Rationale, design, and synthesis of novel phenyl imidazoles as opioid receptor agonists for gastrointestinal disorders

Article abstract of DOI:10.1021/jm030548r

A small series of novel, imidazoles 4 have been prepared that exhibit very good binding affinities for the δ and μ opioid receptors (ORs), as well as demonstrate potent agonist functional activity at the δ OR. Representative imidazole 4a (K_i δ = 0.9 nM; K_i μ = 55 nM; K_{i K} = 124 nM; EC₅₀ δ =13-25 nM) was further profiled for OR related in vivo effects. Compound 4a reduced gastrointestinal (GI) propulsive motility in a dose-dependent and naloxone-reversible manner, based on the results of the mouse glass bead expulsion test (3, 5, and 10 mg/kg, ip) and the mouse fecal pellet output test (1 and 3 mg/kg, ip). Compound 4a showed no analgesic activity as measured by the mouse abdominal irritant test (MAIT) when dosed at 100 mg/kg, sc, but did show significant MAIT activity at doses of both 10 μg (40% inhibition) and 100 μg (100% inhibition) when dosed intracerebroventricularly (icv). Taken together, these in vivo results suggest that 4a acts peripherally when dosed systemically, and that these prototypical compounds may prove promising as medicinal leads for GI indications.

Full text of DOI:10.1021/jm030548r

J . Med. Chem. 2004, 47, 5009-5020
5009
Ra tion a le, Design , a n d Syn th esis of Novel P h en yl Im id a zoles a s Op ioid
Recep tor Agon ists for Ga str oin testin a l Disor d er s
Henry J . Breslin,* Tamara A. Miskowski, Bryan M. Rafferty, Santosh V. Coutinho, J effrey M. Palmer,
Nathaniel H. Wallace, Craig R. Schneider, Edward S. Kimball, Sui-Po Zhang, J ian Li, Raymond W. Colburn,
Dennis J . Stone, Rebecca P. Martinez, and Wei He
J ohnson & J ohnson Pharmaceutical Research & Development, L.L.C., Welsh and McKean Roads,
P.O. Box 776, Spring House, Pennsylvania 19477-0776
Received October 28, 2003
A small series of novel, imidazoles 4 have been prepared that exhibit very good binding affinities
for the δ and µ opioid receptors (ORs), as well as demonstrate potent agonist functional activity
at the δ OR. Representative imidazole 4a (Ki δ ) 0.9 nM; Ki µ ) 55 nM; Ki κ ) 124 nM; EC50
δ )13-25 nM) was further profiled for OR related in vivo effects. Compound 4a reduced
gastrointestinal (GI) propulsive motility in a dose-dependent and naloxone-reversible manner,
based on the results of the mouse glass bead expulsion test (3, 5, and 10 mg/kg, ip) and the
mouse fecal pellet output test (1 and 3 mg/kg, ip). Compound 4a showed no analgesic activity
as measured by the mouse abdominal irritant test (MAIT) when dosed at 100 mg/kg, sc, but
did show significant MAIT activity at doses of both 10 µg (40% inhibition) and 100 µg (100%
inhibition) when dosed intracerebroventricularly (icv). Taken together, these in vivo results
suggest that 4a acts peripherally when dosed systemically, and that these prototypical
compounds may prove promising as medicinal leads for GI indications.
In tr od u ction
Compounds that modulate opioid receptors (ORs) are
well recognized as being useful therapeutic agents for
pain management (e.g., morphine;1 fentanyl;1 Figure 1),
and gastrointestinal (GI) motility regulation (e.g., lop-
eramide;1 Figure 1). These opioid therapeutic effects and
other opioid-mediated pharmacological observations
were relatively well categorized long before the molec-
ular biology of opioids was understood and were largely
tied to the specific class of opioids known as opiates.1 A
molecular mechanism of action for the observed opioid
pharmacology was determined more recently when the
G-protein-coupled ORs were discovered in the 1970s.
The ORs were quickly categorized into three subsets of
receptors (δ, µ, and κ), which more recently were further
subdivided pharmacologically into multiple subtypes.
Subsequent OR studies have documented pharmaco-
logical and functional interactions between the µ and δ
ORs and suggest that modulation of both receptors
simultaneously may be beneficial.2 It has also recently
been postulated that compounds that preferentially
target the peripheral δ OR may provide therapeutic
benefit in gastrointestinal disorders, while circumvent-
ing central µ opioid receptor related side effects.2
Soon after the ORs were characterized, the pentapep-
tide enkephalins were identified as a set of endogenous
OR ligands (e.g.: Tyr-Gly-Gly-Phe-Leu).3 Since the
discovery of the enkephalins, many groups have ex-
plored modifying the parent pentapeptides, yielding
little success in identifying useful medicinal agents. One
approach explored has been the truncation of the
enkephalin pentapeptides. An initial truncation of the
enkephalins led to a set of modified tetrapeptides that
F igu r e 1. Historical opioid receptor agonists.
maintain comparable OR affinities to the enkephalins
[e.g.: Tyr-Tic-Phe-Phe (TIPP; 1a ); Scheme 1].4 The
modified tetrapeptides were subsequently further trun-
cated to a set of modified dipeptides, which exhibit very
good δ OR agonist activities (e.g.: 1b; Scheme 1).5
Besides their potential medicinal value, these truncated
enkephalin analogues began to clarify key chemical and
conformational features of the enkephalins required for
favorable OR binding activity.6 Although these trun-
cated analogues can be viewed as an advancement
toward the design and discovery of new opioids, one
potential drawback noted for some of the truncated
compounds is their inherent instability potential to
cyclize on themselves, extruding an amine 3 while
concurrently forming the considerably less active7 dike-
topiperazine ring 2 (Scheme 1).
Recently we modified a previously reported cholecys-
tokinin (CCK)-related dipeptide (1c;8 Scheme 1) that has
an identical diketopiperazine cyclization liability9 (2c;
Scheme 1) as noted for the truncated enkephalins.
* To whom correspondence should be addressed. Fax: (215) 628-
4985. E-mail: HBreslin@prdus.jnj.com.
10.1021/jm030548r CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/04/2004

5010 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
Sch em e 1
Ta ble 1. Final Target Compounds 4 and 5
in vivo profile of these novel imidazole structures 4,
should they demonstrate favorable in vitro OR activity.
Based on these above rationales and questions, syn-
thetic efforts were initiated toward imidazole-opioids 4.
Following are described our initial experimental results
for the preparation and activities of 4.
no.
R
R′
R′′
A
m
4a
4b
4c
4d
4e
4f
CH2-Ph-4-OH
Et
CH2-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
H
H
Me
Br
Me
Ph
n-Pr
Ph
Ph
n-Pr
fused Ph
fused Ph
fused Ph
fused Ph
fused Ph
fused Ph
1
0
1
1
1
1
fused Ph
Ch em istr y
(i.e. benzimidazole)
4ga
CH2-2,6-di-
fused Ph
fused Ph
1
Novel imidazoles 4 were prepared via one of two
routes. The stereospecific preparation of disubstituted
imidazole 4a (Scheme 2) was begun with a high yielding
1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide/1-hy-
droxybenzotriazole (EDC/HOBT) amide coupling reac-
tion between commercial reagents 6a and 7a , to gen-
erate 8a . Amide 8a was converted to imidazole 9a in
modest yield following an imidazole ring-forming pro-
cedure,10 by warming 8a with ammonium acetate (NH4-
OAc) in acetic acid (AcOH). The tert-butoxycarbonyl (t-
BOC) protecting group of 9a was readily cleaved with
trifluoroacetic acid (TFA) at 0 °C to give 10a . Crude 10a
was reacted directly with an in situ-generated isobutyl
chloroformate mixed anhydride of commercially avail-
able N-t-BOC-O-tert-butyl-L-tyrosine (BOC-Tyr(tBu)-
OH), with a good overall two-step yield for 11a . Note-
worthy is that this stereospecific route was epimerization
free after these four initial steps, with 11a proving to
be a pure compound as determined by high-pressure
liquid chromatography (HPLC) and thin-layer chroma-
tography (TLC). Treatment of 11a with TFA at 0 °C
quickly removed both the N-BOC and O-t-Bu protecting
groups to give initial target compound 4a . Intermediate
11a was alternatively treated with bromine (Br2) in
chloroform (CHCl3) at 0 °C to generate 5-bromo-imida-
zole 11d in good yield. Intermediate 11d was also
readily di-deprotected with TFA leading to the isolation
of product 4d . The synthetic sequence of Scheme 2 also
proved fruitful in the preparation of trisubstituted
imidazoles 4h and 4i. A couple of variances were
incorporated into the scheme for piperidine 4h and
pyrrolidine 4i, based on knowledge gained in the
preparation of 4a . We had found that the imidazole ring-
forming cyclization of 8a had resulted in a modest
return of 9a at best. We assumed that the modest yield
might be in part a result of some instability of the acid
sensitive BOC protecting group of starting material 8a ,
and/or of product 9a , under the somewhat harsh acidic
reaction conditions (AcOH warmed to 100 °C). On the
basis of this hypothesis, we altered the nitrogen-
protecting group in our followup piperidine and pyrro-
lidine sequences to the more acid stable benzyloxycar-
bonyl (CBZ) protecting group. We also altered our
Me-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
CH2-Ph-4-OH
(i.e. benzimidazole)
4h
4i
5a
5b
5c
Me
Me
Me
Me
Ph
Ph
Ph
-
1
0
1
1
1
-
fused Ph
fused Ph
fused Ph
n-Pr
fused Ph
(i.e. benzimidazole)
Previously reported compound.14
a
Except for the potential cyclization liability, the CCK
related dipeptides are structurally distinct from the
opioids, possessing a different scaffold (indoline for 1c;
tetrahydroisoquinoline for 1a ,b) and different append-
ages (i.e. R, R′, and R′′). For the CCK series of com-
pounds, we substituted an imidazole moiety (e.g.: 4b;9
Table 1) as a bioisostere for the susceptible amide
moiety of 1c and experimentally demonstrated that we
could circumvent the diketopiperazine cyclization li-
ability noted for the amide series. We also demonstrated
that these related imidazoles and amides maintained
similar H bonding potentials and good conformational
overlap, based on molecular modeling calculations. Most
significant was our experimental finding of identical
biological activities for the imidazoles as for the parent
amide structures.
