Medetomidine analogs as α2-adrenergic ligands. 2. Design, synthesis, and biological activity of conformationally restricted naphthalene derivatives of medetomidine

Article abstract of DOI:10.1021/jm9506074

A new series of naphthalene analogs of medetomidine have been prepared and evaluated for their α-adrenergic activities. The methylnaphthyl analog 5a showed significant selectivity for α₂-adrenoceptors and behaved as a partial α₁-agonist in rat aorta preparations. In contrast, the Z-ethylene analog 8c was α₁-selective and behaved as a potent α₁-antagonist. Two rigid analogs (6 and 7) exhibited large differences in binding affinities at α₁- vs α₂-receptors, indicating that the conformational flexibility of 5a is important for the fulfillment of the α-adrenergic activities. Molecular modeling studies began with conformational analysis of classical phenethylamines and medetomidine analogs. Superimposition of medetomidine conformations with those of phenethylamines provided a tentative explanation for the α₂-adrenergic activity of the new imidazoles. A common binding mode for phenethylamines and imidazoles with α₂-adrenoceptors is proposed. Knowledge of the biological properties of the 4-substituted imidazoles, integrated with the information derived from computer-assisted molecular modeling, has provided new insights for the structural and conformational requirements of this class as new adrenergic drugs.

Full text of DOI:10.1021/jm9506074

J . Med. Chem. 1996, 39, 3001-3013
3001
Med etom id in e An a logs a s r2-Ad r en er gic Liga n d s. 2. Design , Syn th esis, a n d
Biologica l Activity of Con for m a tion a lly Restr icted Na p h th a len e Der iva tives of
Med etom id in e
Xiaoyan Zhang,† Xiao-Tao Yao,‡ J ames T. Dalton,‡ Gamal Shams,† Longping Lei,† Popat N. Patil,†
Dennis R. Feller,† Fu-Lian Hsu,§ Cliff George,∇ and Duane D. Miller*,‡
Division of Medicinal Chemistry & Pharmacology, College of Pharmacy, The Ohio State University, Columbus, Ohio 43210,
Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee, Memphis, Tennessee 38163,
U.S. Army Edgewood Research Development and Engineering Center, Aberdeen P.G., Maryland 21010-5423, and
Laboratory of the Structure of Matter, Naval Research Laboratory, Code 6030, Washington, D.C. 20375
Received August 14, 1995X
A new series of naphthalene analogs of medetomidine have been prepared and evaluated for
their R-adrenergic activities. The methylnaphthyl analog 5a showed significant selectivity
for R2-adrenoceptors and behaved as a partial R1-agonist in rat aorta preparations. In contrast,
the Z-ethylene analog 8c was R1-selective and behaved as a potent R1-antagonist. Two rigid
analogs (6 and 7) exhibited large differences in binding affinities at R1- vs R2-receptors, indicating
that the conformational flexibility of 5a is important for the fulfillment of the R-adrenergic
activities. Molecular modeling studies began with conformational analysis of classical
phenethylamines and medetomidine analogs. Superimposition of medetomidine conformations
with those of phenethylamines provided a tentative explanation for the R2-adrenergic activity
of the new imidazoles. A common binding mode for phenethylamines and imidazoles with
R2-adrenoceptors is proposed. Knowledge of the biological properties of the 4-substituted
imidazoles, integrated with the information derived from computer-assisted molecular modeling,
has provided new insights for the structural and conformational requirements of this class as
new adrenergic drugs.
In tr od u ction
anesthetic in animals.6 As such, medetomidine provides
a unique template for the design of R2-adrenoceptors
ligands.
R2-Adrenergic agonists have therapeutic applications
in a variety of disease states.1 Clonidine (1) is the most
familiar antihypertensive drug which possesses agonist
activity at R2-adrenoceptors. However, the use of cloni-
dine and other R2-agonists in the treatment of hyper-
tension has been limited by the presence of sedation.
This side effect has been linked to the stimulation of
the central R2-adrenoceptors. Since the discovery of
clonidine, separation of the centrally mediated antihy-
pertension and sedation has been sought. On the other
hand, rilmenidine (2)2 and moxonidine (3)3 have been
used as new therapeutic agents for the treatment of
hypertension which exhibit little sedation. On the other
hand, the sedative effects of R2-agonists have therapeu-
tic application as adjuncts to general anesthesia.4
Compared to barbiturates and benzodiazepines as pre-
anesthetic medications, R2-agonists provide an advan-
tage in that their anesthetic action can be readily
reversed by R2-antagonist administration.
We previously described the synthesis and pharma-
cological evaluation of a series of novel naphthyl analogs
(5a -e) in selected R1- and R2-adrenoceptor systems.7
Analog 5a possessed an equal potency and higher
selectivity for R2-adrenoceptors than medetomidine.
Moreover, the enantiomers of 5a exhibit a stereoselec-
tive interaction [e.g., (R)-(-)-5a < (S)-(+)-5a ] for the
blockade of the R2A-adrenergic-mediated responses in
human platelets and qualitative differences in activity
on R2-adrenoceptors in electrically stimulated guinea pig
ileum. Replacement of the benzylic methyl group with
other functional groups resulted in decreased activities
at both R1- and R2-adrenoceptors. From the structure-
activity relationships (SAR) derived from this series, we
concluded that the benzylic substituent of the naphthyl
analogs plays a critical role in their potent and selective
interactions with R2-adrenergic binding sites.
Medetomidine (4) is a new synthetic drug which
shows a higher degree of the sedative-analgesic effect
than clonidine.5 It represents a new class of R-agonists
that contains a 4-substituted imidazole ring and is
highly selective at R2-adrenoceptors. Dexmedetomidine,
the S-(+)-enantiomer of medetomidine, was identified
as the active component and has been used as an
† The Ohio State University.
‡ University of Tennessee.
§ U.S. Army Edgewood Research Development and Engineering
Center.
∇ Naval Research Laboratory.
X Abstract published in Advance ACS Abstracts, J une 15, 1996.
S0022-2623(95)00607-8 CCC: $12.00 © 1996 American Chemical Society

3002 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
Further investigations are needed to determine con-
formational preferences of analog 5a that can account
for its high selectivity on R2-adrenoceptors. A knowl-
edge of the active conformation of medetomidine-like
analogs, and therefore of the spatial relationship of their
presumed pharmacophoric group (the aromatic moiety
and the imidazole ring), is basic for the rational design
of new molecules capable of interacting with R2-adreno-
ceptors. Various studies have dealt with the pharma-
cophoric conformational requirements of adrenergic
drugs.8 One of the most frequently used methods is to
design conformationally restricted analogs derived from
a flexible molecule. Such a method has been success-
fully carried out to study the conformation-activity
relationships of the phenethylamine class.9-13 The
present report includes two novel conformationally
restricted analogs of 5a which incorporate the benzylic
methyl group at either the 8-position (compound 6) or
2-position (compound 7) of the naphthyl ring (Figure 1).
The major drawback of this design is that in order to
construct these rigid molecules, additional atoms and
bonds have to be added to the parent compound (5a )
which could potentially affect the chemical and physical
properties of the drug, particularly its hydrophilic
properties in this case. Thus, analogs 5f and 8a -c were
proposed to further explore the SAR with regard to the
benzylic modifications of the naphthyl series. We
envisioned that this approach would provide compounds
with higher potency and afford a better understanding
of the spatial and steric requirements for drug activity
and selectivity at R2-adrenoceptors.
F igu r e 1.
mediate of analogs 5a ,b,f and 8a -c. Ketone 11 was
treated with methylmagnesium bromide in THF fol-
lowed by simultaneous dehydroxylation and deproto-
nation to give analog 8a .
The initial synthetic route employed for the prepara-
tion of analogs 5f and 8b,c is shown in Scheme 2. A
Grignard reaction was applied to introduce an ethyl
group to the benzylic position. Since Grignard additions
are sensitive to steric effects, a competing process
involving reduction of the carbonyl group of 11 was
observed. This competing reduction was minimized by
using benzene as a solvent. The two ethylene analogs
8b,c were obtained by elimination of tertiary alcohol 10c
in 60% of trifluoroacetic aqueous solution. Flash chro-
matography of the crude product afforded the two
geometric isomers (E and Z) which were distinguishable
by their H-NMR spectra. The structure of the Z-isomer
8c was further confirmed by the X-ray crystallography
(Figure 3). These two isomers exhibited great differ-
ences in the hydrogenation reaction. The alkene group
of 8c was easily hydrogenated with Pd/C under moder-
ate pressure, whereas compound 8b was resistant to
hydrogenation under various conditions, most likely due
to steric effects. The overall yields of 8c and 5f were
very low using this approach due to the Z-isomer 8c
being a minor product produced from alcohol 10c. To
circumvent this problem, ketone 11 was converted to
the ethylene derivatives 12a ,b by the Wittig reaction
(Scheme 3). The ratio of Z/E-isomers was dependent
on the reaction temperature, with the Z-isomer 12b as
the major product when the reaction was carried out at
80 °C. N-Deprotection of 12b followed by hydrogenation
gave the saturated analog 5f.
Enormous difficulties were encountered in the prepa-
ration of analog 6. Initially, we attempted to synthesize
analog 6 from the commercially available material
perinaphthenone (13a ) (Scheme 4). Grignard coupling
with 13a gave exclusively the 1,2-addition product 14a
in 49% yield. However, deoxygenation of 14a using
either hydrogenation or triethylsilane under acidic
conditions or radical reduction was not successful.15
Mixtures were usually obtained after the reactions. A
plausible explanation for the failure of these reactions
is due to the high symmetry of the perinaphthenyl
cation (14b) or radical leading to a considerable amount
of resonance stabilization of these entities.16
Ch em istr y
The 4-substituted imidazole analogs were synthesized
in a straightforward manner by utilizing a direct
introduction of an imidazole ring, as recently reported
by Turner and Lindell.14 In this approach, an imidazol-
4-yl anion (9b) was generated at ambient temperature
by addition of ethylmagnesium bromide to the N-
protected 4-iodoimidazole 9a . The imidazol-4-yl anion
was then reacted with a variety of aldehydes and
ketones using CH2Cl2 as a solvent to prevent formation
of 2-alkylated products. Different strategies were ap-
plied in the last step of the synthesis to convert tertiary
alcohols to the corresponding target compounds (Figure
2).
Following Turner’s procedure, 1-naphthaldehyde was
allowed to react with N-tritylimidazol-4-yl anion 9b to
give secondary alcohol 10a in 85% yield (Scheme 1). This
alcohol was easily oxidized by manganese dioxide to
afford ketone 11, which served as the common inter-
An alternate route to 6 included introduction of the
imidazole ring to the saturated ketone 13b (Scheme 5).

