1H NMR
;
(DMSO-d6, 500 MHz) d 1.25 (s, 3H, CH3), 1.37 (s,
3H, CH3), 1.84 (m, 1H, HA of CH2), 2.08 (m, 1H, HB
of CH2), 2.23 (s, 3H, 20-CH3), 5.86 (m, 1H, CH), 6.70
(d, 1H, 8-H), 7.16–7.28 (m, 5H, 7-H, 30-H, 40-H, 50-H,
60-H), 7.35 (s, 1H, 5-H), 7.82 (bd, 1H, NH(CH)), 9.25
(s, 1H, NH(C6H4)). Anal. (C19H21BrN2OS) Theoretical:
56.30% C, 5.22% H, 6.91% N, 7.91% S; found: 56.42%
C, 5.29% H, 7.06% N; 7.59% S.
NH(CH)),
9.44
(s,
1H,
NH(C6H4)).
Anal.
(C19H21ClN2O2S) theoretical: 60.55% C, 5.62% H,
7.43% N, 8.51% S; found: 60.64% C, 5.95% H, 7.58%
N, 8.44% S.
5.1.13. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(4-meth-
oxyphenylaminothiocarbonylamino)-2H-1-benzopyran
(24). The title compound was obtained as described for
13 starting from 12c21 (0.4 g, 1.6 mmol) and 4-methoxy-
phenyl isothiocyanate (0.26 mL, 1.9 mmol) (0.33 g,
51%): mp 153.5–155 ꢁC; IR (KBr) t 3239 (N–H), 3039
(C–H aromatic), 2975, 2924 (C–H aliphatic), 1251
5.1.17. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(3-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (28).
The title compound was obtained as described for 13
starting from 12a23 (0.4 g, 2 mmol) and 3-methylphenyl
isothiocyanate (0.32 mL, 2.4 mmol) (0.45 g, 64%): mp
137–139 ꢁC; IR (KBr) t 3376, 3171 (N–H), 3010 (C–H
aromatic), 2977, 2922 (C–H aliphatic), 1197 (C@S)
1
(CH3–O–R), 1195 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.26 (s, 3H, CH3), 1.38 (s, 3H, CH3), 1.82
(m, 1H, HA of CH2), 2.13 (m, 1H, HB of CH2), 3.74
(s, 3H, CH3O), 5.83 (m, 1H, CH), 6.71 (d, 1H, 8-H),
6.91 (d, 2H, 30-H, 50-H), 7.28 (m, 3H, 7-H, 20-H, 60-
H), 7.36 (s, 1H, 5-H), 7.88 (bd, 1H, NH(CH)), 9.44 (s,
1H, NH(C6H4)). Anal. (C19H21BrN2O2S) theoretical:
54.16% C, 5.02% H, 6.65% N, 7.61% S; found: 54.10%
C, 5.32% H, 6.76% N, 7.27% S.
1
cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.26 (s, 3H,
CH3), 1.38 (s, 3H, CH3), 1.80 (m, 1H, HA of CH2),
2.18 (m, 1H, HB of CH2), 2.29 (s, 3H, 30-CH3), 5.80
(m, 1H, CH), 6.76 (dd, 1H, 8-H), 6.94–7.06 (m, 3H, 5-
H, 7-H, 40-H), 7.19–7.26 (m, 3H, 20-H, 50-H, 60-H),
8.02 (d, 1H, NH(CH)), 9.55 (s, 1H, NH(C6H4)). Anal.
(C19H21FN2OS) theoretical: 66.25% C, 6.14% H,
8.13% N, 9.31% S; found: 66.55% C, 6.45% H, 8.28%
N, 8.99% S.
5.1.14. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(2-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (25).
The title compound was obtained as described for 13
starting from 12a23 (0.4 g, 2 mmol) and 2-methylphenyl
isothiocyanate (0.32 mL, 2.4 mmol) (0.40 g, 57%): mp
163.5–165 ꢁC; IR (KBr) t 3340, 3130 (N–H), 2972 (C–
5.1.18. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(3-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (29).