Noting the similar chemical liabilities between the
CCK analogue 1c and enkephalin amides 1a /1b, we
questioned whether our earlier CCK findings could be
extrapolated to the structurally dissimilar truncated
enkephalins (Scheme 1). If successful, this extrapolation
would yield potent imidazole-opioid compounds 4 (Scheme
1) with improved inherent stability over their parent
amides 1, as well as additional stability against poten-
tial protease degradation. Another potential benefit of
imidazoles 4, if they retained good OR activities, would
be the added insight for the conformational pharma-
cophore requirements for OR ligands. This postulate
was proposed as a promising possibility since imidazole
4 has three additionally constrained bonds when over-
laid on the freely rotating amide appendage of 1; thus,
4 has considerably fewer degrees of freedom than 1.
Finally, and most importantly, we wanted to assess the

Phenyl Imidazole Opioid Receptor Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21 5011
Sch em e 2
solvent for the imidazole ring forming reaction to a
mixed solvent system of refluxing xylene/AcOH, which
has been reported in the literature as being favorable
for such cyclizations.11 Following these alternate reac-
tion conditions, we obtained improved yields of imida-
zoles 9b and 9c. Subsequently we easily removed the
CBZ protecting groups of 9b and 9c with iodotrimeth-
ylsilane (TMSI) to give nor-amines 10b and 10c, re-
spectively. Amines 10b and 10c were carried on through
the remainder of the sequence in a similar manner as
described above for 10a , ultimately yielding respective
desired target 4h and 4i.
Another sequence explored in generating trisubsti-
tuted imidazoles 4c and 4e is outlined in Scheme 3. The
previously reported aldehyde 1212 was reacted with
commercially available diketones 13 under Radzisze-
wski reaction13 conditions to generate the desired
trisubstituted imidazole intermediates 9d and 9e in
reasonable yield. As in our prior sequence, intermedi-
ates 9d and 9e were readily deprotected with TFA at 0
°C to yield amines 10d and 10e, respectively. Coupling
of these amines with the in situ generated isobutyl
chloroformate mixed anhydride of BOC-Tyr(tBu)-OH led
to desired products 11. However, we were disappointed
to find that with this synthetic route a mixture of
epimers (11 and 14) were observed by HPLC/MS for
both amines during this coupling step. This discrepancy
between Schemes 2 and 3 suggests that the partial
isomerization for Scheme 3 likely occurs during the
heterocyclic ring forming reaction to generate imidazoles
9. Fortunately, separation of both sets of epimers from
Scheme 3 proved cooperative via preparative HPLC, i.e.
separation of 14a from 11e and 14b from 11f, respec-
tively. The resulting sets of pure compounds were easily
deprotected with TFA, as already mentioned for similar
adducts discussed in Scheme 2. Although Scheme 3
proved generally unattractive due to the noted partial
epimerizations, we did benefit from this synthetic effort
since it yielded the ready opportunity of measuring
relative biological activities for respective epimers 4c
with 5a and 4e with 5b.
Benzimidazole targets 4f and 5c were prepared as
depicted in Scheme 4. Acid 6d was cleanly coupled with
1,2-phenylenediamine via a standard EDC/HOBT amide
coupling reaction to give amide 15. Compound 15 was
internally cyclized and dehydrated in warm AcOH,
generating benzimidazole intermediate 9f in good yield.
Unfortunately, complete epimerization of the chiral
center of compound 15 occurred during this cyclization,
as determined by chiral chromatography. Similar to our
earlier examples, TMSI efficiently removed the CBZ
protecting group of 9f to yield noramine 10f. Noramine
10f was coupled with BOC-Tyr(tBu)-OH via an EDC/
HOBT amide coupling reaction. The resulting crude

5012 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
Sch em e 3
Sch em e 4
amide 11g was carried on and deprotected to give
approximately equal isolated quantities of products 4f
and 5c. The respective products were characterized by
comparing NMR data with related known imidazole
diastereomers.
hypothesis would be validated if an imidazole 4 main-
tained consistent OR activity as its analogous amide 1.
Therefore as our initial gauge, we prepared imidazole
4a , based on the knowledge that its directly related
amide, 1b, has been reported as a potent δ OR agonist.5
We began our biological evaluation of 4a by looking at
its in vitro OR binding affinities for both the δ and µ
receptors and were pleased to find that 4a (Ki δ ) 0.9
nM; Ki µ ) 55 nM) maintained comparable binding
affinities to that described for 1b (Ki δ ) 5.2 nM; Ki µ )
Resu lts a n d Discu ssion
Our preliminary biological goal was to evaluate
whether imidazoles 4 would be bioequivalent mimics of
the amide opioid ligands 1. We reasoned that this

Phenyl Imidazole Opioid Receptor Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21 5013
Ta ble 2. δ and µ Opioid Receptors in Vitro Binding and Functional Activities
binding data selectivity
functional data
no.
Ki δa (nM)
Ki µa (nM)
ratio (µ/δ)
EC50 δb (nM)
EC50 µb (nM)
1b
4a
4b
4c
4d
4e
4f
4h
4i
5a
5b
5c
5.2c
69c
13c
63
-
75
82d
25 (13)
NT
2120d
2400 (2490)
NT
0.9 ( 0.1
54.7 ( 12.5
>100e
>100e
f
0.30 ( 0.04
0.11 ( 0.04
62.7 ( 20.4
15.1 ( 0.7
21.5 ( 8.4
>100e
20.7 ( 3.4
11.6 ( 2.3
>100e
1.4 (3.5)
1.9 (5.8)
85 (43)
37(32)
542 (457)
NT
127 (182)
NT
440 (261)
156 (780)
>10000e
>10000e
NT
14
>1
>6
0.1
<0.02
4
>100e
>10000e
153 (130)
1300
NT
NT
1.5 ( 0.8
23.1 ( 2.4
85.6 ( 10.1
>100e
19.4 ( 5.4
>100e
-
49.6 ( 27.7
50.1 ( 15.2
>100e
>2
0.003
NT
58 (44)
Loperamide
0.16 ( 0.1
a
b
Average (n ) 3) ( SE. n ) 2; i.e. Assay 1 (Assay 2). c Reported value;5 values determined by displacement of selective radioligands
from rat brain membrane preparations. Reported value;5 activities based on inhibition of electrically stimulated contractions of mouse
d
vas deferens (MVD) for δ and of guinea pig ileum (GPI) for µ. e Tested twice. f NT ) Not tested.
69 nM). Compound 4a showed the least OR binding
affinity at the κ OR (Ki κ ) 124 nM). We subsequently
evaluated 4a for its δ OR functional activity via a
GTPγS assay and were further encouraged to find that
our parent imidazole 4a behaved as a full δ OR agonist
(EC50 δ ) 13-25 nM). Compound 4a possessed only
modest µ OR functional agonist activity (EC50 µ ) 2.4
µM via a GTPγS assay), which again was consistent
with the profile reported for amide 1b. Compound 4a
was also assessed for κ OR functional activity in a field
stimulated rabbit vas deferens assay and was found to
cause no inhibition of twitch contractions (EC50 > 30
µM). On the basis of 4a ’s preliminary in vitro results,
which supported our hypothesis, we embarked on a
cursory structure-activity-relationship (SAR) evalua-
tion around our novel imidazoles 4. These preliminary
imidazoles are outlined in Table 1, and their corre-
sponding in vitro OR results are compiled in Table 2;
related experimental details for the in vitro assays are
described in the Experimental Section.