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3003
F igu r e 2. Retrosynthesis of 4-substituted imidazoles.
Sch em e 1a
a
(a) I2, KI (20%), NaOH (2 N); (b) Na2SO3, EtOH/H2O (3:7), reflux; (c) TrCl, Et3N, THF; (d) EtMgBr, CH2Cl2; (e) 1-naphthaldehyde; (f)
MnO2, CH2Cl2, reflux; (g) CH3MgBr, THF; (h) TFA, H2O (60%); (i) maleic acid, CH3OH.
Sch em e 2a
F igu r e 3. Thermal ellipsoid plot of 8c (as a maleate salt).
Dashed lines are intramolecular hydrogen bonds, and the open
bonds are linear intramolecular hydrogen bonds. Atom labeled
H13A is a symmetry-generated atom.
ing agents, such as K-Selectride,18 diimide,19 or Wilkin-
son’s hydrogenation,20 were tried and either resulted in
the isolation of unreacted 13a or generated complex
mixtures. An alternate route to 13b was explored using
a multistep synthesis. The synthesis of 3-(1-naphthyl)-
propionic acid (15) was conveniently achieved by allow-
ing R-(chloromethyl)naphthalene to react with a malonic
ester and then subjecting the product to hydrolysis and
facile decarboxylation.21 Intramolecular Friedel-Crafts
a
(a) EtMgBr, benzene; (b) (i) TFA/H2O (60%); (ii) flash chro-
reaction of 15 gave ketone 13b using stannic chloride
as a Lewis acid. Although the yield of the cyclization
step was not significantly higher than that of LAH
reduction, the purification procedure was much easier.
Isolation of ketone 13b from the reaction mixture must
be carried out quickly due to its instability, and 13b had
to be used in the Grignard addition immediately after
preparation.
matography; (c) maleic acid, CH3OH; (d) H2 (40 psi), Pd/C (5%),
EtOH.
Conversion of perinaphthenone 13a to perinaphthanone
13b was described in the 1950s using lithium aluminum
hydride (LAH) with a reported 65% yield.17 However,
we were unable to duplicate the yield, especially for a
large reaction scale. In addition, isolation of the desired
compound 13b from the crude product was tedious, and
the overall yield of the material was low. Other reduc-
After the Grignard reaction, an attempt to eliminate
the hydroxy group of 14c was not successful by using

3004 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
Sch em e 3a
a
(a) Ethyltriphenylphosphonium bromide, NaH, DMSO, 80 °C; (b) AcOH/H2O (90%), 60 °C; (c) H2 (40 psi), Pd/C (5%), EtOH; (d) maleic
acid, CH3OH.
Sch em e 4a
Sch em e 6a
a
(a) Triphenylphosphine, 150 °C; (b) 2-naphthaldehyde, NaH,
THF/DMSO (1:1), 0 °C f room temperature; (c) H2 (15 psi), Pd/C
(10%), EtOH; (d) (i) PCl5, (ii) SnCl4; (e) 9b, CH2Cl2; (f) Et3SiH,
TFA, 50-55 °C; (g) maleic acid, CH3OH.
a
(a) 9b, CH2Cl2; (b) hydrogenation of Et3SiH, TFA or (i) PTC-
Cl, pyridine, CH2Cl2, (ii) nBu3SnH, AIBN, benzene, reflux.
Sch em e 5a
Crafts acylation of butanoic acid 18 into 2,3-dihydro-
4(1H)-phenanthrone (19) produced the phenanthrene
carbon skeleton exclusively.25,26 This ketone was then
reacted with 9b to give tertiary alcohol 20 in 72% yield.
Deoxygenation and N-deprotection was performed in
triethylsilane-trifluoroacetic acid solution at 50-55 °C
in 56% yield.
Biologica l Resu lts
The R1- and R2-adrenoceptor binding affinities of the
4-substituted imidazoles were determined in membrane
fractions of rat brain. Results for the newly synthesized
4-substituted imidazoles are summarized in Table 1.
Extension of the side chain with one more carbon (5f)
resulted in a 13-fold reduction in binding affinity on R2-
adrenoceptors but only a slight decrease (3-fold) at R1-
adrenoceptors as compared to the methyl derivative 5a .
The methylene analog 8a retains good binding affinity
as well as selectivity at R2-adrenoceptors. In contrast,
the ethylene analogs 8b,c exhibited reversed selectivity
on R-adrenoceptors. For example, the Z-ethylene analog
8c was 33-fold more selective for R1-adrenoceptors in
the rat brain membrane, and it is the most potent and
selective R1-binding ligand that we have found in the
4-substituted imidazole series.
The two rigid molecules 6 and 7 exhibited signifi-
cantly different binding affinities at both R1- and R2-
adrenoceptors. Molecule 6 was nearly equipotent to the
parent compound 5a at R2-adrenoceptors; however, it
also displayed high affinity at R1-receptors and thus a
lost of selectivity. Molecule 7 showed a 111-fold reduc-
tion in binding affinity at R2-adrenoceptors but was only
3-fold less potent at R1-adrenoceptors, as compared to
a
(a) CH2(CO2C2H5)2, NaOC2H5, HOC2H5, reflux; (b) NaOH/H2O,
reflux; (c) 180 °C; (d) (i) SOCl2, (ii) SnCl4; (e) 9b, CH2Cl2; (f)
TMSCl-NaI-CH3CN, CH2Cl2; (g) maleic acid, CH3OH.
triethylsilane in a large excess of hot TFA. Purification
of the reaction mixture was complicated by the presence
of unidentified byproducts. The synthesis of target
compound 6 was eventually accomplished successfully
by using Me3SiCl-NaI-CH3CN. This reagent has been
recognized as a Me3SiI equivalent and has been reported
to be an efficient reduction agent of tertiary benzylic
alcohols, although the mechanism of the reaction has
not been elucidated.22
The preparation of analog 7 is outlined in Scheme 6.
The Wittig reagent 16 was obtained from the reaction
of â-chloropropionic acid and triphenylphosphine.23
Treatment of 2-naphthylaldehyde and the Wittig re-
agent 16 with 2 equiv of base gave butenoic acid 17 with
trans configuration.24 Saturated acid 18 was then
produced by hydrogenation. Intramolecular Friedel-

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3005
Ta ble 1. R1- and R2-Adrenergic Binding Affinity of
4-Substituted Imidazoles in Membrane Preparations of Rat
Brain
Ki (nM)a
compd
R1
R2
R2-selectivity ratiob
4
5a
5f
8a
8b
8c
6
1109 ( 36
224 ( 20
574 ( 413
1734 ( 400
387 ( 87
57 ( 19
25 ( 13
13 ( 11
44.36
17.23
3.48
14.21
0.26
0.03
1.67
0.45
44.98
165 ( 52
122 ( 27
1474 ( 586
1890 ( 923
33 ( 5
F igu r e 4. Rotatable bonds defined in the conformational
search of 21 and (S)-5a .
55 ( 12
7
21
655 ( 150
9221 ( 1196
1449 ( 658
205 ( 55
a
Rat brain membrane preparations were incubated with 0.1
nM [3H]prazosin and 0.2 nM [3H]rauwolscine for R1- and R2-
adrenoceptors, respectively. Phentolamine (10 µM) was used to
determine the fraction of nonspecific binding in both assays. Ki
values were determined using the equation as follows: Ki (nM) )
IC50(1 + [L]/Kd), where IC50 ) concentration (nM) of analog which
reduces binding by 50%, [L] ) concentration of radioligand, and
Kd ) equilibrium dissociation constant of the radioligand. Values
b
are average ( standard deviation (n ) 3-9). R2-Selectivity ratio
) Ki(R1)/Ki(R2).
Ta ble 2. Functional Studies of 4-Substituted Imidazoles for
R1- and R2-Adrenoceptor Activities in Rat Aorta (R1-Type) and
Human Platelets (R2-Type)
R1-agonist
EC50
Emax
(nM)b
R2-antagonist
R1-antagonist
KB (nM)c
IC50
potency
ratioe
a
compd
(µM)d
4
33 ( 14 281 ( 43
39 ( 12 153 ( 60
2.9
4.3
215.4
14.3
97.7
1.00
0.69
0.01
0.20
0.03
5a
8c
6
3388
136
38 ( 11 156 ( 54
7
a
Data are expressed as the percent maximal analog response
b
relative to phenylephrine ) 100%. EC50 ) molar 50% effective
concentration. c KB of the antagonist was calculated by the fol-
lowing equation: KB ) [B]/(DR - 1), where [B] is in the presence
d
F igu r e 5. (a) Definition of D and H in phenethylamines. (b)
Low-energy conformations of (-)-R-MeNE (21).
and absence of the antagonist. pIC50 ) -log IC50, where IC50
)
molar 50% inhibitory concentration. e Potency ratio ) IC50(medeto-
midine)/IC50(analog).
partial agonist with a maximal response of 33% and
EC50 value of 281 nM. The parent compound 5a showed
greater potency (EC50 ) 153 nM) and exhibited a higher
maximal contractile response (Emax ) 39%) than me-
detomidine on rat aorta. The Z-isomer (8c) of the
ethylene analogs, the most selective R1-binding ligand,
was an antagonist on R1-receptors in this tissue with a
KB value of 3.4 µM. The two rigid analogs displayed
opposite effects on rat aorta. Analog 6 was a partial
agonist with potency and intrinsic activity equal to that
of the parent compound 5a , whereas analog 7 was a
potent antagonist on R1-receptors with a KB value of 136
nM.
Ta ble 3. Computational Data for 21 from Conformational
Analysis
energya
conformations of 21 (kcal/mol) angleb (deg)
torsion
distancec heightd
(Å)
(Å)
anti-
R-rotamer
â-rotamer
0.01
0.01
0
0
0.12
0.09
178.8
179.0
57.8
57.9
-64.4
-64.4
5.175
5.174
3.896
3.898
4.032
4.030
1.385
1.388
2.409
2.408
2.137
2.147
gauche+ R-rotamer
â-rotamer
gauche- R-rotamer
â-rotamer
a
The lowest-energy conformation is set arbitrarily to 0 kcal/
mol. Torsion angle is defined as C-Câ-CR-N. c The distance is
measured from the center of the phenyl to the nitrogen (see Figure
5). The height is measured from the nitrogen to the plane of the
b
d
Selected 4-substituted imidazoles were incubated in
human platelet-rich plasma in the presence of varying
epinephrine concentrations (30-100 mM). All com-
pounds were concentration-dependent inhibitors of hu-
man platelet aggregation induced by epinephrine. Com-
pared to medetomidine, rigid molecule 6 showed
moderate inhibitory activity at R2-receptors, whereas 7
was a relatively poor inhibitor of platelet aggregation.
These results are consistent with the binding studies
on R2-adrenoceptors in membrane fractions of rat brain.
phenyl ring (see Figure 5).
5a . Therefore, similar to the two ethylene analogs 8b,c,
7 is more selective at R1-receptors. Results obtained
with the two rigid molecules suggest that the confor-
mational flexibility of 5a is important for the fulfillment
of the R-adrenergic activities.
Selected 4-substituted imidazoles were also evaluated
for R-adrenergic functional activities on rat aorta (R1)
and human platelets (R2), respectively, by methods that
were previously reported.7,27 Results are summarized
in Table 2. On rat aorta, the R1-adrenoceptor responses
were expressed as a percentage of the maximum con-
traction to 1 µM of phenylephrine. Medetomidine is a
Molecu la r Mod elin g Stu d ies
In order to obtain a better understanding of our
experimental findings and to support our efforts to
optimize biological activity and R2-receptor selectivity,

