Synthesis and vasorelaxant activity of new 1,4-benzoxazine derivatives potassium channel openers

Article abstract of DOI:10.1016/S0968-0896(02)00091-3

As part of a search for new potassium channel openers, the synthesis and vasorelaxant activity of new 1,4-benzoxazine derivatives derived from transformation of the benzopyran skeleton of cromakalim were described. Several new 1,4-benzoxazine derivatives were provided with significant vasorelaxant activity with an overall pharmacological behavior similar to CRK (1f, 1i, 2d, 2e, 2f and 2i).

Full text of DOI:10.1016/S0968-0896(02)00091-3

Bioorganic & Medicinal Chemistry 10 (2002) 2663–2669
Synthesis and Vasorelaxant Activity of New 1,4-benzoxazine
Derivatives Potassium Channel Openers
Giuseppe Caliendo,a,* Elisa Perissutti,a Vincenzo Santagada,a Ferdinando Fiorino,a
Beatrice Severino,a Roberta d’Emmanuele di Villa Bianca,b Laura Lippolis,c
Aldo Pintoc and Raffaella Sorrentinob
a
`
Dipartimento di Chimica Farmaceutica e Tossicologica, Universita di Napoli ‘Federico II’ Via D. Montesano 49, 80131 Naples, Italy
b
`
Dipartimento di Farmacologia Sperimentale- Universita di Napoli ‘Federico II’ Via D. Montesano 49, 80131 Naples, Italy
c
`
Dipartimento di Scienze Farmaceutiche- Universita di Salerno via Ponte Don Melillo, Fisciano (SA), Italy
Received 22 January 2002; accepted 8 March 2002
Abstract—As part of a search for new potassium channel openers, the synthesis and vasorelaxant activity of new 1,4-benzoxazine
derivatives derived from transformation of the benzopyran skeleton of cromakalim were described. Several new 1,4-benzoxazine
derivatives were provided with significant vasorelaxant activity with an overall pharmacological behavior similar to CRK (1f, 1i, 2d,
2e, 2f and 2i). # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
modifying cromakalim, the synthesis and the biological
activity of a new series of 2-[3,4-dihydro-3-oxo-2H-1,4-
benzoxazin-4-yl]-ethylacetate derivatives, which exhib-
ited a good channel opening activity, was previously
reported.13
Potassium channel openers are recently discovered as a
novel class of compounds.1,2 Their pharmacological
action involves the relaxation of smooth muscle by the
opening of potassium channels, suggesting a number of
therapeutic targets for this class of compounds such as
hypertension, angina pectoris, coronary heart disease,
asthma, urinary incontinence and alopecia.3ꢀ6 Addi-
tionally, these agents may afford cells a measure of
protection against ischemia, independent of their vaso-
dilating actions7 and have antilipidemic effects, lowering
low-density lipoprotein (LDL) cholesterol and trigly-
cerides while increasing high density lipoprotein (HDL)
cholesterol.8
On the basis of these results we decide to evaluate the
effect on their activity by introducing different alkyl
chain between benzoxazine nucleus and the ester group.
These modifications provided the new 1,4-benzoxazine-
ethyl propionate (1) and ethylbutyrate derivatives (2)
(Table 1). The substituents inserted at the 6-position of
the 1,4-benzoxazine ring are those previously reported
displayng a good discrimination between electronic and
hydrophobic properties. In this way we could more
easily compare the effects of the different alkyl chain on
biological activity.
There are several prototypes of this class of compounds
with a diverse range of chemical structure such as croma-
kalim,9 pinacidil,10 nicorandil,11 and aprikalim12 that act
as openers of ATP-dependent potassium channel (KATP).
Among these compounds, many structural modification
studies have focused on cromakalim, a benzopyran
derivative, because it possesses the most potent activity.
As part of a program to develop new compounds by
The pharmacological study was performed to investi-
gate firstly a vasorelaxant activity of new compounds
and then to elucidate a putative mechanism of action
using pharmacological tools. For this purpose, we per-
formed in vitro assays using rat thoracic aorta rings pre-
contracted with phenylephrine (PE). Since several out-
ward potassium channels have been identified such as
ATP-dependent potassium channels (KATP), calcium-
dependent potassium channel (KCa), voltage-dependent
potassium channels (KV), we tested the active analogues
*Corresponding author. Tel.: +39-081-678648; fax: +39-081-678649;
e-mail: caliendo@unina.it
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00091-3

2664
G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
in presence of selective channel blockers. We used Glib-
enclamide (GLY) for the KATP, tetraethylammonium
(TEA) for the KCa, and 4-aminopyridine (4-AMP) as
unselective potassium channel blockers. Compounds
that showed a vasorelaxation higher than 50% but with
a different behavior to CRK were tested to investigated
on probable calcium blocking activity.
derivatives 1e, 1m, 2e and 2m respectively, with stoi-
chiometric amounts of 2-aminoethanethiol hydrochlor-
ide in absolute ethanol and triethylamine solution. All
products were isolated by column chromatography,
further purified by crystallization from appropriate sol-
1
vents and characterized by H NMR spectroscopy and
GLC-MS. Structures, physicochemical data and % of
relaxation activity of all synthesized compounds was
summarized in Table 1.
