Unique structure-activity relationship for 4-isoxazolyl-1,4-dihydropyridines

Article abstract of DOI:10.1021/jm020354w

A series of 4-isoxazolyl-1,4-dihydropyridines (IDs) were prepared and characterized, and their interaction with the calcium channel was studied by patch clamp analysis. The structureactivity relationship (SAR) that emerges is distinct from the 4-aryldihydropyridines (DHPs), and affinity increases dramatically at higher holding potentials. Thus, among the 3′-arylisoxazolyl analogues p-Br > p-Cl ? p-F, and p-Cl > m-Cl > o-Cl ? o-MeO. Four of the analogues were examined by single-crystal X-ray diffractometry, and all were found to adopt an O-exo conformation in the solid state. The calculated barrier to rotation, however, suggests that rotation about the juncture between the heterocyclic rings is plausible under physiological conditions. A variable-temperature NMR study confirmed the computation. With Striessnig's computational sequence homologation procedure, a working hypothesis was derived from the data that explains the unique SAR for IDs.

Full text of DOI:10.1021/jm020354w

J . Med. Chem. 2003, 46, 87-96
87
Un iqu e Str u ctu r e-Activity Rela tion sh ip for 4-Isoxa zolyl-1,4-d ih yd r op yr id in es
Gerald W. Zamponi,† Stephanie C. Stotz,† Richard J . Staples,‡ Tina M. Andro,§ J ared K. Nelson,§
Victoria Hulubei,§ Alex Blumenfeld,§ and Nicholas R. Natale*,§
Department of Pharmacology and Therapeutics, University of Calgary, 3330 Hospital Drive NW,
Calgary, Alberta T2N 4N1, Canada, Department of Chemistry and Chemical Biology, Harvard University,
12 Oxford Street, Cambridge Massachusetts 02138, and Department of Chemistry, 301 Renfrew Hall,
University of Idaho, P.O. Box 442343, Moscow, Idaho 83844-2343
Received August 12, 2002
A series of 4-isoxazolyl-1,4-dihydropyridines (IDs) were prepared and characterized, and their
interaction with the calcium channel was studied by patch clamp analysis. The structure-
activity relationship (SAR) that emerges is distinct from the 4-aryldihydropyridines (DHPs),
and affinity increases dramatically at higher holding potentials. Thus, among the 3′-
arylisoxazolyl analogues p-Br > p-Cl . p-F, and p-Cl > m-Cl > o-Cl . o-MeO. Four of the
analogues were examined by single-crystal X-ray diffractometry, and all were found to adopt
an O-exo conformation in the solid state. The calculated barrier to rotation, however, suggests
that rotation about the juncture between the heterocyclic rings is plausible under physiological
conditions. A variable-temperature NMR study confirmed the computation. With Striessnig’s
computational sequence homologation procedure, a working hypothesis was derived from the
data that explains the unique SAR for IDs.
In tr od u ction
promise. A recent milestone in this arena is the solved
structure of the K+ channel.4 The K+ channel in
particular has several general features analogous to the
calcium channel, though two major differences arise: (1)
At 400 kDa, the calcium channel is much larger than
the 77 kDa potassium channel, and (2) while the four
motifs comprising the R1 subunit are homologous, they
are not identical, complicating diffractometry. The bind-
ing site of DHPs to the L-type Ca2+ channel has been
approached from several directions. Covalent photoaf-
finity labeling provided the first evidence that the DHP
binding site was located on the transmembrane strands
IIIS6 and IVS6 and on the S5/S6 helices. Site-directed
mutagenesis and chimera studies have been used to
define the binding site of the 4-aryl-1,4-dihydropyridines
to the calcium channel. Site-directed mutagenesis has
been approached from both the perspectives of (1)
changing the residues, which then abolish DHP bind-
ing,5 and (2) adding residues that confer DHP binding
to a non-L-type channel.6 On the basis of the latter (or
constructive approach), Schleifer performed 3D-QSAR
pseudoreceptor modeling based on the nine residues
that are essential determinants of DHP binding and
common to both antagonist and agonist binding sites.
Striessnig displayed the utility of the potassium chan-
nel’s crystal structure by performing in silico site-
directed mutagenesis of the 12 critical amino acid
residues and modeling interactions of those residues
with a DHP to understand binding characteristics of the
hypothesized binding domain. This seems to be a usable
working mimic of the L-type calcium channel. We have
recreated and employed Striessnig’s model in order to
study these interactions with 4-isoxazolyl-1,4-dihydro-
pyridines (IDs) and interpret structure-activity rela-
tionship (SAR) results in order to optimize our picture
of the Ca2+ channel (vide infra). Our own working
hypothesis for the 4-isoxazolyl-1,4-dihydropyridine bind-
ing site,7 developed prior to the reports of site-directed
Small molecules that selectively modulate ion chan-
nels have important therapeutic applications. Calcium
channel blockers (CCBs), including the 4-aryl-1,4-dihy-
dropyridines (DHP) exemplified by “nifedipine”, have
been used in general medical practice worldwide for the
treatment of hypertension and vasospastic angina for
over 2 decades.1 Recently, controversy concerning side
effects has emerged. The first was an increased inci-
dence of myocardial infarction among patients on a
regimen of “nifedipine”,2a,b which led the National
Institutes of Health to issue a warning and to the use
of time release formulations. The second was a retro-
spective correlation of increased incidence of cancer
among older patients taking calcium channel blockers.2c,d
The current evidence on CCBs largely precludes the
adverse effects of the magnitude suggested by the scare
in the mid-1990s but attests to their safety and efficacy.
It is also consistent with the recommendation in the
Sixth Report of the J oint National Committee on
Prevention, Detection, Evaluation, and Treatment of
High Blood Pressure that puts â blockers and diuretics
as first-line agents for the treatment of uncomplicated
hypertension, with DHPs and ACE inhibitors second.3
Despite a rocky stretch, CCBs retain an important role
in hypertension therapy and their improvement remains
an important endeavor.
As a consequence, the search for analogues with novel
binding properties has attracted renewed interest.
Understanding the structure and function of membrane-
bound ion channels remains a challenge to chemists,
and the design of molecules as ligands to selectively
modulate ion channels holds significant therapeutic
* To whom correspondence should be addressed. Phone: 208-885-
6778. Fax: 208-885-6173. E-mail: nrnatale@uidaho.edu.
† University of Calgary.
‡ Harvard University.
§ University of Idaho.
10.1021/jm020354w CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/04/2002

88 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1
Zamponi et al.
Ta ble 1. Structures of 4-Isoxazolyl-1,4-dihydropyridines (IDs), 1, and IC50 from Patch Clamp Studies
mutagenesis, is similar to the Schleifer pseudoreceptor
model in many respects.8 However, none of the models
to date are entirely satisfactory, and the elucidation of
how the channel conformational changes relate to
function and how subtle structural features might be
exploited for therapeutic benefit represents a formidable
task.
Sch em e 1. Synthesis of the New
4-Isoxazolyl-1,4-dihydropyridines,1.j-k a
We herein report the first patch clamp studies of the
4-isoxazolyl-1,4-dihydropyridines, which include a new
series of 3-substituted IDs designed to further system-
atically delineate the SAR of this class of compounds
(Table 1) and which has revealed a unique SAR for the
4-IDs. We also report an extension of the model devel-
oped by Striessnig to develop a working hypothesis of
ID drug-receptor interaction, which provides a reason-
able explanation of our experimental SAR observations.