The title compound was obtained as described for 13
starting from 12b21 (0.4 g, 1.9 mmol) and 3-methyl-
phenyl isothiocyanate (0.31 mL, 2.3 mmol) (0.41 g,
60%): mp 151–152 ꢁC; IR (KBr) t 3367, 3164 (N–H),
1
H aliphatic), 1191 (C@S) cmꢀ1; H NMR (DMSO-d6,
500 MHz) d 1.25 (s, 3H, CH3), 1.37 (s, 3H, CH3), 1.83
(m, 1H, HA of CH2), 2.08 (m, 1H, HB of CH2), 2.22
(s, 3H, 20-CH3), 5.84 (m, 1H, CH), 6.74 (dd, 1H, 8-H),
2976 (C–H aliphatic), 1199 (C@S) cmꢀ1 1H NMR
;
(DMSO-d6, 500 MHz) d 1.27 (s, 3H, CH3), 1.39 (s,

5716
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
3H, CH3), 1.81 (m, 1H, HA of CH2), 2.19 (m, 1H, HB of
CH2), 2.29 (s, 3H, 30-CH3), 5.81 (m, 1H, CH), 6.77 (d,
1H, 8-H), 6.95 (d, 1H, 40-H), 7.17–7.26 (m, 5H, 5-H,
7-H, 20-H, 50-H, 60-H), 8.07 (bd, 1H, NH(CH)), 9.58
(s, 1H, NH(C6H4)). Anal. (C19H21ClN2OS) theoretical:
63.23% C, 5.86% H, 7.76% N, 8.88% S; found: 62.94%
C, 5.94% H, 7.83% N, 8.83% S.
starting from 12c21 (0.4 g, 1.6 mmol) and 4-methyl-
phenyl isothiocyanate (0.28 mL, 1.9 mmol) (0.29 g,
46%): mp 176–176.5 ꢁC; IR (KBr) t 3373, 3189 (N–H),
3033 (C–H aromatic), 2975, 2923 (C–H aliphatic),
1
1198 (C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d
1.27 (s, 3H, CH3), 1.39 (s, 3H, CH3), 1.81 (m, 1H, HA
of CH2), 2.16 (m, 1H, HB of CH2), 2.28 (s, 3H, 40-
CH3), 5.82 (m, 1H, CH), 6.72 (d, 1H, 8-H), 7.14 (d,
2H, 30-H, 50-H), 7.28–7.31 (m, 3H, 7-H, 20-H, 60-H),
7.37 (s, 1H, 5-H), 7.98 (bd, 1H, NH(CH)), 9.52 (s, 1H,
NH(C6H4)). Anal. (C19H21BrN2OS) theoretical:
56.30% C, 5.22% H, 6.91% N, 7.91% S; found: 56.02%
C, 5.43% H, 6.96% N, 7.70% S.
5.1.19. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(3-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (30).
The title compound was obtained as described for 13
starting from 12c21 (0.4 g, 1.6 mmol) and 3-methyl-
phenyl isothiocyanate (0.26 mL, 1.9 mmol) (0.29 g,
46%): mp 160–161 ꢁC; IR (KBr) t 3363, 3163 (N–H),
2976 (C–H aliphatic), 1199 (C@S) cmꢀ1
;
1H NMR
5.1.23. R/S-4-(2-Chlorophenylaminothiocarbonylamino)-
3,4-dihydro-2,2-dimethyl-6-fluoro-2H-1-benzopyran (34).
The title compound was obtained as described for 13
starting from 12a23 (0.4 g, 2 mmol) and 2-chlorophenyl
isothiocyanate (0.31 mL, 2.4 mmol) (0.43 g, 57%): mp
160–162 ꢁC; IR (KBr) t 3379, 3219 (N–H), 3025 (C–H
aromatic), 2978, 2928 (C–H aliphatic), 1196 (C@S)
(DMSO-d6, 500 MHz) d 1.27 (s, 3H, CH3), 1.39 (s,
3H, CH3), 1.80 (m, 1H, HA of CH2), 2.18 (m, 1H, HB
of CH2), 2.30 (s, 3H, 30-CH3), 5.82 (m, 1H, CH), 6.72
(d, 1H, 8-H), 6.95 (d, 1H, 40-H), 7.20–7.31 (m, 4H, 7-
H, 20-H, 50-H, 60-H), 7.38 (s, 1H, 5-H), 8.05 (bd, 1H,
NH(CH)),
9.56
(s,
1H,
NH(C6H4)).
Anal.
1
(C19H21BrN2OS) theoretical: 56.30% C, 5.22% H,
6.91% N, 7.91% S; found: 56.53% C, 5.30% H, 7.06%
N, 7.64% S.
cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.26 (s, 3H,
CH3), 1.38 (s, 3H, CH3), 1.80 (m, 1H, HA of CH2),
2.14 (m, 1H, HB of CH2), 5.80 (m, 1H, CH), 6.76 (dd,
1H, 8-H), 6.99 (m, 1H, 7-H), 7.05 (d, 1H, 5-H), 7.26
(m, 1H, 40-H), 7.34 (m, 1H, 50-H), 7.50 (d, 1H, 30-H),
7.64 (d, 1H, 60-H), 8.26 (d, 1H, NH(CH)), 9.31 (s, 1H,
NH(C6H4)). Anal. (C18H18ClFN2OS) theoretical:
59.25% C, 4.97% H, 7.68% N, 8.79% S; found: 59.04%
C, 5.04% H, 7.74% N, 8.61% S.
