Synthesis and pharmacologic evaluation of 2-endo-amino-3-exo- isopropylbicyclo[2.2.1]heptane: A potent imidazoline1 receptor specific agent

Full text of DOI:10.1021/jm960012o

© Copyright 1996 by the American Chemical Society
Volume 39, Number 6
March 15, 1996
Communications to the Editor
During the course of our studies, we discovered AGN
Syn th esis a n d P h a r m a cologic Eva lu a tion
of 2-en d o-Am in o-3-exo-
isop r op ylbicyclo[2.2.1]h ep ta n e: A P oten t
Im id a zolin e1 Recep tor Sp ecific Agen t
192403 (3), the first agent that is equipotent to mox-
onidine in imidazoline receptor binding assays and
nonpotent in adrenergic binding assays. To our sur-
prise, this potent, I1 specific agent was devoid of all
physiologic response. Compound 3 did not affect blood
pressure, intraocular pressure, or induce sedation in our
animal models. Further, 3 did not antagonize the
effects of agonists thought to mediate their physiologic
effects through imidazoline receptors nor did 3 potenti-
ate the effects of putative imidazoline receptor agonists.
This may suggest that we have identified a new binding
site. Alternatively, this may imply that the imidazoline
site defined by this binding assay is not a functional
receptor but rather simply a nonfunctional binding site.
Stephen A. Munk,* Ronald K. Lai, J ames E. Burke,
Premilla N. Arasasingham, Alexander B. Kharlamb,
Cynthia A. Manlapaz, Edwin U. Padillo,
Mercy K. Wijono, Dain W. Hasson,
Larry A. Wheeler, and Michael E. Garst
Allergan Pharmaceuticals, 2525 Dupont Drive,
Irvine, California 92715
Received J anuary 3, 1996
The imidazoline (I receptors) receptor family has
received extensive attention since its identification by
Bousquet et al., Coupry et al., and Ernsberger et al. in
the 1980s.1 Recently, agmatine (1, Figure 1) has been
proposed to be the endogenous ligand for the I recep-
tors.2 There is some question whether agmatine is the
sole endogenous agent.3 These receptors do not interact
with catecholamines including norepinephrine (2), the
natural ligand for adrenoceptors. The imidazoline fam-
ily of receptors does, however, recognize many imida-
zoline-containing R2 adrenergic agents including cloni-
dine (4) and moxonidine (5). The I1 receptor subtype
in particular has been suggested to mediate a variety
of physiologic functions including the reduction of blood
pressure.4 To date, the evidence for the imidazoline
binding site functioning as a receptor has been indirect
as all of the ligands described interact with other
receptors including the R2 receptors. Thus, agents
including clonidine and moxonidine interact with both
receptor families precluding confirmation of the specific
receptor responsible for the observed physiologic re-
sponse. Our interest in this area arose from our search
for selective R2 agonists possessing no imidazoline
receptor activity. Our hypothesis was that enhanced
receptor selectivity could be achieved through prepara-
tion of a conformationally restrained framework. An
R2 agonist without imidazoline receptor activity should
reduce elevated intraocular pressure, a condition often
associated with glaucoma, without untoward side effects
including reduction of blood pressure.
Compound 3 was prepared as shown in Scheme 1.
2-Methylpropionaldehyde was condensed with nitro-
methane in the presence of potassium hydroxide5 and
then treated with methanesulfonyl chloride in the
presence of triethylamine6 to afford (E)-1-nitro-3-meth-
yl-1-butene as the sole isomer in 50% yield. Diels-
Alder reaction of the nitro olefin and freshly cracked
cyclopentadiene in dichloroethane in a sealed tube at
80 °C afforded the [2.2.1] system bearing an endo nitro
group and an exo isopropyl group in 60% yield as the
major product.7 This was confirmed by a single-crystal
X-ray analysis of the subsequent reduction product, 3.
The reduction was conducted in ethanol and an atmo-
sphere of hydrogen using catalytic Pd-C to afford a
quantitative yield of the amine. Those conditions
reduced both the nitro group and the double bond in a
single step.