Resu lts a n d Discu ssion
A series of 4-isoxazolyl-1,4-dihydropyridines were
prepared by nitrile oxide cycloaddition, and after adjust-
ment of the isoxazole C-4 ester to a carbaldehyde
functional group, Hantzsch synthesis provided new
compounds 1.j-1.n (Scheme 1). Products were charac-
terized, and their interaction with the calcium channel
was studied by patch clamp analysis. The R1C, â1B, and
R2-δ calcium channels were transfected transiently into
tsa-210 cells and characterized via whole-cell patch
clamp recordings, thus allowing us to work with a
defined and pure population of channels. The data
shown in Figure 1 were performed at a typical neuronal
resting potential of -80 mV, but the affinity increased
significantly in a dramatic manner at more depolarized
holding potentials of -30 mV (data not shown). A
representative experiment is illustrated in Figure 1
(inset). The lipophilic derivatives previously examined
by radioligand assay (1.i and 1.f.) were the best antago-
nists in this series, and overall, the radioligand binding
previously studied follows a trend similar to that of the
patch clamp electrophysiology reported here. The struc-
a
(a) NH2OH‚HCl, EtOH reflux; (b) NCS, DMF, room temp; (c)
Et3N, 8, room temp; (d) LiAlH4, THF 0 °C; (e) PCC, CH2Cl2, room
temp; (f) 2CH3C[O]CH2CO2Et, aqueous NH3, EtOH, reflux.
ture-activity relationship that emerges for IDs is
distinct in some critical details from the 4-aryldihydro-
pyridines. Regarding state dependence, we suggest that
not only do different side groups interact differently with
the varying kinetic states of the calcium channel but
also within a given state (e.g., open channel block) the
blocking affinity should depend on specific interactions
between side groups on the drug molecule and certain
residues on the channel (vide infra).
Among the 3′-arylisoxazolyl analogues, the larger
halogens were more active in the series p-Br > p-Cl .
p-F (entries 1.b -1.d ). The position of a specific
substituent is related to the biological activity in
the progression p-Cl > m-Cl > o-Cl . o-MeO (entries

4-Isoxazolyl-1,4-dihydropyridines
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1 89
C-3 and C-5 methyl groups of the ethyl esters, reaches
coalescence at -53 °C (Figure 2). At temperatures below
this, a nuclear Overhauser effect (NOE) is observed
between the isoxazole methyl and the DHP C-4 methine
and between the phenyl protons and the C-2 and C-6
methyl groups of the DHP, which unequivocally indi-
cates that one low-temperature conformation is similar
to that observed in the X-ray. However, a strong NOE
is also observed between the isoxazole C-5 methyl and
these same C-2 and C-6 methyl groups of the DHP. This
is reasonable only if both conformers at the ring
juncture are present. At room temperature, the 1H NMR
spectrum at 500 Mz exhibits a single signal for the C-2
and C-6 methyls groups of the dihydropyridine, which
would only be plausible if the average conformation in
solution at this temperature reflected free rotation about
the ring juncture between the isoxazole and dihydro-
pyridine moieties. The orthogonal relationship between
isoxazole and 3′-aryl observed by crystallography, in
contrast, would be expected to be maintained in solution
for the ortho-substituted examples and is above the
threshold that Oˆ ki has suggested for isolation of atro-
poisomers.19
F igu r e 1. IC50 values for L-type calcium channel block
obtained with the isoxazolyl DHP series. The experiments were
carried out at a holding potential of -80 mV with 5 mM Ba2+
as the charge carrier. The IC50 values were obtained from fits
to individual dose-response curves. The numbers in paren-
theses reflect the numbers of experiments. Inset: Representa-
tive dose-response curve and whole-cell current records
obtained with compound 1.i. The current traces shown were
obtained in the absence of drug and after application of 30
nM, 300 nM, 1 µM, and 3 µM 1.i. The Hill coefficient obtained
from the dose-response curve was 1.32, and the IC50 was 320
nM. Error bars are standard errors, and eight experiments
were included in the figure.
Subsequent molecular modeling of these compounds
showed that a reasonable pocket exists that would
accommodate a DHP or ID and provide significant
electronic and hydrogen-bonding interactions. Docking
calculations were carried out using INSIGHT 2000 on
a SGI Indigo II workstation. Calculations utilized the
cvff force field. Sequence homologation was performed
in accordance with previously published work of Striess-
nig.13 The closed conformation of the K+ channel (KcsA)
was used as a scaffold to mimic the Ca2+ channel in
silico. The coordinates for the KcsA crystal structure
was downloaded from the Brookhaven Protein Database
(accession number 1BL8).4 Alterations to the KcsA
sequence are as follows (numbered according to R1C-
II18): in IIIS5, Thr-1039, Gln-1043; in IIIS6, Tyr-1152,
Ile-1153, Ile-1156, Phe-1158, Phe-1159, Met-1161; in
IVS6, Tyr-1463, Met-1464, Ile-1471, Asn-1472.
1.j-1.n .) (Figure 1); thus, the ID SAR is essentially the
opposite of the well-known DHP SAR for the aryl group.
We felt that this observation required some further
examination. An explanation for our newly observed
SAR would be more convincing if it supported both the
domain interface binding sites proposed by Catterall
and the computational model recently reported by
Schleifer yet also encompassed the apparent fundamen-
tal differences between the typical 4-aryl-DHPs exem-
plified by “nifedipine” and our new synthetic analogues,
specifically 1.j-1.n .
Of the analogues that were examined by single-crystal
X-ray diffractometry, all were found to adopt an O-exo
conformation in the solid state (Chart 1). The barrier
to rotation of 1.l was estimated using INSIGHT 2000
on a Silicon Graphics Indigo II workstation. The rota-
tional barrier calculations were performed using the
torsion force constraint in the Discover module. The
bond in question was rotated through 360 intervals (1°
increments) using a force constant of 10. After each
rotation, the structure was subjected to 1000 steps of
minimization, or until an rms value of 0.01 was reached,
using the VA09A algorithm. The results of the confor-
mational searches were examined with the Analysis
module by constructing a graph of total energy vs
dihedral angle. Rotation about three features were
considered: (i) the barrier about the ester, (ii) the ring
juncture between the heterocyclic rings, and (iii) the ring
juncture between the isoxazole and the C-3 aryl moiety.
These barriers were calculated to be 6.0, 15.0, and 24.6
kcal/mol, respectively. The magnitude of the barrier to
rotation about the heterocyclic ring juncture would
suggest that rotation would be plausible under physi-
ological conditions. To experimentally address this
point, we examined the variable-temperature (VT) NMR
behavior of 1.l. The signal for the C-2 and C-6 methyl
groups of the dihydropyridine, as well as that for the
We considered numerous modes of docking in the
putative receptor site. We first considered the lowest
energy ID O-exo conformer but reluctantly discarded
this notion based on two lines of reasoning: (1) the para
halogens, especially the larger ones, would appear to
disrupt the N-H hydrogen bonding, a feature critical
to all known DHP SAR; (2) none of the binding modes
we considered for O-exo IDs provided any reasonable
explanation for the lower activity of the o-Cl analogue
1.l. Since our VT NMR data suggested that the O-endo
conformers would be reasonably attainable at body
temperature, Figure 3 shows two views of molecule 1.l
in this conformation interacting with Striessnig’s do-
main spanning interface of the calcium channel. Note
the H bond formed between Gln-1043 and the amino
hydrogen, the electronic interactions between Met-1464
and the isoxazole ring, and the π-π interactions be-
tween Tyr-1463 and the C-5′ aryl ring.
Scheme 2 schematically illustrates that meta and
para substituents of 1.l would easily be accommodated
within the lipophilic channel as oriented, but the 3′-aryl
ortho groups, on being forced from planarity from the
isoxazolyl group, would interfere with either the critical
Y or M residue of IVS6.13 It is worthy of note that this

90 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1
Ch a r t 1. X-ray Crystal Structures for Four IDsa
Zamponi et al.
a
The isoxazolyl- group approximately bisects the 1,4-dihydropyridine ring in each case (87.2-101.8°). All were found to be in the
O-exo conformation, that is, presenting the C-3′ aryl group roughly parallel to the 1,4-dihydropyridine ring. This forces the C-3′ aryl to
adopt an approximately orthogonal relationship to the isoxazole (88.6-108.4°). The esters were ap/sp for 1.k , 1.m , and 1.n and were
sp/sp for 1.l. Neither ∑F nor ∑τ correlated with IC50
.
F igu r e 2. Variable-temperature 500 MHz NMR study of 1.l.
picture does indeed support both the domain interface
spanning picture proposed by Catterall and the com-
putational model recently reported by Schleifer. Figure
3b indicates that in this binding mode, the 4-isoxazolyl
is easily accommodated. A superimposed “nifedipine”
analogue (not shown) would have ample room for ortho
and meta substituents but very limited tolerance for
para.