3006 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
F igu r e 6. (a) R-Rotamer of 21 (yellow). (b) (S)-5a (green). (c) â-Rotamer of 21 (yellow). (d) Relaxed stereoview of superimposition
of a and b. The rms value is 0.47. (e) Relaxed stereoview of superimposition of b and c. The rms value is 0.30. Nitrogen and
oxygen atoms are blue and red, respectively. The R-methyl group of 21 and the benzylic methyl group of 5a are pink. In d and
e, all hydrogen atoms were hidden for better visualization except for the one of N1 of the imidazole ring and the â-hydroxy group
of 21.
we concluded molecular modeling studies on the me-
detomidine-like analogs.
(-)-R-MeNE (21) is the most active and selective
isomer of R-MeNE at R2-adrenoceptors.31 Addition of a
R-methyl group to NE in the correct configuration
significantly enhances activity at R2-adrenoceptors.
Thus, R-methyl-substituted phenethylamines exhibit
significant R2- vs R1-adrenoceptors selectivity. In the
R-adrenergic binding assay, 21 was 8-fold less potent
at the R2-receptor than medetomidine, but its selectivity
of R2- vs R1-subtypes was similar to that of medetomi-
dine. Since our ultimate goal was to develop highly
selective and potent R2-agonists, we chose 21 as an R2-
agonist template for superimposition with the imidazole
analogs.
(a ) Con for m a tion a l An a lysis of (-)-r-MeNE. Us-
ing SYBYL software, (-)-R-MeNE was subjected to a
full systematic search with three rotatable bonds as
defined in Figure 4. Due to the flexibility of molecule
21, many conformations were energetically possible. The
conformational analysis reflected this by yielding nu-
merous structures within a reasonable energy range.
Concerning the aromatic ring to the amino group
orientation, three different low-energy conformations,
termed anti-, gauche+, and gauche-, were obtained from
the search (Figure 5). Rotation of the phenyl group
Phenethylamines, represented by norepinephrine (NE),
are a well-known class of drugs that interact with
R-adrenoceptors. Numerous studies on the structural
and conformational requirements of phenethylamines
binding at R-adrenoceptors have been published.28,29 The
"bioactive conformation” of phenethylamines interacting
with R-adrenoceptors has been proposed based on the
theoretical calculations, X-ray crystallography, NMR
studies, and conformation-activity studies of confor-
mationally restricted analogs.28,30 Although NE and
medetomidine are chemically different, both consist of
an aromatic moiety some distance from a basic nitrogen
atom and exhibit R-adrenergic activity. We hypoth-
esized that these compounds present their presumed
pharmacophores (aromatic moiety and basic amino
group) situated in a similar spatial relationship, comple-
mentary to that of the active centers of the correspond-
ing receptors. On the basis of this hypothesis, we
examined the molecular superimposability of phenethyl-
amines and imidazoles by using R-methylnorepineph-
rine (R-MeNE) as a model compound of phenethyl-
amines.

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3007
relative to the C1-CR side chain yielded two rotamers
for each conformation, termed R- and â-rotamers. Thus,
six conformational families were revealed from the
computations. Valid conformations derived from the
search were individually minimized, and their param-
eters are summarized in Table 3. The energy differ-
ences of these conformers are not significant (<0.2 kcal/
mol); however, the distances D and H are remarkably
different between the anti-, gauche+, and gauche-
conformations. It has been proposed that phenethyl-
amines adopt a fully extended trans conformation when
interacting with R-adrenoceptors, and the distances D
and H are in the range of 5.1-5.2 and 1.2-1.4 Å,
respectively.30 The anti- conformations of 21, both R-
and â-rotamers obtained from the computations, are in
agreement with the published model. Therefore, both
rotamers of the anti- conformation of 21 were consid-
ered as the template for the superimposition with
imidazoles.
F igu r e 7. Relaxed stereoview of superimposition of R-rotamer
of 21 (yellow) and (S)-medetomidine (green). The rms value
is 0.45. Nitrogen and oxygen atoms are blue and red,
respectively. The R-methyl group of 21 and the benzylic
methyl group of medetomidine are pink. All hydrogen atoms
were hidden for better visualization.
(b) Con for m a tion a l An a lysis of Im id a zoles. (S)-
Medetomidine (4), (S)-5a , and 8c were constructed from
their X-ray atomic coordinates. Since the S-isomer of
medetomidine and 5a are the active isomers for R2-
adrenoceptors, we assumed that the S-isomer of other
imidazole analogs is the optically potent isomer for R2-
adrenoceptors if its racemate is active. The S-isomers
of 5f, 6, and 7 were constructed based on the X-ray
structure of (S)-5a . The structures were first energy
minimized and then subjected to systematic searches
with the rotatable bonds as exemplified in Figure 4.
After the search, the conformations were sorted accord-
ing to energy, and the low-energy conformations were
examined for best fit with 21.
decreased, the affinity at R-adrenoceptors.7 In addition,
the stereochemical configuration of these two classes at
the benzylic carbon does not match.33 This result
suggests that the â-rotamer of 21 may not be the
appropriate conformation of phenethylamines, which is
in agreement with hypotheses advanced by other au-
thors.12,13,34,35
More examples of superimposition of the R-rotamer
of 21 and imidazole analogs are shown in Figures 7-10
with rms values indicated for each fit. From these
superimpositions, we observed the following interesting
features about how the imidazole derivatives may
interact with R2-adrenoceptors. (1) N1 of the imidazole
ring projects to the same direction of the â-hydroxy
group of 21, although they do not overlap with each
other (e.g., Figure 6d). A recent study of site-directed
mutagenesis experiments with R2-adrenoceptors re-
vealed that Ser90 at the second transmembrane domain
could be the potential binding site of the â-hydroxy
group of NE.36 If we assume that N3 of the imidazole
ring serves as the potential binding component with
Asp113 of the receptor, the orientation of N1 of the
imidazole ring makes it possible to play a role like the
â-hydroxy group of NE. (2) The benzylic methyl group
of (S)-medetomidine and (S)-5a , which have high bind-
ing affinity at R2-adrenoceptors, projects into the same
region as the R-methyl group of 21 (Figures 6d and 7).
However, the benzylic substituent of (R)-5a and 8c does
not overlap with the R-methyl group of 21 (Figure 8)
and may account for the poor binding affinity at R2-
adrenoceptors. Such a pattern is also observed when
the two rigid molecules 6 (high R2-affinity) and 7 (low
R2-affinity) were superimposed with 21 (Figures 9 and
10). Therefore, this model provides an approach to
explain the R2-activity of imidazoles.
(c) Su p er im p osition of (-)-r-MeNE a n d Im id a -
zole An a logs. The FIT ATOM function of SYBYL
permitted least-squares fitting of two different mol-
ecules. Appropriate atoms (i.e., atoms that are part of
the potential pharmacophore) were selected for fitting,
and best fits were represented by low rms (root mean
square) distance (typically <0.5 Å) between the fitted
atoms.
More efficient overlaps (i.e., smaller rms value) were
consistently observed when N3, instead of N1, of the
imidazole ring was fitted with the nitrogen of 21. This
finding suggested that N3 on the imidazole ring of
medetomidine-like analogs may play the same impor-
tant role as the amine function in phenethylamines. On
the basis of site-directed mutagenesis experiments with
R2A-adrenoceptors, it has been proposed that the amine
function in phenethylamines may interact with the
carboxy group of Asp113 in the third transmembrane
domain of R2A-adrenoceptors.32 This model suggests
that a similar interaction may exist between the N3 on
the imidazole ring of medetomidine-like analogs and the
carboxy group of Asp113 of R2A-adrenoceptors.
A better fit (i.e., smaller rms values) was obtained
when an imidazole analog was superimposed with the
â-rotamer of 21 rather than the R-rotamer, as exempli-
fied in Figure 6. In such a superimposition, the two
benzylic carbons of the molecules are overlapped indi-
cating that the benzylic substitution may mimic the
â-hydroxy group of phenethylamines. However, this is
not consistent with the experimental data which showed
that replacement of the benzylic methyl group of 5a with
a hydroxy group did not increase, but significantly
Due to their different steric characteristics, the ben-
zylic substitution of 5a and 8c may have a profound
influence on the conformation of the whole molecule and
hence on the interaction with the receptor. Comparing
the lowest-energy conformation of (S)-5a and 8c, a
significant difference was found in the dihedral angle
C1-C2-CR-Câ (Figure 11). This dihedral angle is

