(S)-4-Methyl-2-(methylamino)pentanoic Acid [4,4-Bis(4-fluorophenyl)butyl]amide hydrochloride, a novel calcium channel antagonist, is efficacious in several animal models of pain [1]

Full text of DOI:10.1021/jm000134n

3474
J . Med. Chem. 2000, 43, 3474-3477
Communications to the Editor
Ch a r t 1. Structure of Compound 1
(S)-4-Meth yl-2-(m eth yla m in o)p en ta n oic
Acid [4,4-Bis(4-flu or op h en yl)bu tyl]a m id e
Hyd r och lor id e, a Novel Ca lciu m Ch a n n el
An ta gon ist, Is Effica ciou s in Sever a l
An im a l Mod els of P a in
Yuntao Song,*,† S. Scott Bowersox,‡
David T. Connor,† David J . Dooley,†
Susan M. Lotarski,† Thomas Malone,†
George Miljanich,‡ Elizabeth Millerman,‡
Michael F. Rafferty,† David Rock,† Bruce D. Roth,†
J oann Schmidt,† Sally Stoehr,† Balazs G. Szoke,‡
Charles Taylor,† Mark Vartanian,† and
Yong-Xiang Wang‡
channel can lead to hypotension. It is believed that a
rapid and profound lowering of arterial pressure tends
to counteract the neuroprotective effects of L-type
calcium channel antagonists. Therefore, it is desirable
to have an antagonist that is selective for N-type
calcium channels over L-type calcium channels to avoid
the potential hypotensive effects.
Neuronal sodium channel antagonists have also been
shown to have neuroprotective and analgesic effects.12
To avoid side effects, it is usually desirable to have
channel modulators that are selective for one subtype
of certain ion channel over other subtypes of the same
ion channel and other ion channels. However, in prac-
tice, it has been difficult to achieve selectivity at a
meaningful level in vitro for small molecule ion channel
modulators. Furthermore, in terms of in vivo efficacy,
it could be beneficial to have a balanced ion channel
antagonist that blocks several type ion channels such
as Na+ and Ca2+ channels.
A number of small-molecule N-type calcium channel
antagonists have been reported in the literature.1 The
studies with those antagonists have been focused on the
aspect of neuroprotection; some of them have been
shown to be neuroprotective in stroke models.1 System-
atic studies on analgesic properties of small-molecule
N-type calcium channel antagonists have not been
reported to our knowledge.
We embarked on a project searching for small mol-
ecules that have potent activity against neuronal N-type
calcium channels and show analgesic and neuroprotec-
tive effects in vivo without cardiovascular side effects.
The SAR studies of several chemical series were pri-
marily guided by IMR32 assay, and all synthetic
analogues were also routinely tested in a L-type channel
calcium flux assay using A10 cells (A10 assay), monitor-
ing the activity against L-type calcium channels found
in smooth muscle cells. Selected compounds were evalu-
ated in electrophysiology to confirm that potent com-
pounds identified in a functional assay are interfering
with the channels’ activity. To assess the CNS bioavail-
ability the compounds were then tested in the audio-
genic seizure model in DBA/2 mice (DBA/2 assay).
Compounds with the best in vitro and in vivo profiles
and desirable physical properties were evaluated in the
formalin assay, Chung model, and other pain models
for analgesic activity and in a head trauma model for
neuroprotection. (S)-4-Methyl-2-(methylamino)pentan-
oic acid [4,4-bis(4-fluorophenyl)butyl]amide hydrochlo-
Pfizer Global Research and Development,
Ann Arbor Laboratories, Pfizer Inc., 2800 Plymouth Road,
Ann Arbor, Michigan 48105, and Elan Pharmaceuticals,
Inc., 3760 Haven Avenue, Menlo Park, California 94025
Received March 21, 2000
In tr od u ction . There is considerable unmet medical
need for treatments of pain and stroke. Voltage-depend-
ent ion channels are attractive drug targets for anal-
gesia and neuroprotection.1,2 Voltage-dependent calcium
channels (VDCC) are involved in regulation of many
physiological functions of excitable cells, such as neuro-
transmitter and hormone release, gene expression, and
muscle contraction.1 Molecular cloning and pharmaco-
logical studies have shown that there are six known
subtypes for VDCC: T, L, N, P, Q, and R.1,3,4 Recently,
the N-type calcium channel has attracted much atten-
tion as a drug target for analgesia and neuroprotection.