The 4-isoxazolyl-1,4-dihydropyridines reported in this
work represent a potentially new series of antihyper-

4-Isoxazolyl-1,4-dihydropyridines
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1 91
F igu r e 3. SGI InsightII docking study of 1.l with the binding pocket of the calcium channel, constructed according to Striessnig
(ref 13). (a) This view best illustrates the 4-aryl moiety of 1.l flanked by M and Y. (b) The isoxazole bears no “para” group and
thus is a good bioisostere for the o-nitrobenzene group of “nifedipine”. In this view, it is clear that a superimposed para-substitiuted
4-aryl (not shown) would not easily be accommodated.
tensives given their unique structure-activity relation-
ship, which in turn probably gives rise to their previ-
ously observed lack of negative ionotropic activity,12
which points to their potential advantage in circum-
venting the serious issue of myocardial infarction as-
sociated with the 4-aryl DHPs. Furthermore, we report
a working hypothesis that reconciles some aspects of
known SAR of classic DHPs, with certain apparent
discrepancies exhibited by our own new synthetic IDs.
There are numerous important issues to be explained
further; prominent among them are the role of all of
the essential binding amino acid residues and the
explanation of the formidable changes in conformation
of the channel during opening and closing. We present
this drug-receptor picture as a starting point for the
elucidation and, hopefully, eventual better understand-
ing of the DHP sensitive L-type calcium channel.
Our future research will explore these hypotheses,
and we will report on our progress in due course.
Exp er im en ta l Section
All reagents were purchased from Aldrich Chemical Co. and
purified by recrystallization or distillation, as appropriate,
before use. All chromatography solvents were distilled prior
to use. Radial chromatography was performed on a Harrison
Associates chromatotron, on silica gel, unless otherwise noted.
Flash chromatography was performed using Baker flash gel,
and in-house compressed air. NMR spectra were recorded on
Bruker AC 200 (200 MHz 1H, 50 MHz 13C) and IBM Bruker
1
AF300 (300 MHz H, 75 MHz 13C) spectrometers at ambient
temperature in CDCl3 unless otherwise noted. Chemical shifts

92 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1
Zamponi et al.
dried over anhydrous sodium sulfate, and concentrated in
vacuo to give a yellow oil, 8.32 g (94%). 1H NMR: δ 6.99-7.59
(m, 4H). The other hydroximinoyl chlorides were prepared in
a similar manner, with stirring for varying lengths of time,
and were pure enough to carry on to the next step.
Sch em e 2. Schematic Representation of the Domain
Interface Model for the DHP Binding Site of Catterall5,6
in Which the 4-Isoxazolyl-DHP Conformationally Adapts
To Orient the Lipophilic Aryl Moiety toward the
Intracellular Pore of the Ca2+ Channela
3-Ch lor oben zoh yd r oxim in oyl Ch lor id e, 4.k . Yield 90%,
1
mp 65-67 °C (lit.14 mp 65-67 °C). H NMR: δ 8.87 (bs, 1 H),
7.85-7.38 (m, 4H).
In the case of the methoxy analogues, no entraining HCl(g)
was added.
2-Meth oxyben zoh yd r oxim in oyl Ch lor id e, 4.n . Yield
78%, mp 135-139 °C (lit.14 mp 112-112.5 °C, so possibly a
1
syn-anti mixture). H NMR: δ 9.61 (bs, 1 H), 7.58-6.73 (m,
4H), 3.89 (s, 3H), 3.42 (s, <1 H), 3.32 (s, <1 H).
4-Meth oxyben zoh yd r oxim in oyl Ch lor id e. Yield 98%,
1
mp 64-73 °C (lit.30 mp 87.5-88.5 °C). H NMR: δ 9.69 (bs, 1
H), 7.72 (d,2H), 6.91 (d,2H), 3.78 (s,3H).
Nitr ile Oxid e Cycloa d d ition . P r ep a r a tion of C-3 Su b-
stitu ted Isoxa zole Ester s 5. Eth yl 3-(2-Ch lor op h en yl)-5-
m eth ylisoxa zole-4-ca r boxyla te, 5.l. To a stirred solution of
the enamine of ethyl acetoacetate (7.61 g, 41.4 mmol) and
triethylamine (3.5 mL) in absolute ethanol (60 mL) at O °C
under nitrogen was added a solution of 2-chlorobenzohydrox-
iminoyl chloride (7.89 g, 41.5 mmol) in absolute ethanol (65
mL) dropwise via an addition funnel. The resulting yellow
solution was warmed to room temperature and stirred over-
night. The ethanol was then removed under vacuum. The
residue was then taken up in ethyl ether (150 mL), and this
solution was washed with 1 M HCl (75 mL) and water (3 × 50
mL), dried over anhydrous sodium sulfate, filtered, and
concentrated to give a dark-orange liquid that was distilled
using the Kugelrohr apparatus (0.04 mmHg, oven temp 100
°C) to give the ester as a yellow liquid, 8.27 g (75%). Esters
prepared in this manner were of sufficient purity to carry on
to the next step. An analytical sample was obtained by radial
chromatography (2 mm silica plate; eluent, 97% hexane, 2%
a
The orientation shown schematically below is based on the
DHP binding model of Schliefer,8 and the approximate 40° tilt of
the strands is based on the potassium channel structure of
MacKinnon.4
are reported downfield from an internal tetramethylsilane
(TMS) standard. Mass spectra were obtained on a VG Micro-
mass 70/70 HS mass spectrometer using electron impact (EI)
or chemical ionization (CI) as indicated. Combustion analyses
were performed by Desert Analytics, Tucson, AZ.
Syn th esis of C-3 Su bstitu ted Isoxa zole Dih yd r op y-
r id in es. P r ep a r a tion of Oxim es 3: 2-Ch lor oben za ld e-
h yd e Oxim e, 3.l. To a stirred mixture of 2-chlorobenzalde-
hyde, 2.l (7.24 g, 51.5 mmol), in 13 mL of water, 13 mL of 95%
ethanol, and 25 g of ice was added hydroxylamine hydrochlo-
ride (3.90 g, 56 mmol). Next, 5 mL of 50% NaOH was added,
turning the mixture strongly alkaline as indicated by litmus
paper. The mixture was stirred for 1 h, acidified with concen-
trated HC1 (to strongly acidic as indicated by litmus paper),
and extracted with CH2Cl2 (3 × 50 mL). The combined organic
extracts were washed with water (2 × 50 mL), dried over
anhydrous sodium sulfate, filtered, and concentrated to give
a white solid: yield 7.70 g (96%, crude), mp 70-74 °C. 1H
NMR: δ 9.10 (bs, 1H, NOH), 8.57 (s, 1H, CHNOH), 7.81-7.23
(m, 4H, aryl). Oxime obtained in this manner was sufficiently
pure for use in the next step. The other oximes were prepared
using the same method.
1
CH2Cl2, 1% EtOAc; Rf ) 0.41). H NMR: δ 7.38-7.17 (m, 4H,
aryl), 4.02 (q, 2H, CO2CH2CH3), 2.62 (s, 3H, CH3), 0.95 (t, 3H,
CO2CH2CH3). 13C NMR: δ 174.9, 161.3, 160.5, 133.89, 130.7,
130.5, 129.0, 128.5, 126.3, 109.6, 60.3, 13.4, 12.9. IR (NaCl,
CDC13) cm-1: 3130, 3040, 2960, 2915, 2880, 1700, 1585, 1420,
1300, 1240, 1150, 1090, 1050, 1020, 900. MS (CI) m/z: 266 (M
+ 1). Anal. Calcd for C13H1 2ClNO3: C, 58.77; H, 4.5 5; N, 5.27.
Found: C, 58.82; H, 4.41; N, 5.09.
The other esters were prepared in the same manner.