Chemistry
The synthesis strategy for the preparation of the new
1,4-benzoxazine derivatives 1 and 2 was outlined in
Scheme 1. The starting compounds 2-amino-4-sub-
stituted-phenols (3) for the synthesis of general
compounds (5) were commercially available, only
the 2-amino-4-cyanophenol was obtained by a proce-
dure previously reported.13 Acylation of 2-amino-4-X-
phenol (3) with bromoacetyl bromide or 2-bromoisobu-
tyryl bromide in CHCl3 in the presence of sodium
bicarbonate gave the corresponding amides 4. Cycli-
zation in the presence of potassium carbonate in DMF
afforded benzoxazine derivatives 5.
Results and Discussion
The vasorelaxant activities of compounds (series 1 and 2)
were evaluated in vitro using rat aorta rings pre-con-
tracted with PE (1 mM) in a concentration range 1–100
mM. Results obtained with compounds at concentration
of 100 mM, expressed as percentage of relaxation, are
summarized in Table 1. Data indicate that several of the
new ester derivatives are provided with significant
vasorelaxant activity profile.
Figure 1 shows the concentration–response curves of
compounds, which exhibited a vasorelaxation activity
higher than 50%. Vasorelaxations induced by the active
compounds are concentration-dependent. The con-
centration–response curve of analyzed compounds was
not superimposable to those obtained with cromakalim.
Indeed, all the active compounds show an EC50 (mM)
higher than cromakalim (Table 2). Cromakalim (0.1–
100 mM), employed as reference drug, produced a
relaxation of 83.1ꢁ2% (n=20) when concentration of
These latters were converted into the desired com-
pounds 1a–e, 1g–m and 2a–e, 2g–m by condensation
with ethyl-3-bromopropionate (series 1) or ethyl-4-
bromobutyrate (series 2) in sodium hydride/DMF at
room temperature.
The 6-(Á2-thiazolin-2-yl) derivatives 1f, 1n, 2f and 2n
were obtained by reaction of the corresponding 6-cyano
Scheme 1. Reagents: (a) bromoacethyl bromide or 2-bromoisobutyryl bromide, NaHCO3, CHCl3, room temperature; (b) anhydrous K2CO3, DMF,
80 ꢂC; (c) ethyl-3-bromopropionate or ethyl 4-bromobutyrate, NaH, DMF, room temperature; (d) 2-aminoethanethiol hydrochloride, triethylamine/
absolute EtOH, reflux.

G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
2665
Table 1. Physicochemical properties and vasorelaxant activity of 1,4-benzoxazine derivatives (1a–n) and (2a–n).