5.1.20. R/S-3,4-Dihydro-2,2-dimethyl-6-fluoro-4-(4-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (31).
The title compound was obtained as described for 13
starting from 12a23 (0.4 g, 2 mmol) and 4-methylphenyl
isothiocyanate (0.35 mL, 2.4 mmol). The product was
then crystallised in a mixture of ethyl acetate/petroleum
ether 40:60 (1:3) (0.40 g, 57%): mp 170–170.5 ꢁC; IR
(KBr) t 3181 (N–H), 3031 (C–H aromatic), 2977, 2926
5.1.24. R/S-6-Chloro-4-(2-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (35).
The title compound was obtained as described for 31
starting from 12b21 (0.4 g, 1.9 mmol) and 2-chloro-
phenyl isothiocyanate (0.30 mL, 2.3 mmol) (0.31 g,
42%): mp 158–159.5 ꢁC; IR (KBr) t 3360, 3164 (N–H),
1
(C–H aliphatic), 1195 (C@S) cmꢀ1; H NMR (DMSO-
d6, 500 MHz) d 1.26 (s, 3H, CH3), 1.38 (s, 3H, CH3),
1.80 (m, 1H, HA of CH2), 2.16 (m, 1H, HB of CH2),
2.27 (s, 3H, 40-CH3), 5.80 (m, 1H, CH), 6.75 (dd, 1H,
8-H), 6.96–7.05 (m, 2H, 5-H, 7-H), 7.13 (d, 2H, 30-H,
50-H), 7.30 (d, 2H, 20-H, 60-H), 7.96 (bd, 1H, NH(CH)),
9.51 (s, 1H, NH(C6H4)). Anal. (C19H21FN2OS) theoret-
ical: 66.25% C, 6.14% H, 8.13% N, 9.31% S; found:
65.95% C, 6.11% H, 8.17% N, 9.09% S.
2975 (C–H aliphatic), 1195 (C@S) cmꢀ1 1H NMR
;
(DMSO-d6, 500 MHz) d 1.27 (s, 3H, CH3), 1.39 (s,
3H, CH3), 1.82 (m, 1H, HA of CH2), 2.14 (m, 1H, HB
of CH2), 5.82 (m, 1H, CH), 6.77 (d, 1H, 8-H), 7.17 (d,
1H, 7-H), 7.26–7.36 (m, 3H, 5-H, 40-H, 50-H), 7.51 (d,
1H, 30-H), 7.62 (d, 1H, 60-H), 8.26 (bd, 1H, NH(CH)),
9.33 (s, 1H, NH(C6H4)). Anal. (C18H18Cl2N2OS) theo-
retical: 56.70% C, 4.76% H, 7.35% N, 8.41% S; found:
56.39% C, 4.77% H, 7.37% N, 8.29% S.
5.1.21. R/S-6-Chloro-3,4-dihydro-2,2-dimethyl-4-(4-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (32).
The title compound was obtained as described for 13
starting from 12b21 (0.4 g, 1.9 mmol) and 4-methyl-
phenyl isothiocyanate (0.34 mL, 2.3 mmol) (0.41 g,
60%): mp 172–173.5 ꢁC; IR (KBr) t 3376, 3191 (N–H),
3035 (C–H aromatic), 2975, 2924 (C–H aliphatic),
5.1.25. R/S-6-Bromo-4-(2-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (36).
The title compound was obtained as described for 13
starting from 12c21 (0.4 g, 1.6 mmol) and 2-chlorophenyl
isothiocyanate (0.24 mL, 1.9 mmol) (0.36 g, 54%): mp
162–163 ꢁC; IR (KBr) t 3368, 3164 (N–H), 2977, 2949,
1
1199 (C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d
1.27 (s, 3H, CH3), 1.39 (s, 3H, CH3), 1.81 (m, 1H, HA
of CH2), 2.17 (m, 1H, HB of CH2), 2.27 (s, 3H, 40-
CH3), 5.82 (m, 1H, CH), 6.77 (d, 1H, 8-H), 7.13–7.18
(m, 3H, 7-H, 30-H, 50-H), 7.25 (s, 1H, 5-H), 7.30 (d,
2H, 20-H, 60-H), 7.98 (bd, 1H, NH(CH)), 9.53 (s, 1H,
NH(C6H4)). Anal. (C19H21ClN2OS) theoretical:
63.23% C, 5.86% H, 7.76% N, 8.88% S; found: 62.91%
C, 6.07% H, 7.79% N, 8.78% S.