Eth yl 3-(3-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r box-
1
yla te, 5.k . Yield 75%, mp 45-46 °C. H NMR: δ 7.65-7-33
(m,4H), 4.25 (q,2H), 2.75 (s,3H), 1.25 (s,3H). 13C NMR: δ 176.1,
161.6, 161.2, 133.7, 130.1, 129.7, 129.5, 129.2, 127.5, 108.3,
60.8, 13.8, 13.5. IR (NaCl, CDC13) cm-1: 2980, 2910, 1715,
1595, 1565, 1435, 1310, 1250, 1145, 1105, 915. MS (El) m/z:
265 (M+). Anal. Calcd for C13Hl2ClNO3: C, 58.77; H, 4.55; N,
5.27. Found: C, 58.95; H, 4.49; N, 5.13.
3-Ch lor oben za ld eh yd e Oxim e, 3.k . Yield 69% (crude), mp
62-66 °C (lit.14 mp 68-69.5 °C). H NMR: δ 9.12 (br. s, 1 H),
1
8.11 (s, 1 H), 7.77-7.44 (m, 4H).
Eth yl 3(4-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r box-
yla te, 5.j. Yield 74% crude, mp 53-57 °C (lit.15 mp 59-600
°C). 1H NMR: δ 7.52 (d, 2H), 7.40 (d,2H), 4,22 (q,2H), 2,71
(s,3H), 1.22 (s,3H).
Eth yl 3-(2-Meth oxyp h en yl)-5-m eth ylisoxa zole-4-ca r -
boxyla te, 5.n . Yield 51%, mp 70-76 °C. 1H NMR: δ 7.43-
7.39 (m, 2H, aryl), 7.04-6.92 (m, 2H, aryl), 4.13 (q, 2H,
CO2CH2CH3), 3.75 (s, 3H, OCH3), 2.69 (s, 3H, CH3), 1.09 (t,
3H, CO2CH2CH3).
2-Meth oxyben zaldeh yde Oxim e, 3.n . Yield 100% (crude),
mp 90-91.5 °C (lit.14 mp 91.5-93 °C). 1 H NMR: δ 9.81 (bs, 1
H), 8.47 (s, 1 H), 7.65-6.88 (m, 4H), 3.85 (s, 3H).
3-Meth oxyben za ld eh yd e Oxim e. Yellow liquid. Yield
100% (crude). 1H NMR: δ 8.14 (s, 1 H), 7.32-6.92 (m, 4H),
3.80 (s, 3H).
4-Met h oxyben za ld eh yd e Oxim e. Dark-orange liquid.
Yield 100% (crude). 1H NMR: δ 9.36 (bs, 1H), 8.12 (s,1 H),
7.50 (d, 2H), 6.90 (d, 2H), 3.74 (s, 3H).
Eth yl 3-(4-Meth oxyp h en yl)-5-m eth ylisoxa zole-4-ca r -
boxyla te. Yield 66%, mp 58-64 °C. 1 H NMR: δ 7.57 (dq 2H),
6.93 (dq 2H), 4.23, (qq 2H), 3.83 (s, 3H), 2.70 (s, 3H), 1.24 (t,
3H).
P r ep a r a tion of C-3 Su bstitu ted Isoxa xole Alcoh ols 6.
3-(2-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r bin ol, 6.l. A
solution of the isoxazole ester (5.31 g, 20 mmol) in 125 mL of
THF (freshly distilled over Na/benzophenone) was stirred
under nitrogen at room temperature in a 250 mL three-neck
round-bottom flask fitted with a reflux condenser. LiA1H4 (0.76
g, 20 mmol) was added in portions via a dry addition tube.
The resulting mixture was stirred overnight and warmed to
P r ep a r a tion of Hyd r oxim in oyl Ch lor id es, 4. 2-Ch lo-
r oben zoh yd r oxim in oyl Ch lor id e, 4.l. To a stirred solution
of the oxime (7.27 g, 46.7 mmol) in 40 mL of DMF was added
one-fifth (6.24 g, 46.7 mmol) of N-chlorosuccinimide (NCS).
After 10 min, the temperature of the solution did not rise, so
a small amount of HCl(g) was added using an H2SO4/NaCl
generator. The temperature of the solution rose to 35 °C and
was kept below this by periodic ice bath cooling. The rest of
the NCS was added slowly in portions, and the solution was
stirred overnight. The solution was then poured into 250 mL
of ice-water and extracted with ethyl ether (3 × 100 mL). The
ether extracts were washed with water (2 × 100 mL), filtered,

4-Isoxazolyl-1,4-dihydropyridines
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1 93
1
room temperature. Sodium sulfate decahydrate (6.40 g, 20
mmol) and 50 mL of THF were added, and the mixture was
stirred for 1 h. It was then filtered through Celite and
concentrated in vacuo to give the isoxazole alcohol as a honey-
like material, 3.93 g (93%). An analytical sample was obtained
by radial chromatography (2 mm silica gel plate; eluent, 70%
92% yield, mp 85-86 °C. H NMR: δ 9.66 (s, 1H, CHO), 7.52
(t, 2H, aryl), 7.06 (d, 1H, aryl), 7.00 (d, 1H, aryl), 3.81 (s, 3H,
OCH3), 2.75 (s, 3H, CH3). 13C NMR: δ 185. 6, 174.5, 160.1,
156.8, 132.0, 130.9, 121.1, 116.1, 115.7, 111.1, 55.4, 12-9. IR
(NaCl, CDC13) cm-1: 2900, 2820, 1685, 1590, 1510, 1470, 1440,
1385, 1295, 1270, 1250, 1185, 1165, 1110, 1050, 1025, 975, 910.
3-(4-Meth oxyp h en yl)-5-m eth ylisoxa zole-4-ca r boxa ld e-
h yd e. Tan solid (white solid after chromatography), 55% yield,
1
hexane, 15% CH2C12, 15% EtOAc; TLC Rf ) 0. 14). H NMR:
δ 7.50-7.31 (m, 4H, aryl), 4.36 (s,2H, CH2OH), 2.47 (s, 3H,
CH3), 2.32 (bs, 1H, OH). 13C NMR: δ 168.2, 161.0, 133.3, 131.7,
130.9, 129.8, 128.3, 126.9, 114.6, 53.9, 11.3. IR (NaCI, CDC13)
cm-1: 3370, 3130, 3040, 2900, 2860, 1605, 1430, 1365, 1270,
1240, 1195, 1115, 1080, 1050, 1020, 1005, 895. All other
isoxazole alcohols were prepared in the same manner and were
sufficiently pure to carry on to the next step.
1
mp 59-63 °C. H NMR: δ 9.96 (s, 1 H), 7.66 (d, 2H), 7.03 (d,
2H), 3.87 (s, 3H), 2.76 (s, 3H). 13C NMR: δ 184.6, 176.8, 161.8,
161.4, 130.4, 119.3, 115.1, 114.4, 55.4, 13.0. IR (NaCl, CDC13)
cm-1: 3030, 2980, 2940, 2920, 2820, 1670, 1595, 1560, 1510,
1420, 1285, 1170, 1100, 1020, 965, 900.
P r ep a r a t ion of C-3 Ar ylisoxa zolyl-1,4- d ih yd r op y-
r id in es 1. All new isoxazole dihydropyridines were prepared
in a manner similar to that described in detail below for 1.l.
For the 4-chloro, dimethyl ester 1.c, methyl acetoacetate was
used instead of ethyl acetoacetate.
3-(3-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r bin ol, 6.k .
1
Yield 81%, mp 63-66 °C. H NMR: δ 7.78 (m, 1H, aryl), 7.65
(m, 1 H), 7.39-7.33 (m, 2 H), 4.48 (s, 2 H), 3.22 (bs, 1 H), 2.40
(s, 3 H). 13C NMR: δ 168.9, 161.49, 134.79, 130.79, 130.19,
129.8, 128.29, 126.49, 112.99, 53.29, 11.0. IR (NaCl, CDC13)
Dim eth yl 2,6-Dim eth yl-(5-m eth yl-3-[p-ch lor op h en yl]-
isoxa zol-4-yl)-1,4-d ih yd r op yr id in e-3.5-d ica r boxyla te. 1.c.