Compound 3 was evaluated in a variety of binding
assays. Binding studies were conducted using adren-
ergic membrane preparations to determine the subtype
selectivity. The R1 preparation was from human brain
and has a mixed population of subtypes. The R2
membranes were all from CHO cells transfected with
appropriate constructs: the R2A was the human C-10
receptor; the R2B was the rat RNG receptor; the R2C was
the human C-2 receptor. The imidazoline I1 preparation
was from the bovine ventrolateral medulla.8 Binding
affinity for the I1 binding site was determined by the
displacement of [3H]clonidine binding with concurrent
0022-2623/96/1839-1193$12.00/0 © 1996 American Chemical Society

1194 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 6
Communications to the Editor
monkeys, 3 proved to have no affect on blood pressure
at a dose as high as 5000 µg/kg while clonidine reduced
blood pressure at a dose of 17 µg/kg and moxonidine
reduced blood pressure at a dose of 167 µg/kg. When
the animals were pretreated with 3 and then treated
with clonidine, there was no affect on clonidine’s ability
to reduce blood pressure. The agent also had no affect
on blood pressure upon direct administration of the
agent to the central nervous system of rabbit.9 We
demonstrated that the agent penetrated the cornea.10
We therefore attempted to reduce elevated intraocular
pressure with topical administration of this agent to the
rabbit.11 Again, there was no observed physiologic
response to the agent. The agent also proved to be
nonsedating in both monkey and rat.
We have discovered an agent that is specific for
binding to the I1 binding site. On the basis of the
potency and selectivity of this agent in binding assays,
we evaluated it using in vivo models for reduction of
blood pressure, reduction of intraocular pressure, and
sedation. Compound 3 was devoid of both agonist and
antagonist activity in all functional assays used. Our
data suggest that the I1 site is a unique binding site. It
may not, however, be a functional receptor. These
conclusions may be supported by recent pharmacologic
studies suggesting that the effects of rilmenidine,
another putative I1 agent, may also be mediated by R2
adrenoceptors.12
F igu r e 1.
Sch em e 1a
Ack n ow led gm en t. We thank Professor Paul Erns-
berger (Division of Hypertension, Case Western Reserve
School of Medicine) for confirming our binding selectiv-
ity and Dr. Marilyn Olmstead (Department of Chemis-
try, U. C. Davis) for the X-ray analysis.
3
Su p p or tin g In for m a tion Ava ila ble: Details of the agent
synthesis, X-ray crystallographic analysis, and binding assays
(12 pages) are available. Ordering information is given on any
current masthead page.
a
Reagents: (i) CH3NO2, 3 N methanolic KOH; (ii) MsCl, Et3N;
(iii) cyclopentadiene, Cl(CH2)2Cl; (iv) Pd-C, H2, EtOH.
Refer en ces
Ta ble 1. Binding Affinities
Ki (nM, ( SEM)a
(1) (a) Bousquet, P.; Feldman, J .; Schwartz, J . Central Cardiovas-
cular Effects of Alpha-Adrenergic Drugs: Differences Between
Catecholamines and Imidazolines. J . Pharmacol. Exp. Ther.
1984, 230, 232-236. (b) Coupry, I.; Podevin, R. A.; Dausse, J .-
P.; Parini, A. Evidence for Imidazoline Binding Sites in Baso-
lateral Membranes from Rabbit Kidney Biochem. Biophys. Res.
Commun. 1987, 147, 1055-1060. (c) Ernsberger, P.; Meeley, M.
P.; Mann, J . J .; Reis, D. J . Clonidine Binds to Imidazole Binding
Sites as well as R2-Adrenoceptors in the Ventrolateral Medulla.
Eur. J . Pharmacol. 1987, 134, 1-13.
agent
R1
2600 ( 290 66 ( 4.2 78 ( 6.5
>100000 >20000 >20000
R2A
R2B
R2C
I1
2
3
4
5
63 ( 7.2
>20000
>100000
42 ( 17
8.9 ( 2.2
510 ( 110 3.8 ( 0.4 8.3 ( 0.2 30 ( 2
>30000
150 ( 25 1000 ( 34 2000 ( 200 56 ( 8.7
a
In all cases where the value is listed as “greater than”, the
SEM is not reported as all measurements were greater than the
number reported.
(2) Li, G.; Regunathan, S.; Barrow, C. J .; Eshraghi, J .; Cooper, R.;
Reis, D. J . Agmatine: An Endogenous Clonidine-Displacing
Substance in the Brain. Science 1994, 263, 966-969.