3008 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
F igu r e 8. (a) Relaxed stereoview of superimposition of
R-rotamer of 21 (yellow) and (R)-5a (green). The rms value is
0.40. (b) Relaxed stereoview of superimposition of R-rotamer
of 21 and 8c. The rms value is 0.49. Nitrogen and oxygen
atoms are blue and red, respectively. The R-methyl group of
21 and the benzylic side chains of 5a and 8c are pink. All
hydrogen atoms were hidden for better visualization.
F igu r e 9. Relaxed stereoview of superimposition of R-rotamer
of 21 (yellow) and 6 (green). The rms value is 0.53: (a) model
with cylindrical bonds and (b) model of space filling. Nitrogen
and oxygen atoms are blue and red, respectively. The R-meth-
yl group of 21 and the benzylic bridge of 6 are pink. All
hydrogen atoms were hidden for better visualization.
adrenoceptors as Ruffolo proposed. Consequently, only
imidazoles with an appropriate orientation of the ben-
zylic substituent (e.g., 5a and 6) may exhibit high R2-
activity. Additional biological evidence to support this
hypothesis is that demethylated analogs of medetomi-
dine and 5a were less potent at R2-adrenoceptors than
their parent compounds.7
In conclusion, from the results of this study, we have
gained a new insight into the preferred conformation
of imidazoles for interaction with R2-receptors. We
propose that the nitrogen N3 on the imidazole ring binds
with Asp113 of the receptor. The benzylic methyl group
of (S)-medetomidine and its naphthyl analog (S)-5a is
superimposable with the R-methyl group of 21, indicat-
ing that both methyl groups of the two different classes
of drugs may interact at the same site of R2-adrenocep-
tors. By contrast, the activity of analog 8b,c would
appear to indicate that the steric hindrance created by
the ethylene group and the limited freedom of rotation
of the double bond (CR-Câ) are a hindrance to the
indicative of the twist out of plane of the benzylic side
chain relative to the aromatic ring and is equal to 74.5°
for (S)-5a and 127.3° for 8c. The two rigid analogs 6
and 7 were designed in a way that forces the side chain
into different orientations so that the dihedral angle of
6 (42.2°) is close to that of (S)-5a , while the angle of 7
(119.4°) is close to that of 8c. This could account for
the high R2-affinities of (S)-5a and 6 and the low R2-
affinities of 8c and 7.
Ruffolo et al.31 have proposed a four-point attachment
for the interaction between R-MeNE (21) and R2-
adrenoceptors. In their model, the R-methyl group of
21 binds to an additional recognition site only at R2-
adrenoceptors, with the other three sites the same as
described in Easson-Stedman hypotheses.37 In this
study, we were encouraged that the benzylic methyl
group of medetomidine and the R-methyl group of 21
were superimposable, indicating that these two methyl
groups may fit into the same binding pocket at R2-

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3009
were dried by distillation from CaH2. Tetrahydrofuran (THF)
was dried by distillation from sodium metal with benzophe-
none as an indicator. Unless specifically indicated otherwise,
amine maleate salts were obtained and purified by the
dropwise addition of a molar equivalent solution of maleic acid
in anhydrous Et2O to a solution of the amine in absolute
methanol.
4-[(1-Na p h th yl)h yd r oxym eth yl]-N-(tr ip h en ylm eth yl)-
im id a zole (10a ). A 3.0 M solution of EtMgBr (4 mL, 12
mmol) in Et2O was added to a 0.25 M solution of 4-iodo-N-
(triphenylmethyl)imidazole38 (9a ; 4.36 g, 10 mmol) in dry CH2-
Cl2 at ambient temperature. After stirring for 45 min,
1-naphthylaldehyde (2 mL, 15 mmol) was added, and stirring
was continued overnight. Saturated NH4Cl solution was
added to quench the reaction. The aqueous phase was
extracted with an equal amount of CH2Cl2 (2×). The combined
organic extracts were washed with brine, dried over MgSO4,
and concentrated. The crude product was crystallized in CH2-
Cl2/hexane to yield 3.95 g (85%) of 10a as white crystals: mp
235-236 °C; 1H NMR (250 MHz, CDCl3) δ 4.52 (bs, 1H, OH),
6.33 (s, 1H), 6.51 (s, 1H), 7.04-6.99 (m, 6H, Ar-H), 7.18-7.29
(m, 9H, Ar-H), 7.36-7.46 (m, 4H, Ar-H), 7.72-7.82 (m, 3H,
Ar-H), 7.99-8.02 (m, 1H, Ar-H); MS m/e 244 (TrH+). This
compound was used in the synthesis of 11 without further
characterization.
4-[(1-Na p h t h yl)ca r b on yl]-N-(t r ip h en ylm et h yl)im id a -
zole (11). 4-[(1-Naphthyl)hydroxymethyl]-N-(triphenylmeth-
yl)imidazole (10a ; 4.18 g, 9 mmol) was partially dissolved in
130 mL of dry CH2Cl2 with activated MnO2 (7.82 g, 90 mmol).
The mixture was brought to reflux for 4-5 h. The mixture
was filtered through Celite, and the CH2Cl2 solution was
evaporated under reduced pressure. The crude residue was
triturated with Et2O to give 3.98 g (96%) of 11 as a white
1
solid: mp 187-188 °C; H NMR (300 MHz, CDCl3) δ 7.13-
7.17 (m, 6H), 7.35-7.37 (m, 9H), 7.47-7.54 (m, 4H), 7.64 (d,
J ) 1.34 Hz, 1H), 7.85-7.87 (m, 2H), 7.95 (d, J ) 8.24 Hz,
1H), 8.28-8.31 (m, 1H); IR (KBr) 1631 cm-1; MS m/e 244
(TrH+). This compound was used in the synthesis of 5f and
8a ,b without further characterization.
F igu r e 10. Relaxed stereoview of superimposition of R-rota-
mer of 21 (yellow) and 7 (green). The rms value is 0.43: (a)
model with cylindrical bonds and (b) model of space filling.
Nitrogen and oxygen atoms are blue and red, respectively. The
R-methyl group of 21 and the benzylic bridge of 7 are pink.
All hydrogen atoms were hidden for better visualization.
4-[1-(1-Na p h th yl)-1-h yd r oxyeth yl]-N-(tr ip h en ylm eth -
yl)im id a zole (10b). To a cold solution of 11 (0.46 g, 1 mmol)
in 10 mL of dry THF was added a 1.4 M solution of CH3MgBr
(1 mL, 1.5 mmol) in toluene/THF (75:25). After the mixture
stirred overnight at room temperature, the reaction was
quenched with saturated NH4Cl aqueous solution. The aque-
ous layer was extracted with CH2Cl2 (3 × 10 mL), and the
combined organics were washed with brine, dried over MgSO4,
and concentrated. The crude residue was triturated with Et2O
to give 0.47 g (97%) of 10b as a white solid: mp 179-180 °C;
1H NMR (300 MHz, CDCl3) δ 2.08 (s, 3H, CH3), 3.33 (bs, 1H,
OH), 6.52 (d, J ) 1.35 Hz, 1 H, Im-H), 7.06-7.11 (m, 6H),
7.25-7.31 (m, 9H), 7.36-7.44 (m, 4H), 7.73-7.82 (m, 3H), 8.17
(d, J ) 8.67 Hz, 1H); IR (KBr) 3246, 1490, 1444 cm-1; MS m/e
calcd for C34H28N2O 481, found 481. Anal. (C34H28N2O) C, H,
N.
4-[1-(1-Na p h th yl)eth ylen e]-1H-im id a zole Ma lea te (8a ).
A solution of 10b (0.96 g, 2 mmol) in 30 mL of TFA aqueous
solution (60%) was stirred overnight. The solvents were
evaporated, and the residue was partitioned between 30 mL
of CH2Cl2 and 30 mL of HCl solution (10%). The organic
portion was washed with HCl solution (10%, 3 × 30 mL). The
combined acidic solutions were neutralized to pH ∼ 10 and
then extracted with CH2Cl2 (4 × 100 mL). The organics were
dried over Na2SO4 and concentrated to yield 0.15 g (33%) of
8a (free base) as white crystals: 1H NMR (300 MHz, acetone-
d6) δ 5.06 (d, J ) 2.21 Hz, 1H), 6.20 (s, 1H), 6.40 (s, 1H), 7.34-
7.53 (m, 4H), 7.65 (d, J ) 0.83 Hz, 1H), 7.86-7.91 (m, 3H).
F igu r e 11. Defined dihedral angle of (S)-5a and 8c.
expression of the activity on R2-adrenoceptors. We are
currently working on a more detailed study of the
conformation-binding relationships of medetomidine-
like compounds in a 3D-R2-adrenoceptor model which
will more clearly define the receptor-ligand binding
mode at R2-adrenoceptors.
Exp er im en ta l Section
Ch em istr y. Melting points were determined on a Thomas-
Hoover capillary melting point apparatus and are uncorrected.
NMR (1H and 13C) spectra were obtained on a Bruker 300
spectrometer, and chemical shift values are reported as parts
per million (δ) relative to tetramethylsilane (TMS) as an
internal reference. IR spectra were obtained on a System
2000-FTIR instrument. Elemental analyses were performed
by Atlantic Microlab, Inc., and the obtained analytical results
are within (0.4% of the theoretical values. Routine thin-layer
chromatography (TLC) was performed on silica gel UNIPLATE
(250 µm, 2.5 × 10 cm; Analtech Inc., Newark, DE). Flash
chromatography was performed on silica gel (Merck; grade
9385, 230-400 mesh, 60 Å). Acetonitrile (CH3CN), benzene,
ethyl ether (Et2O), methylene chloride (CH2Cl2), and toluene
A portion of the free base was converted to the maleate
which was crystallized from absolute EtOH to give 8a as a
white solid: mp 140-141 °C; 1H NMR (300 MHz, acetone-d6)
δ 5.47 (s, 1H), 6.27 (s, 2H, CHdCH), 6.50 (s, 1H), 6.90 (s, 1H,
Im-H), 7.42-7.59 (m, 4H), 7.86 (d, J ) 8.46 Hz, 1H), 7.96 (d,
J ) 7.40 Hz, 2H), 8.71 (s, 1H, Im-H); IR (neat) 1456 cm-1
;
HRMS m/e calcd for C15H12N2 (free base) 220.1001, found
220.0997. Anal. (C19H16N2O4‚1/4H2O) C, H, N.