Fluffy, pale-yellow crystals, 57% yield, mp 161-162 °C. 1H
NMR: δ 7.42-7.23 (m, 4H), 5.45 (bs, 1 H), 4.99 (s, 1 H), 3.48
(s, 3H), 2.39 (s, 6H, CO2CH3), 2.01 (s,6H). 13C NMR: δ 167.6,
166.8, 162.4, 144.1, 134.8, 130.9, 130.8, 129.6, 128.0, 127.9,
cm-1
: 3450, 30509, 29109, 2860, 1605, 1585, 15509 1420,
13809, 1360, 1270, 1250, 1200, 1120, 10859, 10709, 900.
3-(4-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r bin ol, 6.j.
Yellow solid, 98% yield, mp 68-71 °C. (lit.16 mp 70-71 °C).
1H NMR: δ 7.73 (d, 2H), 7.41 (dq 2H), 4.52 (s, 2H), 2.47 (s,
3H).
119.7, 101.0, 50.9, 29.3, 19.1, 11.2. IR (NaCl, CDC13) cm-1
:
3-(2-Meth oxyph en yl)-5-m eth ylisoxazole-4-car bin ol, 6.n .
Yellow oil, 87% yield. 1H NMR: δ 7.47-7.36 (m, 2H, aryl),
7.06-6.98 (m, 2H, aryl), 4.30 (s, 2H, CH2OH), 3.79 (s, 3H,
OCH3), 2.67 (bs, 1H, OH), 2.45 (s, 3H, CH3). 13C NMR: δ
167.49 160.29 156.59 130.9, 120.8, 117.8, 114.69 111.1, 55.4,
53.7, 10.8. IR (NaCl, CDC13) cm-1: 3390, 2935, 2915, 2860,
2815, 1610, 1590, 1570, 1500, 1450, 1430, 1280, 1250, 1200,
1170, 1150, 1095, 1040, 900.
3-(4-Meth oxyph en yl)-5-m eth ylisoxazole-4-car bin ol. Yel-
low solid, 94% yield, mp 123-126 °C. 1H NMR: δ 7.72 (dd,
2H), 6.97 (dd, 2H), 4.53 (s, 2H), 3.82 (s, 3H), 2.43 (s, 3H). 13C
NMR (acetone-d6): δ 168.0, 161.9, 160.9, 129.7, 122.2, 114.2,
113.4, 54.9, 52.7, 10.2. IR (NaCl, CDC13) cm-1: 3300, 2940,
2900, 1595, 1510, 1450, 1420, 1370, 1285, 1240, 1170, 1125,
1100, 1025, 900.
3420, 3125, 2970, 2930, 1675, 1600, 1460, 1420, 1290, 1270,
1205, 1180, 1130, 1090, 1040, 1020, 900. MS (FAB) m/z: 416
(M+), 418 (M + 2). Anal. Calcd for C21H21ClN2O5: C, 60.51; H,
5.08; N, 6.72. Found: C, 60.55; H, 4.87; N, 6.61.
Dieth yl 2,6-Dim eth yl-(5-m eth yl-3-[o-ch lor oph en yl]isox-
a zol-4-yl)-l,4-d ih yd r op yr id in e-3.5-d ica r boxyla te, 1.l. A
stirred solution of the isoxazole aldehyde 7.l (2.50 g, 11.28
mmol), ethyl acetoacetate (2.94 g, 22.56 mmol, 2 equiv), 4 mL
of NH3(aq), and 35 mL of 95% ethanol was refluxed for ca. 48
h. It was then concentrated under vacuum to give a yellow
solid that was recrystallized from absolute ethanol (rinsing
the crystals with cold ethyl ether) to give the dihydropyridine
1.l as pale-yellow cubic crystals: 2.26 g (45%), mp 169-170
1
°C. H NMR: δ 7.20 (m, 4H, aryl), 5.96 (bs, 1H, NH), 4.93 (s,
1H, methine), 4.11 (qd, 4H, CO2CH2CH3), 2.58 (s, 3H, CH3),
1.87 (s, 6H, CH3), 1.24 (t, 6H, CO2CH2CH3). 13C NMR: δ 167.2,
165.29, 161.3, 145.0, 134.6, 131.7, 130.0, 129.7, 128.8, 125.6,
P r ep a r a tion of C-3 Su bstitu ted Isoxa zole Ald eh yd es
7. 3-(2-Ch lor op h en yl)-5-m eth ylisoxa zole-4-ca r boxa ld e-
h yd e, 7.l. To a stirred solution of pyridinium chlorochromate
(PCC) (6.90 g, 32.0 mmol, 2 equiv) and anhydrous magnesium
sulfate (10 g) in 100 mL of CH2C12 was added a solution of
the isoxazole alcohol (3.58 g, 16.0 mmol) in 75 mL of CH2Cl2.
Upon this addition, the color of the solution turned im-
mediately from orange to brown. The solution was stirred for
2 h and then taken up in 200 mL of ethyl ether, filtered
through a 3 in. pad of silica gel in a sintered glass funnel, and
concentrated in vacuo to give a green viscous liquid that was
Kugelrohr distilled (0.07 mmHg, oven temp 100-110 °C) to
give a yellow waxy solid, 2.76 g (82%). An analytical sample
was obtained by radial chromatography (2 mm silica gel plate;
eluent, 70% hexane, 15% CH2C12, 15% EtOAc; TLC Rf ) 0.17),
118.8, 99.6, 59.4, 29.1, 18.8, 14.3, 11.3. IR (NaC1, CDC13) cm-1
:
3320, 3090, 2980, 2930, 2900, 1650, 1450, 1380, 1325, 1300,
1275, 1210, 1170, 1110, 1055, 1025, 910. MS (EI) 444 (M+).
Anal. Calcd for C23H25ClN2O5: C, 62.09; H, 5.66; N, 6.36.
Found: C, 62.25; H, 5.58; N, 6.30.
Diet h yl 2,6-Dim et h yl-(5-m et h yl-3-[m -ch lor op h en yl]-
isoxa zol-4-yl)-1,4-d ih yd r op yr id in e-3,5-d ica r boxyla te, 1.k .
1
Greenish cubic crystals, 39% yield, mp 160-161 °C.
H
NMR: δ 7.41-7.2 7 (ms, 4 H), 5.3 6 (bs, 1 H), 5.02 (s, 1 H),
4.17-3.93 (m, 4 H), 2.5 0 (s, 3H), 1.99 (s, 6H), 1.20 (t, 6H). 13
C
NMR: δ 167.3, 166.1, 162.5, 144.3, 133.5, 132.7, 129.6, 128.9,
128.6, 127.8, 119.5, 100.7, 59.8, 29.3, 19.1, 14.4, 11.4. IR (NaCl,
CDC13) cm-1: 3410, 3310, 3070, 2960, 2910, 2880,1660, 1600,
1550, 1465, 1430, 1370, 1315, 1260, 1200, 1160, 1085, 1040,
1
giving a white solid, mp 62-66 °C. H NMR: δ 9.71 (s, 1H,
CHO), 7.57-7.37 (m, 4H, aryl), 2.80 (s, 3H, CH3). 13C NMR: δ
184.9, 175.5, 160.6, 133.5, 131.7, 131.6, 129.0, 127.2, 126.6,
115.8, 13.0. IR (NaCl, CDC13) cm-1: 3075, 2840, 2760, 1675,
1575, 1430, 1305, 1365, 1270, 1235, 1040, 1015, 955, 890. All
other aldehydes were prepared in the same manner and were
of sufficient purity to carry on to the next step.
1010, 900. MS (EI) m/z: 444 (M+) Anal. Calcd for C23H25
-
ClN2O5: C, 62.09; H, 5.66; N, 6.36. Found: C, 61.99; H, 5.57;
N, 6.16.