Such interests stem in part from the following observa-
tions: First, N-type calcium channels are located at
presynaptic termini of neurons, where they are directly
involved in the regulation of neurotransmitter release.
It is thought that N-type calcium channels play a key
role in a pathological process called excitotoxicity due
to ischemia or hypoglycemia.5 Second, the transmission
of pain signals from periphery to the central nervous
system (CNS) is mediated by N-type calcium channels
located in the spinal cord.5 Third, ziconotide (SNX-111),
a selective N-type calcium channel blocker, was found
to have analgesic activity in animal models6,7 and
neuroprotective activity in focal and global ischemia
models.8-10 It has also shown marked effects in anal-
gesia during clinical trials,6,7 strongly supporting the
above-mentioned concepts. Much has been learned
about L-type calcium channels through the studies with
dihydropyridines (DHPs), a class of L-type calcium
channel antagonists.11 It has been shown that inhibition
of neuronal L-type VDCCs is beneficial for neuropro-
tection. However, inhibition of cardiac L-type calcium
* To whom correspondence should be addressed. Phone: 734-622-
5476. Fax: 734-622-1407. E-mail: yuntao.song@wl.com.
† Pfizer Inc.
‡ Elan Pharmaceuticals, Inc.
10.1021/jm000134n CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/01/2000

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 19 3475
Sch em e 1. Synthesis of Compound 1a
a
(i) HBTU, DMF, (i-Pr)2NEt, rt; (ii) (a) TFA, DCM, (b) HCl/Et2O.
Ta ble 1. In Vitro Properties of Compound 1
ride, compound 1 (Chart 1), was identified as one of the
most interesting compounds in terms of its overall
profile, especially its pharmacological properties. In this
Communication, we report the in vitro studies and the
pharmacological properties of compound 1.
Ch em istr y. Leucine derivative 2 (Scheme 1) was
coupled with primary amine 3 to gave amide 4. Depro-
tection followed by treatment of ethereal HCl provided
compound 1 in 89% yield over two steps.
Biologica l Meth od s. 1. IMR32 Assa y.13 IC50 values
for N-type calcium channel blockade were measured
using the fluorescent Ca2+ indicator Indo-1 in IMR32
human neuroblastoma cells in the presence of 5 µM
nitrendipine to block L-type channels. Fluorescence
measurements were carried out on a Photon Technology
International (PTI) spectrofluorometer.
2. A10 Assa y.14 IC50 values for L-type calcium chan-
nel blockade were measured using Oregon Green 488
Bapta-1 dye in the A10 smooth muscle cell line. Fluo-
rescence measurements were carried out on a fluores-
cent image plate reader (FLIPR).
3. Electr op h ysiology.15 Percent inhibition or EC50
values were determined by the voltage clamp assay
using superior cervical ganglion neurons (SCG neurons).
4. Acetic Acid -In d u ced Wr ith in g Mod el.16 Male
CF-1 mice (23-30 g) were injected intraperitoneally
with acetic acid (0.6%) to induce writhing behavior. The
number of writhing movements in 5 min (starting 7 min
after injection) was recorded.
IMR32 (N-type)
IC50 ( SEM (µM)
A10 (L-type)
IC50 ( SEM (µM)
electrophysiology
IC50a (µM)
1.5 ( 0.071
0.40 ( 0.080
SCGCa: 1.3
SCGNa: 5.1
SCGK: 9.9
a
The IC50 values are the average of two determinations.
housed individually. The number of flinching behaviors
was counted for 60 s during each 10-min interval
afterward for a total of 90 min.
8. Kim & Ch u n g Mod el of Neu r op a th ic P a in .20
Rats were anesthetized, and a tight ligature was placed
on the L5 and L6 dorsal roots distal to the dorsal root
ganglion. 14-21 days later, 50% withdrawal thresholds
were measured using calibrated vonFrey filaments.
