Synthesis and voltage-clamp studies of methyl 1,4-dihydro-2,6-dimethyl- 5-nitro-4-(benzofurazanyl)pyridine-3-carboxylate racemates and enantiomers and of their benzofuroxanyl analogues

Article abstract of DOI:10.1021/jm980623b

Racemic methyl 1,4-dihydro-2,6-dimethyl-5-nitro-4- (benzofurazanyl)pyridine-3-carboxylates (±)10 and (±)-11 and their benzofuroxanyl analogues (±)-12 and (±)-13 were prepared using a modified Hantzsch reaction that involved the condensation of nitroaceton

Full text of DOI:10.1021/jm980623b

1422
J . Med. Chem. 1999, 42, 1422-1427
Syn th esis a n d Volta ge-Cla m p Stu d ies of Meth yl
1,4-Dih yd r o-2,6-d im eth yl-5-n itr o-4-(ben zofu r a za n yl)p yr id in e-3-ca r boxyla te
Ra cem a tes a n d En a n tiom er s a n d of Th eir Ben zofu r oxa n yl An a logu es
Sonja Visentin,† Pascale Amiel,† Roberta Fruttero,† Donatella Boschi,† Christian Roussel,§ Laura Giusta,‡
Emilio Carbone,‡ and Alberto Gasco*,†
Department of Scienza e Tecnologia del Farmaco, Facolta` di Farmacia, Universita` degli Studi di Torino, via P. Giuria 9,
10125 Torino, Italy, Department of Neuroscienze, Unita` di ricerca INFM, Universita` degli Studi di Torino, Corso Raffaello 30,
10125 Torino, Italy, and ENSSPICAM, University Aix Marseille III, Av. Escadrille Normandie Niemen, F-13397 Marseille
Cedex 20, France
Received November 4, 1998
Racemic methyl 1,4-dihydro-2,6-dimethyl-5-nitro-4-(benzofurazanyl)pyridine-3-carboxylates (()-
10 and (()-11 and their benzofuroxanyl analogues (()-12 and (()-13 were prepared using a
modified Hantzsch reaction that involved the condensation of nitroacetone with methyl
3-aminocrotonate and the appropriate aldehydes. The racemic mixtures were resolved into
the corresponding enantiomers. Whole-cell voltage-clamp studies on L-type Ca2+ channels
expressed in a rat insulinoma cell line (RINm5F) showed that all the dextrorotatory antipodes
were effective agonists of L-type Ca2+ currents, while the levorotatory ones were weak Ca2+
entry blockers. The (+)-enantiomer of benzofurazan-5′-yl derivative 11 demonstrated unusual
activity in that, in addition to producing a potentiation of L-type currents, it interfered with
the voltage-dependent gating of L-type channels by producing a net delay of their activation
at low voltages. This compound represents an interesting tool to probe L-type Ca2+ channel
structure and function.
Ch a r t 1
In tr od u ction
1,4-Dihydropyridines (DHPs) are an important class
of drugs which exert potent blocking activities on
calcium (Ca2+) currents through voltage-dependent L-
type channels.1 Several DHPs are clinically used in the
treatment of a number of cardiovascular diseases.
Classical 1,4-dihydropyridines have the structure A
(Chart 1). Structure-activity relationships of this class
of compounds have been the subject of several studies.1,2
The ester groups in the 3- and 5-positions are of crucial
importance for activity. If one of the two ester moieties
is replaced by appropriate groups such as nitro, lactone,
and thiolactone, the symmetry of the molecule is lost
and the resulting derivative exists as a pair of enanti-
omers, which have opposite pharmacological profiles.
Indeed one of the antipodes is a calcium entry blocker
(calcium antagonist), while the other is a calcium entry
activator (calcium agonist).1-3 Classical examples of this
behavior are the two enantiomeric pairs Bay K8644 (B)
and 202-791 (C) (Chart 1). Ca2+ channel activators are
interesting pharmacological tools. They represent po-
tential new drugs for the management of congestive
heart failure, in view of their positive inotropic effects.
Derivatives able to act as dual cardioselective calcium
channel agonists and smooth muscle-selective calcium
channel antagonists would be of great utility for the
treatment of cardiovascular diseases.4
benzofurazanyl- and benzofuroxanyl-1,4-dihydropyri-
dines.5 All the compounds displayed potent calcium
antagonist properties, evaluated in isolated rabbit basi-
lar artery by measuring relaxation of calcium-induced
contractions in high K+-depolarizing solutions. The
activity of benzofurazan derivatives was not substan-
tially changed by N-oxidation. On these grounds, as a
logical progression of our work on benzofurazanyl- and
benzofuroxanyl-1,4-dihydropyridines, we synthesized
derivatives 10-13 (Chart 2) and studied the action of
the racemic mixtures and the single enantiomers on
L-type Ca2+channels expressed by a rat insulinoma cell
line (RINm5F cells). Here we show that this class of
molecules, besides causing the typical Ca2+current
enhancement of 1,4-DHP agonists, is also able to
interfere with the activation kinetics of L-type channels,
In a recent paper we reported the synthesis, structure,
and calcium entry blocker activity of a number of
* To whom correspondence should be addressed. Phone: 0039-11-
6707670. Fax: 0039-11-6707659. E-mail: gasco@pharm.unito.it.
† Department of Scienza e Tecnologia del Farmaco.