3010 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
4-[1-(1-Na p h th yl)-1-h yd r oxyeth yl]-N-(tr ip h en ylm eth -
yl)im id a zole (10c). To a solution of 11 (1.56 g, 3.35 mmol)
in 25 mL of dry benzene was added a 3.0 M solution of EtMgBr
(2 mL, 6 mmol) in Et2O at 0 °C. The mixture was allowed to
stir for 25 h at room temperature. The reaction mixture was
cooled in an ice bath, and excess EtMgBr was destroyed by
addition of saturated NH4Cl solution. The aqueous layer was
extracted with Et2O (3 × 25 mL). The combined organics were
washed with brine, dried over Na2SO4, and concentrated. The
crude residue was first crystallized with CH2Cl2/Et2O to yield
the secondary alcohol 10a (0.33 g, 21%) and then crystallized
in Et2O to give the desired product 10c (0.88 g) contaminated
with a small amount of 10a . This mixture was oxidized with
MnO2 in refluxed CH2Cl2 for 4 h and filtered through Celite.
The resulting CH2Cl2 solution was concentrated, and the crude
residue was crystallized in Et2O to give 0.66 g (40%) of 10c as
white crystals: mp 178-179 °C; 1H NMR (300 MHz, CDCl3) δ
0.81 (t, J ) 7.35 Hz, 3H, CH3), 2.40-2.62 (m, 2H, CH2), 3.17
(b, 1H, OH), 6.59 (d, J ) 1.40 Hz, 1H, Im-H), 7.06-7.14 (m,
6H), 7.24-7.34 (m, 10H), 7.39-7.44 (m, 3H), 7.74 (d, J ) 8.00
Hz, 1H), 7.81 (d, J ) 8.09 Hz, 2H), 8.13 (d, J ) 8.65 Hz, 1H);
IR (KBr) 3184, 1492, 1446 cm-1; MS m/e calcd for C35H30N2O
495, found 495. Anal. (C35H30N2O‚3/2H2O) C, H, N.
4-[1-(1-Na p h th yl)-1-p r op en yl]-N-(tr ip h en ylm eth yl)im i-
d a zole (12a ,b). Sodium hydride (95%, 0.27 g, 11.1 mmol) was
suspended in 10 mL of dry DMSO and heated to 80 °C for 0.5
h. The resulting deep-green solution of dimsyl sodium was
cooled to 0 °C. A solution of the ethyltriphenylphosphonium
bromide (4.115 g, 11.1 mmol) in 10 mL of dry DMSO was
added, and the resulting deep-red solution was maintained for
a further 10 min at room temperature. A solution of ketone
11 (3.43 g, 7.39 mmol) in 10 mL of dry DMSO was added, and
the mixture was heated to 80 °C for 3 h. After cooling to 0 °C,
the mixture was poured into 50 mL of cold water. The
heterogeneous mixture was extracted with EtOAc (3 × 50 mL).
The combined organics were washed with H2O (5 × 100 mL),
dried over MgSO4, and concentrated. The crude residue was
triturated with Et2O, and the white precipitate (triphenyl
phosphoxide) was removed by filtration. The Et2O solution
was concentrated, and the resulting mixture of two alkene
isomers was separated by flash chromatography on silica gel,
eluting with EtOAc/hexane (1:4).
4-[1-(1-Na p h th yl)-(1E)-p r op en yl]-1H-im id a zole (8b): 1H
NMR (300 MHz, DMSO-d6) δ 1.40 (d, J ) 7.00 Hz, 3H, CH3),
6.07 (b, 1H, Im-H), 6.62 (q, J ) 7.02 Hz, 1H, HCdC), 7.30 (dd,
J ) 1.05, 6.95 Hz, 1H), 7.41 (m, 1H), 7.49 (m, 1H), 7.55 (m,
1H), 7.61 (s, 1H), 7.68 (d, J ) 8.20 Hz, 1H), 7.93 (m, 2H), 12.02
(bs, 1H, NH).
4-[1-(1-Na p h th yl)-(1Z)-p r op en yl]-1H-im id a zole (8c): 1H
NMR (300 MHz, CD3OD) δ 2.12 (d, J ) 7.12 Hz, 3H, CH3),
5.84 (q, J ) 7.12 Hz, 1H, HCdC), 6.90 (s, 1H, Im-H), 7.26-
7.32 (m, 1H), 7.35-7.47 (m, 3H), 7.54 (s, 1H, Im-H), 7.73 (d, J
) 8.46 Hz, 1H), 7.79-7.83 (m, 2H).
The free bases were converted to the maleates, respectively.
4-[1-(1-Na p h th yl)-(1E)-p r op en yl]-1H-im id a zole m a lea te
1
(8b): mp 159-160 °C; H NMR (300 MHz, DMSO-d6) δ 1.46
(d, J ) 6.90 Hz, 3H, CH3), 6.07 (d, J ) 0.77 Hz, 2H, HCdHC),
6.70-6.67 (m, 2H), 7.36 (d, J ) 6.96 Hz, 1H), 7.46-7.66 (m,
4H), 7.97-8.01 (m, 2H), 8.74 (s, 1H); 13C NMR (300 MHz,
DMSO-d6) δ 14.72 (CH3), 167.11 (CdO); IR (KBr) 1507, 1457
cm-1; MS m/e calcd for C16H14N2 (free base) 234, found 234.
Anal. (C20H18N2O4) C, H, N.
4-[1-(1-Na p h th yl)-(1Z)-p r op en yl]-1H-im id a zole m a le-
1
a te (8c): mp 139.0-140 °C; H NMR (300 MHz, DMSO-d6) δ
2.09 (d, J ) 7.13 Hz, 3H, CH3), 6.01 (q, J ) 7.14 Hz, 1H,
HCdC), 6.08 (s, 2H, HCdHC), 7.38-7.46 (m, 3H), 7.48-7.55
(m, 2H), 7.68 (d, J ) 8.35 Hz, 1H), 7.93 (t, J ) 6.23 Hz, 2H),
8.55 (s, 1H); 13C NMR (300 MHz, DMSO-d6) δ 15.63 (CH8),
166.99 (CdO); IR (KBr) 1576, 1474, 1442 cm-1; MS m/e calcd
for C16H14N2 (free base) 234, found 234. Anal. (C20H18N2O4)
C, H, N.
4-[1-(1-Na p h th yl)p r op yl]-1H-im id a zole Ma lea te (5f).
To a solution of 8c (0.45 g, 2 mmol) in absolute EtOH (25 mL)
was added 5% Pd/C (0.05 g). The mixture was subjected to
40 psi of H2 for 24 h. The mixture was filtered through Celite,
and the solvent was evaporated to give 0.43 g (95%) of 5f (free
base) as a white solid: 1H NMR (300 MHz, CD3OD) δ 0.95 (t,
J ) 7.33 Hz, 3H, CH3), 2.07-2.25 (m, 2H, CH2), 4.68 (t, J )
7.46 Hz, 1H, CH), 6.80 (t, J ) 0.87 Hz, 1H, Im-H), 7.37-7.39
(m, 2H), 7.41-7.48 (m, 2H), 7.55 (d, J ) 1.13 Hz, 1H, Im-H),
7.69 (dd, J ) 4.55, 2.36 Hz, 1H), 7.80-7.83 (m, 1H), 8.14-
8.18 (m, 1H).
A portion of the free base was converted to the maleate and
recrystallized in CH2Cl2/hexane to give the title compound as
white crystals: mp 112-113 °C; 1H NMR (300 MHz, CDCl3) δ
1.00 (t, J ) 7.28 Hz, 3H, CH3), 2.26-2.39 (m, 2H, CH2), 4.78
(t, J ) 7.66 Hz, 1H, CH), 6.39 (s, 2H, CHdCH), 6.80 (s, 1H,
Im-H), 7.45-7.54 (m, 4H), 7.79 (dd, J ) 2.54, 6.78 Hz, 1H),
7.86-7.89 (m, 1H), 8.00-8.03 (m, 1H), 9.09 (d, J ) 1.35 Hz,
1H, Im-H), 14.30 (bs, 2H, CO2H); 13C NMR (300 MHz, DMSO-
d6) δ 12.25 (CH3), 27.61 (CH2), 38.11 (CH), 167.08 (CdO); IR
(KBr) 1696, 1575, 1514, 1475 cm-1; MS m/e calcd for C16H16N2
(free base) 236, found 236. Anal. (C20H20N2O4) C, H, N.
4-[1-(1-Na p h th yl)-(1E)-p r op en yl]-N-(tr ip h en ylm eth yl)-
1
im id a zole (12a ): 1.0 g, 28%; mp 130-131 °C; H NMR (300
MHz, CDCl3) δ 1.53 (d, J ) 7.05 Hz, 3H, CH3), 5.99 (s, 1H,
Im-H), 6.89 (q, J ) 7.09 Hz, 1H, HCdC), 7.00-7.03 (m, 6H),
7.17-7.34 (m, 10H), 7.36-7.45 (m, 4H), 7.73 (d, J ) 8.16 Hz,
1H), 7.78-7.85 (m, 2H); IR (KBr) 1493, 1445 cm-1; MS m/e
calcd for C35H28N2 476, found 476. Anal. (C35H28N2) C, H, N.
4-[1-(1-Na p h th yl)-(1Z)-p r op en yl]-N-(tr ip h en ylm eth yl)-
im id a zole (12b): 1.90 g, 54%; mp 154-155 °C; 1H NMR (300
MHz, CDCl3) δ 2.22 (d, J ) 7.19 Hz, 3H, CH3), 5.82 (q, J )
7.03 Hz, 1H, HCdC), 6.36 (d, J ) 1.06 Hz, 1H, Im-H), 7.04-
7.07 (m, 6H), 7.21-7.29 (m, 9H), 7.34-7.43 (m, 5H), 7.71 (t, J
4-(1-Hyd r oxy-1-p er in a p h th en yl)-N-(tr ip h en ylm eth yl)-
im id a zole (14a ). A 3.0 M solution of EtMgBr (0.80 mL, 2.4
mmol) in Et2O was added to a 0.25 M solution of 4-iodo-N-
(triphenylmethyl)imidazole (9a ; 0.87 g, 2 mmol) in dry CH2-
Cl2 at ambient temperature. After 1 h, a solution of perinaph-
thenone (13a ; 0.40 g, 2.2 mmol) in 1 mL of dry CH2Cl2 was
added, and stirring was continued overnight. The reaction
mixture was worked up in a manner identical with that
previously described for the synthesis of 10a . The resulting
solid was recrystallized from CH2Cl2/hexane to provide 0.49 g
(49%) of the titled compound as an orange solid: mp 199-
200 °C; 1H NMR (300 MHz, CDCl3) δ 5.44 (d, J ) 3.78 Hz,
1H), 6.00 (dd, J ) 4.40, 9.77 Hz, 1H), 6.59 (s, 1H, Im-H), 6.68
(dd, J ) 9.81, 1.67 Hz, 1H), 6.98-7.16 (m, 8H), 7.21 (s, 1H),
7.27-7.33 (m, 9H), 7.40 (d, J ) 1.32 Hz, 1H), 7.52 (d, J ) 7.93
) 4.08 Hz, 1H), 7.76-7.85 (m, 2H); IR (KBr) 1494, 1445 cm-1
;
MS m/e calcd for C35H28N2 476, found 476. Anal. (C35H28N2)
C, H, N.
4-[1-(1-Na p h t h yl)-1-p r op en yl]-1H -im id a zole Ma lea t e
(8b,c). Compound 10c (1.50 g, 3.03 mmol) was deoxygenated
and deprotected in TFA aqueous solution (60%, 25 mL) as
previously described in the preparation of 8a to afford a
mixture of E/Z-isomers with a ratio of 3:1. The isomers were
separated by flash chromatography, eluting with EtOAc/
hexane (5:2), to yield 0.38 g (51%) of 8b and 0.13 g (17%) of
8c.
Alternatively, 8c could be prepared from the deprotection
of 12b. A solution of 12b (1.95 g, 4.1 mmol) in 60 mL of AcOH
aqueous solution (90%) was stirred at 60 °C for 2 h. The
solvents was removed under reduced pressure, and the residue
was partitioned between 50 mL of EtOAc and 20 mL of H2O.
The water layer was made basic to pH ∼ 10 and extracted
with EtOAc (3 × 30 mL). The organics were washed with
brine and dried over Na2SO4. Removal of the solvent and flash
chromatography, eluting with EtOAc/hexane (2:1), gave 0.69
g (97%) of 8c (free base) as white needles.
Hz, 1H), 7.58 (d, J ) 8.72 Hz, 1H); IR (KBr) 3413, 1617 cm-1
;
MS m/e 243 (Tr+). Anal. (C35H26N2O) C, H, N.
2,3-Dih yd r o-1H-p h en a len on e (13b). A solution of 3-(1-
naphthyl)propionic acid21 (15; 1.0 g, 5 mmol) in 4 mL of SOCl2
was brought to reflux for 2 h. Removal of the solvent under
reduced pressure gave 3-(1-naphthyl)propionic acid chloride
as a light brown oil: 1H NMR (300 MHz, acetone-d6) δ 3.46-
3.54 (m, 4H, 2CH2), 7.41-7.47 (m, 2H), 7.49-7.60 (m, 2H),
7.80-7.83 (m, 1H), 7.91-7.94 (m, 1H), 8.12-8.15 (m, 1H).