Compd
X
n
R=R0
Formulaa
Mr
Yield (%) Mp (ꢂC) Recrystallization
solventsb
MS (m/z)
% of relaxation
(10ꢀ4 M)
1a
1b
1c
1d
1e
H
CH3
Cl
NO2
CN
2
2
2
2
2
H
H
H
H
H
C13H15NO4
C14H17NO4
249.26
263.29
C13H14ClNO4 283.71
45
47
50
65
38
Oil
—
249; 220; 205; 177
263; 218; 188; 91
283; 183; 174
294; 193; 165; 107
274; 159; 145
12.3
14.3
42.7
43.4
25.9
42–44
87–88
103–104
85–87
a+b
a+d
a+b
b+c
C13H14N2O6
C14H14N2O4
294.26
274.26
1f
2
H
C16H18N2O4S 334.37
63
96–97
a+b
334; 289; 261; 174
87.1
1g
1h
1i
1l
1m
H
CH3
Cl
NO2
CN
2
2
2
2
2
CH3
CH3
CH3
CH3
CH3
C15H19NO4
C16H21NO4
C15H18ClNO4 311.76
277.27
291.34
73
45
71
40
41
Oil
Oil
Oil
66–67
63–64
—
—
—
a+b
a+c
277; 234; 162; 134
291; 248; 230; 176
311; 211; 196; 101
322; 220; 187
29.7
28.3
63.2
20.5
43.0
C15H18N2O6
C16H18N2O4
322.31
302.35
302; 187; 173
1n
2
CH3
C18H22N2O4S 362.44
35
92–93
a
362; 307; 279; 192
69.0
2a
2b
2c
2d
2e
H
CH3
Cl
NO2
CN
3
3
3
3
3
H
H
H
H
H
C14H17NO4
C15H19NO4
263.29
277.30
C14H16ClNO4 297.72
43
54
55
59
68
41–43
39–40
66–67
94–96
92–94
b
b
a
a
a
263; 218; 190; 134
277; 204; 148; 29
297; 197; 177; 88
308; 207; 179; 121
288; 201; 187; 145
48.0
42.0
69.5
71.1
62.8
C14H16N2O6
C15H16N2O4
308.27
288.30
2f
3
H
C17H20N2O4S 348.24
60
108–109
a
348; 261; 174; 115
73.7
2g
2h
2i
2l
2m
H
CH3
Cl
NO2
CN
3
3
3
3
3
CH3
CH3
CH3
CH3
CH3
C16H21NO4
C17H23NO4
C16H20ClNO4 325.77
291.33
305.37
53
30
55
48
80
69–71
Oil
79–81
88–89
89–90
a+b
—
e
a+b
a+b
291; 230; 177; 162
305; 232; 176; 87
325; 225; 205; 110
336; 236; 208; 140
316; 229; 180; 124
77.3
81.8
70.7
56.9
80.0
C16H20N2O6
C17H20N2O4
336.32
316.35
2n
3
CH3
C19H24N2O4S 376.48
58
60–61
a
376; 289; 247; 202
83.9
aAll compounds were analyzed for C, H, Cl, N and S and the analytical results were within ꢁ0.4% of the calculated values for the formulae shown.
bCrystallization solvents: (a) diethyl ether; (b) n-hexane; (c) ethyl alcohol; (d) dichloromethane; (e) petroleum ether.
100 mM was reached, while the EC50 calculated was 0.4
mM and 0.25–0.65 as 95% confidence limits, hence the
considered compounds are less active than cromakalim.
derivatives.13 Note compounds 2g and 2h with a methyl or
no substituent also exhibited high value of vasorelaxant
activity (81.8 and 77.3%, respectively). Moreover in the
serie 1 the two methyl groups at the 2-position of the
1,4-benzoxazine ring did not seem greatly influenced on
the vasorelaxant activity. However, in the series 2 the
2,2-dimethyl derivatives (2g–2n) revealed to be more
active than their corresponding demethyl derivatives
(2a–2f) except for 2l that showed a low vasorelaxant
activity than corresponding demethyl derivative 2d.
The 1,4-benzoxazine ethylbutyrate derivatives (series 2)
resulted generally more actives than corresponding
ethylpropionate derivatives (series 1). In fact only two
compounds of series 2 produced vasorelaxation less
than 50% while nine compounds of series 1 showed a
relaxation less than 50%. These results shows that the
length of the alkylchain plays an important role in the
vasorelaxant activity and allowed us to conclude that
the propylene chain was the better to furnish analogues
with a good activity.
To evaluate the activity of these compounds as putative
openers of potassium channels were performed experi-
ments using Glibenclamide (GLY; 10 mM) and tethra-
ethylammonium (TEA, 1 mM) as inhibitors of ATP-
and calcium- potassium channels respectively and 4-
aminopyridine (4-AMP, 10 mM) as unselective inhibitor.
Experiments with the inhibitors above stated, were per-
formed only for compounds that produced vasorelaxation
greater than 50%. In Table 3 are summarized all the
results obtained for series 1 and 2.
Subsequently, we attempted the modification of the
nature of the substituents at the 6-position on the 1,4-
benzoxazine ring. Analysis of the structures endowed
with highest vasorelaxant activity suggest that a thiazoline
ring is favorable for the activity (1f, 1n, 2f, 2n). This
finding is consistent with data relative to other analogue

2666
G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
channels because TEAdid not block significantly the
relaxation observed. We can exclude an activity on vol-
tage potassium channels since 4-AMP did not inhibit
the relaxation induced by these compounds. Conversely,
4-AMP and TEA enhanced significantly the relaxation
of compound 1n.
As stated above, series 2 compounds were all more
active if compared to series 1 except for the compound
2f, which exhibited a low vasorelaxant activity respect
to its analogue 1f. GLY and 4-AMP significantly
inhibited the relaxation induced by compounds 2d, 2e,
2f and 2i. TEAat a concentration of 1 mM, that as been
shown to block only calcium activated potassium chan-
nel, inhibited significantly the relaxation mediated by
compound 2h. In contrary TEAenhanced significantly
the relaxation induced by compounds 2c, 2d and 2l.