2926 (C–H aliphatic), 1196 (C@S) cmꢀ1 1H NMR
;
(DMSO-d6, 500 MHz) d 1.27 (s, 3H, CH3), 1.39 (s,
3H, CH3), 1.82 (m, 1H, HA of CH2), 2.13 (m, 1H, HB
of CH2), 5.83 (m, 1H, CH), 6.72 (d, 1H, 8-H), 7.26–
7.30 (m, 2H, 7-H, 40-H), 7.35 (m, 1H, 50-H), 7.42 (s,
1H, 5-H), 7.51 (d, 1H, 30-H), 7.60 (d, 1H, 60-H), 8.25
(bd, 1H, NH(CH)), 9.33 (s, 1H, NH(C6H4)). Anal.
(C18H18BrClN2OS) theoretical: 50.78% C, 4.26% H,
6.56% N, 7.53% S; found: 50.40% C, 4.24% H, 6.58%
N, 7.34% S.
5.1.22. R/S-6-Bromo-3,4-dihydro-2,2-dimethyl-4-(4-meth-
ylphenylaminothiocarbonylamino)-2H-1-benzopyran (33).
The title compound was obtained as described for 13

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5717
5.1.26. R/S-4-(3-Chlorophenylaminothiocarbonylamino)-
3,4-dihydro-2,2-dimethyl-6-fluoro-2H-1-benzopyran (37).
The title compound was obtained as described for 31
starting from 12a23 (0.4 g, 2 mmol) and 3-chlorophenyl
isothiocyanate (0.31 mL, 2.4 mmol) (0.36 g, 48%): mp
171–172 ꢁC; IR (KBr) t 3380, 3221 (N–H), 3043 (C–H
aromatic), 2977, 2931 (C–H aliphatic), 1196 (C@S)
(0.35 g, 46%): mp 193–194 ꢁC; IR (KBr) t 3237 (N–H),
3041 (C–H aromatic), 2978, 2929 (C–H aliphatic),
1
1197 (C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d
1.26 (s, 3H, CH3), 1.39 (s, 3H, CH3), 1.80 (m, 1H, HA
of CH2), 2.19 (m, 1H, HB of CH2), 5.78 (m, 1H, CH),
6.77 (dd, 1H, 8-H), 6.99 (m, 1H, 7-H), 7.05 (d, 1H, 5-
H), 7.37 (d, 2H, 30-H, 50-H), 7.51 (d, 2H, 20-H, 60-H),
8.18 (bs, 1H, NH(CH)), 9.68 (bs, 1H, NH(C6H4)). Anal.
(C18H18ClFN2OS) theoretical: 59.25% C, 4.97% H,
7.68% N, 8.79% S; found: 58.86% C, 5.00% H, 7.65%
N, 8.50% S.
1
cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.27 (s, 3H,
CH3), 1.39 (s, 3H, CH3), 1.80 (m, 1H, HA of CH2),
2.21 (m, 1H, HB of CH2), 5.78 (m, 1H, CH), 6.77 (dd,
1H, 8-H), 7.00 (m, 1H, 7-H), 7.05 (d, 1H, 5-H), 7.16
(d, 1H, 40-H), 7.33–7.37 (m, 2H, 50-H, 60-H), 7.73 (s,
1H, 20-H), 8.27 (bd, 1H, NH(CH)), 9.72 (s, 1H,
NH(C6H4)). Anal. (C18H18ClFN2OS) theoretical:
59.25% C, 4.97% H, 7.68% N, 8.79% S; found: 58.87%
C, 5.03% H, 7.69% N, 8.47% S.
5.1.30. R/S-6-Chloro-4-(4-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (41).
4-Chlorophenyl isothiocyanate (0.39 mL, 2.3 mmol)
was added to a solution of 12b21 (0.4 g, 1.9 mmol) in
methylene chloride (5 mL). After 30 min, the resulting
precipitate was collected by filtration, washed with
petroleum ether and dried. The crude product was tritu-
rated with ethyl acetate. The insoluble was collected by
filtration and petroleum ether was added to the filtrate.
The resulting precipitate was collected by filtration,
washed with petroleum ether and dried (0.33 g, 46%):
mp 186–187.5 ꢁC; IR (KBr) t 3240 (N–H), 3042 (C–H
aromatic), 2983, 2933 (C–H aliphatic), 1198 (C@S)
5.1.27. R/S-6-Chloro-4-(3-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (38).
The title compound was obtained as described for 13
starting from 12b21 (0.4 g, 1.9 mmol) and 3-chloro-
phenyl isothiocyanate (0.30 mL, 2.3 mmol) (0.41 g,
57%): mp 161.5–163 ꢁC; IR (KBr) t 3339, 3159 (N–H),
3006 (C–H aromatic), 2976, 2925 (C–H aliphatic),
1
1199 (C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d
1
1.28 (s, 3H, CH3), 1.40 (s, 3H, CH3), 1.80 (m, 1H, HA
of CH2), 2.22 (m, 1H, HB of CH2), 5.79 (m, 1H, CH),
6.79 (d, 1H, 8-H), 7.16–7.20 (m, 2H, 7-H, 40-H), 7.27
(s, 1H, 5-H), 7.33–7.38 (m, 2H, 50-H, 60-H), 7.73 (s,
1H, 20-H), 8.30 (bd, 1H, NH(CH)), 9.74 (s, 1H,
NH(C6H4)). Anal. (C18H18Cl2N2OS) theoretical:
56.70% C, 4.76% H, 7.35% N, 8.41% S; found: 56.68%
C, 4.73% H, 7.40% N, 8.25% S.
cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.27 (s, 3H
CH3), 1.40 (s, 3H, CH3), 1.80 (m, 1H, HA of CH2),
2.20 (m, 1H, HB of CH2), 5.80 (m, 1H, CH), 6.78 (d,
1H, 8-H), 7.18 (d, 1H, 7-H), 7.26 (s, 1H, 5-H), 7.37 (d,
2H, 30-H, 50-H), 7.51 (d, 2H, 20-H, 60-H), 8.19 (bd,
1H, NH(CH)), 9.68 (s, 1H, NH(C6H4)). Anal.
(C18H18Cl2N2OS) theoretical: 56.70% C, 4.76% H,
7.35% N, 8.41% S; found: 56.35% C, 4.78% H, 7.37%
N, 8.40% S.
5.1.28. R/S-6-Bromo-4-(3-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (39).
The title compound was obtained as described for 13
starting from 12c21 (0.4 g, 1.6 mmol) and 3-chlorophenyl
isothiocyanate (0.24 mL, 1.9 mmol) (0.45 g, 68%): mp
162–163 ꢁC; IR (KBr) t 3337, 3155 (N–H), 3001 (C–H
aromatic), 2975, 2924 (C–H aliphatic), 1199 (C@S)
5.1.31. R/S-6-Bromo-4-(4-chlorophenylaminothiocarbon-
ylamino)-3,4-dihydro-2,2-dimethyl-2H-1-benzopyran (42).
The title compound was obtained as described for 13
starting from 12c21 (0.4 g, 1.6 mmol) and 4-chlorophenyl
isothiocyanate (0.32 mL, 1.9 mmol) (0.40 g, 60%): mp
173.5–174.5 ꢁC; IR (KBr): t: 3239 (N–H), 3088, 3041
(C–H aromatic), 2984, 2933 (C–H aliphatic), 1197
1
cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.28 (s, 3H,
1
CH3), 1.40 (s, 3H, CH3), 1.80 (m, 1H, HA of CH2),
2.21 (m, 1H, HB of CH2), 5.80 (m, 1H, CH), 6.73 (d,
1H, 8-H), 7.16 (d, 1H, 40-H), 7.30–7.39 (m, 4H, 5-H,
7-H, 50-H, 60-H), 7.73 (s, 1H, 20-H), 8.30 (d, 1H,
(C@S) cmꢀ1; H NMR (DMSO-d6, 500 MHz) d 1.27
(s, 3H, CH3), 1.39 (s, 3H, CH3), 1.80 (m, 1H, HA of
CH2), 2.19 (m, 1H, HB of CH2), 5.80 (m, 1H, CH),
6.73 (d, 1H, 8-H), 7.30 (d, 1H, 7-H), 7.37–7.39 (m, 3H,
5-H, 30-H, 50-H), 7.51 (d, 2H, 20-H, 60-H), 8.19 (bd,
1H, NH(CH)), 9.68 (s, 1H, NH(C6H4)). Anal.
(C18H18BrClN2OS) theoretical: 50.78% C, 4.26% H,
6.58% N, 7.53% S; found: 50.87% C, 4.21% H, 6.63%
N, 7.44% S.
NH(CH)),
9.74
(s,
1H,
NH(C6H4)).
Anal.
(C18H18BrClN2OS) theoretical: 50.78% C, 4.26% H,
6.58% N, 7.53% S; found: 50.81% C, 4.34% H, 6.62%
N, 7.25% S.
5.1.29. R/S-4-(4-Chlorophenylaminothiocarbonylamino)-
3,4-dihydro-2,2-dimethyl-6-fluoro-2H 1-benzopyran (40).
4-Chlorophenyl isothiocyanate (0.40 mL, 2.4 mmol) was
added to a solution of 12a23 (0.4 g, 2 mmol) in methy-
lene chloride (5 mL). After 30 min, the resulting precip-
itate was collected by filtration, washed with petroleum
ether and dried. The crude product was triturated with
ethyl acetate. The insoluble was collected by filtration
and petroleum ether was added to the filtrate. The
resulting precipitate was collected by filtration, washed
with petroleum ether and dried. The product was recrys-
tallised in ethyl acetate/petroleum ether 40:60 (1:3)
5.2. Biological assays
( )-Cromakalim (Beecham Pharmaceutical, UK), diaz-
oxide (Sigma Chemical, USA), BPDZ 73 and ( )-pinac-
idil (Laboratory of Medicinal Chemistry, ULg,
Belgium) were tested as reference compounds.