[4c (R′ ) Me, R′′ ) Ph; Ki δ ) 0.3 nM; Ki µ ) 21 nM) >
4f (R′, R′′ ) fused Ph; Ki δ ) 15 nM; Ki µ > 100 nM)].
The associated δ OR GTPγS functional activities were
reflective of the binding data, with the phenyl imidazole
4c showing enhanced potency relative to benzimidazole
4f [4c (EC50 δ ) 1.4-3.5 nM) > 4f (EC50 δ ) 32-37
nM)].14
In addition to varying substitutions off the imidazole
moiety of generic structure 4, we also altered the
isoquinoline core scaffold of 4c to piperidine (4h ) and
pyrrolidine (4i). Both 4h and 4i had inverted δ to µ OR
selectivity affinities relative to 4c, with 4h possessing
considerable better OR affinities than 4i [4h (piperidine;
Ki δ ) 21 nM; Ki µ ) 2 nM); 4c (isoquinoline; Ki δ ) 0.3
nM; Ki µ ) 21 nM); 4i (pyrrolidine; Ki δ >100 nM; Ki µ
) 23 nM)]. Although the relative binding affinities for
piperidine 4h compared to pyrrolidine 4i were rather
informative, the inverted µ/δ selectivity ratio for both
nonfused heterocycles 4h and 4i relative to isoquinoline
4c was less enlightening, as it’s consistent with previous
reports for similar scaffold variations in amide-associ-
ated opioids.15 Consistent with the previous OR binding
to function activity comparisons, the related functional
activities for this subset of compounds were reflective
of their binding affinities [4h (EC50 δ ) 542 nM) < 4c
(EC50 δ 1.4-3.5 nM); 4h (EC50 µ ) 130-153 nM) > 4i
(EC50 µ ) 1300 nM) > 4c (EC50 δ >10 µM)].
To initiate our followup work around 4a , we varied
the 4,5-substituents of the imidazole as the first subset
of analogues. We found that a simple addition of some
lipophilicity at the R′ position of the imidazole, in place
of the H of 4a , slightly enhanced δ OR binding affinities.
This conclusion is supported by comparing monovaried
analogues 4c and 4d relative to 4a [4d (R′ ) Br; Ki δ )
0.11 ( 0.04 nM) > 4c (R′ ) Me; Ki δ ) 0.3 ( 0.04 nM)
> 4a (R′ ) H; Ki δ ) 0.9 ( 0.1 nM)]. The δ OR functional
SAR supported the same conclusion, using the δ OR
The last subset of compounds evaluated for in vitro
activity was comprised of 5a , 5b, and 5c, the respective
epimers of 4c, 4e, and 4f. In all comparable cases, the
S,S configuration was the preferred stereochemistry for
optimized δ OR affinity binding [4c (S,S; Ki δ ) 0.3 nM;
Ki µ ) 21 nM) > 5a (R,S; Ki δ ) 19 nM; Ki µ ) 86 nM);
4e (S,S; Ki δ ) 63nM; Ki µ >100 nM) > 5b (R,S; Ki δ
>100 nM; Ki µ >100 nM); 4f (S,S; Ki δ ) 15 nM; Ki µ
>100 nM) > 5c (R,S; Ki δ ) 50 nM; Ki µ > 100 nM)], as
GTPγS functional activities as the metric [4c (EC50
δ
) 1.4-3.5 nM) ∼ 4d (EC50 δ ) 1.9-5.8 nM) > 4a (EC50
δ ) 13-25 nM)]. An additional comparison of trisub-
stituted imidazoles 4e with 4a clearly demonstrated
that at the R′′ position an aromatic moiety was prefer-
able relative to an n-Pr aliphatic group for both δ and
µ OR binding [4e (R′′ ) n-Pr; Ki δ ) 63 nM; Ki µ >100
nM) , 4c (R′′ ) Ph; Ki δ ) 0.3 nM; Ki µ ) 21 nM)] and
OR function [4e (EC50 δ ) 43-85 nM) , 4c (EC50 δ 1.4-
3.5 nM)]. We also prepared a trisubstituted fused phenyl
imidazole, i.e., benzimidazole 4f, whose activity could
be directly compared to its respective nonfused trisub-
stituted phenyl imidazole analogue 4c. We found that
the 4-phenyl substituted imidazole 4c proved signifi-
cantly better in terms of δ (∼45-fold) and µ (>5-fold)
OR binding affinities than the fused benzimidazole 4f
well as for optimized OR functional activity [4c (EC50
δ
1.4-3.5 nM) > 5a (EC50 δ 127-182 nM); 4f (EC50 δ 32-
37 nM) > 5c (EC50 δ 261-440 nM)].
We also tested our previously reported potent TPPII
imidazole inhibitor 4b for OR in vitro activity. Not
surprisingly, 4b was found void of any significant OR
affinity (Ki δ >100 nM; Ki µ >100 nM).
Having found that compound 4a was a potent OR
ligand, we were curious how it would overlay in silico
relative to a previously reported peptidyl compound

5014 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
F igu r e 2. Two lowest energy conformers of compound 4a . The intramolecular hydrogen bonds are shown in red dash lines.
F igu r e 3. Complimentary orthogonal views for superimposition of compound 4a over 1b (in orange).
such as 1b. To initiate this modeling work, we did a
conformational search and energy minimization of imi-
dazole 4a , using the methodology and software de-
scribed in the Experimental Section. This initial mod-
eling exercise yielded conformers I and II among the
lowest energy conformers (Figure 2). Conformer-I con-
tains an intramolecular hydrogen bond between the
NH2 amine nitrogen and the imidazole NH hydrogen,
while Conformer-II possesses an intramolecular hydro-
gen bond between the NH hydrogen of the imidazole
and the oxygen of the carbonyl group. A computation
of relative energies between these two conformers
revealed that Conformer-I is about 0.5 kcal/mol more
stable than Conformer-II, based on Merck Molecular
Force Field (MMFF)16 calculations. To further validate
that Conformer-I was preferred over Conformer-II, high-
level quantum chemical calculations were carried out
on the two conformers. This additional study confirmed
Conformer-I as the preferred conformer by 2.7 kcal/mol
relative to Conformer-II. Having identified Conformer-I
as an energy-minimized structure of 4a , we then gener-
ated a lowest energy conformation of modified dipeptide
1b . We were encouraged to see that our energy-
minimized structure for 1b was very similar to a
previously calculated conformation17 for related ana-
logue H-Dmt-Tic-NH-CH2-BID, whose conformation
was based on a recent X-ray crystal structure of N,N-
(Me)2-DMT-Tic-OH.18 To complete our modeling exer-
cise, we overlaid the energy-minimized structures 4a
and 1a , which readily demonstrates the favorable
overlap of all key pharmacophore elements for the two
structures (Figure 3). Although the in vitro and in silico
results for compounds 4 proved quite favorable, the key
and foremost question remained, i.e. how would imida-
zoles 4 behave in vivo.
we focused on a preliminary set of classical in vivo
analgesic and GI assays to ascertain an initial profile
for this new series. The mouse abdominal irritant test
(MAIT) was utilized to assess central analgesic activity,
while the mouse glass bead expulsion test (MGBET) and
the mouse fecal pellet output test (MFPOT) were
employed to appraise GI propulsive motility effects.
Compound 4a was chosen as a representative compound
for the preliminary in vivo profiling. It was predeter-
mined that to fairly judge 4a as a practical medicinal
chemistry lead, related in vivo activities would have to
be observed from systemic administration. Although
peptidyl-like compounds are often reported in the
literature as extremely potent in vivo when administra-
tion is intracerebroventricular (icv), we felt this route
of administration should be utilized simply as a proof
of principle (POP) exercise.
Compound 4a possessed no activity (0% inhibition of
abdominal constriction) in the MAIT at a dose of 100
mg/kg when dosed subcutaneously (sc). As a POP
followup, compound 4a was dosed icv in the MAIT and
was found to exhibit a 40% inhibition of abdominal
constriction at a dose of 10 µg, and a 100% inhibition
at a dose of 100 µg. These results suggest that 4a does
possess OR-mediated analgesic activity, but that 4a
does not cross the blood-brain barrier. With regard to
the in vivo GI screens, compound 4a inhibited reflex-
stimulated colonic propulsion in a dose-dependent man-
ner in the MGBET when dosed intraperitoneal (ip) at
3, 5, and 10 mg/kg (Figure 4). Subsequently we showed
that 4a ’s inhibition of colonic propulsion in the MGBET
was blocked by the OR antagonist naloxone (ip), which
is supportive evidence that 4a ’s MGBET in vivo effects
are OR mediated (Figure 5). Compound 4a ’s GI ef-
fectiveness was also corroborated with the MFPOT,
where it was found that 4a significantly attenuated fecal
pellet output at 1 and 3 mg/kg for at least 6 h following
ip administration (Figure 6).