These results suggest that structural modification at
2- and 6-positions of the 1,4-benzoxazine nucleus lead
to relevant changes in the mechanism of action. In fact,
the presence of some electrons with-drawing substituents
such as nitro and cyano groups or thyazoline ring asso-
ciated to the absence of the geminal methyl groups at the
2-position lead to compounds (1f, 2f, 2d and 2e) which,
although less active than CRK (EC50=0.4mM), displayed
overall pharmacological behavior similar to that of the
reference drug, since their vasorelaxant activity was
inhibited by GLY and 4-AMP but not by TEA. It may
be hypothesized that these compounds interact mainly
with the KATP channels.
Figure 1. Concentration–response curve of compounds 1f, 1i, 1n, 2c–
2n (1–100 mM) on rat aorta rings pre-contracted with phelylephrine.
Data are expressed as meanꢁSEM of 6–12 at aorta rings.
Results of the corresponding 2,2-dimethyl analogues (1n,
2n, 2l, and 2m) suggest that these compounds may act by
the involvement of a different mechanism of action.
Table 2. Calculated EC50 (mM) of active compounds and croma-
kalim on phenylephrine-induced contraction in rat aorta rings
The influence of the geminal methyl groups was inverse
with compounds 1i and 2i characterized by a presence
of a chlorine atom at the 6-position which displayed a
behavior similar to CRK. Results of the corresponding
demethyl derivative 2c showed that the blockers of
potassium channels GLY and 4-AMP did not inhibit
the vasorelaxation whereas the effect observed in pre-
sence of TEAwas increased, indicating a possible
involvement of another mechanism of action. Also the
unsubstituted or methyl substituted at the 6-position led
to compounds 2g and 2h with a different behavior to
CRK; compound 2h seems to act inhibiting the calcium-
dependent potassium channels.
Compound
EC50 (95% confidence limits)
1f
1i
1n
2c
2d
2e
35 (21–59)
82 (41–162)
77 (69–86)
71 (45–113)
48 (31–73)
63 (23–140)
32 (10–93)
59 (31–110)
53 (30–97)
27 (14–51)
72 (41–126)
20 (11–37)
19 (11–33)
0.4 (0.25–0.65)
2f
2g
2h
2i
2l
2m
2n
CRKa
These results suggest that some structural modification
operated have caused a shift in the biological activity.
Indeed, blockers of potassium channels did not inhibit
the vasorelaxation and in some cases increased the
vasorelaxation activity, indicating a possible involve-
ment of another mechanism of action. In this regard we
have investigated on a possible involvement of calcium
activated channels. We choose some compounds that
showed a vasorelaxant activity higher than 50% (1n, 2c,
2g, 2h, 2l, 2m and 2n), that were not inhibited by
potassium channels blockers, to investigate on a prob-
able ability to block calcium voltage-dependent chan-
nels. The calcium blocking activity was investigated as
Results are expressed as mean (95% confidence limits) of several
experiments (3–16).
aCromakalim.
In our experimental model cromakalim, induced relax-
ation of aorta rings was significantly inhibited by GLY
and 4-AMP (p<0.001) but not by TEAat concentration
used (Table 3).