Dieth yl 2,6-Dim eth yl-(5-m eth yl-3-[p-ch lor oph en yl]isox-
azol-4-yl)-1,4-dih ydr opyr idin e-3.5-dicar boxylate, 1.j. Pale-
1
yellow crystals, 47% yield, mp 149-151 °C. H NMR: δ 7.35
3-(3-Ch lor op h en yl)-5-m et h ylisoxa zole-4-ca r b oxa ld e-
h yd e, 7.k . Tan solid (white solid after chromatography), 93%
(s, 4 H), S. 4 8 (bs, 1 H), 5.0 2 (s, 1 H), 4.13-3.96 (m, 4 H), 2.4
8 (s, 3 H), 1. 9 8 (s, 6H), 1.18 (t, 6H). 13C NMR: δ 166.4, 165.2,
161.6, 143.1, 133.7, 129.7, 128.4, 126.8, 118.7, 99.9, 58.8, 28.3,
18.0, 13.4, 10.4. IR (NaCl, CDC13) cm-1: 3410. 3300, 3200,
3070, 2960, 2810, 1655, 1595, 1465, 1410, 1365, 1320, 1260,
1200, 1160, 1080, 1040, 1020, 900. MS (EI) m/z: 445 (M + 1),
447 (M + 3). Anal. Calcd for C23H25ClN2O5: C, 62.09; H, 5.66;
N, 6.36. Found: C, 62.07; H, 5.62; N, 6.28.
1
yield, mp 81-82 °C. H NMR: δ 9.96 (s, 1 H), 7.72-7.41 (m,
4H), 2.80 (s, 3H). 13C NMR: δ 183.8, 177.3, 160.9, 134.9, 130.6,
130.2, 129.0, 128.9, 127.2, 115.0, 12.9. IR (NaCl, CDC13) cm-1
:
3130, 3040, 2810, 2730, 1670, 1550, 1425, 1370, 1280, 900.
3-(4-Ch lor op h en yl)-5-m et h ylisoxa zole-4-ca r b oxa ld e-
h yd e, 7.j. Yellow solid, 50% yield, mp 94-96 °C (lit.16 mp 95
1
°C). H NMR: δ 9.85 (s, I H), 7.58 (dd, 2H), 7.37 (dd, 2H),
Dieth yl 2,6-Dim eth yl-(5-m eth yl-3-[-o-m eth oxyp h en yl]-
isoxa zol-4-yl)-1,4-d ih yd r op yr id in ed ica r boxyla te, 1.n . Yel-
2.67 (s, 3H).
1
3-(2-Meth oxyp h en yl)-5-m eth ylisoxa zole-4-ca r boxa ld e-
h yd e, 7.n . Yellow solid (white solid after chromatography),
low crystals, 35% yield, mp 204-205 °C. H NMR: δ 7.28 (td,
1H, aryl), 7.06 (dd, 1H, aryl), 6.87 (q, 2H, aryl), 4.92 (s, 1H,

94 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1
Zamponi et al.
methine), 4.87 (bs, 1H, NH), 4.17-4.03 (qd, 4H, CO2CHCHA),
3.65 (s, 3H, OCHA), 2.54 (s, 3H, CH3), 1.90 (s, 6H, CH3), 1.25
(t, 6H, CO2CH2QH3). 13C NMR (300 MHz, CDC13): δ 15 167.5,
164.7, 161.2, 157.7, 144.4, 131.5, 130.2, 119.6, 119.5, 119.2,
109.9, 100.2, 59.6, 55.4, 29.4, 19.1, 14.6, 11.5. IR (NaCl, CDC13)
cm-1: 3390, 3250, 3035, 2930, 2880, 2850, 2780, 1640, 1440,
1340, 1275, 1220, 1180, 1140, 1070, 1025, 1000, 880. MS (CI)
m/z: 441 (M + 1). MS (EI) m/z: 439 (M - 1). Anal. Calcd for
3-(5-Ch lor o-2-m eth oxyph en yl)-5-m eth ylisoxazole-4-car -
boxa ld eh yd e, 7.m . To a stirred solution of PCC (3.93 g, 18.24
mmol, 2 equiv) and anhydrous magnesium sulfate (15 g) in
CH2C12 (100 mL) was added a solution of the isoxazole alcohol
6.m (2.00 g, 9.12 mmol) in CH2C12 (40 mL). Upon this addition,
the solution turned immediately from orange to brown. The
mixture was stirred for 2 h and then taken up in ethyl ether
(250 mL), filtered through a 3 in. silica gel pad in a sintered
glass funnel, and concentrated under vacuum to give a tan
solid, 1.71 g (86%). An analytical sample was obtained by
radial chromatography (2 mm silica gel plate; eluent, 98%
C
24H28N2O6: C, 65.44; H, 6.41; N, 6.36. Found: C, 65.58; H,
6.48; N, 6.28.
Ring Chlorination was observed when methoxyaryls were
entrained with HCl(g).
1
hexane, 2% EtOAc), giving a white solid, mp 124-125 °C. H
NMR: δ 9.66 (s, 1H, CHO), 7.51-6.95 (m, 3H, aryl), 3.81 (s,
3H, OCH3), 2.76 (s, 3H, CH3). 13C NMR: δ 184.0, 174.0, 158.0,
154.6, 130.7, 129.6, 125.1, 116.8, 114.8, 111.5, 54.9, 12.0. IR
(NaCl, CDC13) cm-1: 3000, 2950, 2920, 2820, 2725, 1675, 1490,
1450, 1395, 1365, 1270, 1235, 1170, 1110, 1015, 900.
5-Ch lor o-2-m eth oxyben zoh ydr oxim in oyl Ch lor ide 4.m .
To a stirred solution of the 2-methoxybenzaldehyde oxime (4.50
g, 29.8 mmol) in 25 mL of DMF was added about one-fifth of
(3.98 g, 29.8 mmol) NCS. After 10 min, the temperature of
the solution did not rise, so a small amount of HCl(g) from an
H2SO4/NaCl generator was added, causing the temperature
of the solution to rise to 40 °C. The rest of the NCS was added
in portions, with intermittant ice bath cooling such that the
temperature of the solution was kept below 35 °C. The solution
was stirred overnight, then poured into 300 mL of ice-water
and extracted with ethyl ether (2 × 100 mL). The combined
ether extracts were washed with water (3 × 50 mL), dried over
anhydrous sodium sulfate, filtered, and concentrated under
vacuum to give a yellow solid, 4.38 g (78%), mp 135-139 °C.
1H NMR: δ 9.61 (bs, 1H, NOH), 7.58-6.73 (ms, 3H, aryl), 3.89
(s, 3H, OCH3), 3.42 (s, <1H), 3.32 (s, <1H) (likely a syn-anti
mixture).
Dieth yl 2,6-Dim eth yl-(5-m eth yl-3-(2′m eth oxy-5′-ch lo-
r o)p h en ylisoxa zol-4-yl)-1,4-d ih yd r op yr id in e-3.5-d ica r -
boxyla te, 1.m . A stirred solution of the isoxazole aldehyde
7.m (1.50 g, 6.91 mmol), ethyl acetoacetate (1.80 g, 13.82 mmol,
2 equiv), and NH3(aq) (3 mL) in 20 mL of 95% ethanol was
refluxed for approximately 48 h, upon which time crystalliza-
tion began to take place. The ethanol was then removed under
vacuum, and the solid was recrystallized from absolute ethanol
to give a yellow-green powder: 1.18 g (3 9%), mp 248-250 °C.
1
H NMR: δ 7.32-6.78 (m, 3H, aryl), 4.93 (s, 1H, methine),
4.71 (br. s, 1H, NH), 4.11 (qd, 4H, CO2CH2CH3), 3.66 (s, 3H,
OCH3), 2.5 4 (s, 3H, CH3), 1.99 (s, 6H, CH3), 1.25 (t, 6H,
CO2CH2CH3). MS (EI) m/z: 475 (M+), 477 (M + 2).