It is well-known that opioids can induce in vivo effects
centrally and/or peripherally, eliciting analgesic and/
or GI responses. To measure 4’s potential in vivo effects,

Phenyl Imidazole Opioid Receptor Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21 5015
activity is limited to peripheral modes of action when
administered systemically, i.e., 4a affected GI motility
while demonstrating no apparent central analgesic
affects. Further chemical and biological endeavors
around these opioid efforts are currently underway.
Exp er im en ta l Section
Novel compounds listed in Table 1, i.e. final products 4a ,
4c-f, 4h , 4i, and 5a -c, were all characterized by 300-MHz
1H NMR (Bruker-Biospin, Inc. DPX-300), mass spectra (Finne-
gan 3300), and HPLC (>99% purity at 214 nm and 254 nm;
Hewlett-Packard Series 1050 HPLC with a 3 µm, 3.3 mm ×
50 mm Supelco AZB+ C18 column using a gradient mobile
phase of 4:96:0.1 acetonitrile/water/TFA to 100:0:0.1 acetoni-
trile/water/TFA over 8.5 min, at a flow rate of 1.20 mL/min
and a total run time of 9.5 min). These final products were
also assayed for homogeneity by TLC on Whatman MK6F (1
in. x 3 in. × 250 um) silica gel plates. These final products
were also evaluated by high-resolution mass spectrum (HRMS)
analyses (Autospec E high resolution magnetic sector mass
spectrometer). The HRMS analyses were carried out internally
by our Spectroscopy Department in Spring House, PA. Rep-
resentative final product 4a was also analyzed by combustion
analyses for C, H, and N, which were carried out by Robertson
Microlit Laboratories, Inc, Madison, NJ . The results from this
analysis suggest that the final products are diaddition TFA
salts, and related biological data are reflected as such. The
F igu r e 4. Compound 4a (J NJ -10335572) inhibits reflex-
stimulated colonic propulsion in a dose-dependent manner. *
P < 0.05 vs vehicle.
1
related H NMR spectra data for all final products, as well as
representative intermediates, are included in the Experimen-
tal Section. Intermediates were also assayed by HPLC, HPLC/
MS, and/or TLC, with related data also included in the
Experimental Section. All reagents were commercially avail-
able unless otherwise specified, and all reactions were run
under an inert atmosphere of Ar or N2. Where required,
preparative purifications were performed on a Gilson semi-
preparative HPLC unit (Column: YMC ODS-A 30 × 100 mm,
5 µm; temperature: ambient; flow rate: 35 mL/min; mobile
phase: (A) 10/90 acetonitrile/ water with 0.1% trifluoroacetic
acid, (B) 90/10 acetonitrile/ water with 0.1% trifluoroacetic
acid; gradient: linear gradient from A to B over 9 min;
wavelength: 214 and 254 nm).
F igu r e 5. Compound 4a (J NJ -10335572) inhibits reflex-
stimulated colonic propulsion in a naloxone-sensitive manner.
* P < 0.05 vs vehicle; # P < 0.05 vs compound 4a .
These in vivo results demonstrate that 4a possesses
OR activities, and that these effects are restricted to
the periphery. Consequently, these results suggest that
closely related compounds of 4a might prove useful for
GI indications. As a result, most followup synthetic
efforts have been steered in that direction.
In conclusion, we have identified novel imidazoles 4
as extremely potent δ OR agonists, and their in vitro
OR activities are reflective of the activities reported for
amides 1. These related in vitro results correlate well
with our molecular modeling work, where we showed
that an energy-minimized structure of imidazole 4a
overlays nicely with an energy-minimized conformer of
amide 1b. Our preliminary in vivo evaluation of a
representative imidazole, 4a , revealed that its OR
Molecu la r Mod elin g. Conformational searching for 1b and
4a were carried out by molecular dynamics (MD) simulations
using Sybyl 6.8 software19 of Tripos. Merck Molecular Force
Field (MMFF)16 and charges were used for molecular dynamic
(MD) simulations and the following energy minimizations. The
MD simulations were run for 500 ps in a vacuum with a
dielectric constant ꢀ ) 1 at 700 K in order to sample various
conformations for the six member aliphatic rings in all these
three molecules. Conformations were saved at each 1 ps along
the MD trajectories and then were energy minimized using
conjugated gradient method for 1000 iterations. Conformer-I
and -II were optimized using density functional method at
F igu r e 6. Compound 4a (J NJ -10335572) reduces fecal pellet output in a dose-dependent manner. * P < 0.05 vs vehicle.

5016 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
B3LYP/6-31G** level in the quantum chemistry package
J aguar 5.5., Schrodinger L.L.C.(1500 SW First Ave, Portland,
OR 97201).
butyl ester (12)12 (1.83 g, 7 mmol) was combined with AcOH
(25 mL). To this mixture were immediately added 13a (Ald-
rich) (3.11 g, 21 mmol) and NH4OAc (13.49 g, 175 mmol). The
reaction mixture was then placed on a steam bath and heated
for 20 min. The reaction mixture was cooled in an ice bath
and then added to an ice slurry (400 g). The resulting mixture
was basified by addition of concentrated NH4OH (50 mL) and
then extracted twice with Et2O (150 mL each). The combined
organic phases were dried over MgSO4, filtered and concen-
trated to yield 5.25 g of crude product. This material was
purified by preparative HPLC to yield 1.40 g (51%) of 9d as a
white lyophil.(TLC:80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.65).
Syn th eses. (S)-3-(2-Oxo-2-p h en yl-eth ylca r ba m oyl)-3,4-
d ih yd r o-1H -isoqu in olin e-2-ca r b oxylic Acid ter t-Bu t yl
Ester (8a ). Compounds 6a (Aldrich) (2.77 g, 10 mmol), 7a
(Aldrich) (1.71 g, 10 mmol), and HOBT (2.70 g, 20 mmol) were
dissolved in dichloromethane (CH2Cl2) (100 mL). The solution
was cooled to 0 °C and then EDC (2.29 g, 12 mmol) was added
followed by N-methyl-morpholine (NMM) (1.31 g, 13 mmol).
The reaction mixture was then warmed to room temperature
(RT). After 72 h the mixture was extracted with water, and
the organic phase was extracted consecutively with aqueous
solutions of saturated NaHCO3, 2 N citric acid, and saturated
NaHCO3 once again. The organic phase was then dried over
MgSO4, filtered, and concentrated to yield 3.95 g (T.W. 3.94
g) of 8a as a yellow foam, which was used without further
purification (HPLC: 86% at 214 nm; HPLC/MS: m/z 395
(S)-3-(5-Meth yl-4-p r op yl-1H-im ida zol-2-yl)-3,4-d ih yd r o-
1H-isoqu in olin e-2-ca r boxylic Acid ter t-Bu tyl Ester (9e).
Starting from 13b (Aldrich), 9e was prepared in a similar
manner as 9d (Yield: 54%; HPLC: 98% at 214 nm).
3-(1H-Ben zoim idazol-2-yl)-3,4-dih ydr o-1H-isoqu in olin e-
2-ca r boxylic Acid Ben zyl Ester (9f). Intermediate 15 (2.0
g, 5 mmol) was added to AcOH (70 mL) and warmed to reflux
for 1.5 h. The solvent was then removed under reduced
pressure and the residue partitioned between Et2O and
saturated aqueous NaHCO3. The organic phase was dried over
MgSO4, filtered, and concentrated under reduced pressure to
give 1.44 g (75%) of 9f as an off white solid, which was used
without further purification (HPLC: 91% at 254 nm; HPLC/
MS: m/z 384 (MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH,
Rf ) 0.73). Noteworthy is that compound 9f gave two distinct
peaks of similar intensity on three separate chiral columns -
Daicel Chiralcel OD-H, Daicel Chiralcel AD-H, and Daicel
Chiralcel OJ -H. Each column run utilized the same condi-
tions: 95/5 to 60/40 hexane/IPA with an equilibrium time of
13 min, an initial hold time of 5 min, a gradient of 1.75%/min,
a gradient time of 20 min and a final hold time of 5 min. In
addition, the data from column Daicel Chiralcel AD-H included
a distinct polarimeter trace indicating the separated peaks had
opposite polarities. Conversely, starting material 15 showed
no separation on the chiral columns.
(S)-3-(4-P h en yl-1H -im id a zol-2-yl)-1,2,3,4-t et r a h yd r o-
isoq u in olin e (10a ). Intermediate 9a (0.75 g, 2 mmol) was
added to 0 °C neat TFA (4 mL). The reaction mixture was
allowed to warm to room temperature over about 45 min.
Excess TFA was removed under a stream of N2. The residue
was partitioned between CH2Cl2 (15 mL) and saturated
aqueous NaHCO3. The aqueous phase was then re-extracted
with a second portion of CH2Cl2, and the organic phases were
combined, dried over MgSO4, and filtered, to yield 10a as a
solution in CH2Cl2. This solution was used in the next step
without further purification or isolation. A small portion was
concentrated under reduced pressure for NMR purposes.