Resu lts a n d Discu ssion s. Compound 1 inhibited the
function of neuronal N-type calcium channels with an
IC50 value of 1.5 µM as measured in the IMR32 assay
(Table 1). Its antagonistic activity toward neuronal
N-type calcium channels was confirmed by electrophysi-
ology studies. It had an IC50 value of 1.3 µM against
N-type calcium channels in SCG neurons. It was also
5. DBA/2 Au d iogen ic Seizu r e Mod el.17 Male DBA/
2J mice 21-26 days old were exposed to a sinusoidal
tone (SPL ) 100 dB), swept in frequency between 8 and
16 kHz once each 10 ms for a total of 60 s. Vehicle-
treated mice experienced wild running behavior, fol-
lowed by clonic seizures and a tonic extension seizure
of forelimbs and hindlimbs within 20 s. Absence of tonic
extension was scored as an anticonvulsant effect.
6. F ootp a d In cision Mod el of Hyp er a lgesia .18
Rats were anesthetized, and the footpads and plantaris
muscles were incised on one hindlimb. After the wound
was sutured and the mice had recovered from anesthe-
sia, withdrawal threshold was measured with vonFrey
filaments and heat hyperalgesia was measured by the
method of Hargreaves (score in seconds, 12 s under
control conditions).
active, to a lesser extent, against Na+ channels (IC50
)
5.1 µM) and K+ channels (IC50 ) 9.9 µM) in SCG
neurons. Therefore, compound 1 is a balanced ion
channel antagonist.
Since known potent L-type calcium channel antago-
nists, such as nitrendipine, exhibited hypotensive prop-
erties in vivo which is undesirable for neuron protection
in the case of stroke, compound 1 was evaluated for its
inhibitory activity toward the function of L-type calcium
channels found in smooth muscles. As measured in the
A10 assay, it had an IC50 value of 0.4 µM indicating that
compound 1 is active against L-type calcium channels
but considerably less potent than nitrendipine under the
assay conditions. To assess this compound’s potential
cardiovascular side effects in vivo, compound 1 was
dosed in rat and it did not decrease blood pressure
dramatically following doses up to 30 mg/kg.
7. F or m a lin F ootp a d Test.19 Rats were injected
with 50 µL of 5% formalin into a single hindpaw and

3476 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 19
Ta ble 2. Pharmacological Properties of Compound 1
Communications to the Editor
model
DBA/2a
iv
antiwrithingb
iv
Chung modelc
formalin testd
oral
formalin testd
incisional pain modele
route of
admin
ED50
intrathecal bolus
injection 12.5-100 µg
23 µg
iv infusion 15 and
30 mg/kg over 10 min
16 mg/kg
iv infusion 30 mg/kg
over 10 min
active in mechanical
allodynia
4.05 mg/kg
(95% CL)f
4.46 ( 0.23
87 mg/kg
(95% CL)i
mg/kgg
(95% CL)h
(95% CL)i
a
3 doses, 5 mice/dose. ED50 value is the dose of the compound estimated by statistical methods to prevent tonic seizures in 50% of the
animals. Phenytoin was tested in this assay as a positive control and gave a similar result as reported in the literature.22 3 doses, 6
b
mice/dose. ED50 value is the dose of the compound resulting in 50% reduction in the number of writhes. Morphine was tested in this
assay as a positive control and gave a similar result as reported in the literature.23 c 4 doses, 6 rats/dose. ED50 value is the dose of the
compound resulting in 50% reduction in the amount of foot withdraw responding to repeated mechanical stimuli. Ziconotide was tested
d
in this assay as a positive control and gave a similar result as reported in the literature.24 2 doses, 8 rats/dose. ED50 value is the dose
of the compound resulting in 50% reduction in the number of flinches/minute. Ziconotide was tested in this assay as a positive control
and gave a similar result as reported in the literature.24 e 8 rats/dose. Ziconotide was tested in this assay as a positive control and gave
g
a similar result as reported in the literature.25 f ED50 values were calculated by Probit analysis. ED50 value was calculated by logistic
h
fit of data using nonlinear regression analysis. Statistical analyses were made with one-way and RM ANOVA and post hoc two-tailed
i
t-tests. ED50 (95% CL) was calculated by a nonlinear least-squares fit to the logistic equation.