The results obtained testing the series 1 showed that
only the compounds 1f and 1i had displayed a behaviour
similar to CRK. All the other compounds did not
shown an activation of calcium-dependent potassium

G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
2667
Table 3. Relaxation % of compounds (100 mM), with or without potassium channels inhibitors on rat aorta rings pre-contracted by PE 1 mM
Compd
Control (vehicle)
Glibenclamide (10 mM)
Tetraethylammonium (1 mM)
4-Aminopyridine (10 mM)
1f
1i
1n
2c
2d
2e
87.1ꢁ3.2 (6)
63.2ꢁ6.1 (4)
69ꢁ1.8 (4)
59.7ꢁ10.5* (5)
48.7ꢁ1.5* (4)
84.8ꢁ9.3 (3)
59.9ꢁ4.6 (3)
52.5ꢁ6.6**(4)
25ꢁ18.7* (3)
52.7ꢁ3.4* (4)
62.4ꢁ8.7 (8)
79.7ꢁ3.8 (5)
54.6ꢁ4* (4)
61.9ꢁ14.8 (3)
55.9ꢁ17 (3)
90.1ꢁ5.7**(3)
100ꢁ0y (3)
34.7ꢁ6.8{ (5)
35.6ꢁ8.3* (3)
95.7ꢁ2.3{ (3)
58.4ꢁ5.2 (3)
39.2ꢁ7y (5)
69.5ꢁ5.6 (13)
71.1ꢁ4 (11)
62.8ꢁ8 (12)
73.7ꢁ8.8 (6)
77.3ꢁ9.5 (5)
81.8ꢁ7.2 (5)
70.7ꢁ5.8 (13)
56.9ꢁ4 (8)
87.9ꢁ6.6* (4)
88.7ꢁ6.6 (3)
66.2ꢁ19.4 (4)
76.5ꢁ9.0 (3)
58.2ꢁ2.1* (3)
84.6ꢁ5.2 (4)
97.4ꢁ2.6{ (3)
89.2ꢁ7.3 (3)
92.5ꢁ6.8 (4)
73.0ꢁ10.2 (3)
18.1ꢁ5.2**(3)
31.7ꢁ15* (4)
73.9ꢁ4.1 (3)
60.5ꢁ8.7* (4)
50.2ꢁ3.9* (4)
49.4ꢁ4.5 (4)
67.2ꢁ4.5 (7)
70.3ꢁ7.9 (4)
43.5ꢁ1.9D (3)
2f
2g
2h
2i
2l
67.7ꢁ1.0 (3)
65.9ꢁ8.3 (5)
66.6ꢁ5.9 (5)
38.5ꢁ7.9D (14)
2m
2n
CRKa
80.0ꢁ4.6 (13)
83.9ꢁ6.2 (13)
83.1ꢁ2 (20)
Results are expressed as meanꢁSEM and (n). Statistical analysis was performed comparing the % of relaxation (100 mM) in presence of inhibitor
versus control value.
p<0.05 was considered significant. *=p<0.05; **p<0.01; y=p<0.005; {=p<0.0001.
Figure 2. Inhibition effect of compounds 1n, 2c, 2g, 2h, 2l, 2m, 2n (30 mM) of KCl (60 mM) induced contraction. Data are expressed as meanꢁSEM
of 3–4 rat aorta rings.
ability to inhibit the KCl (60 mM) induced contraction.
Compounds 1n, 2c, 2g, 2h, 2l, 2m and 2n, at concentra-
tion of 30 mM significantly reduced KCl induced con-
traction of rat aorta rings (Fig. 2). Vehicle (DMSO
1:1000) did not modify KCl induced contraction; the %
of inhibition value observed was 0.6ꢁ0.4 (n=7).
act as template for deriving new compounds with a dif-
ferent profile.
The complexity of the pharmacological data, mostly
resulting from the involvement of different receptors
each to different extent, impedes the development of
reliable SARs for the present series of compounds.
In conclusion, our data indicates that the new ester
derivatives are provided with significant vasorelaxant
activity profile. Compounds 1f, 1i, 2d, 2e, 2f and 2i have
a similar activity to that observed for cromakalim.
Subtile structural modifications at the 2- and 6-positions
of the 1,4-benzoxazine nucleus lead to relevant changes
in the mechanism of action. In fact, pharmacological
data led us to hypothesize that structural changes of
some compounds have brought the synthesis of drugs
that can act as calcium voltage potassium channels
without exclude the possibility to act on other calcium
channels. Moreover the increase in vasorelaxation
observed for some compounds in presence of potassium
channels inhibitors, such as TEAor 4-AMP, could be
justify by the fact that probably these compounds have
still preserved some ability to interact with potassium
channels without explicate the pharmacological activity.
Hence, in presence of potassium channels inhibitors the
concentration reached is higher to act as calcium chan-
nels blockers than in absence of these inhibitors. Our
results implying that these structural modification may
Experimental
Chemistry
General. All melting points were determined on a capil-
lary melting points apparatus and are uncorrected.
Nuclear magnetic resonance (NMR) spectra were
recorded on a Bruker WM 500 spectrometer using tet-
ramethylsilane as an internal standard. The following
abbreviations are used: brs=broad single, d=doublet,
t=triplet, m=multiplet, s=singlet. Combined GLC–
MS analyses were performed on Hewlett-Packard and
5890 Gas Chromatograph with a mass selective detector
MDS HP 5970. Acolumn 25 m Â0.20 mm HP-5 (cross-
linked PhMe silicone 5%) with a 0.33 mm film tickness
was employed. Silica gel F254 (Merck) plates were used
for thin-layer chromatography (TLC). Column chro-
matography was performed using Carlo Erba silica-gel
(0.05–0.20 mm). Elemental analyses were carried out on

2668
G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
a Carlo Erba Model 1106. Elemental analyses (C, H, Cl,
N, S) and the results were withinꢁ0.4% of the theoretical
values. Anhydrous Na2SO4 was used as drying agent for
organic extraction. All solvent evaporation was per-
formed under vacuum. Reagent grade materials were
purchased from Aldrich Chemical Co. and were used
without further purification. The following experimental
methods represent general procedures for the synthesis
of each of the compounds presented in the text.