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3011
The oily residue was dissolved in 8 mL of dry benzene, and
SnCl4 (0.6 mL, 5.1 mmol) was added via syringe at 0 °C. After
15 min, the mixture was hydrolyzed by dumping into a slurry
of ice water and aqueous HCl solution (10%). The water layer
was extracted with an equal amount of Et2O. The combined
organics were washed successively with water, saturated
NaHCO3, water, and brine and then dried and concentrated.
The residue was purified over aluminum oxide (activated,
neutral, Brockmann I), eluting with CH2Cl2/hexane (5:1) to
yield 0.33 g (36%) of the title compound as a light yellow
solid: mp 78-81 °C (lit.17 mp 85 °C); 1H NMR (300 MHz,
CDCl3) δ 2.90 (m, 2H, CH2), 3.34 (t, J ) 7.09 Hz, 2H, CH2),
7.36-7.45 (m, 2H), 7.52 (t, J ) 7.67 Hz, 1H), 7.72 (d, J ) 7.61
Hz, 1H), 8.00 (d, J ) 2.11 Hz, 1H), 8.29 (dd, J ) 7.18, 0.92
Hz, 1H); 13C NMR (300 MHz, CD3OD) δ 29.24 (CH2), 39.34
(CH2), 200.59 (CdO); IR (KBr) 1681 cm-1. The spectral data
were consistent with the structure of the product; the product
was immediately used in the synthesis of 14c without further
characterization.
4-(1-H yd r oxy-2,3-d ih yd r o-1-p er in a p h t h a n yl)-N-(t r i-
p h en ylm eth yl)im id a zole (14c). A 3.0 M solution of EtMgBr
(0.55 mL, 1.65 mmol) in Et2O was added to a 0.25 M solution
of 4-iodo-N-(triphenylmethyl)imidazole (9a ; 0.72 g, 1.65 mmol)
in dry CH2Cl2 at ambient temperature. After 1 h, a solution
of perinaphthanone (13b; 0.20 g, 1.1 mmol) in 1 mL of dry
CH2Cl2 was added, and stirring was continued overnight. The
reaction mixture was worked up in a manner similar to that
previously described for the synthesis of 10a . The crude
product was recrystallized from CH2Cl2/hexane to provide 0.40
g (73%) of the title compound as a white solid: mp 206-208
°C; 1H NMR (300 MHz, CDCl3) δ 2.30-2.37 (m, 1H), 2.61-
2.69 (m, 1H), 2.91-2.94 (m, 1H), 3.23-3.31 (m, 1H), 3.34 (bs,
1H, OH), 6.25 (s, 1H, Im-H), 7.05-7.08 (m, 6H), 7.26-7.28 (m,
9H), 7.34-7.44 (m, 4H), 7.55 (d, J ) 7.20 Hz, 1H), 7.66 (d, J
) 8.16 Hz, 1H), 7.72 (d, J ) 8.18 Hz, 1H); IR (KBr) 3208, 1492,
1446 cm-1; HRMS m/e calcd for C35H28N2O 492.2202, found
292.2211. Anal. (C35H28N2O) C, H, N.
4-(1-P er in a p h th a n yl)-1H-im id a zole Ma lea te (6). To a
mixture of Me3SiCl (1.54 mL, 12 mmol), NaI (1.8 g 12 mmol),
and dry CH3CN (0.6 mL, 12 mmol) was added a solution of
14c (0.5 g, 1 mmol) in dry CH2Cl2 (6 mL). The mixture was
stirred for 24 h at room temperature. Dilution with H2O,
extraction with CH2Cl2, and subsequent flash chromatography,
eluting with CH2Cl2/CH3OH (20:1), gave 0.15 g (63%) of 6 (free
base) as a white solid: 1H NMR (300 MHz, CD3OD) δ 2.18-
2.27 (m, 1H), 2.40-2.47 (m, 1H), 2.99-3.06 (m, 2H, Ar-CH2),
4.46 (t, J ) 4.63 Hz, 1H, CH), 6.63 (s, 1H, Im-H), 7.11-7.14
(m, 1H), 7.23 (dd, J ) 6.95, 1.17 Hz, 1H), 7.34-7.39 (m, 2H),
7.66-7.73 (m, 3H).
A portion of the free base was converted to the maleate and
recrystallized in CH2Cl2 to afford the title compound as white
crystals: mp 143-145 °C; 1H NMR (300 MHz, CD3OD) δ 2.34-
2.41 (m, 2H, CH2), 3.05-3.09 (m, 1H), 3.14-3.19 (m, 1H), 4.69
(t, J ) 5.48 Hz, 1H, CH), 6.24 (s, 2H, HCdCH), 6.98-6.99 (m,
1H), 7.12 (dt, J ) 7.09 Hz, 1H), 7.31 (dd, J ) 7.00, 1.16 Hz,
1H), 7.39-7.45 (m, 2H), 7.73-7.76 (m, 1H), 7.81 (d, J ) 8.27
Hz, 1H), 8.80 (d, J ) 1.39 Hz, 1H, Im-H); 13C NMR (300 MHz,
CD3OD) δ 28.98 (CH2), 30.03 (CH2), 38.42 (CH), 170.75 (CdO);
IR (KBr) 1571, 1507, 1465 cm-1; MS m/e calcd for C16H14N2
(free base) 234, found 234. Anal. (C20H18N2O4) C, H, N.
4-(1,2,3,4-Tetr a h yd r o-1-h yd r oxy-1-p h en a n th r en yl)-N-
(tr ip h en ylm eth yl)im id a zole (20). A 3.0 M solution of
EtMgBr (1.6 mL, 4.8 mmol) in Et2O was added to a 0.25 M
solution of 4-iodo-N-(triphenylmethyl)imidazole (9a ; 2.09 g, 4.8
mmol) in dry CH2Cl2 at ambient temperature. After 1 h, a
solution of 2,3-dihydro-4(1H)-phenanthrone25 (19; 0.47 g, 2.4
mmol) in 2 mL of dry CH2Cl2 was added, and stirring was
continued overnight. The reaction mixture was worked up in
a manner similar to that previously described for the synthesis
of 10a . The crude product was recrystallized from CH2/Cl2/
hexane to provide 0.87 g (72%) of the title compound as a white
solid: mp 195-196 °C; 1H NMR (300 MHz, CDCl3) δ 1.71 (m,
1H), 1.94 (m, 1H), 2.30 (t, 4.63H, 2H, CH2), 2.83-2.88 (m, 1H),
2.90-3.00 (m, 1H), 3.88 (bs, 1H, OH), 6.04 (d, J ) 1.34 Hz,
1H, Im-H), 6.95-7.07 (m, 6H), 7.12-7.25 (m, 9H), 7.28-7.38
(m, 3H), 7.41 (d, J ) 1.24 Hz, 1H, Im-H), 7.61 (d, J ) 8.41 Hz,
1H), 7.71 (m, 1H), 8.17 (d, J ) 8.70 Hz, 1H); IR (KBr) 3218,
1493, 1445 cm-1; HRMS m/e calcd for C36H30N2O 506.2358,
found 506.2356. Anal. (C36H30N2O‚1/4H2O) C, H, N.
4-(1,2,3,4-Te t r a h yd r o-1-p h e n a n t h r e n yl)-1H -im id a -
zole Ma lea te (7). To a mixture of triethylsilane (3 mL, 20
mmol) and 20 (1.015 g, 2 mmol) was added TFA (3 mL, 40
mmol). After stirring at 50-55 °C for 28 h, the mixture was
cooled to room temperature and diluted with 25 mL of EtOAc
and 15 mL of NaOH aqueous solution (10%) at 0 °C. The
layers were separated; the pH of water layer was adjusted to
8-9. The aqueous portion was extracted with EtOAc (15 × 3
mL). The combined organics were washed with H2O and brine
and dried over Na2SO4. The crude product was purified by
flash chromatography, eluting with EtOAc/CH3OH (10:1), to
yield 0.28 g (56%) of the product (free base): 1H NMR (300
MHz, DMSO-d6) δ 1.67-1.75 (m, 2H, CH2), 1.89-1.98 (m, 1H),
2.15-2.20 (m, 1H), 2.90 (t, J ) 7.21 Hz, 2H, CH2), 4.69 (d, J
) 3.22 Hz, 1H, CH), 6.03 (s, 1H, Im-H), 7.25 (d, J ) 8.42 Hz,
1H), 7.32-7.38 (m, 2H), 7.50 (s, 1H, Im-H), 7.68-7.76 (m, 2H),
7.78-7.81 (m, 1H), 11.65 (b, 1H, NH).
A portion of the free base was converted to the maleate, and
the salt was crystallized in Et2O to give 7: mp 169-170 °C;
1H NMR (300 MHz, DMSO-d6) δ 1.56-1.64 (m, 1H), 1.78-
7.82 (m, 1H), 2.05-2.14 (m, 2H, CH2), 2.92-2.97 (m, 2H, CH2),
4.90 (d, J ) 2.45 Hz, 1H, CH), 6.04 (s, 2H, HCdCH), 6.69 (s,
1H, Im-H), 7.32 (d, J ) 9.47 Hz, 1H), 7.39-7.43 (m, 2H), 7.66-
7.69 (m, 1H), 7.79 (d, J ) 8.46 Hz, 1H), 7.85-7.88 (m, 1H),
8.82 (d, J ) 1.09 Hz, 1H); 13C NMR (300 MHz, DMSO-d6) δ
17.37 (CH2), 28.91 (CH2), 29.23 (CH2), 31.22 (CH), 167.07
(CdO); IR (KBr) 1702 cm-1; MS m/e calcd for C17H16N2 (free
base) 248, found 248. Anal. (C21H20N2O4‚1/4H2O) C, H, N.
R ecep t or Isola t ion a n d P r ep a r a t ion . R1- And R2-
adrenergic receptors were isolated using established meth-
ods.7,39 Briefly, frozen rat brain were purchased from Harlan
Bioproducts for Science, Inc. and stored at -70 °C until use
(Indianapolis, IN). Brains were rapidly thawed, frontal cor-
tices dissected, and the cortices homogenized in approximately
30 vol/wet wt of ice-cold buffer containing 50 mM Tris-HCl
and 10 mM MgCl2, pH 7.4. The homogenate was first
centrifuged at 48000g for 30 min. The resulting pellet,
representing the membrane fraction, was then resuspended
in buffer and recentrifuged. This process was repeated twice.
The final pellet was then resuspended in approximately 10
vol and stored at -70 °C until use in competitive binding
studies.
Ra d ioliga n d Bin d in g Stu d ies. All radioligand binding
assays were performed in triplicate in polypropylene tubes.
R1-Adrenergic receptors were labeled using 0.1 nM [3H]-
prazosin (78.7 Ci/mL) dissolved in 50 mM Tris buffer, pH 7.4,
containing 10 mM MgCl2. R2-Adrenergic receptors were
labeled using 2.0 nM [3H]rauwolscine (80.5 Ci/mL) in 50 nM
Tris buffer, pH 7.4, containing 10 mM EDTA. The final
protein concentrations were approximately 300 and 500 µg/
mL for the R1- and R2-assays, respectively. Phentolamine (10
µM) was used to determine the fraction of nonspecific binding
in both assays. Samples were incubated at 30 °C for 30 min
and then rapidly filtered (Whatman GF/C glass fiber filters
presoaked in 10 µM phentolamine) and washed with 20 mL
of ice-cold assay buffer on a Brandel cell harvester.40 Indi-
vidual filters were inserted into scintillation vials, shaken, and
allowed to equilibrate for 5 h with 5 mL of aqueous scintillation
fluid before radioactivity counting on a Beckman LS 6800
liquid scintillation counter. The IC50 for each compound in
each of the receptor systems was then determined by plotting
the percentage specific binding at each sigmoidal Emax model.
The equilibrium dissociation constant of each compound (Ki)
was then calculated using the standard equation.
Aor ta Con tr a ction Activities. Rats, weighing 150-300
g, of either sex were killed by a sharp blow to the head and
thoracic. Abdominal aorta was removed and placed in a Petri
dish containing oxygenated Krebs solution of the following
composition (mM): NaCl, 94.8; KCl, 4.7; MgSO4, 1.2; CaCl2,
2.5; KH2PO4, 1.2; NaHCO3, 25.0; and glucose, 11.7. Fat and
connective tissues were trimmed away using fine curved
scissors, and helical strips were prepared. The specimen was
mounted vertically in a 10-mL organ bath containing the