Compound 1 has exhibited analgesic efficacy in
several animal models of pain. It was bioavailable to
CNS in mice as indicated by the anticonvulsant activity
in the DBA/2 model (ED50 ) 4.05 mg/kg). Compound 1
gave an ED50 value of 4.46 mg/kg in the antiwrithing
assay, suggesting analgesic property for visceral inflam-
matory pain. Compound 1 also showed analgesic efficacy
for neuropathic pain: intrathecal administration of
compound 1 completely reversed mechanical allodynia
in the Chung model (ED50 ) 23 µg). When given by iv,
compound 1 prevented the early phase and more
importantly the late phase formalin-evoked nocifensive
behavior in rats (ED50 ) 16 mg/kg), suggesting efficacy
particularly for delayed inflammatory pain. Further-
more, in an acute postsurgical pain model, compound 1
was shown as an efficacious postoperative analgesic
agent in rats.
pharmacological properties observed with compound 1
are mainly due to N-type calcium channel blockade.
Su m m a r y. Compound 1 was identified as a small-
molecule, nonpeptide, balanced Ca2+, Na+, and K+
antagonist; it is bioavailable to CNS when given iv or
orally. It is the first example of small-molecule N-type
calcium channel antagonists showing potent analgesic
properties in several animal models for different types
of pain, and it did not exhibit hypotensive effects in rat.
The in vivo activity profile of compound 1 differs from
Na+ channel blocking drugs such as phenytoin or
lidocaine. The results shown in this paper extend those
with ziconotide and suggest that a small molecule with
specific neuronal N-type calcium channel blocking activ-
ity could be a novel and useful analgesic. Further
evaluations of PK properties and neuroprotective prop-
erties are in progress.
Both compound 1 and ziconotide have been shown to
be efficacious in the Chung model, formalin test, and
incisional pain model. Ziconotide was far more potent
than compound 1 in these pain models.6 However,
compound 1, being a small, nonpeptide molecule, would
have some advantage over ziconotide in terms of physi-
cal and pharmacokinetic properties and access across
the blood-brain barrier after systemic dosing.
Since compound 1 is a balanced ion channel inhibitor
with activities against N-type and L-type calcium chan-
nels and Na+ and K+ channels, it is almost certain that
the pharmacological properties observed in vivo are not
solely due to its antagonistic activity toward calcium
channels. It is difficult to assess how much its calcium
channel blocking activity has contributed toward the
overall in vivo pharmacological properties observed.
However, comparison of compound 1 and known Na+
channel antagonists such as phenytoin or lidocaine did
reveal some major differences in their pharmacological
profiles in pain models.21 For example, compound 1 was
active in the antiwrithing model, whereas phenytoin
and lidocaine were inactive at a dose of 30 mg/kg (data
not shown). Therefore, it is likely that some of the
Ack n ow led gm en t. We thank Dr. Gary McClusky
and staff at Pfizer Global Research and Development,
Ann Arbor Laboratories, Pfizer Inc., for microanalytical
and spectral data.
Refer en ces
(1) Cox, B.; Denyer, J . C. N-type calcium channel blockers in pain
and stroke. Exp. Opin. Ther. Patents 1998, 8, 1237-1250.
(2) Williams, M.; Kowaluk, E. A.; Arneric, S. P. Emerging Molecular
Approaches to Pain Therapy. J . Med. Chem. 1999, 42, 1481-
1500.
(3) Hofmann, F.; Biel, M.; Flockerzi, V. Molecular basis for Ca2+
channel diversity. Annu. Rev. Neurosci. 1994, 17, 399-418.
(4) Tsien, R. W.; Lipscombe, D.; Madison, D.; Bley, K.; Fox, A.
Reflections on Ca2+-channel diversity, 1988-1994. Trends Neu-
rosci. 1995, 18, 52-54.
(5) Marangos, P. J . U.S. Patent 5,677,288.
(6) Bowersox, S.; Tich, N.; Mayo, M.; Luther, R. SNX-111. Anti-
nociceptive agent. N-type voltage-sensitive calcium channel
blocker. Drugs Future 1998, 23, 152-160.
(7) Mathur, V. S.; McGuire, D.; Bowersox, S. S.; Miljanich, G. P.;
Luther, R. R. Neuronal N-type calcium channels: new prospect
in pain therapy. Pharm. News 1998, 5, 25-29.
(8) Gaur, S.; Newcomb, R.; Rivnay, B.; Bell, J . R.; Yamashiro, D. et
al. Calcium channel antagonist peptides define several compo-
nents of transmitter release in hippocampus. Neuropharmacol-
ogy 1994, 33, 1211-1219.