‡ epartment of Neuroscienze.
§ ENSSPICAM.
10.1021/jm980623b CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/03/1999

New Class of Drugs for Probing L-Type Ca2+ Channel
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1423
Ch a r t 2
F igu r e 1. Temperature dependence of the C(4)H 1,4-DHP
protons of 12 and 13.
Sch em e 2a
Sch em e 1a
a
(a) iPrOH, ∆.
suggesting specific interactions with the charged groups
of the protein responsible for the voltage dependence of
channel opening. These compounds may represent an
interesting new class of drugs for probing L-type Ca2+
channel structure and function.
a
(a) CH3COOH, Thiele reagent (CrO3, (CH3CO)2O, H2SO4); (b)
EtOH, HCl; (c) iPrOH, 3, 4, ∆; (d) hυ, 24 h, distilled THF.
known that substituted benzofuroxans exhibit the isomer-
ization (or tautomerism) equilibrium (Figure 1A). De-
pending on both the nature and the position of R
substituents, as well as on the solvent and the temper-
ature, the equilibrium can assume different positions.6,7
The racemic methyl 1,4-dihydro-2,6-dimethyl-5-nitro-
4-(benzofuroxan-5(6)-yl)pyridine-3-carboxylate (13) was
obtained in a manner similar to the benzofurazanyl
analogues 10 and 11 (Scheme 1).
Synthesis of derivative 12 required the more complex
strategy shown in Scheme 2, since attempts to prepare
it through the route used for the analogue 10 failed. The
intermediate 2-azido-3-nitrobenzaldehyde (8) was ob-
tained by oxidation of 2-azido-3-nitrotoluene (6), dis-
solved in acetic acid, with Thiele reagent (CrO3-acetic
Resu lts
Ch em istr y. The racemic methyl 1,4-dihydro-2,6-
dimethyl-5-nitro-4-(benzofurazan-4-yl)pyridine-3-car-
boxylate (10) and the benzofurazan-5-yl analogue 11
were prepared by a modified Hantzsch reaction. The
condensation of nitroacetone (4) with methyl 3-amino-
crotonate (3) and either 4-benzofurazanaldehyde (1) or
5-benzofurazanaldehyde (2) afforded the expected com-
pounds in fair yield (Scheme 1). 1H NMR spectra are in
agreement with the proposed structures. Synthesis and
structural characterization of benzofuroxan derivatives
12 and 13 require more detailed consideration. It is

1424 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8
Visentin et al.
anhydride-H2SO4). The resulting 2-azido-3-nitrobenzal-
diacetate (7) was hydrolyzed by hydrochloric acid to the
expected aldehyde 8. This latter compound was trans-
formed into the desired methyl 1,4-dihydro-2,6-di-
methyl-5-nitro-4-(2-azido-3-nitrophenyl)pyridine-3-carbox-
ylate (9). Irradiation for 24 h of this derivative in
distilled THF solution with a medium-pressure Hg lamp
in an immersion well-photochemical apparatus af-
forded derivative 12 in fair yield.
The 1H NMR spectra in CD3COCD3 of the two
benzofuroxan DHPs, at room temperature, show broad
peaks, indicating extensive benzofuroxan tautomerism.
On cooling, the spectrum of derivative 12 sharpens into
a complex trace in the aromatic region, while the two
broad signals due to C(4)H 1,4-DHP protons were
transformed into a sharp biased doublet at 5.99 and 5.78
ppm, the signal at higher field being the more intense
(Figure 1B). This latter peak is attributed to the
4′-tautomer 12a , for steric reasons.7 At 273 K the
intensity ratio was 83:17. Therefore ∆G° value at 273
K for the 12a / 12b equilibrium is ca. 3.5 kJ /mol in
favor of 12a . The separation of the two peaks, which
did not change on further cooling, was δ ) 43 Hz.