from appropriate solvents gave final products 1a–e and
1g–m (yields ranging 38–73%). Spectral data of title
compound 1b: 1H NMR (CDCl3) d 6.95 (d, 1H8, Ar–H,
J=8.2 Hz), 6.80 (d, 1H7, Ar–H, J=8.4 Hz), 6.52 (s,
1H5, Ar–H), 4.61 (s, 2H, O-CH2C¼O), 4.28 (t, 2H,
CH2N J=7.3 Hz), 4.15 (m, 2H, OCH2), 2.70 (t, 2H,
CH2C¼O) 2.30 (s, 3H, CH3) and 1.25 ppm (t, 3H, CH3,
1
J=7.3 Hz). Similar H NMR data occur in all deriva-
tives of general formula 1.
2-Amino-4-cyanophenol. To a stirred solution of tin (II)
chloride (0.433 mol) in 37% hydrogen chloride (60 mL)
was added 4-hydroxy-3-nitrobenzonitrile. After the addi-
tion, the resulting mixture was allowed to stir at room
temperature and carefully monitored by TLC. After 3 h,
the reaction mixture was cooled to 5 ꢂC, alkalinized to
pH=8 with 30% NaOH and than with NaHCO3 solution
until to white precipitate was formed and finally extrac-
ted with ethyl acetate. The organic layer was separated
and dried over anhydrous Na2SO4. The solvents were
evaporated and the crude residue was purified using a
silica gel column chromatography and acetone–NH4OH
(9.5:0.5 v/v) as eluent. Recrystallization from ethanol
give the required compound as brown pale solid (yield
General procedure for the preparation of 4-[3,4-dihydro-
3-oxo-2H-1,4-benzoxazin-4-yl]-ethyl butyrate (2a–e and
2g–m). Compounds 2a–e and 2g–m were prepared from
5 (0.11 mol) with NaH (60% in oil dispersion, 0.16 mol)
and ethyl-4-bromobutyrate (0.17 mol) in DMF (70 mL)
by the same procedure used for the preparation of 1a–e
and 1g–m from 5. The products were crystallized from
appropriate solvents. Spectral data of title compound 2b:
1H NMR (CDCl3) d 7.10 (d, 1H8, A r–H,J=8.2 Hz), 6.90
(d, 1H7, A r–H,J=8.4 Hz), 6.86 (s, 1H5, A r–H), 4.91 (s,
2H, O–CH2C¼O), 4.55 (t, 2H, CH2N J=7.3 Hz), 4.15
(m, 2H, OCH2), 4.01(m, 2H, CH2), 2.44 (t, 2H,
CH2C¼O), 2.36 (s, 3H, CH3) and 1.26 ppm (t, 3H, CH3,
1
J=7.3 Hz). Similar H NMR data occur in all deriva-
84%), mp 158–159 ꢂC. IR (KBr): 1295, 1517, 2224 cmꢀ1
.
tives of general formula 2.
1H NMR (DMSO-d): d 10.35 (brs, 2H, NH2), 7.01 (s, 1H,
Ar–H), 6.95 (d, 1H, Ar–H, J=8.8 Hz) and 6.87 ppm (d,
1H, Ar–H, J=8.1 Hz).
General procedure for the preparation of 3-[3,4-dihydro-3-
oxo-6-(D2-thiazolin-2-yl)-2H-1,4-benzoxazin-4-yl]ethyl
propionate (1f, 1n). Amixture of appropriate 2-(3,4-
dihydro-3-oxo-6-ciano-1,4-benzoxazin-4-yl)-ethyl pro-
pionate 1e or 1m (0.1 mol) and 2-aminoethanethiol
hydrochloride (0.1 mol) in absolute ethanol and tri-
ethylamine (0.1 mol) solution was heated to reflux for 3
h and monitored by TLC. After cooling the ethanol was
removed under reduced pressure and the residue was
purified by silica gel column chromatography (ethyl
acetate/n-hexane 6:4 v/v, as eluent). Fractions contain-
ing the product were combined, dried in vacuo, and
recrystallized from appropriate solvents to give, respec-
tively, analytically pure products 1f (yield 63%) and 1n
(yield 35%) as white crystals.