Eth yl 3-(5-Ch lor o-2-m eth oxyph en yl)-5-m eth ylisoxazole-
4-ca r boxyla te, 5.m . To a stirred solution of the enamine of
ethyl acetoacetate (3.80 g, 20.75 mmol) and 2 mL of trieth-
ylamine in 30 mL of absolute ethanol was added a solution of
the hydroximinoyl chloride 4.m (3.85 g, 20.75 mmol) in 30 mL
of absolute ethanol dropwise via an addition funnel, with
stirring under nitrogen at 0 °C. The orange-brown solution
was warmed to room temperature and stirred, followed by
reflux for approximately 20 h. It was then concentrated under
vacuum, taken up in 100 mL of ethyl ether, washed with 1 M
HCI (1 × 75 mL) and water (3 × 50 mL), dried over anhydrous
sodium sulfate, filtered, and concentrated under vacuum to
give a brown semisolid residue that was Kugelrohr distilled
to give a light-yellow solid: 3.48 g (64%), mp 79-81 °C. An
analytical sample was obtained by radial chromatography (2
mm silica gel plate; eluent, 70% hexane, 15% CH2C12, 15%
Tr a n sien t Tr a n sfection of Ca lciu m Ch a n n el Su bu n its
in to tsa -201 Cells. Human embryonic kidney tsa-201 cells
were grown in standard DMEM medium, supplemented with
10% fetal bovine serum and 0.5 mg/mL penicillin streptomycin
(5% CO2, 37 °C). Cells were grown to 85% confluency, split
with trypsin-EDTA and plated on round glass coverslips at
10% confluency 12 h before transfection. J ust prior to trans-
fection, the medium was replaced with fresh DMEM, and a
standard calcium phosphate protocol was used to transiently
transfect the cells with cDNA constructs encoding for calcium
channel R1C, â1b, and R2-δ subunits and green fluorescent
protein as an expression marker (7, 7, 7, and 4 µg, respec-
tively). After 12 h, cells were washed with fresh medium,
recover for 12 h and then incubated at 28 °C in 5% CO2 for
2-4 days prior to recording.
1
EtOAc), giving a white solid, mp 88 °C. H NMR: δ 7.29
Electr op h ysiology. Immediately prior to recording, indi-
vidual coverslips were transferred to a 3.5 cm polystyrene
culture dish containing external recording solution comprising
5 mM BaCl2, 1 mM MgCl2, 10 mM HEPES, 40 mM TEACl, 10
mM glucose, and 87 mM CsCl (pH 7.2). Patch pipets (Sutter
borosilicate glass, BF150-86-15) were pulled using a Sutter
P-87 microelectrode puller, fire-polished using a Narashige
microforge, and showed a typical resistance of about 3-4 MΩ.
Recording pipets were filled with 108 mM CsMS, 4 mM MgCl2,
9 mM EGTA, and 9 mM HEPES (pH 7.2). Cells were selected
on the basis of their GFP expression as visualized by UV-
induced fluorescence. Seals were formed directly in the
external recording solution, and after seal rupture, cells were
dialyzed for 5-10 min prior to recording. Whole-cell patch
clamp recordings were performed using an Axopatch 200B
amplifier (Axon Instruments, Foster City, CA) linked to a
personal computer equipped with pCLAMP v 6.0. Series
resistance was compensated by 85% to minimize voltage
errors. To assess the DHP block of noninactivated channels,
currents were typically elicited from a holding potential of -80
mV to various test potentials using Clampex software (Axon
Instruments). To assess inactivated channel block, cells were
held at a more depolarized voltage (-30 mV) or steady-state
inactivation curves were recorded in the presence and absence
of the blockers. Drugs were dissolved in DMSO at 10 mM stock
concentrations and diluted into the external recording solution.
The drug-containing recording solutions were perfused directly
into the vicinity of the cells using a gravity-driven home-built
(dd, 1H, aryl), 7.09 (d, 1H, aryl), 6.80 (d, 1H, aryl), 4.15 (qd,
2H, CO2CH2CH3), 3.75 (s, 3H, OCH3), 2.71 (s, 3H, CH3),
1.12 (t, 3H, CO2CH2CH3). 13C NMR: δ 174.4, 161.9, 159.1,
156.2, 130.7, 130.1, 125.3, 119.9, 111.8, 110.1, 60.4, 55.7,
13.9, 13.0. IR (NaCl, CDC13) cm-1: 2960, 2910, 2890, 2820,
1700, 1585, 1490, 1435, 1370, 1275, 1235, 1120, 1090, 1025,
900. MS (EI) m/z: 295 (M+), 297 (M + 2). Anal. Calcd for C1
4H1 5NO4: C, 56.86; H, 4.77; N, 4.74. Found: C, 56.66; H, 4.66;
N, 4.62.
3-(5-Ch lor o-2-m et h oxyp h en yl)-5-m et h ylisoxa zole-4-
ca r bin ol, 6.m . To a stirred solution of the isoxazole ester 5.m
(3.29 g, 12.59 mmol) in THF (40 mL, freshly distilled over Na/
benzophenone) under nitrogen at 0 °C was added LiAlH4 (500
Mg, 13-18 mmol) in portions via a dry addition funnel. The
resulting mixture was warmed to room temperature and
stirred overnight. Sodium sulfate decahydrate (5.16 g, 16
mmol) and 40 mL of THF were added, and the mixture was
stirred for 2 h. It was then filtered through Celite and
concentrated under vacuum to give a thick orange oil, 2.40 g
(87%). An analytical sample was obtained by radial chroma-
tography (2 mm silica gel plate; eluent, 90% hexane, 5%
CH2C12, 5% EtOAc), giving a yellow liquid. 1H NMR:
δ
7.42-7.36 (m, 2H, aryl), 6.97-6.92 (m, 1H, aryl), 4.33 (s, 2H,
CH2OH), 3.81 (s, 3H, OCH3), 2.48 (s, 3H, CH3), 2.36 (bs, 1H,
OH). 13C NMR: δ 168.0, 159.3, 155.4, 130.9, 130.7, 126.0, 119.6,
114.7, 112.6, 59.1, 53.9, 11.0. IR (NaCl, CDC13) cm-1: 3360,
3050, 2920, 2820, 1610, 1490, 1445, 1425, 1250, 1235, 1110,
1010, 900.

4-Isoxazolyl-1,4-dihydropyridines
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1 95
microperfusion system. Data were usually filtered at 1 kHz
and recorded directly into a personal computer.
parameters, and hydrogen coordinates. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
Da ta An a lysis. Data are analyzed using Clampfit (Axon
Instruments) and Sigmaplot 4.0 (J andel Scientific). Steady-
state inactivation curves were fitted with the Boltzman
equation
Refer en ces
(1) (a) Zamponi, G. W. Antagonist Sites of Voltage-Dependent
Channels. Drug. Dev. Res. 1997, 42, 131-143. (b) Striessnig, J .
Pharmacology, Structure and function of cardiac L-type calcium
channels. Cell Physiol. Biochem. 1999, 9, 242-269. (c) 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. (d) Natale, N. R. Learning from
the Hantzsch Synthesis. In Chemical Innovation; American
Chemical Society: Washington, DC, 2001.
(2) (a) New Analyses Regarding the Safety of Calcium-Channel
Blockers: A Statement for Health Professionals from the Na-
tional Heart, Lung, and Blood Institute, September 1, 1995.
<http://www.nhlbi.nih.gov/new/press/cutlrccb.txt>. (b) Psaty, B.
M.; Heckbert, S.; Koepsell, T.; Siscovick, D. S. Raghunathan, T.
E.; Weiss, N. S.; Rosendal, F. R.; Lemaitre, R. N.; Smith, N. L.;
Wahl, P. W.; Wagner, E. H.; Furberg, C. D. The Risk of
Myocardial Infarction Associated with Antihypertensive Drug
Therapy. J . Am. Med. Assoc. 1995, 274, 620-625. (c) Pahor, M.;
Guralnik, J . M.; Ferruci, L.; Corti, M. C.; Salive, M. E.; Cerhan,
J . R.; Wallace, R. B.; Havlik, R. J . Calcium Channel Blockade
and Incidence of Cancer in Aged Populations. Lancet 1996, 348,
493-497. (d) Fitzpatrick, A. L.; Daling, J . R; Furberg, C. D.;
Kronmal, R. A.; Weissfeld, J . L. Use of Calcium Channel
Blockers and Breast Carcinoma Risk in Postmenopausal Women.