Coalescence of these signals was observed at approxi-
mately 313 K (Tc). Because of the large difference in
abundance of the two tautomers and the consequent
difficulties in evaluating the coalescence temperature,
it was impossible to obtain a reliable value for the free
energy of activation ∆G*. From the temperature at
which separate spectra were observed, we believe that
this energy is ca. 63-67 kJ /mol. On cooling, the other
regions of the spectrum also sharpen and remain
substantially unchanged. The spectrum of derivative 13
behaves similarly. On lowering the temperature, the
aromatic region sharpens into a first-order trace and
the broad signal at 5.40 ppm due to C(4)H 1,4-DHP
protons is transformed into a sharp doublet at 5.36 and
5.33 ppm (Figure 1B). The higher field signal is tenta-
tively attributed to the 6′-tautomer 13b. At 243 K the
intensity ratio was ca. 1:1, indicating a ∆G° value for
benzofuroxan tautomerism close to zero. The separation
of the two peaks, δ ) 4.9 Hz, did not change on further
cooling. Coalescence of the signals was observed at 263
K (Tc), indicating a free energy of activation ∆G* )58.6
kJ /mol for the 13a / 13b interconversion.
F igu r e 2. Action of benzofurazanyl- and benzofuroxanyl-1,4-
DHPs on L-type Ba2+ currents of rat insulinoma RINm5F cells.
Each panel consists of two current recordings: one control and
one during application of the 1,4-DHP at 3 µM concentration
as indicated. The chemical structure of each racemate is drawn
to the left. Pulse protocols are given at the head of each
column. In the first and second column are reported the
agonistic effects of the racemates and their (+)-enantiomers.
On the third column are illustrated the antagonist effects of
the (-)-enantiomers. Note the current enhancements at -20
mV and the prolonged tails at -40 mV induced by the agonists
and the small antagonistic effects of the (-)-enantiomers. Each
panel is obtained from a different RINm5F cell except for (()-
11 and (+)-11 which are from the same cell.
tested for their action on L-type Ca2+ channels ex-
pressed by the rat insulinoma cell line RINm5F.9,10 Cells
were bathed in 10 mM Ba2+ and held at a relatively
low resting potential (-60 mV) to increase the percent-
age of available L-type channels. Under these condi-
tions, saturating concentrations of nifedipine (3 µM)
reversibly blocked about 70% of the total Ba2+ current
(66.7 ( 2.8%, n ) 14), indicating a predominance of L-
over the non-L-type channels in these cells.
Figure 2 summarizes the typical action of the various
DHPs on the Ba2+ currents of RINm5F cells when the
compounds were applied at the same concentration (3
µM). The test potential was -20 mV for the molecules
with agonistic effects and -10 mV for those with
antagonistic action. These conditions were chosen since
-20 mV represents a membrane potential at which the
open probability of L-type channels is sufficiently high
to reveal maximal agonistic effects of the 1,4-DHPs,
while -10 mV is a potential value that gives maximal
activation for the L-type channel and is thus sufficiently
high to test the blocking action of the antagonists. To
better visualize the agonistic effects of the compounds,
the return potential from test was set at -40 mV, which
was sufficiently negative to produce fast decaying tail
currents at control and slow tails with the agonist. As
shown in Figure 2, all four racemates produced a
current increase at -20 mV and a remarkable slowing
of tail currents on return to -40 mV (left column). The
The separation of racemic mixtures of compounds 10-
13 was successfully carried out by HPLC, using an
amylose (S)-1-phenylethyl carbamate (Chiralpak AS)
column,8 on analytical as well as semipreparative scales.
The single enantiomers were obtained in 97-100%
optical purity.
As expected, theoretical studies using an AM-1 Hamil-
tonian algorithm show the same conformational domain
for all the compounds. For both substances two energy
minimum rotamers are possible; the benzoheterocycle
system is in a pseudoaxial position and orthogonal to
the dihydropyridine flat boat ring. In one conformation
the heteropentatomic ring lies on the same side as the
C(4)H 1,4-DHP hydrogen, while in the other it is
oriented on the opposite side. This picture is typical of
4-phenyl-1,4-dihydropyridines ortho or meta substituted
at the phenyl ring.2
Volta ge-Cla m p Stu d ies. The four racemates il-
lustrated in Chart 2 and their single enantiomers were

New Class of Drugs for Probing L-Type Ca2+ Channel
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1425
rate of rise of the current, while (+)-11 caused a net
prolongation of the current which had hardly reached
a steady-state level by the end of the pulse. It should
be noted that Bay K 8644 and (+)-11 (or (()-11) had
nearly identical effects on the current/voltage charac-
teristics. They both shifted the current/voltage curve
toward more negative potentials (Figure 3B, left). The
voltage shift in the ramp of several millivolts was dose-
dependent and saturated at concentrations of (()-11
above 10 µM (Figure 3B, right). The origin of the slow
activation phase induced by (()-11 and its enantiomer
(+)-11 on L-type currents was studied in a series of
experiments in which the action of the two agonists was
tested at increasing voltages (Figure 3C). The slowing
down of activation was most evident below -20 mV,
while at more positive voltages (>-10 mV) the slow
phase was less pronounced and progressively faster. At
-30 mV the slow phase had an average time constant
of 26.3 ms (τslow) and contributed to about 50% of the
total DHP-modified current, while at -10 mV the
average τslow was 9.4 ms and contributed to approxi-
mately 23% of the total amplitude. At 0 mV there was
almost no difference between the rate of rise of the
current in the control and with the agonist despite the
marked current enhancement. In eight cells τfast was 1.1
( 0.1 ms in the control and 1.3 ( 0.1 ms in the presence
of 11. This “voltage-dependent” modulation of Ca2+
channel activation is reminiscent of the modulating
effects of G proteins on N-type channels and is generally
interpreted as a reversible inhibition of Ca2+ channel
gating.14 The opening of N-type channels is markedly
delayed at negative membrane potential, but the delay
is progressively removed with increasing positive po-
tential (facilitation),15 although through different mech-
anisms of action this seems to hold true also for the
“voltage-dependent” effects of (()-11 (or (+)-11) on
L-type channels. We tested this by applying the classical
double-pulse protocol in which the test pulse to -20 mV
is preceded by a strong and brief prepulse to +80 mV
(Figure 3D). In the absence of the prepulse, the activa-
tion of the Ba2+ current is markedly delayed with (()-
11, but the delay is removed following the facilitatory
prepulse (filled circle). It should be noted that under
control conditions the prepulse does not alter the
activation kinetics of Ba2+ currents at -20 mV (top
traces in Figure 3D).