General procedure for the preparation of 3,4-dihydro-3-
oxo-2H-1,4-benzoxazines (5). Bromoacetyl bromide or
2-bromoisobutyryl bromide (0.15 mol) was added
dropwise to an ice-bath cooled solution of 2-amino-
phenol 3 (0.1 mol) and 350 mL of saturated solution of
sodium carbonate in 600 mL of CHCl3. The reaction
mixture was stirred at room temperature for 3 h and
monitored by TLC (diethylether/n-hexane 1:1 v/v, as
eluent). The layers were separated and the organic phase
was (washed with H2O) dried over anhydrous Na2SO4
and concentrated in vacuo to provide crude product 4
as brown oil, which was used without further purifica-
tion. The solution of crude product 4 and anhydrous
K2CO3 (0.1 mol) in 250 mL of DMF was heated at
80 ꢂC and stirred for 3 h (TLC diethylether/n-hexane 1:1
v/v, as eluent). After cooling the reaction mixture was
poured into H2O (250 mL) and extracted several times
with CHCl3. The combined organic extracts were dried
over anhydrous Na2SO4 and evaporated in vacuo.
Recrystallization from appropriate solvents gave com-
pound 5 as white solid (yields ranging from 62 to 80%).
1H NMR data are in according with the proposed
structures.
Spectral data of title compound 1f: 1H NMR (CDCl3) d
7.60 (s, 1H5, Ar–H), 7.50 (d, 1H8, Ar–H, J=8.2 Hz),
7.00 (d, 1H7, Ar–H, J=8.4 Hz), 4.70 (s, 2H, O–
CH2C¼O), 4.55 (s, 2H, CH2N), 4.30 (t, 2H, CH2N¼,
J=8.2 Hz), 4.10 (m, 2H, OCH2, J=7.3 Hz), 3.43 (t, 2H,
CH2S, J=8.2 Hz), 2.70 (t, CH2C¼O) and 1.25 ppm (t,
1
3H, CH3, J=7.3 Hz). Similar H NMR data occur in
compound 1n.
General procedure for the preparation of 4-[3,4-dihydro-3-
oxo-6-(D 2-thiazolin-2-yl)-2H-1,4-benzoxazin-4-yl]ethyl
butyrate (2f and 2n). Compounds 2f and 2n were pre-
pared from appropriate 2-(3,4-dihydro-3-oxo-6-ciano-
1,4-benzoxazin-4-yl)ethyl butyrate 2e or 2m (0.1 mol)
with 2-aminoethanethiol hydrochloride (0.1 mol) in
absolute ethanol and triethylamine (0.1 mol) according
to the procedure used for the preparation of 1f and 1n.
The products were crystallized from appropriate solvents
General procedure for the preparation of 3-[3,4-dihydro-
3-oxo-2H-1,4-benzoxazin-4-yl]-ethyl propionate (1a–e
and 1g–m). To a solution of compound 5 (0.11 mol) in
70 mL of DMF cooled with an ice bath was added NaH
(60% in oil dispersion 0.16 mol) in portions and after 10
min ethyl-3-bromopropionate (0.17 mol) was added.
The reaction was stirred at room temperature for 3 h
and then poured into cold water (250 mL) and extracted
with CHCl3. The organic phase was washed several
times with H2O, dried and evaporated. Recrystallization
1
as reported in Table 1. 2f: H NMR (CDCl3) d 7.70 (s,
1H5, Ar–H), 7.55 (d, 1H8, A r–H,J=8.2 Hz), 7.10 (d, 1H7,
Ar–H, J=8.4 Hz), 4.81 (s, 2H, O–CH2C¼O), 4.60 (s,

G. Caliendo et al. / Bioorg. Med. Chem. 10 (2002) 2663–2669
2669
2H, CH2N), 4.35 (t, 2H, CH2N, J=8.2 Hz), 4.21 (m,
2H, CH2), 4.15 (m, 2H, OCH2, J=7.3 Hz), 3.52 (t, 2H,
CH2S, J=8.2 Hz), 2.44 (t, CH2C¼O) and 1.26 ppm (t,
experiment was performed in presence of vehicle (DMSO
0.1%) as control. Inhibition of calcium channels activity
was expressed as% of inhibition of first KCl induced
contraction.
1
3H, CH3, J=7.3 Hz). Similar H NMR data occur in
compound 2n.
Acknowledgements
Pharmacology
This work was supported by a grant from Murst, Rome.