(HPLC: 99% at 214 nm; HPLC/MS: m/z 276 (MH+); TLC: 80:
20:5 CHCl3:CH3OH:HCOOH, Rf) 0.36, homogeneous) 1H NMR
(DMSO-d6) δ 3.15-3.25 (d, 2H), 4.1-4.25 (dd, 2H), 4.3-4.4 (t,
1H), 7.1-7.25 (t, 4H), 7.3-7.4 (t, 3H), 7.55-7.6 (s, 1H), 7.75-
7.8 (d, 2H)).
(S )-2-(4-Me t h yl-5-p h e n yl-1H -im id a zol-2-yl)-p ip e r i-
d in e (10b). Intermediate 9b (0.56 g, 1.5 mmol) was dissolved
in CHCl3 (20 mL) at 20 °C, and then TMSI (0.64 mL, 4.5 mmol)
was added neat. After 16 h the reaction was concentrated
under reduced pressure and the residue was treated with CH3-
OH (7 mL) followed by 2 N HCl (aq) (2 mL). The resulting
mixture was warmed on a steam bath for 3 h. The mixture
was then concentrated to remove the majority of CH3OH, and
to the resulting mixture was added 1 N HCl (aq) (6 mL). This
aqueous mixture was extracted twice with Et2O and then
cooled and basified with 3 N NaOH (aq). The basic solution
was extracted thrice with CHCl3 (15 mL each). The combined
CHCl3 layers were dried over NaSO4, filtered, and concen-
trated under reduced pressure to give 284 mg (79%) of 10b,
which was used without further purification (HPLC: 99% at
214 nm; HPLC/MS: m/z 242 (MH+); 80:20:5 CHCl3:CH3OH:
HCOOH, Rf ) 0.27, homogeneous).
1
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.82) H
NMR (CDCl3) δ 1.4-1.65 (d, 9H), 3.0-3.2 (m, 1H), 3.2-3.45
(m, 1H), 4.45-4.8 (m, 4H), 6.8-7.0 (s, 0.5H), 7.05-7.3 (m,
3.5H), 7.4-7.5 (t, 2H), 7.6-7.7 (t, 1H), 8.9-8.95 (d, 2H).
(2S)-2-[(1RS)-1-Meth yl-2-oxo-2-ph en yl-eth ylcar bam oyl]-
p ip er id in e-1-ca r boxylic Acid Ben zyl Ester (8b). Starting
from 6b (Aldrich) and 7b,20 compound 8b was prepared in a
similar manner as 8a and was used for its subsequent reaction
without further purification (crude yield: ∼100%; HPLC: 99%
at 214 nm; HPLC/MS: m/z 395 (MH+)).
(2S)-2-[(1RS)-1-Meth yl-2-oxo-2-ph en yl-eth ylcar bam oyl)-
p yr r olid in e-1-ca r boxylic Acid Ben zyl Ester (8c). Starting
from 6c (Aldrich) and 7b,20 compound 8c was prepared in a
similar manner as 8a and was used for its subsequent reaction
without further purification (crude yield: ∼100%; HPLC: 92%
at 214 nm; HPLC/MS: m/z 381 (MH+)).
(S)-3-(4-P h en yl-1H-im id a zol-2-yl)-3,4,-d ih yd r o-1H-iso-
qu in olin e-2-ca r boxylic Acid ter t-Bu tyl Ester (9a ). Inter-
mediate 8a (3.55 g, 9 mmol), NH4OAc (20.8 g, 270 mmol), and
AcOH (30 mL) were combined at RT, and the reaction mixture
was warmed on a steam bath for about 3 h. The reaction
mixture was then cooled to RT and poured into an ice slurry
mix (400 g). To this mixture were added concentrated am-
monium hydroxide (NH4OH; 50 mL) and ethyl ether (Et2O).
The layers were separated, and the aqueous phase washed
with a second portion of Et2O. The organic phases were
combined, dried over MgSO4, filtered, and concentrated under
reduced pressure to yield 3.51 g of brown foam. This crude
sample was purified by preparative HPLC to yield 0.85 g (25%)
of 9a as a white lyophile (HPLC: 96% at 214 nm; HPLC/MS:
m/z 376 (MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf )
1
0.77) H NMR (CDCl3) δ 1.2 (br s, 3H), 1.45 (br s, 6H), 3.1-
3.45 (br m, 2H), 4.45-4.7 (br m, 2H), 5.3-5.7 (br m, 1H), 6.9-
7.6 (bm, 9H).
(S)-2-(4-Meth yl-5-p h en yl-1H-im id a zol-2-yl)-p ip er id in e-
1-ca r boxylic Acid Ben zyl Ester (9b). Intermediate 8b (0.59
g, 1.5 mmol) was dissolved in RT xylenes (20 mL), then NH4-
OAc (2.89 g, 37.5 mmol) and AcOH (1 mL) were added. This
mixture was placed in an oil bath and warmed to 160 °C with
water being azeotroped into a Dean-Stark trap. After 2 h the
mixture was cooled to RT and extracted with brine. The
resulting organic phase was dried over MgSO4, filtered, and
concentrated under reduced pressure to give 0.61 g (T.W. 0.56
g) of 9b as a brown oil, which was used without further
purification (HPLC: 98% at 254 nm; HPLC/MS: m/z 376
(MH+)).
(S)-2-(4-Meth yl-5-ph en yl-1H-im idazol-2-yl)-pyr r olidin e-
1-ca r boxylic Acid Ben zyl Ester (9c). Starting from 8c,
compound 9c was prepared in a similar manner as 9b and
was used for its subsequent reaction without further purifica-
tion (crude yield: ∼100% of brown oil; HPLC: 98% at 254 nm;
HPLC/MS: m/z 362 (MH+); TLC: 80:20:5 CHCl3:CH3OH:
HCOOH, Rf ) 0.55, homogeneous).
(S)-3-(5-Meth yl-4-ph en yl-1H-im idazol-2-yl)-3,4-dih ydr o-
1H-isoqu in olin e-2-ca r boxylic Acid ter t-Bu tyl Ester (9d ).
3-Formyl-3,4-dihydro-1H-isoquinoline-2-carboxylic acid tert-
(S )-4-Me t h yl-5-p h e n yl-2-p yr r olid in -2-yl-1H -im id a -
zole (10c). Starting from 9c, 10c was prepared in a similar
manner as 10b (Yield: 77%; HPLC: 100% at 214 nm; HPLC/

Phenyl Imidazole Opioid Receptor Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21 5017
MS: m/z 228 (MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH,
Rf ) 0.13, homogeneous).
(S)-3-(5-Meth yl-4-p h en yl-1H-im id a zol-2-yl)-1,2,3,4-tet-
r a h yd r o-isoqu in olin e (10d ). Starting from 9d , 10d was
prepared in a similar manner as 10a . As with 10a , the CH2-
Cl2 solution of 10d was used directly without further manipu-
lation.
(S)-3-(4-Met h yl-5-p r op yl-1H-im id a zol-2-yl)-1,2,3,4-t et -
r a h yd r o-isoqu in olin e (10e). Starting from 9e, 10e was
prepared in a similar manner as 10a . As with 10a , the CH2-
Cl2 solution of 10e was used directly without further manipu-
lation (TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.4, homo-
geneous).
{(1S)-1-(4-ter t-Bu toxy-ben zyl)-2-[(3S)-3-(4-m eth yl-5-pr o-
p yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-yl]-2-
oxo-eth yl}-ca r ba m ic Acid ter t-Bu tyl Ester (11f). Starting
from 10e, 11f was prepared in a similar manner as 11a . The
final crude product isolated (270 mg; T.W. 242 mg) proved to
be an isomeric mixture of epimers, based on HPLC and HPLC/
MS analysis. By HPLC, there proved to be 48% (5.33 min) of
desired 11f in the crude reaction mixture, and 33% (5.48 min)
of its related epimer 14b. These compounds were separated
by preparative HPLC purification yielding 100 mg (41%) of
desired 11f and 10 mg (4%) of 14b. (11f: HPLC/MS: m/z 574
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.75,
homogeneous).
[2-[3-(1H-Ben zoim id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in -
olin -2-yl]-1-(4-ter t-b u t oxy-b en zyl)-2-oxo-et h yl]-ca r b a m -
ic Acid ter t-Bu tyl Ester (11g). Intermediate 10f (0.160 g,
0.6 mmol), BOC-Tyr(tBu)-OH (0.22 g, 0.6 mmol), and HOBT
(2.70 g, 1.28 mmol) were dissolved in DMF, and then EDC
(0.15 g, 0.77 mmol) was added at RT. After 16 h the mixture
was partitioned between water and EtOAc. The organic phase
was re-extracted with water, dried over MgSO4, filtered, and
concentrated to yield 0.39 g (T.W. 0.34 g) of 11g as an orange
foam, which was used without further purification.