F igu r e 3. (A) Action of Bay K 8644 (3 µM) and (+)-11 (3 µM)
applied sequentially to the same cell. Note the marked
prolongation of the rising phase of the L-type current induced
by (+)-11 in comparison to the fast activation induced by Bay
K 8644. The two currents reach nearly the same maximum
value at the end of the pulse and produce identical tail currents
on return to -40 mV, suggesting that both compounds produce
an equal enhancing effect after about 60-80 ms. The pulse
protocol is indicated below. (B) Current/voltage (I/V) charac-
teristics at control and with either 3 µM (()-11 or 3 µM Bay
K 8644 measured by ramp commands of 1.4 V/s slope. To the
left is reported the dose dependence of the voltage shifts
produced by increasing doses of (()-11. The shifts in millivolts
were calculated at the half-maximal current amplitude of the
I/V curves in the voltage region of negative slope. (C) Action
of 3 µM (+)-11 on L-type currents at potentials between -30
and 0 mV. Note the slow phase of activation at -30 and -20
mV, which becomes progressively faster and more attenuated
at -10 and 0 mV. (D) Slowing down of the L-type current
induced by (()-11 is removed by strong conditioning prepulses.
The current traces were recorded sequentially with control and
during application of (()-11. In each case the recordings
consisted of two stimulations separated by a 10-s interval in
which a test depolarization to -20 mV was preceded (filled
circles) or not by a facilitatory prepulse of 50 ms to +80 mV.
Notice the absence of any change of the current kinetics in
the control and the complete removal of the current prolonga-
tion in the presence of (()-11.
action of the racemates was comparable to that of the
(+)-enantiomers (middle column) but opposite to that
of the corresponding (-)-enantiomers which had little
or no antagonist effects (right column).
Of the compounds tested, the benzofurazan derivative
(()-11 and its enantiomer (+)-11 were by no means the
most potent molecules of the series studied. They had
very similar action with average current increases at
-20 mV of 247 ( 21% (n ) 8) and 224 ( 27% (n ) 5),
respectively. The second benzofurazan derivative (()-
10 and its enantiomer (+)-10 produced much smaller
current enhancements (164 ( 4%, n ) 12), while the
benzofuroxan derivatives (()-13 and (()-12 were even
less effective and, thus, not further investigated.
An interesting feature of (()-11 and (+)-11 was their
ability to slow the rate of rise of the L-type current, an
effect which has not previously been reported for other
well-known DHP agonists, such as Bay K 8644 and 202-
791.11-13 Direct comparison of the action of (+)-11 with
respect to Bay K 8644 on the same cell clarified how
the two DHPs act on L-type current activation (Figure
3A). Bay K 8644 produced a nearly 2-3-fold increase
of the control current without significantly altering the
Discu ssion
Our data show that enantiomers of methyl 1,4-
dihydro-2,6-dimethyl-5-nitro-4-(benzofurazanyl)pyridine-
3-carboxylates 10 and 11 and their benzofuroxanyl
analogues 12 and 13 display opposite effects on Ca2+
currents through voltage-dependent L-type calcium
channels. Dextrorotatory antipodes are potent calcium
entry activators, while the levorotatory ones are weak
calcium entry blockers. Consequently the corresponding
racemic mixtures behaved preferentially as Ca2+ ago-
nists. Noteworthy is the behavior of the racemic mixture
of derivative (()-11 and its (+)-enantiomer, which was
the most potent of the series. This compound, besides
causing a potentiation of L-type currents, is capable of
interfering with the voltage-dependent gatings of L-type
channels, by producing a net delay in the channel
activation. To our knowledge, this latter property was
observed and only partially described for the 4-(2-

1426 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8
Visentin et al.