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 19 3477
(9) Olivera, B. M.; Milijanich, G. P.; Ramachandran, J .; Adams, M.
E. Calcium channel diversity and neurotransmitter release: The
w-conotoxins and w-agatoxins. Annu. Rev. Biochem. 1994, 63,
823-867.
(10) Takizawa, S.; Matsushima, K.; Fujita, H.; Nanri, K.; Ogawa, S.;
et al. A selective N-type calcium channel antagonist reduces
extracellular glutamate release and infarct volume in focal
cerebral ischemia. J . Cereb. Blood Flow Metab. 1995, 15, 611-
618.
(19) Dubuisson D.; Dennis, S. G. The formalin test: a quantitative
study of the analgesic effects of morphine, meperidine, and brain
stem stimulation in rats and cats. Pain 1977, 4, 161-174.
(20) Kim, S. H.; Chung, J . M. An experimental model for peripheral
neuropathy produced by segmental spinal nerve ligation in the
rat. Pain 1992, 50, 355-363.
(21) Lotarski, S. M.; Wang, Y.-X.; Bowersox, S. S.; Song, Y.; Talor,
C. P. PD 176078, a potent blocker of voltage-gated Ca2+ channels
prevents pain-related behaviors in mice and rats. 18th Annual
Scientific Meeting, American Pain Society, Ft. Lauderdale, FL,
Oct 21-24, 1999; Poster 873.
(11) Scriabine, A. Calcium channel antagonists as neuroprotective
agents. Neuroprot. CNS Dis. 1997, 27-51.
(12) Taylor, C. P. Voltage-gated Na+ channels as targets for anti-
convulsant, analgesic and neuroprotective drugs. Curr. Pharm.
Des. 1996, 2, 375-388.
(22) Loescher, W. Genetic animal models of epilepsy as a unique
resource for the evaluation of anticonvulsant drugs. Methods
Findings Exp. Clin. Pharmacol. 1984, 6, 531-547.
(13) Newcomb, R.; Szoke, B.; Palma, A.; Wang, G.; Chen, X. et al.
Selective peptide antagonist of the class E calcium channel from
the venom of the tarantula Hysterocrates gigas. Biochemistry
1998, 37, 15353-15362.
(23) Koster, R.; Anderson, M.; Debeer, E. Acetic Acid for Analgesic
Screening. Fed. Proc. 1959, 18, 412.
(14) Kongsamut, S.; Freedman, S. B.; Miller, R. J . Dihydropyridine
sensitive calcium channels in a smooth muscle cell line. Biochem.
Biophys. Res. Commun. 1985, 127, 71-79.
(24) Bowersox, S. S.; Gadbois, T.; Singh, T.; Pettus, M.; Wang, Y. X.;
et al. Selective N-type neuronal voltage-sensitive calcium chan-
nel blocker, SNX-111, produces spinal antinociception in rat
models of acute, persistent and neuropathic pain. J . Pharmacol.
Exp. Ther. 1996, 279, 1243-1249.
(25) Wang, Y. X.; Pettus, M.; Gao, D.; Phillips, C.; Scott Bowersox,
S. Effects of intrathecal administration of ziconotide, a selective
neuronal N-type calcium channel blocker, on mechanical allo-
dynia and heat hyperalgesia in a rat model of postoperative pain.
Pain 2000, 84, 151-158.
(15) Stoehr, S. J .; Campbell, G. W.; Rock, D. M. Evaluation of N-type
Ca2+ channel currents in cultured rat superior cervical ganglion
neurons. Drug Dev. Res. 1997, 41, 85-90.
(16) Koster, R.; Anderson, M.; Debeer, E. Acetic acid for analgesic
screening. Fed. Proc. 1959, 18, 412.
(17) J ackson, H. C.; Scheideler, M. A. Behavioural and anticonvulsant
effects of Ca2+ channel toxins in DBA/2 mice. Psychopharma-
cology 1996, 126, 85-90.
(18) Brennan, T. J .; Vandermeulen, E. P.; Gebhart, G. F. Charac-
terization of a rat model of incisional pain. Pain 1996, 64, 493-
501.
J M000134N