Cancer 1997, 80, 1438-1447.
1
Ipeak (normalized) ) 1 + exp(
)z
V - Vh
25.6
where V and Vh are respectively the conditioning and the half-
inactivation potential and z is a slope factor. Current-voltage
relations were fitted according to the equation
1
Ipeak ) (V - Erev)G
Va - V
(
)
)
1 + exp
(
S
where Erev is the reversal potential, Va is the half-activation
potential, G is the maximum slope conductance, and S is a
slope factor that is inversely proportional to the effective gating
charge.
1
NMR. The H and 13C NMR high-resolution and solid-state
spectra were obtained with a Bruker DRX500 spectrometer.
The signal assignments were performed on the basis of a series
of 2D experiments with z-gradient selection: 1H-1H DQF
COSY, 1H-13C HMQC, and 1H-13C HMBC.17 The low-tem-
perature NOE experiments were performed in the rotating
frame (2D ROESY) with a mixing time of 250 ms.17 The solid-
state 13C spectra were obtained under cross-polarization magic
angle spinning (CP-MAS) conditions with a Bruker 5 mm CP-
MAS probe. The spinning rates were 8 and 14.5 kHz. These
spectra were referenced to the external secondary standard
(solid adamantane, 38.6 and 28.8 ppm from TMS).
X- r a y Cr ysta llogr a p h y. Data were collected on a Siemens
SMART CCD (charge-coupled device) diffractometer equipped
with an LT-2 low-temperature apparatus operating at 193 K.
A suitable crystal was chosen and mounted on a glass fiber.
Data were measured using ω scans of 0.3° per frame for 10 s
such that a hemisphere was collected. A total of 1271 frames
were collected with a final resolution of 0.90 Å. The first 50
frames were re-collected at the end of the collection to monitor
and correct for decay. Cell parameters were retrieved using
SMART software and were refined using SAINT software on
all reflections. Data reduction was performed using the using
SAINT software, which corrects for Lp and decay. Absorption
corrections were applied using XEMP supplied by Siemens’
SHEXTL-PC software.
(3) J oint National Committee. The Sixth Report of the J oint
National Committee on Prevention, Detection, Evaluation, and
Treatment of High Blood Pressure (J NC-VI). Arch. Intern. Med.
1997, 157, 2413-2446.
(4) Doyle, D. A.; Cabral, J . M.; Pfuetner, R. A.; Kuo, A.; Gulbis, J .
M.; Cohen, S. L.; Chait, B. T.; MacKinnon, R. The Structure of
the Potassium Channel: Molecular Basis of K+ Conduction and
Selectivity. Science 1998, 280, 69-77.
(5) Peterson, B. Z.; Tanada, T. N.; Catterall, W. A. Molecular
Determinants of High Affinity Binding in L-Type Calcium
Channels. J . Biol. Chem. 1996, 271, 5293-5296.
(6) Hockerman, G. H.; Peterson, B. Z.; Sharp, E.; Tanada, T. N.;
Scheuer, T.; Catterall, W. A. Construction of a High-Affinity
Receptor Site for Dihydropyridine Agonist and Antagonists by
Single Amino Acid Substitutions in a Non-L-Type Ca2+ Channel.
Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 14906-14911.
(7) Natale, N. R.; Rogers, M. E.; Staples, R. J .; Triggle, D. J .;
Rutledge, A. Lipophilic 4-Isoxazolyl-1,4-dihydropyridines: Syn-
thesis and Structure-Activity Relationships. J . Med. Chem.
1999, 42, 3087-3093.
(8) Schleifer, K.-J . Stereoselective Characterization of the 1,4-
Dihydropyridine Binding Site at L-Type Calcium Channels in
the Resting State and the Opened/Inactivated State. J . Med.
Chem. 1999, 42, 2204-2211.
(9) Spaetgens, R. L.; Zamponi, G. W. Multiple Structural Domains
Contribute to Voltage-Dependent Inactivation of Rat Brain R1E
Calcium Channels. J . Biol. Chem. 1999, 274, 22428-22436.
(10) Natale, N. R.; Triggle, D. J .; Palmer, R. B.; Lefler, B. J .; Edwards,
W. D. 4-Isoxazolyl-1,4-dihydropyridines: Biological, Theoretical,
and Structural Studies. J . Med. Chem. 1990, 33, 2255-2259.
(11) Palmer, R. B.; Andro, T. M.; Natale, N. R.; Anderson, N. H.
Conformational Preferences and Dynamics of 4-Isoxazolyl-1,4-
dihydropyridine Calcium Channel Antagonists As Determined
by Variable Temperature NMR and NOE Experiments. Magn.
Reson. Chem. 1996, 34, 495-504.
(12) McKenna, J . I.; Schlicksupp, L.; Natale, N. R.; Willett, R. D.;
Maryanoff, B. E.; Flaim, S. F. Cardioactivity and solid-state
structure of two 4-isoxazolyldihydropyridines related to the
4-aryldihydropyridine calcium-channel blockers. J . Med. Chem.
1988, 31, 473-476.
Ack n ow led gm en t. This work was funded in part
by an operating grant to G.W.Z. from the Heart and
Stroke Foundation of Alberta and Northwest Ter-
ritories. S.C.S. holds studentship awards from the
Alberta Heritage Foundation for Medical Research
(AHFMR) and the Canadian Institutes for Health
Research (CIHR). G.W.Z. is an AHFMR Senior Scholar
and a CIHR Investigator. N.R.N. thanks the National
Institute of Neurological Disease and Stroke for Grant
2R15-NS38444. T.M.A., V.H., and J .K.N. acknowledge
support from the Malcolm and Carol Renfrew Scholar-
ship endowment, University of Idaho. We thank the
Environmental Biotechnology Institute, University of
Idaho for use of the SGI computational facility.
(13) Huber, I.; Wappl, E.; Herzog, A.; Mitterdorfer, J .; Glossman, H.;
Langer, T.; Striessnig, J . Conserved Ca2+ Antagonist Binding
Properties and Putative Folding Structure of a Recombinant
High Affinity Dihydropyridine Binding Domain. Biochem. J .
2000, 347, 829-826.
(14) Liu, K.; Shelton, B. R.; Howe, R. K. A particularly convenient
preparation of benzohydroximinoyl chlorides (nitrile oxide pre-
cursors). J . Org. Chem. 1980, 45, 3916-3918.
(15) Doyle, F. P.; Hanson, J . C.; Long, A. A.; Nayler, J . H. C.; Stove,
E. R. Derivatives of 6-Aminopenicillanic Acid. Part IV. Penicillins
from 3- and 5-phenylisoxazole-4-carboxylic Acids and Their Alkyl
and Halogen Derivatives. J . Chem. Soc. 1963, 5838-5845.
(16) Heck, R.; Oflenloch, R.; Wolf, R. Eine Verbesserte Methode zur
Synthese Substituierter Isoxazol-4-carbaldehyde. Synthesis 1990,
62-64.
Su p p or tin g In for m a tion Ava ila ble: Low-temperature
NOE of 1.l, X-ray crystallographic files for 1.k , 1.l, and 1.m
and 1.n , including crystal data and structure refinement,
atomic coordinates, and equivalent isotropic displacement
parameters, bond lengths and angles, anisotropic displacement

96 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 1
Zamponi et al.
(17) Claridge, T. D. High Resolution NMR Techniques in Organic
Chemistry; Pergamon: New York, 1999.
(18) Snutch, T. P.; Tomlinson, W. J .; Leonard, J . P.; Gilbert, M. M.
Distinct Calcium Channels Are Generated by Alternative Splic-
ing and Are Differentially Expressed in the Mammalian CNS.
Neuron 1991, 7, 45-57.
(19) (a) Oˆ ki, M. The Chemistry of Rotational Isomers; Springer: New
York, 1993. (b) Oˆ ki, M. Recent Advances in Atropisomerism. Top.
Stereochem. 1983, 14, 1. Oˆ ki’s definition for atropisomers sug-
gests 22.3 kcal/mol at 300 K and 26.2 kcal/mol at 350 K as values
where the isomers exist as isolable entities.
J M020354W