5.2.1. Measurement of insulin release from incubated rat
pancreatic islets. Experiments were performed with pan-
creatic islets isolated from adult fed Wistar rats (Charles
River Laboratories, Belgium).

5718
S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
Groups of 10 islets, each derived from the same batch of
islets, were preincubated for 30 min at 37 ꢁC in 1 mL of
a physiological salt medium (in mM: NaCl 115, KCl 5,
CaCl2 2.56, MgCl2 1, NaHCO3 24) supplemented with
2.8 mM glucose, 0.5% (w/v) dialysed albumin (fraction
V, Sigma) and equilibrated against a mixture of O2
(95%) and CO2 (5%). The islets were then incubated at
37 ꢁC for 90 min in 1 mL of the same medium contain-
ing 16.7 mM glucose and, in addition, the reference
compound or the required chroman derivative.
iological medium containing 16.7 mM glucose and
either 86Rb (0.15–0.25 mmol/L:50 lCi/mL) or 45Ca
(0.02–0.04 mmol/L:100 lCi/mL). The validity of 86Rb
as a tracer for the study of K+ handling in the islets
has been assessed previously.33 After incubation, the is-
lets were washed three times with non-radioactive med-
ium and then placed in a perfusion chamber. The
perifusate was delivered at a constant rate (1.0 mL/
min). From the 31st to the 90th min, the effluent was
continuously collected over successive periods of 1 min
each. An aliquot of the effluent (0.6 mL) was used for
scintillation counting while the remainder was stored
at ꢀ20 ꢁC for insulin radioimmunoassay. At the end
of the perifusion, the radioactive content of the islets
was also determined. The outflow of 86Rb or 45Ca
(cpm/min) was expressed as a fractional outflow rate
(% of instantaneous islet content per min; FOR).
The release of insulin was measured radioimmunologi-
cally using rat insulin as a standard.18
Residual insulin secretion was expressed as a percentage
of the value recorded in control experiments (100%);
that is in the absence of drug and presence of
16.7 mM glucose.
Some media contained no CaCl2 and were enriched with
0.5 mM EGTA (Sigma).
5.2.2. Measurement of tension in rat aorta rings. Experi-
ments were performed with aortae removed from adult
fed Wistar rats (Charles River Laboratories, Belgium).
Results are expressed as means ( SEM) together with
the number of individual experiments (n).
A section of the thoracic aorta was cleared of adhering
fat and connective tissue and was cut into transverse
rings (3–4 mm long). The endothelium was removed
by rubbing the intimal surface with forceps. The seg-
ments were suspended under 1.5 g tension by means of
steel hooks in an organ bath containing 20 mL of a
physiological solution (in mM: NaCl 118, KCl 4.7,
CaCl2 2.5, NaHCO3 25, KH2PO4 1.2, MgSO4 1.2, glu-
cose 5). The physiological solution was maintained at
37 ꢁC and continuously bubbled with a mixture of O2
(95%) and CO2 (5%). Isometric contractions of the aor-
tic rings were measured with a force–displacement trans-
ducer. After 60 min of equilibration, the rings were
exposed to KCl (30 mM). When the tension had stabi-
lized, the drug was added to the bath at increasing con-
centrations until maximal relaxation (or until 300 lM).
5.2.4. Measurement of cytosolic Ca2+ concentration from
single rat pancreatic islet cells. Pancreatic islets were dis-
rupted in a Ca2+-deprived medium and then centri-
fuged through an albumin solution to remove debris
and dead cells. Cells were seeded onto glass coverslips
and maintained in tissue culture for 72 h before use.
The cells were then incubated with fura-2 AM
(2 lmol/L) (Molecular Probes) for 1 h. The medium
used to perifuse the cells contained (in mM) NaCl
115, KCl 5, CaCl2 2.56, MgCl2 1, NaHCO3 24, glucose
2.8 and was gassed with O2 (95%) and CO2 (5%).
When high concentrations (50 mM) of extracellular
K+ were used, the concentration of extracellular NaCl
was lowered to keep the osmolarity constant. Fura-2
fluorescence of single-loaded cells was measured by
use of dual-excitation microfluorimetry with a Carl
Zeiss ratio imaging system (Carl Zeiss Belgium). The
excitation and emission wavelengths were set at 340/
380 and 510 nm, respectively. Data are presented as ra-
tio signals (F340/F380). The experiments were repeated
on different cell populations.