3012 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Zhang et al.
oxygenated Krebs solution kept at 37 °C and connected to an
isometric tension transducer (Grass FT-03C) with initial loads
of 1 g. The experiment was begun after a 90-min equilibration
period. The responsiveness of each preparation was first
tested by applications of 30 nM of phenylephrine (PE). Then,
PE was added cumulatively to the organ bath. After 90 min,
the compound was added to the organ bath, cumulatively. The
responses were expressed as percentage of the maximum
contraction to 1 µM of PE. When the compound caused more
than 20% of the contraction, it was considered as an agonist;
however, when the contraction was less than 20%, the com-
pound was considered as an antagonist. When the compound
was considered to be an antagonist, it was added 60 min before
a second application of PE. A maximal response to PE in the
tissue was 100%, and the percentage of response for each
agonist was adjusted from this value. The EC50 values of
agonists and pKB values were determined as described previ-
ously.27
P la telet An tia ggr ega tor y Activities. All aggregation
studies were performed using human platelet-rich plasma,
according to methods reported previously.27 Various concen-
trations of inhibitors were added 1 min prior to activation of
platelets by epinephrine (30-100 µM). Aspirin (1 mM) was
routinely added to platelet preparations in experiments to
examine drug effects on the primary wave aggregation re-
sponse to epinephrine. IC50 values of drugs were analyzed by
computer programs as described previously.27
X-r a y An a lysis of 8c. A clear colorless 0.36 × 0.36 × 0.52
mm data crystal of 8c (C16H15N2+‚C4H3O4-), fw ) 350.4) was
crystallized from CHCl3/hexane. Data were collected on a
computer-controlled diffractometer with an incident beam
graphite monochromator (Siemens R3m/V with Cu KR radia-
tion, λ ) 1.541 78 Å, T ) 295 K). A least-squares refinement
using 25 centered reflections within 60° < 2θ < 720° gave the
P21/c monoclinic cell, a ) 7.050(1) Å, b ) 34.165(5) Å, c )
7.899(1) Å, and â ) 110.80(1)°, with V ) 1776.3(4) Å3, Z ) 4,
and Dcalcd ) 1.310 gm/cm3. A total of 2627 reflections were
measured in the θ/2θ mode to 2θmax ) 1150°, of which 2432
were unique (Rint ) 1.66%). The scan width was [2θ(KR1) -
1.1]-[2θ(KR2) + 1.1]°, and scan rate was a function of count
rate (7.5°/min minimum, 30.0°/min maximum in ω). Correc-
tions were applied for Lorentz, polarization, and absorption
effects. The structure was solved by direct methods with the
aid of the program SHELXTL (version 5.0; Siemens Analytical
Instruments, Madison, WI) and refined on Fo2 with full matrix
least-squares. The 255 parameters refined included the
coordinates and anisotropic thermal parameters for all non-
hydrogen atoms. Carbon hydrogens used a riding model of
C-H ) 0.96 Å and idealized angles. Hydroxyl hydrogens and
those on the heterocyclic ring were refined isotropically with
fixed thermal parameters. Atomic scattering factors are from
the International Tables for X-ray Crystallography (1974). The
maximum excursions for the final Fourier difference map were
0.20 and -0.29 eÅ-3. The final R values for the 2115 observed
reflections with Fo > 4σ(|Fo|) were R ) 0.044 and R2w ) 0.121.
Tables of coordinates, bond distances, and bond angles have
been deposited with the Crystallographic Data Center, Cam-
bridge CB2 1EW, England.
Molecu la r Mod elin g. Molecular modeling studies were
conducted using the program SYBYL 5.32 (Tripos Associates,
St. Louis, MO) running on a Silicon Graphics Iris workstation.
The structures of model compounds were constructed from
their X-ray atomic coordinates using the Crysin interface of
SYBYL. The X-ray crystal coordinates of (S)-medetomidine41
were obtained from the published atomic coordinates. The
X-ray analyses of (S)-5a 33 and 8c were provided by the Naval
Research Laboratory. Energy minimizations were performed
using Maximin2, a molecular mechanics method available in
SYBYL. Atomic charges were calculated by the Gasteiger-
Hu¨ckel method. Conformational analysis of each compound
was performed by using systematic search, rotating bonds and
distance constraints as depicted in Figure 4. The low-energy
conformations of each molecule were minimized prior to
superimposition. Molecules were superimposed using the FIT
ATOM command in SYBYL. The selected 3D-features were
read into the program CSC Chem3D Plus for printed repre-
sentations. Hydrogen atoms were hidden in Figures 7-10 for
better visualization.
Ackn owledgm en t. We thank the U.S. Army Chemi-
cal Research, Development and Engineering Center,
Aberdeen Proving Ground, for partial support of this
research project. We also thank the National Institutes
of Health for financial assistance in the form of USPHS
Grant GM-29358-12.
Su p p or tin g In for m a tion Ava ila ble: Elemental analyses
of the new compounds and X-ray data for 8c (5 pages).
Ordering information can be found on any current masthead
page.
Refer en ces
(1) Ruffolo, R. R., J r.; Nichols, A. J .; Stadel, J . M.; Hieble, J . P.
Pharmacologic and Therapeutic Applications of R2-Adrenoceptor
Subtypes. Annu. Rev. Pharmacol. Toxicol. 1993, 32, 243-279.
(2) Urban, R.; Szabo, B.; Starke, K. Is the Sympathoinhibitory Effect
of Rilmenidine Mediated by Alpha-2 Adrenoceptors or Imida-
zoline Receptors? J . Pharmacol. Exp. Ther. 1994, 270, 572-
578.
(3) Ernsberger, P.; Damon, T. H.; Graff, L. M.; Schafer, S. G.;
Christen, M. O. Moxonidine, a Centrally Acting Antihypertensive
Agent, Is a Selective Ligand for I1-Imidazoline Sites. J . Phar-
macol. Exp. Ther. 1993, 264, 172-182.
(4) Hayashi, Y.; Maze, M. Alpha2 Adrenoceptor Agonists and
Anaesthesia. Br. J . Anaesth. 1993, 71, 108-118.
(5) Aantaa, R.; Kallio, A.; Virtanen, R. Dexmedetomidine, a Novel
R2-Adrenergic Agonist. A Review of Its Pharmacodynamic
Characteristics. Drugs Future 1993, 18, 49-56.
(6) Savola, J .-M.; Virtanen, R. Central R2-Adrenoceptors are Highly
Stereoselective for Dexmedetomidine, the Dextro Enantiomer of
Medetomidine. Eur. J . Pharmacol. 1991, 195, 193-199.
(7) Hong, S.-S.; Romstedt, K. J .; Feller, D. R.; Hsu, F.-L.; Cupps, T.
L.; Lyon, R. A.; Miller, D. D. A Structure-Activity Relationship
Study of Benzylic Modifications of 4-[1-(1-Naphthyl)ethyl]-1H-
imidazoles on R1- and R2-Adrenergic Receptors. J . Med. Chem.
1994, 37, 2328-2333.
(8) Mutschler, E.; Lambrecht, G. Stereoselectivity and Conforma-
tion: Flexible and Rigid Compounds. In Stereochemistry and
Biological Activity of Drugs; Ariens, E. J ., Soudijn, W., Timmer-
mans, P. B. M. W. M., Eds.; Blackwell: Oxford, 1983; pp 63-
79.
(9) Erhardt, P. W. Conformational Analogues of Dopamine. Syn-
thesis and Pharmacological Activity of (E)- and (Z)-2(3,4-Dihy-
droxyphenyl)cyclopropylamine Hydrochlorides. J . Med. Chem.
1979, 22, 907-911.
(10) Ruffolo, R. R., J r.; Anderson, K. S.; Miller, D. D. Conformational
Requirements of Alpha2-Adrenergic Receptors. Mol. Pharmacol.
1982, 21, 259-261.
(11) Hicks, P. E.; Waldron, C.; Bum, P.; Crooks, P. A. R2-Adrenoceptor
Agonist Properties of Exo- and Endo-isomers of 2-Amino-6,7-
dihydroxybenzonorbornene Designed as Rigid Catecholamines.
J . Pharm. Pharmacol. 1983, 35, 94-99.
(12) DeBernardis, J . F.; Kerkman, D. J .; Winn, M.; Bush, E. N.;
Arendsen, D. L.; McClellan, W. J .; Kyncl, J . J .; Basha, F. Z.
Conformationally Defined Adrenergic Agents. 1. Design and
Synthesis of Novel R2 Selective Adrenergic Agents: Electrostatic
Repulsion Based Conformational Prototypes. J . Med. Chem.
1985, 28, 1398-1404.
(13) Macchia, B.; Balsamo, A.; Breschi, M. C.; Lapucci, A.; Lucacchini,
A.; Macchia, F.; Mamera, C.; Martinelli, A.; Martini, C.; Mar-
tinotti, E.; Nencetti, S. Conformational Effects on the Activity
of Drugs. 13. A Revision of Previously Proposed Models for the
Activation of R- and â-Adrenergic Receptors. J . Med. Chem.
1992, 35, 1009-1018.
(14) Turner, R. M.; Lindell, S. D. A Facile Route to Imidazol-4-yl
Anions and Their Reaction with Carbonyl Compounds. J . Org.
Chem. 1991, 56, 5739-5740.
(15) Hsu, F.-L.; Zhang, X.; Hong, S.-S.; Berg, F. J .; Miller, D. D.
Imidazole-Assisted Intramolecular Phenoxythiocarbonylation of
Tertiary Alcohols. A Key Reaction for the Deoxygenation of
R-trifluoromethylarylmethyl Alcohols. Heterocyclic 1994, 39,
801-809.
(16) Prinzbach, H.; Freudenberger, V.; Scheidegger, U. Cyclische,
Gekreuzt-Konjugierte Bindungssysteme. XIII. NMR.-Unter-
suchungen am Phenafulven-System. (Cyclic Cross-conjugated
Binding Systems. XIII. NMR Investigations with the Phenaful-
vene System.) Helv. Chim. Acta 1967, 50, 1087-1107.
(17) Boekelheide, V.; Larrabee, C. E. An Investigation of the Prepa-
ration and Some Properties of Perinaphthene. J . Am. Chem.
Soc. 1950, 72, 1245-1249.
(18) Ganem, B. Conjugate Reduction and Reductive Alkylation of R,â-
Unsaturated Cyclohexenones Using Potassium Tri-sec-butyl-
borohydride. J . Org. Chem. 1975, 40, 146-147.