{(1S)-1-(4-ter t-Bu t oxy-b en zyl)-2-[(3R)-3-(4-m et h yl-5-
p h en yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-
yl]-2-oxo-eth yl}-ca r ba m ic Acid ter t-Bu tyl Ester (14a ). The
isolation of this product is described under conditions for 11e.
(14a : HPLC/MS: m/z 574 (MH+); TLC: 80:20:5 CHCl3:CH3-
OH:HCOOH, Rf ) 0.68).
{(1S)-1-(4-ter t-Bu toxy-ben zyl)-2-[(3R)-3-(4-m eth yl-5-pr o-
p yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-yl]-2-
oxo-eth yl}-ca r ba m ic Acid ter t-Bu tyl Ester (14b). The
isolation of this product is described under conditions for 11f.
(14b: HPLC/MS: m/z 574 (MH+); TLC: 80:20:5 CHCl3:CH3-
OH:HCOOH, Rf ) 0.65).
(S)-3-(2-Am in o-p h en ylca r ba m oyl)-3,4-d ih yd r o-1H-iso-
qu in olin e-2-ca r boxylic Acid Ben zyl Ester (15). Starting
material 6d (1.56 g, 5 mmol) was dissolved in CH2Cl2 (40 mL)
and the reaction mixture cooled to about 0 °C. To the solution
was added NMM (0.51 g, 5 mmol) followed by isobutyl
chloroformate (0.68 g, 2 mmol), and the solution was allowed
to stand for 1 h. 1,2-Phenylenediamine (0.54 g, 5 mmol) in CH2-
Cl2 (10 mL) was then added and the reaction mixture stirred
for 3 h. The reaction mixture was then extracted with water,
dried over MgSO4, filtered, and concentrated to yield 2.11 g
(T.W.2.01 g) of 15 as an off white solid, which was used without
further purification (HPLC: 100% at 214 nm; TLC: 80:20:5
CHCl3:CH3OH:HCOOH, Rf ) 0.73, homogeneous).
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(3S)-3-(5-p h en yl-
1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-yl]-p r o-
p a n -1-on e Tr iflu or oa ceta te Hyd r a te (1:2:1) (4a ). TFA (4
mL) was cooled to about 0 °C, and then 11a (1.10 g, 1.85 mmol)
was added. The reaction mixture sat for about 0.5 h. Excess
TFA was then removed under a stream of N2 to yield a brown
oil. The oil was purified via preparative HPLC to yield 600
mg of 4a as a white lyophile (HPLC: 100% at 214 and 254
nm; HPLC/MS: m/z 439 (MH+); TLC: 80:20:5 CHCl3:CH3OH:
HCOOH, Rf ) 0.32, homogeneous). 1H NMR (MeOH-d4) δ 2.1-
2.25 (dd, 0.3H), 3.0-3.6 (m, 3.7H), 4.6-4.85 (m, 3H), 5.2-5.35
(t, 0.3H), 5.6-5.7 (t, 0.7H), 6.7-6.75 (d, 1.4H), 6.8-6.85 (d,
0.6H), 7.05-7.1 (d, 1.4H), 7.15-7.2 (d, 1.6H), 7.35-7.65 (m,
6H), 7.75-7.8 (d, 2H), 7.85 (s, 1H). HRMS calcd for C27H26N4O2
(MH+): 439.2134. Found: 439.2146. Anal. (C27H26N4O2/2C2-
HF3O2/H2O): Calc′d: C: 54.39; H: 4.42; N: 8.18, Found: C:
54.13; H: 4.45; N: 7.91.
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(3S)-3-(4-m eth yl-
5-p h en yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -
2-yl]-p r op a n -1-on e Tr iflu or oa ceta te (1:2) (4c). Starting
from 11e, 4c was prepared in a similar manner as 4a (Yield:
71%; HPLC: 100% at 214 and 254 nm; HPLC/MS: m/z 453
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.37,
homogeneous). 1H NMR (MeOH-d4) δ 2.25 (s, 1H), 2.45 (s, 2H),
3.0-3.6 (m, 4H), 4.4-4.6 (m, 1H), 4.7-5.0 (m, 2H), 5.15-5.25
(m, 0.3H), 5.65-5.75 (t, 0.7H), 6.7-6.75 (m, 2H), 7.0-7.2 (m,
3-(1H-Ben zoim id a zol-2-yl)-1,2,3,4-tetr a h yd r o-isoqu in o-
lin e (10f). Starting from 9f, 10f was prepared in a similar
manner as 10b (Yield: 74%; HPLC: 93% at 214 nm; HPLC/
MS: m/z 250 (MH+); 80:20:5 CHCl3:CH3OH:HCOOH, Rf
)
0.30, homogeneous).
[(1S)-1-(4-ter t-Bu toxy-ben zyl)-2-oxo-2-[(3S)-3-(4-p h en -
yl-1H-im idazol-2-yl)-3,4-dih ydr o-1H-isoqu in olin -2-yl]-eth -
yl]-ca r ba m ic Acid ter t-Bu tyl Ester (11a ). 2-tert-Butoxy-
carbonylamino-3-(4-tert-butoxy-phenyl)-propionic acid (BOC-
Tyr(tBu)-OH; Advanced ChemTech) (0.74 g, 2.2 mmol) was
dissolved in CH2Cl2 (40 mL) and the reaction mixture cooled
to about 0 °C. To the solution was added NMM (0.21 g, 2.1
mmol) followed by isobutyl chloroformate (0.27 g, 2 mmol, 0.26
mL), and the solution was allowed to stand for about 1.25 h.
To the reaction mixture was then added the solution of 10a
(0.55 g, 2 mmol) and the reaction mixture stirred for about 16
h. The reaction mixture was then extracted with water,
saturated aqueous NaHCO3, 2 N citric acid, and saturated
aqueous NaHCO3 again, dried over MgSO4, filtered, and
concentrated to yield 1.10 g (93%) of 11a as a foam, which
was used without further purification (HPLC: 67% pure (5.30
min) at 254 nm; HPLC/MS: m/z 595 (MH+); TLC: 80:20:5
CHCl3:CH3OH:HCOOH, Rf ) 0.87, homogeneous).
{(1S)-1-(4-ter t-Bu t oxy-b en zyl)-2-[(2S)-2-(4-m et h yl-5-
p h en yl-1H -im id a zol-2-yl)-p ip er id in -1-yl]-2-oxo-et h yl}-
ca r ba m ic Acid ter t-Bu tyl Ester (11b). Starting from 10b,
11b was prepared in a similar manner as 11a (Yield: 97%;
HPLC: 71% pure (5.15 min) at 254 nm; HPLC/MS: m/z 561
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.77,
homogeneous).
{(1S)-1-(4-ter t-Bu t oxy-b en zyl)-2-[(2S)-2-(4-m et h yl-5-
p h en yl-1H -im id a zol-2-yl)-p yr r olid in -1-yl]-2-oxo-et h yl}-
ca r ba m ic Acid ter t-Bu tyl Ester (11c). Starting from 10c,
11c was prepared in a similar manner as 11a (Yield: 82%;
HPLC: 92% pure (4.98 min) at 254 nm; HPLC/MS: m/z 547
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.64,
homogeneous).
[2-[(3S)-3-(4-Br om o-5-p h en yl-1H-im id a zol-2-yl)-3,4-d i-
h yd r o-1H-isoqu in olin -2-yl]-(1S)-1-(4-ter t-bu toxy-ben zyl)-
2-oxo-eth yl]-ca r ba m ic Acid ter t-Bu tyl Ester (11d ). To a 0
°C solution of 11a (0.18 g, 0.3 mmol) in CHCl3 (5 mL) was
added a solution of Br2 (0.015 mL, 0.3 mmol) in CHCl3 (1 mL)
over 1 min. After 1h at 0 °C the mixture was extracted with
cold saturated aqueous NaHCO3. The organic phase was dried
over MgSO4, filtered, and concentrated under reduced pressure
to give 220 mg (T.W. 202 mg) of 11d as a brown foam, which
was used without further purification (HPLC: 79% pure (7.22
min.) at 254 nm; HPLC/MS: m/z 673 (MH+)).
{(1S)-1-(4-ter t-Bu t oxy-b en zyl)-2-[(3S)-3-(4-m et h yl-5-
p h en yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-
yl]-2-oxo-eth yl}-ca r ba m ic Acid ter t-Bu tyl Ester (11e).
Starting from 10d , 11e was prepared in a similar manner as
11a . The final crude product isolated (1.55 g; 90%) proved to
be an isomeric mixture of epimers, based on HPLC and HPLC/
MS analysis. By HPLC, there proved to be 49% (5.47 min) of
desired 11e in the crude reaction mixture and 23% (5.60 min)
of its related epimer 14a . These compounds were separated
by preparative HPLC purification yielding 310 mg (18%) of
desired 11e and 50 mg (3%) of 14a . (11e: HPLC/MS: m/z 609
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.79,
homogeneous).