EGTA, 8 glucose, 10 HEPES, 2 Mg-ATP, 0.25 cAMP, 0.5 GTP,
and 15 phosphocreatine (pH 7.3 with CsOH). The four race-
mates and their enantiomers as well as Bay K 8644 (Bayer
AG, Wuppertal, Germany) were dissolved in ethanol (95%) and
stored in the dark at 4 °C as 1 mM stock solutions. Dilution
to the final concentration was performed daily under light
protection. The external solutions were exchanged by fast
superfusion using a multibarrelled glass pipet held close to
the cell.21 Experiments were carried out at room temperature
(22 °C).
Wh ole-Cell Cu r r en t Recor d in gs. Membrane currents
were measured in the whole-cell configuration using a List
EPC7 (Darmstadt, Germany) patch-clamp amplifier and pipets
of borosilicate glass with a resistance of 3-4 MΩ. Stimulation,
acquisition, filtering, and data analysis were performed as
described elsewhere.10 Cells were clamped at -60 mV holding
potential (Vh). Step depolarizations of 60-150 ms from Vh were
applied at intervals of 10 s to minimize Ca2+ channel “run-
down”. Capacitative and leakage currents were compensated
electronically and by subtracting residual Cd2+-insensitive
currents remaining after addition of 200 µM Cd2+. Data are
expressed as means ( SEM for n ) number of cells.
Gen er a l Meth od of P r ep a r a tion of DHP s 10, 11, a n d
13. A solution of the appropriate aldehyde (Scheme 1: 1, 2, 5)
(5 mmol) and nitroacetone (4) (0.62 g, 6 mmol) in 2-propanol
(15 mL) was refluxed for 1 h and then methyl 3-aminocrotonate
(3) (0.58 g, 5 mmol) was added. Reflux was continued for 5 h.
Solvent removal gave a residue, which was purified by flash
chromatography. Chromatographic eluents, yields, melting
points, and crystallization solvents of the products were as
follows.
benzylbenzoyl)-2,5-dimethyl-1H-pyrrole-3-carboxylic acid
methyl ester (FPL 64176), an L-type Ca2+ channel
activator with a benzoylpyrrole structure.16 The race-
mate (()-10 and its (+)-enantiomer do not display this
activity. This suggests that in 11 the benzofurazan-5-
yl moiety could interact with a receptor region, near the
area fitted by the 1,4-dihydropyridine substructure,
which cannot be occupied by the benzofurazan-4-yl
substructure in 10. Although in 13 the benzofuroxan
ring is properly oriented to interact with the same
receptor region fitted by benzofurazan in 11, this former
compound exerts a strongly reduced capacity to alter
the rate constant of Ca2+channel activation. Benzofura-
zan and benzofuroxan rings have similar lipophilicity
(log P ) 1.62 and 1.43, respectively),17 but different
electronic distribution, as evidenced by their 1H NMR
spectra.18 Thus we can hypothesize that in 11 the
correct orientation of the benzofurazan system and/or
its electronic properties are important determinants in
slowing down the rate of L-type current rise. Probably,
these two characteristics are crucial for the interactions
of 11 with the gating charges which control the voltage-
dependent transition of L-type Ca2+ channels from their
closed to open conformation.19 Anyway, at this stage of
our work, it is difficult to drawn more precise conclu-
sions about structure-activity relationships, owing to
the limited number of structures studied.
In conclusion the results of the present work open up
interesting issues concerning the utility of these mol-
ecules for probing Ca2+ channel structure. In addition
they uncover the possibility to design new 1,4-DHPs
with potential utility as therapeutical tools in the
treatment of pathologies where a modest selectivity and
a significant use-dependent agonistic action on these
channels would be beneficial.
(()-Meth yl 1,4-d ih yd r o-2,6-d im eth yl-5-n itr o-4-(ben zo-
fu r a za n -4-yl)p yr id in e-3-ca r boxyla te (10): eluent CH2Cl2/
EtOAc (95/5, v/v); yield 40%; mp 208-209 °C (EtOAc/
petroleum ether 40-60 °C) (lit. mp 205 °C);22 1H NMR
(CD3COCD3) δ 5.85 (s, 1H, CH-4), 2.55/2.34 (2s, 6H, 2-CH3/6-
CH3), 3.59 (s, 3H, OCH3), 8.92 (br s, 1H, NH), 7.81-7.48 (m,
3H, Arom). Anal. (C15H14N4O5) C, H, N.
(()-Meth yl 1,4-d ih yd r o-2,6-d im eth yl-5-n itr o-4-(ben zo-
fu r a za n -5-yl)p yr id in e-3-ca r boxyla te (11): eluent CH2Cl2/
EtOAc (95/5, v/v); yield 35%; mp 171-172 °C (EtOAc/
1
Exp er im en ta l Section
petroleum ether 40-60 °C); H NMR (CD3COCD3) δ 5.60 (s,
1H, CH-4), 2.71/2.53 (2s, 6H, 2-CH3/6-CH3), 3.77 (s, 3H, OCH3),
8.96 (br s, 1H, NH), 7.98-7.74 (m, 3H, Arom). Anal. (C15H14N4-
O5) C, H, N.