Medetomidine Analogs as R2-Adrenergic Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3013
(19) Hoffman J un, J . M.; Schlessinger, R. H. Sodium Metaperio-
date: a Mild Oxidizing Agent for the Generation of Di-imide from
Hydrazine. Chem. Commun. 1971, 1245-1255.
(20) Djerassi, C.; Gutzwiller, J . Selective Reduction of Steroids by
Homogeneous Catalytic Hydrogenation. J . Org. Chem. 1969,
34, 4537-4538.
(21) Fieser, L. F.; Gates, M. D., J r. Synthetic Experiments Utilizing
Perinaphthanone-7. J . Am. Chem. Soc. 1940, 62, 2335-2341.
(22) Sakai, T.; Miyata, K.; Utaka, M.; Takeda, A. Me3SiCl-NaI-CH3-
CN as an Efficient and Practical Reducing Agent for Benzylic
Alcohols. Tetrahedron Lett. 1987, 28, 3817-3818.
(23) Denney, D. B.; Smith, L. C. Preparation and Reaction of Some
Phosphobetaines. J . Org. Chem. 1962, 27, 3404-3408.
(24) Sally, J . J ., J r.; Glennon, R. A. Studies on Simplified Ergoline
Derivatives. A General Six-Step Synthesis of Phenyl-Substi-
tuted 4-Methyl-3,4,4a,5,6,10b-hexahydrobenzo[f]quinolin-1-(2H)-
one Analogs (1). J . Heterocycl. Chem. 1982, 19, 545-550.
(25) Bachmann, W. E.; Cortes, G. D. Phenanthrene Derivatives. XI.
Accetylation and Succinoylation of 3-Methylphenanthrene. J .
Am. Chem. Soc. 1943, 65, 1329-1334.
esis of R2A-Adrenergic Receptors: Identification of Amino Acids
Involved in Ligand Binding and Receptor Activation by Agonists.
Mol. Pharmacol. 1991, 40, 168-179.
(33) Hong, S.-S.; Romstedt, K. J .; Feller, D. R.; Hsu, F.-L.; George,
C.; Cupps, T. L.; Lyon, R. A.; Miller, D. D. Resolution and
Adrenergic Activities of the Optical Isomers of 4-[1-(1-Naphthyl)-
ethyl]-1H-imidazole. Chirality 1992, 4, 432-438.
(34) Squier, G. J .; van der Schyf, C. J .; Oliver, D. W.; Venter, D. P.
Comparative R- and â-Adrenoceptor Activity of 2- and 6-Ring-
chlorinated Noradrenaline Analogues. Arzneim.-Forsch. Drug
Res. 1986, 36, 457-460.
(35) DeBernardis, J . F.; Arendsen, D. L.; Kyncl, J . J .; Kerkman, D.
J . Conformationally Defined Adrenergic Agents. 4. 1-(Amino-
methyl)phthalans: Synthesis and Pharmacological Conse-
quences of the Phthalan Ring Oxygen Atom. J . Med. Chem.
1987, 30, 178-184.
(36) Li, Y. O.; Bergsma, D. J .; Ganguly, S.; Swift, A. M.; Ruffolo, R.
R., J r.; Hieble, J . P. Presentation at XII International Congress
of Pharmacology, Montreal, Canada, J uly 24-29, 1994.
(37) Easson, L. H.; Stedman, E. Studies on the Relationship Between
Chemical Constitution and Physiological Action. V. Molecular
Dissymetry and Physiological Activity. Biochem. J . 1933, 27,
1257-1266.
(38) Kirk, K. L. 4-Lithio-1-tritylimidazole as a Synthetic Intermedi-
ate. Synthesis of Imidazole-4-carboxyaldehyde. J . Heterocycl.
Chem. 1985, 22, 57-59.
(39) Pitha, J .; Szabo, L.; Szurmai, Z.; Buchowiecki, W.; Kusiak, J .
W. Alkylating Prazosin Analog: Irreversible Label for R1-
Adrenoceptors. J . Med. Chem. 1989, 32, 96-100.
(40) Veenstra, D. M. J .; van Buuren, K. J . H.; Krielaart, M. J .;
Nijkamp, F. P. Errors Introduced in Radioligand Binding Studies
Due to Displaceable, Cation Dependent, [3H]Prazosin Binding
to Glass-Fiber and Glass Surface. J . Recept. Res. 1993, 13, 801-
814.
(41) Kivikoski, J . H.; Rissanen, K. T.; Parhi, S. S. L. Crystal
Structures and Absolute Configurations of Dexmedetomidine
and its Tosyl Derivative. Tetrahedron: Asymmetry 1993, 4,
45-58.
(26) Agranat, I.; Shih, Y.-S. Haworth Synthesis as a Route to the
Anthracene Ring System. Synthesis 1974, 865-866.
(27) Amemiya, Y.; Hong, S.-S.; Venkataraman, B. V.; Patil, P. N.;
Shams, G.; Romstedt, K.; Feller, D. R.; Hsu, F.-L.; Miller, D. D.
Synthesis and R-Adrenergic Activities of 2- and 4-Substituted
Imidazoline and Imidazole Analogs. J . Med. Chem. 1992, 35,
750-755.
(28) Ruffolo, R. R., J r. Medicinal Chemistry of Adrenoceptor Agonists.
Drug. Des. Discovery 1993, 9, 351-367.
(29) Ruffolo, R. R., J r.; Stadel, J . M.; Hieble, J . P. R-Adrenoceptors:
Recent Developments. Med. Res. Rev. 1994, 14, 229-270.
(30) Timmermans, P. B. M. W. M.; Chiu, A. T.; Thoolen, M. J . M. C.
R-Adrenergic receptors. In Comprehensive of Medicinal Chem-
istry; Hansch, C., Ed.; Pergamon Press: Oxford, 1990; pp 133-
185.
(31) Ruffolo, R. R., J r.; Waddell, J . E. Stereochemical Requirements
of R2-Adrenergic Receptors for R-Methyl Substituted Phenethyl-
amines. Life Sci. 1982, 31, 2999-3007.
(32) Wang, C. D.; Buck, M. A.; Fraser, C. M. Site-Directed Mutagen-
J M9506074