5018 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
3.5H), 7.25-7.35 (m, 4H), 7.5-7.65 (m, 3.5H). HRMS calcd
lin -2-yl]-p r op a n -1-on e Tr iflu or oa ceta te (1:2) (5a ). Starting
from 14a , 5a was prepared in a similar manner as 4a (Yield:
47%; HPLC: 100% at 214 and 254 nm; HPLC/MS: m/z 453
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.33,
homogeneous). 1H NMR (MeOH-d4) δ 2.35 (s, 3H), 3.1-3.35
(m, 4H), 4.15-4.25 (d, 1H), 4.75-5.0 (m, 2H), 5.75-5.8 (t, 1H),
6.7-6.75 (d, 2H), 7.05-7.1 (d, 2H), 7.15-7.3 (m, 4H), 7.4-7.5
for C28H28N4O2 (MH+): 453.2290. Found: 453.2291.
(2S)-2-Am in o-1-[(3S)-3-(4-br om o-5-p h en yl-1H-im id a zol-
2-yl)-3,4-dih yd r o-1H-isoqu in olin -2-yl]-3-(4-h yd r oxy-ph en -
yl)-p r op a n -1-on e Tr iflu or oa ceta te (1:2) (4d ). Starting from
11d , 4d was prepared in a similar manner as 4a (Yield: 23%;
HPLC: 99.3% at 214 and 100% at 254 nm; HPLC/MS: m/z
517 (MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.55,
homogeneous). 1H NMR (MeOH-d4) δ 2.0-2.1 (dd, 0.6H), 2.9-
3.4 (m, 3.4H), 4.2-4.3 (d, 0.6H), 4.45-4.6 (m, 1H), 4.65-4.95
(m, 3.2H), 5.45-5.5 (t, 0.4H), 6.5-6.55 (d, 2H), 6.9-7.1 (m,
4H), 7.15-7.4 (m, 5.4H), 7.6-7.65 (d, 0.6H). HRMS calcd for
(m, 5H). HRMS calcd for
Found: 453.2281.
C
28H28N4O2 (MH+): 453.2290.
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(3R)-3-(4-m eth -
yl-5-pr opyl-1H-im idazol-2-yl)-3,4-dih ydr o-1H-isoqu in olin -
2-yl]-p r op a n -1-on e Tr iflu or oa ceta te (1:2) (5b). Starting
from 14b, 5b was prepared in a similar manner as 4a (Yield:
56%; HPLC: 100% at 214 and 254 nm; HPLC/MS: m/z 419
(MH+); TLC: 80:20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.28,
C
27H25BrN4O2 (MH+): 517.1239. Found: 517.1257.
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(3S)-3-(4-m eth yl-
5-p r op yl-1H-im id a zol-2-yl)-3,4-d ih yd r o-1H-isoqu in olin -2-
yl]-p r op a n -1-on e Tr iflu or oa ceta te (1:2) (4e). Starting from
11f, 4e was prepared in a similar manner as 4a (Yield: 17%;
HPLC: 99.3% at 214 and 100% at 254 nm; HPLC/MS: m/z
1
homogeneous). H NMR (DMSO-d6) δ 0.7-0.75 (t, 3H), 1.4-
1.5 (m, 2H), 2.1 (s, 3H), 2.4-2.65 (m, 4H), 3.0-3.05 (d, 2H),
4.25-4.3 (d, 1H), 4.65-4.7 (m, 1H), 4.7-4.75 (d, 1H), 5.65-
5.7 (t, 1H), 6.7-6.75 (d, 2H), 7.05-7.1 (d, 2H), 7.15-7.3 (m,
4H). HRMS calc′d for C25H30N4O2 (MH+): 419.2447. Found:
419.2446.
1
419 (MH+)). H NMR (DMSO-d6) δ 0.65-0.7 (t, 1H), 0.85-0.9
(t, 2H), 1.35-1.45 (m, 0.7H), 1.5-1.6 (m, 1.3H), 2.05 (s, 1H),
2.25 (s, 2H), 2.35-2.45 (m, 0.7H), 2.55-2.6 (m, 1.3H), 2.9-
3.0* (2H), 3.25-3.30* (1H), 3.30-3.35* (1H), 4.5-4.55 (d, 1H),
4.7-4.8 (m, 1H), 4.95-5.05 (d, 1H), 5.25-5.35 (m, 0.65H), 5.4-
5.45 (m, 0.35H), 6.65-6.7 (d, 2H), 7.05-7.1 (d, 2H), 7.15-7.4
(m, 4H), 8.0-8.1 (br s, 1H). Verified closely related peaks are
rotameric by heating NMR sample to 80 °C, which resulted
in the initiation of coalescence of peaks; * peaks were identified
based on a COSY experiment (as they were obscured under a
large H2O peak of initial spectra; quantification of related H’s
is speculated). HRMS calcd for C25H30N4O2 (MH+): 419.2447.
Found: 419.2447.
(2S)-2-Am in o-1-[(3R)-3-(1H-ben zoim id a zol-2-yl)-3,4-d i-
h ydr o-1H-isoqu in olin -2-yl]-3-(4-h ydr oxy-ph en yl)-pr opan -
1-on e Tr iflu or oa ceta te (1:2) (5c). The isolation of this
product is described under conditions for 4f. (5c: HPLC: 100%
at 214 and 254 nm; HPLC/MS: m/z 413 (MH+); TLC: 90:9:1
CHCl3:MeOH:NH4OH Rf ) 0.27. 1H NMR (MeOH-d4) δ 3.05-
3.35 (m, 3H), 3.4-3.5 (dd, 1H), 4.05-4.1 (d, 1H), 4.8-4.85 (d,
1H), 4.85-5.0 (m, 1H), 6.05-6.15 (t, 1H), 6.7-6.75 (d, 2H),
7.1-7.25 (m, 6H), 7.4-7.5 (m, 2H), 7.55-7.65 (m, 2H). HRMS
calc′d for C25H24N4O2 (MH+): 413.1978. Found: 413.1971.)
(2S)-2-Am in o-1-[(3S)-3-(1H-ben zoim id a zol-2-yl)-3,4-d i-
h ydr o-1H-isoqu in olin -2-yl]-3-(4-h ydr oxy-ph en yl)-pr opan -
1-on e Tr iflu or oa ceta te (1:2) (4f). Starting from 11g, 4f was
prepared in a similar manner as 4a . The final crude product
isolated from 11g yielded an isomeric mixture of epimers 4f
and 5c, based on HPLC and HPLC/MS analysis. By HPLC,
desired 4f proved the front running material (2.78 min), while
5c ran slightly behind (2.87 min). These compounds were
separated by preparative HPLC purification yielding 14% of
desired 4f and 18% of product 5c. (4f: HPLC: 100% at 214
and 254 nm; HPLC/MS: m/z 413 (MH+); TLC: 90:9:1 CHCl3:
MeOH:NH4OH Rf ) 0.31, homogeneous, also 80:20:5 CHCl3:
Biologica l Assa ys. In Vitr o Assa ys. Ra t Br a in δ a n d µ
Op ioid Recep tor Bin d in g Assa y.21 Procedure: Male, Wistar
rats (150-250 g, VAF, Charles River, Kingston, NY) were
sacrificed by cervical dislocation and their brains removed and
placed immediately in ice cold Tris HCl buffer (50 mM, pH
7.4). The forebrains were separated from the remainder of the
brain by a coronal transection, beginning dorsally at the
colliculi and passing ventrally through the midbrain-pontine
junction. After dissection, the forebrains were homogenized
in Tris buffer in a Teflon-glass homogenizer. The homogenate
was diluted to a concentration of 1 g of forebrain tissue per 80
mL Tris and centrifuged at 39 000g for 10 min. The pellet was
resuspended in the same volume of Tris buffer containing 5
mM MgCl2 with several brief pulses from a Polytron homo-
genizer. This particulate preparation was used for the δ and
µ opioid receptor binding assays. Following incubation with
the appropriate peptide ligand [∼4 nM [3H]DPDPE for δ and
∼0.8 nM [3H]DAMGO for µ] at 25 °C for 2.5 h in a 96-well
plate with total volume of 1 mL, the plate contents were
filtered through Wallac filtermat B sheets on a Tomtec 96-
well harvester. The filters were rinsed three times with 2 mL
of 10 mM HEPES (pH7.4), and dried in a microwave oven 1:45
min twice. To each sample area 2 × 40 µL of Betaplate Scint
scintillation fluid (LKB) was added and analyzed on a LKB
(Wallac) 1205 BetaPlate liquid scintillation counter.