Melting points were determined on a Bu¨chi 530 apparatus
after introducing the sample into the bath at a temperature
10 °C lower than the melting point. A heating rate of 1 °C
min-1 was used, 3 °C min-1 in the case of decomposition. The
compounds were routinely checked by infrared spectropho-
(()-Meth yl 1,4-d ih yd r o-2,6-d im eth yl-5-n itr o-4-(ben zo-
fu r oxa n -5(6)-yl)p yr id in e-3-ca r boxyla te (13): eluent petro-
leum ether 40-60 °C/EtOAc (7/3, v/v); yield 35%; mp 159-
160 °C dec (EtOAc/petroleum ether 40-60 °C); 1H NMR
(CD3COCD3) δ for 13a ,b 5.32/5.30 (s, 1H, CH-4), 2.56, 2.57/
2.38, 2.38 (2s, 6H, 2-CH3/6-CH3), 3.63/3.64 (s, 3H, OCH3), 9.14/
9.11 (br s, 1H, NH), 7.23-7.60/7.23-7.60 (Arom). Anal.
(C15H14N4O6) C, H, N.
2-Azid o-3-n itr oben za ld eh yd e (8). Concentrated sulfuric
acid (330 mL) was added dropwise to a mechanically stirred
and ice-salt-cooled solution of 6 (25 g, 0.14 mol) in a mixture
of acetic acid (200 mL) and acetic anhydride (200 mL), keeping
the temperature below 20-25 °C. The mixture was cooled
below 10 °C, and then CrO3 (39 g, 0.39 mol) was added
portionwise, keeping the temperature below 10 °C. After this
addition the mixture was stirred, without any control of
temperature, for 3 h and then poured into ice-water. The
resulting yellow precipitate was collected by filtration, washed
with water, then treated with a 2% Na2CO3 solution, and again
filtered and washed with water. The crude product was
partially purified by crystallization (EtOH/H2O) to obtain 7
(40%) sufficiently pure to be used for further reaction. Hy-
drolysis of 7 (17 g) in a boiling mixture of water (100 mL),
EtOH (50 mL), and concentrated hydrochloric acid (75 mL)
afforded after 45 min the title product: yield 75%; mp 60-61
°C dec (EtOH/water) (lit.23 mp 60-61 °C).
1
tometry (Shimadzu FT-IR 8101 M). H and 13C NMR spectra
were recorded on a Bruker AC-200 spectrometer. Resonance
assignments were given by analyzing the spectra at low
temperature to resolve averaged signals of the two tautomers
13a ,b and 12a ,b, respectively. Column chromatography was
performed on silica gel (Merck Kieselgel 60, 230-400 mesh
ASTM) with the indicated solvent system. Anhydrous MgSO4
was used as drying agent. Solvent removal was achieved under
reduced pressure at room temperature. Elemental analyses
of the new compounds were performed by REDOX (Cologno
M.) and the results are within (0.4% of the theoretical values.
Intermediates 1,5 2,5 5,5 and 620 were synthesized according
to the literature. HPLC studies on Chiralpak AS (Daicel Co.,
Tokyo) were performed with a Merck-Hitachi Lichrograph
model L-600 HPLC pump at controlled temperature 22 °C, a
Merck-Hitachi Lichrograph L-4000 UV detector (l ) 254 nm),
and a Merck D-2500 recorder. Dead volume was determined
by co-injection of 1,3,5-tri-tert-butylbenzene. Two columns were
used, one (250 × 4.6 mm) for analytical purposes and one (250
× 10 mm) for semipreparative applications. Specific rotations
were measured on a Perkin Elmer 241 polarimeter.
Cell Cu ltu r e a n d Solu tion s. RINm5F cells were kindly
provided by Dr. E. Sher (Eli Lilly and Co., Windlesham,
Surrey, U.K.) and were cultured as previously described.9 The
recording external solution was (mM) 135 NaCl, 10 BaCl2, 1
MgCl2, 10 HEPES (pH 7.3 with NaOH), and 300 nM TTX. The
standard internal solution was (mM) 110 CsCl, 30 TEACl, 10
(()-Meth yl 1,4-d ih yd r o-2,6-d im eth yl-5-n itr o-4-(2-a zid o-
3-n it r op h en yl)p yr id in e-3-ca r b oxyla t e (9): prepared and
purified according to the method described for the preparation

New Class of Drugs for Probing L-Type Ca2+ Channel
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1427
(6) Gasco, A.; Boulton, A. J . Furoxans and Benzofuroxans. In
Advances in Heterocyclic Chemistry; Katritzky, A. R., Boulton,
A. J ., Eds.; J . Wiley: New York, 1981; Vol. 29, pp 251-340.
(7) Boulton, A. J .; Halls, P. J .; Katritzky, A. R. N-Oxides and Related
Compounds. Part XXXVII. The effect of Methyl and Aza-
substituents on the Tautomeric Equilibrium in Benzofuroxan.