The 1H NMR and MS spectra were performed at Centro
di Ricerca Interdipartimentale di Analisi Strumentale,
Universita di Napoli ‘Federico II’. The assistance of the
staff is gratefully appreciated.
Vasorelaxant activity in vitro assay. Male Wistar rats
(200–250 g; Nossan, Italy) were killed by exsanguin-
ation after exposition to CO2 and the thoracic aorta was
removed, cleaned of adherent connective tissue, and cut
into rings ꢃ3 mm in length. The endothelium was
removed by gently rubbing the intimal surface with
moistened filter paper. Endothelium-denuded rings were
mounted under 0.5 g of tension in 2.5 mL organ baths
containing Krebs salt solution at the following compo-
sition (in mM): NaCl, 118.4; KCl, 4.7; MgSO4, 1.2;
CaCl2, 1.3; KH2PO4, 1.2; NaHCO3, 25.0; and glucose
11.7. The solution was maintained at 37 ꢂC and bubbled
with 95% O2–5% CO2 (pH 7.4). Developed tension was
measured using an isometric force transducer (7003
transducer, Ugo Basile, Comerio, Italy) connected to a
recorder (Graphtec Linearcorder, WR 3310). Rings
were allowed to equilibrate for 60 min and the Krebs
solution was replaced each 15 min. After equilibration
drugs were tested as vasorelaxant on phenylephrine-
induced contraction (1 mM). Drugs were dissolved in
DMSO and added to the organ bath in cumulative
manner in the range 1–25 mL (maximum final con-
centration of DMSO 1%). Drugs that showed a vasor-
elaxation in concentration-dependent manner with a
relaxation over than 50% of contraction, at the highest
used concentration (100 mM), were considered for the
test in presence of inhibitors. Each tissue was used only
for one concentration–response curve of tested com-
pound in presence or absence of inhibitors such as GLY
(0.1 mM), or TEA(1 mM) or 4-AMP (10 mM).
References and Notes
1. Empfield, J. R.; Russel, K. Annu. Rep. Med. Chem. 1996,
30, 81.
2. Atwal, K. S. Curr. Med. Chem. 1996, 3, 227.
3. Garcia, M. L.; Hanner, M.; Knaus, H. G.; Koch, R.;
Schmalhofer, W.; Slaughter, R. S.; Kaczorowski, G. Adv.
Pharmacol. 1997, 39, 425.
4. Lawson, K. Pharmacol. Ther. 1996, 70, 39.
5. Yokoshiki, H.; Sunagawa, M.; Seki, T.; Sperelakis, N. Am.
J. Physiol. 1998 C25.
6. Newgreen, D. T. Emerg. Drugs 1998, 3, 209.
7. Escande, D.; Cavero, I. Trends Pharmacol. Sci. 1992, 13,
269.
8. Hamilton, T. C.; Beerahee, A.; Moen, J. S.; Price, P. K.;
Ramij, J. V.; Clapham, J. C. Cardiovasc. Drug Rev. 1993, 11,
199.
9. Ashowood, V. A.; Buckingham, R. E.; Cassidy, F.; Evans,
J. M.; Faruk, E. A.; Hamilton, T. C.; Nash, D. J.; Stemp, G.;
Willcocks, K. J. Med. Chem. 1986, 29, 2194.
10. Peterson, H. J.; Nielsen, C. K.; Martelli, E. A. J. Med.
Chem. 1978, 21, 773.
11. Sakai, K. Jpn. J. Pharmacol. 1980, 30, 881.
12. Brown, T. J.; Chapman, R. F.; Cook, D. C.; Hart, T. W.;
Mclay, I. M.; Jordan, R.; Masdon, J. S.; Palfreyman, M. N.;
Walsh, R. J. A.; Withnall, M. T.; Aloup, J. C.; Cavero, I.;
Farge, D.; James, C.; Mondot, S. J. Med. Chem. 1992, 35,
3613.
13. Caliendo, G.; Grieco, P.; Perissutti, E.; Santagada, V.;
Fattorusso, C.; Albrizio, S.; Sorrentino, R.; Pinto, A. Eur. J.
Med. Chem. 1998, 33, 957.
To investigate on the ability of some compounds to
interfere with calcium channels, we performed experi-
ments using KCl (60 mM)14 as contracting agent. After
tissue stabilization, KCl was added to the organ bath
(100% contraction) and than washed-out. Compounds,
at concentration of 30 mM, were added to the organ bath;
15 min after, KCl 60 mM was added again. Analogous
14. Kajino, M.; Shibuto, Y.; Nishikawa, K.; Meguro, K.
Chem. Pharm. Bull. 1991, 39, 2896.