The relaxation response was expressed as the percentage
of the contractile response to KCl. The EC50 values
(concentration evoking 50% inhibition of the plateau
phase induced by KCl) were assessed from dose–re-
sponse curves using Datanalyst software (EMKA Tech-
nologies, France).32
5.3. Statistical evaluation
5.2.3. Measurements of 86Rb, 45Ca outflow and insulin
release from perifused rat pancreatic islets. Experiments
were performed with pancreatic islets isolated from
adult fed Wistar rats (Charles River Laboratories,
Belgium).
The statistical significance of difference between mean
data was assessed by using Student’s t-test or by analysis
of variance followed for multiple comparisons by a Bon-
ferroni test procedure. The biological results were con-
sidered as statistically different when p < 0.05.
The media used for incubating, washing and perifusing
the islets consisted of a physiological salt medium (in
mM: NaCl 115, KCl 5, CaCl2 2.56, NaHCO3 24, MgCl2
1) supplemented with 0.5% (w/v) dialysed albumin (frac-
tion V, Sigma) and gassed with O2 (95%) and CO2 (5%).
5.4. Conformational studies
Quantum mechanical calculations using Density Func-
tional Theory at the PBE0/6-31G(d) level have been
used to investigate the intrinsic conformational prefer-
ences of urea and thiourea compounds in the gas phase
in comparison with pinacidil and cromakalim. All calcu-
lations have been performed with the Gaussian03
program.
The methods used to measure 86Rb outflow, 45Ca out-
flow and insulin release from perifused pancreatic islets
have been described previously.28,31Briefly, groups of
100 islets were incubated for 60 min at 37 ꢁC in the phys-

S. Sebille et al. / Bioorg. Med. Chem. 16 (2008) 5704–5719
5719
14. Alemzadeh, R.; Langley, G.; Upchurch, L.; Smith, P.;
Slonim, A. E. J. Clin. Endocrinol. Metab. 1998, 83, 1911.
15. Rasmussen, S. B.; Sorensen, T. S.; Hansen, J. B.;
Mandrup-Poulsen, T.; Hornum, L.; Markholst, H. Horm.
Metab. Res. 2000, 32, 294.
A conformational scan was first performed by varying the
so-called T1 dihedral angle (1–2–3–4; see Fig. 6) from 0 to
360ꢁ. The crystal structure of pinacidil34 and R/S-6-bro-
mo-4-(4-chlorophenylaminocarbonylamino)-3,4-dihy-
dro-2,2-dimethyl-2H-1-benzopyran was used as a starting
point. A QM-minimized structure was first calculated for
thiourea because the crystal structure was not available.
In order to simplify the calculations, only one part of mol-
ecules, in the R configuration, was used.
16. Cosgrove, K.; Antoine, M.-H.; Lee, A.; Barnes, P.; de
Tullio, P.; Clayton, P.; McCloy, R.; De Lonlay, P.;
´ ´ ´
Nihoul-Fekete, C.; Robert, J.; Saudubray, J.-M.; Rahier,
J.; Lindley, K.; Hussain, K.; Aynsley-Green, A.; Pirotte,
B.; Lebrun, P.; Dunne, M. J. Clin. Endocrinol. Metab.
2002, 87, 4860.
17. Hamilton, T. C.; Weir, S. W.; Weston, A. H. Br. J.
Pharmacol. 1986, 88, 103.
In the second set of calculations, the crystal structure of
cromakalim35 was used as a starting point but the cyano
group at the 6-position was removed in order to simplify
the calculations. The halo substituents on the aromatic
rings were also removed from urea and thiourea dimeth-
ylchromans. The three structures were in the R
configuration.
18. Lebrun, P.; Antoine, M.-H.; Devreux, V.; Hermann, M.;
Herchuelz, A. J. Pharmacol. Exp. Ther. 1990, 255, 948.
19. Petit, P.; Hillaire-Buys, D.; Mir, A.-K.; Loubatieres-
Mariani, M.-M. Fundam. Clin. Pharmacol. 1992, 6, 185.
20. Lebrun, P.; Arkhammar, P.; Antoine, M.-H.; Nguyen, Q.-
A.; Hansen, J. B.; Pirotte, B. Diabetologia 2000, 43, 723.
21. Sebille, S.; de Tullio, P.; Becker, B.; Antoine, M.-H.;
Boverie, S.; Pirotte, B.; Lebrun, P. J. Med. Chem. 2005, 48,
614.
Acknowledgements
22. Sebille, S.; Gall, D.; de Tullio, P.; Florence, X.; Lebrun,
P.; Pirotte, B. J. Med. Chem. 2006, 49, 4690.