J . Chem. Soc. B 1970, 636-640. Boulton, A. J .; Katritzky, A.
R.; Swell, M. J .; Wallis, B. N-Oxides and Related Compounds.
Part XXXI. The Nuclear Magnetic Resonance Spectra and
Tautomerism of Some Substituted Benzofuroxans. J . Chem. Soc.
B 1967, 914-919.
(8) Chiralpak AS is available from DAICEL Co., Tokyo, J apan.
(9) Pollo, A.; Lovallo, M.; Biancardi, E.; Sher, E.; Socci, C.; Carbone,
E. Sensitivity to dihydropyridines, ω-conotoxin and noradrena-
line reveals multiple HVA Ca2+ channels in rat insulinoma and
human pancreatic â-cells. Pflu¨gers Archiv. 1993, 423, 462-471.
(10) Magnelli, V.; Pollo, A.; Sher, E.; and Carbone, E. Block of non-
L-, non-N-type Ca2+channels in rat insulinoma RINm5F cells
by ω-conotoxin MVIIC. Pflu¨gers Archiv. 1995, 429, 762-771.
(11) Sanguinetti, M. C.; Krafte, D. S.; Kass, R. S. Voltage-dependent
modulation of Ca2+channel current in heart cells by Bay K 8644.
J . Gen. Physiol. 1986, 88, 369-392.
(12) Kokubun, S.; Prod′hom, B.; Becker, C.; Porzig, H.; Reuter, H.
Studies on Ca2+channels in intact cardiac cells: voltage-depend-
ent effects and cooperative interactions of dihydropyridine
enantiomers. Mol. Pharmacol. 1986, 30, 571-584.
(13) Brown, A. M.; Kunze, D. L.; Yatani, A. The agonist effect of
dihydropyridines on Ca2+channels. Nature 1984, 311, 570-572.
(14) Carbone, E.; Magnelli, V.; Pollo, A.; Carabelli, V.; Albillos, A.;
Zucker, H. Biophysics of voltage-dependent Ca2+channels: gat-
ing kinetics and their modulation. In Ion Channel Pharmacology;
Soria, B., Cen˜a, V., Eds.; Oxford University Press: Oxford,
England, 1998; pp 97-128.
of the analogues 10, 11, and 13; eluent CH2Cl2/EtOAc (95/5,
v/v); yield 28%; mp 188-190 °C dec (EtOAc/petroleum ether
40-60 °C); 1H NMR (DMSO-d6) δ 5.69 (s, 1H, CH-4), 2.31 (2s,
6H, 2-CH3/6-CH3), 3.59 (s, 3H, OCH3), 9.78 (br s, 1H, NH),
7.97-7.41 (m, 3H, Arom). Anal. (C15H14N6O6) C, H, N.
(()-Meth yl 1,4-Dih yd r o-2,6-d im eth yl-5-n itr o-4-(ben zo-
fu r oxa n -4(7)-yl)p yr id in e-3-ca r boxyla te (12). A solution of
9 (0.56 g) in distilled THF (50 mL) in a quartz vessel was
irradiated under nitrogen for 24 h with a medium-pressure
Hg lamp in an immersion well Rayonet-type photochemical
apparatus. The solution was then concentrated, and 2 N
hydrochloric acid (10 mL) was added. The resulting mixture
was stirred for 20 min at room temperature; the organic phase
was separated, dried, and evaporated. The residue was purified
by flash chromatography: eluent petroleum ether 40-60 °C/
THF (7/3, v/v); yield 40%; mp 153-158 °C dec (THF/petroleum
1
ether 40-60 °C); H NMR (CD3COCD3) δ for 12a ,b 5.77/5.99
(s, 1H, CH-4), 2.52, 2.52/2.33, 2.36 (2s, 6H, 2-CH3/6-CH3), 3.60/
3.46 (s, 3H, OCH3), 9.01/8.76 (br s, 1H, NH), 7.24-7.55/7.24-
7.55 (m, 3H, Arom). Anal. (C15H14N4O6) C, H, N.