(3) (a) Atlas, D. Identifying Clonidine-Displacing Substance. Science
1994, 266, 462-463. (b) Reis, D. J .; Li, G.; Regunathan, S.;
Barrow, C. J .; Cooper, R. Response to D. Atlas. Science 1994,
266, 463.
masking of the R2 receptors with norepinephrine. The
results of this evaluation are presented in Table 1.
Compound 3 had an affinity of 42 nM for the I1
preparation. This proved comparable to that of mox-
onidine, an agent thought to mediate its effect on blood
pressure through the I1 site.8 Compound 3 was only
5-fold less potent than clonidine at the I1 binding site.
In sharp contrast, the agent was devoid of binding
activity at any R adrenoceptor evaluated. 3 is the first
ligand reported to bind selectively at the I1 binding site.
While the data are not presented, the agent is devoid
of affinity for the I2 binding sites as well. Clonidine was
potent in all R2 and in I1 binding assays while moxoni-
dine displayed moderate affinity for the R2A receptor and
had a high affinity for the I1 binding site.
(4) (a) Gomez, R. E.; Ernsberger, P.; Feinland, G.; Reis, D. J .
Rilmenidine Lowers Arterial Pressure via Imidazole Receptors
in Brainstem C1 Area. Eur. J . Pharmacol. 1991, 195, 181-191.
(b) Reis, D. J .; Regunathan, S.; Meeley, M. P. Imidazole Recep-
tors and Clonidine-Displacing Substance in Relationship to
Control of Blood Pressure, Neuroprotection, and Adrenomedul-
lary Secretion. Am. J . Hypertens. 1992, 5, 51S-57S. (c) Erns-
berger, P.; Westbrooks, K. L.; Christen, M. O.; Scha¨fer, S. G. A
Second Generation of Centrally Acting Antihypertensive Agents
act on Putative I1-Imidazoline Receptors. J . Cardiovasc. Phar-
macol. 1992, 20 (Suppl. 4), S1-S10. (d) Michel, M. C.; Erns-
berger, P. Keeping an Eye on the I Site: Imidazoline Preferring
Receptors. Trends Pharmacol. Sci. 1992, 13, 369-370.
(5) Noland, W. E. 2-Nitroethanol. In Organic Syntheses; Baumgar-
ten H. E., Ed.; J ohn Wiley and Sons: New York, 1973; Collect.
Vol. 5, pp 833-838.
(6) Melton, J . S.; McMurry, J . E. A New Method for the Dehydration
of Nitro Alcohols J . Org. Chem. 1975, 40, 2138-2139.
Upon intravenous administration to cynomologus

Communications to the Editor
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 6 1195
(7) (a) Poos, G. I.; Kleis, J .; Wittekind, R. R.; Rosenau, J . D. Bicyclic
Bases. III. Isomeric 2-Amino-3-phenylnorbornanes. J . Org.
Chem. 1961, 26, 4898-4904. (b) Hartzler, H. E., 3-Isopropyl-
norbornanamine. U.S. Patent 3,514,486, 1970.
(8) Ernsberger, P.; Damon, T. H.; Graff, L. M.; Schaefer, S. G.;
Christen, M. O. Moxonidine: A Centrally Acting Antihyperten-
sive Agent is a Selective Ligand for I1 Imidazoline Sites. J .
Pharmacol. Exp. Ther. 1993, 264, 172-182.
(9) We have previously used this technique with agents that were
efficacious but could not pass the blood-brain barrier. Munk,
S. A.; Harcourt, D.; Burke, J .; Lai, R.; Roberts, D.; Small, D.;
Gluchowski, C.; Manlapaz, C.; Padillo, E.; Kharlamb, A.; Runde,
E.; Wheeler, L.; Garst, M. Synthesis and Biological Evaluation
of AGN 193080. A Potent and Selective Ocular Antihypertensive
Agent. MEDI 193, 209th American Chemical Society National
Meeting, Anaheim, CA, April 1995.
(10) Chun, T.; Munk, S. A.; Lai, R. Unpublished results.
(11) Campbell, W. R.; Potter, D. E. Potential Role of Imidazoline (I1)
Receptors in Modulating Aqueous Humor Dynamics J . Ocular
Pharm. 1994, 10, 393-402.
(12) Evans, R. G.; Anderson, W. P. Renal Effects of Infusion of
Rilmenidine and Guanabenz in Conscious Dogs: Contribution
of Peripheral and Central Nervous System R2-Adrenoceptors. Br.
J . Pharmacol. 1995, 116, 1557-1570.
J M960012O