23. Khelili, S.; Nguyen, Q.-A.; Lebrun, P.; Delarge, J.; Pirotte,
B. Pharm. Pharmacol. Commun. 1999, 5, 189.
24. de Tullio, P.; Becker, B.; Boverie, S.; Dabrowski, M.;
Wahl, P.; Antoine, M.-H.; Somers, F.; Sebille, S.;
Ouedraogo, R.; Hansen, J. B.; Lebrun, P.; Pirotte, B.
J. Med. Chem. 2003, 46, 3342.
This study was supported by grants from the National
Fund for Scientific Research (F.N.R.S., Belgium) from
which P. Lebrun is Research Director and P. de Tullio is
Research Associate. The technical assistance of
Y. Abrassart, S. Counerotte, F. Leleux, A.-M. Vanbel-
linghen and A. Van Praet is also gratefully acknowledged.
25. de Tullio, P.; Pirotte, B.; Lebrun, P.; Fontaine, J.; Dupont,
L.; Antoine, M.-H.; Ouedraogo, R.; Khelili, S.; Maggetto,
C.; Masereel, B.; Diouf, O.; Podona, T.; Delarge, J.
J. Med. Chem. 1996, 39, 937.
References and notes
26. Ashcroft, F. M.; Rorsman, P. Prog. Biophys. Mol. Biol.
1989, 54, 87.
1. Seino, S. Annu. Rev. Physiol. 1999, 61, 337.
2. Noma, A. Nature 1983, 305, 147.
27. Malaisse, W. J.; Lebrun, P. Diabetes Care 1990, 13, 9.
28. Lebrun, P.; Devreux, V.; Hermann, M.; Herchuelz, A.
J. Pharmacol. Exp. Ther. 1989, 250, 1011.
3. Cook, D. L.; Hales, C. N. Nature 1984, 311, 271.
4. Spruce, A. E.; Standen, N. B.; Stanfield, P. R. Nature
1985, 316, 736–738.
29. Lebrun, P.; Antoine, M.-H.; Ouedraogo, R.; Kane, C.;
Dunne, M.; Hermann, M.; Herchuelz, A.; Masereel, B.;
Delarge, J.; de Tullio, P.; Pirotte, B. J. Pharmacol. Exp.
Ther. 1996, 277, 156.
5. Standen, N. B.; Quayle, J. M.; Davies, N. W.; Brayden, J.
E.; Huang, Y.; Nelson, M. T. Science 1989, 245, 177.
6. Bernardi, H.; Fosset, M.; Lazdunski, M. Proc. Natl. Acad.
Sci. U.S.A. 1988, 85, 9816.
30. Lebrun, P.; Antoine, M.-H.; Herchuelz, A. Minirev. Life
Sci. 1992, 51, 795.
7. Pirotte, B.; Fontaine, J.; Lebrun, P. Curr. Med. Chem.
1995, 2, 573.
31. Lebrun, P.; Malaisse, W. J.; Herchuelz, A. Am. J. Physiol.
1982, 242, E59.
8. Babenko, A. P.; Aguilar-Bryan, J. A.; Bryan, J. Annu. Rev.
Physiol. 1998, 60, 667.
32. Becker, B.; Antoine, M.-H.; Nguyen, Q.-A.; Rigo, B.;
Cosgrove, K. E.; Barnes, P. D.; Dunne, M. J.; Pirotte, B.;
Lebrun, P. Br. J. Pharmacol. 2001, 134, 375.
9. Inagaki, N.; Gonoi, T.; Clement, J. P., 4th; Namba, N.;
Inazawa, J.; Gonzalez, G.; Aguilar-Bryan, L.; Seino, S.;
Bryan, J. Science 1995, 270, 1166.
33. Malaisse, W. J.; Boschero, A. C.; Kawazu, S.; Hutton, J.
C. Pflu¨gers Arch. 1978, 373, 237.
10. Hambrock, A.; Lo¨fer-Walz, C.; Delabar, U.; Horio, Y.;
Kurachi, Y.; Quast, U. Mol. Pharmacol. 1999, 55, 832.
11. Lawson, K. Pharmacol. Ther. 1996, 70, 39.
12. Robertson, D. W.; Steinberg, M. I. J. Med. Chem. 1990,
33, 1529.
34. Pirotte, B.; Dupont, L.; de Tullio, P.; Masereel, B.;
Schynts, M.; Delarge, J. Helv. Chim. Acta 1993, 76, 1311.
35. Cassidy, F.; Evans, J. M.; Smith, D. M.; Stemp, G.; Edge,
C.; Williams, D. J. J. Chem. Soc. Chem. Commun. 1989, 6,
377.
13. Bjo¨rk, E.; Berne, C.; Ka¨mpe, O.; Wibell, P.; Oskarsson, P.;
Karlsson, F. A. Diabetes 1996, 45, 1427.