HP LC Sep a r a tion of Ra cem ic Mixtu r es (()-10, (()-11,
(()-12, a n d (()-13. HPLC separation was performed on a
semipreparative Chiralpak AS column (25 × 1 cm); eluent
hexane/ethanol (85/15); flow rate 3 mL/min, t0 ) 4.7 min: 10
k(-) 1.68, [R]25D ) +37 (c ) 0.50, EtOH), mp 205 °C, k(+) 2.04,
[R]25 ) -36 (c ) 0.55, EtOH), mp 205 °C, R ) 1.22, Res )
D
0.66; 11 k(-) 1.77, [R]25 ) +11 (c ) 0.40, EtOH), mp 170-
D
171 °C, k(+) 2.62, [R]25D ) -11 (c ) 0.40, EtOH), mp 170-171
°C, R ) 1.48, Res ) 1.3; 12 k(-) 3.09, [R]25D ) +36.2 (c ) 0.50,
EtOH), mp 152-158 °C dec, k(+) 3.78, [R]25D ) -38.2 (c ) 0.47,
EtOH), mp 152-158 °C dec, R ) 1.23, Res ) 0.9; 13 k(-) 3.45,
[R]25D ) +8.18 (c ) 0.55, EtOH), mp 159-160 °C dec, k(+) 5.19,
[R]25D ) -8.19 (c ) 0.55; EtOH), mp 159-160 °C dec, R ) 1.50,
Res ) 1.1.
(15) Carabelli, V.; Lovallo, M.; Magnelli, V.; Zucker, H.; Carbone, E.
Voltage-dependent modulation of single N-type channel kinetics
by receptor agonists in IMR32 cells. Biophys. J . 1996, 70, 2144-
2154.
(16) Rampe, D.; Anderson, B.; Rapien-Pryor, V.; Li, T.; Dage, R. C.
J . Pharmacol. Exp. Ther. 1993, 265, 1125-1130.
(17) Calvino, R.; Gasco, A.; Leo, A. An Analysis of Lipophilicity of
Furazan and Furoxan Derivatives using the CLOGP Algorithm.
J . Chem. Soc., Perkin Trans. 1992, 2, 1643-1646.
(18) Harris, R. K.; Katritzky, A. R.; Øksne, S.; Bailey, A. S.; Paterson,
W. G. N-Oxides and Related Compounds. Part XIX. Proton
Resonance Spectra and the Structure of Benzofuroxan and its
Nitro-derivatives. J . Chem. Soc. 1963, 197-203.
(19) Armstrong, C. M.; Hille, B. Voltage-gated ion channels and
electrical excitability. Neuron 1998, 20, 371-380.
(20) Zincke, T.; Schwarz, P. U¨ ber o-Dinitrosoverbindungen der
Benzolreihe. Ann. Chem. 1899, 307, 28-49.
(21) Carbone, E.; Lux, H. D. Kinetics and selectivity of a low voltage-
activated calcium current in chick and rat sensory neurones. J .
Physiol. 1987, 386, 547-570.
(22) Vogel, A.; Bormann, G. Nouveaux Derive´s de la 1,4-Dihydro-
pyridine, Leur Pre´paration et leur Utilisation en The´rapeutique
Comme Me´dicaments. BE 900874, 1985.
(23) Balasubrahmanyam, S. N.; Radhakrishna, A. S.; Boulton, A. J .;
Kan-Woon, T. Furazans and Furazan Oxides. 7. Interconversion
of Anthranils, Benzofurazan Oxides, and Indazoles. J . Org.
Chem. 1977, 42, 897-901.
Ack n ow led gm en t. This work was supported by a
MURST grant.
Refer en ces
(1) Triggle, D. J .; Langs, D. A.; J anis, R. A. Ca2+ Channel Ligands:
Structure-Function Relationships of the 1,4-Dihydropyridines.
Med. Res. Rev. 1989, 9, 123-180.
(2) Goldmann, S.; Stoltefuss, J . 1,4-Dihydropyridines: Effects of
Chirality and Conformation on the Calcium Antagonist and
Calcium Agonist Activities. Angew. Chem., Int. Ed. Engl. 1991,
30, 1559-1578.
(3) Rovnyak, G. C.; Kimball, S. D.; Beyer, B.; Cucinotta, G.; Di
Marco, J . D.; Gougoutas, J .; Hedberg, A.; Malley, M.; McCarthy,
J . P.; Zhang, R.; Moreland, S. Calcium Entry Blockers and
Activators: Conformational and Structural Determinants of
Dihydropyridine Calcium Channel Modulators. J . Med. Chem.
1995, 38, 119-129.
(4) Vo, D.; Matowe, W. C.; Ramesh, M.; Iqbal, N.; Wolowyk, M. W.;
Howlett, S. E.; Knaus, E. E. Modulation Activities, and Voltage-
Clamp Studies of Isopropyl 1,4-Dihydro-2,6-dimethyl-3-nitro-4-
pyridine-5-carboxylate Racemates and Enantiomers. J . Med.
Chem. 1995, 38, 2851-2859.
(5) Gasco, A. M.; Ermondi, G.; Fruttero, R.; Gasco, A. Benzofura-
zanyl- and benzofuroxanyl-1,4-dihydropyridines: synthesis, struc-
ture and calcium entry blocker activity. Eur. J . Med. Chem.
1996, 31, 3-10.
J M980623B