1
CH3OH:HCOOH, Rf ) 0.25, homogeneous. H NMR (MeOH-
d4) δ 2.2-2.3 (dd, 0.5H), 2.9-3.4 (m, 3.5H), 4.4-4.5 (d, 0.5H),
4.55-4.6 (d, 0.5H), 4.65-5.0 (m, 2H), 5.25-5.3 (t, 0.4H), 5.8-
5.9 (t, 0.6H), 6.55-6.6 (d, 1H), 6.75-6.8 (d, 1H), 7.05-7.1 (d,
1H), 7.1-7.65 (m, 8H), 7.7-7.8 (m, 1H). HRMS calc′d for
C
25H24N4O2 (MH+): 413.1978. Found: 413.1965.)
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(2S)-2-(4-m eth yl-
5-p h e n yl-1H -im id a zol-2-yl)-p ip e r id in -1-yl]-p r op a n -1-
on e Tr iflu or oa ceta te (1:2) (4h ). Starting from 11b, 4h was
prepared in a similar manner as 4a (Yield: 27%; HPLC: 99.7%
at 214 and at 254 nm; HPLC/MS: m/z 405 (MH+); TLC: 80:
20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.35, homogeneous). 1H
NMR (MeOH-d4) δ 0.6-0.7 (m, 0.3H), 1.25-2.35 (m, 5.4 H),
2.4 (s, 1.2H), 2.5 (s, 1.8H), 2.55-2.65 (m, 0.3H), 2.95-3.25 (m,
2H), 3.75-3.85 (d, 0.7H), 4.55-4.95 (m, 2.3H), 5.1-5.15 (t,
0.3H), 6.0-6.05 (t, 0.7H), 6.7-6.75 (d, 1.2H), 6.8-6.85 (d,
0.8H), 7.0-7.05 (d, 1.2H), 7.15-7.2 (d, 0.8H), 7.4-7.65 (m, 5H).
HRMS calcd for C24H28N4O2 (MH+): 405.2291. Found: 405.2303.
Da ta An a lysis. The data were used to calculate either the
% inhibition compared to control binding (when only a single
concentration of test compound was evaluated) or a Ki value
(when a range of concentrations was tested). % inhibition was
calculated as: [(total dpm-test compound dpm)/(total dpm-
nonspecific dpm)] × 100. Kd and Ki values were calculated
using GraphPad PRISM data analysis program.
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(2S)-2-(4-m eth yl-
5-p h en yl-1H -im id a zol-2-yl)-p yr r olid in -1-yl]-p r op a n -1-
on e Tr iflu or oa ceta te (1:2) (4i). Starting from 11c, 4i was
prepared in a similar manner as 4a (Yield: 54%; HPLC: 99.4%
at 214 and at 254 nm; HPLC/MS: m/z 391 (MH+); TLC: 80:
20:5 CHCl3:CH3OH:HCOOH, Rf ) 0.08, homogeneous). 1H
NMR (MeOH-d4) δ 2.1-2.35 (m, 3H), 2.5 (s, 3H), 2.5-2.65 (m,
1H), 2.9-3.0 (dd, 1H), 3.15-3.25 (dd, 1H), 3.6-3.7 (m, 1H),
3.9-4.0 (m 1H), 4.45-4.45 (t, 1H), 5.3-5.35 (t, 1H), 6.7-6.75
(d, 2H), 6.95-7.0 (d, 2H), 7.5-7.65 (m, 5H). HRMS calc′d for
[
35S]GTP γS Bin d in g Assa y in CH O-h γ Cell Mem -
br a n es.22 P r ep a r a tion of Mem br a n es. CHO-hγ cell mem-
branes were purchased from Receptor Biology, Inc. (Baltimore,
MD). 10 mg/mL of membrane protein were suspended in 10
mM TRIS-HCl pH 7.2, 2 mM EDTA, and 10% sucrose.
Membranes were maintained at 4-8 °C. 1 mL of membranes
was added into 15 mL of cold binding assay buffer. The assay
buffer contained 50 mM HEPES, pH 7.6, 5 mM MgCl2, 100
mM NaCl, 1 mM DTT, and 1 mM EDTA. The membrane
suspension was homogenized with a Polytron two times and
centrifuged at 3000 rpm for 10 min. The supernatant was then
C
23H26N4O2 (MH+): 391.2134. Found: 391.2139.
(2S)-2-Am in o-3-(4-h yd r oxy-p h en yl)-1-[(3R)-3-(5-m eth -
yl-4-p h en yl-1H -im id a zol-2-yl)-3,4-d ih yd r o-1H -isoqu in o-

Phenyl Imidazole Opioid Receptor Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21 5019
centrifuged at 18 000 rpm for 20 min. The pellet was saved in
a tube and 10 mL assay buffer was added into the tube. The
pellet and buffer were mixed with a Polytron.
Gordon, Dennis Hlasta, Scott Dax, Ellen Codd, and
Pamela Hornby for fruitful discussions during the
course of this work.
In cu ba tion P r oced u r e. The pellet membranes (20 µg/mL)
were preincubated with SPA (10 mg/mL) at 25 °C for 45 min
in the assay buffer. The SPA (5 mg/mL) coupled with mem-
branes (10 µg/mL) was then incubated with 0.5 nM [35S]GTPγS
in the same HEPES buffer containing 50 µM GDP in total
volume of 200 µL. Increasing concentrations of receptor
agonists were used to stimulate [35S]GTPγS binding. The basal
binding was tested in the absence of agonists and no specific
binding was tested in the present 10 µM unlabeled GTPγS.
The data were analyzed on a Top counter.
Su p p or t in g In for m a t ion Ava ila b le: Analytical purity
data has been tabulated for all targeted compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Refer en ces
(1) (a) Fries, D. S. Analgesics. In Principles of Medicinal Chemistry,
4th ed.; Foye, W. O., Lemke, T. L., Williams, D. A., Eds.;
Willimams and Wilkins: Baltimore, MD, 1995; pp 247-269; (b)
Aldrich, J . V. “Analgesics” Burger’s Medicinal Chemistry and
Drug Discovery, 5th ed., Volume 3: Therapeutic Agents, J ohn
Wiley & Sons: New York, 1996; pp 321-441.
DATA analysis: % of basal ) (stimulate - nonspecific) ×
100/(basal - nonspecific). EC50 values were calculated using
a GraphPad PRISM data analysis program.
(2) Riviere, J . M.; J unien, J .-L. Opioid Receptors: Targets for New
Gastrointestinal Drug Development. In Drug Development:
Molecular Targets for GI Diseases; Gaginella, T. S., Guglietta,
A., Eds.; Humana Press Inc., Totowa, NJ , pp 203-238.
(3) Hughes, J .; Smith, T. W.; Kosterlitz, H. W.; Fothergill, L. A.;
Morgan, B. A.; Morris, H. R. Identification of two related
pentapeptides from the brain with potent opiate agonist activity.
Nature (London) 1975, 258 (5536), 577-579.
(4) Schiller P. W.; Weltrowska G.; Nguyen T. M.; Wilkes B. C.;
Chung N. N.; Lemieux C. TIPP[psi]: a highly potent and stable
pseudopeptide delta opioid receptor antagonist with extraordi-
nary delta selectivity. J . Med. Chem. 1993, 36 (21), 3182-
3187.
Κ Bin d in g a n d F u n ction a l Assa ys. Both the κ binding
and functional studies were outsourced to Cerep (www.cerep.
com). Guinea pig cerebellum was the source for the κ binding
study, and rabbit vas deferens tissue was employed for the κ
functional study.
In Vivo Assa ys. Male CD1 mice (18-40 g) were used
throughout.
Mou se Abd om in a l Ir r ita n t Test (MAIT).23 Mice (n ) 10)
were dosed with the test compound either sc (100 mg/kg at a
dosing volume of 10 mL/kg) or icv24 (10 µg or 100 µg). Following
drug treatment, the mice were injected ip with a challenge
dose of acetylcholine bromide. The acetylcholine was com-
pletely dissolved in distilled water at a concentration of 5.5
mg/kg and injected at the dose of 0.20 mL/20 g. The mice were
observed for 10 min for the presence or absence of abdominal
constriction response beginning immediately after administra-
tion of the acetylcholine. For scoring purposes, an “abdominal
constriction” was defined as a contraction of the abdominal
musculature accompanied by arching of the back and extension
of the limbs. The percent of inhibition of this response was
calculated as follows: % inhibition ) 100 x (number of
nonresponders)/number of animals in group).
Mou se Gla ss Bea d Exp u lsion Test.25 A single, colored
glass bead (3 mm diameter) was inserted through the anus
into the distal colon of nonfasted mice to a distance of
approximately 2 cm, with the aid of a small glass rod
lubricated with 0.5% methylcellulose. Test compound(s) or
vehicle was administered ip (0.1 mL per 10 g body weight) 15
min before bead insertion. After bead insertion, mice were
placed in Pyrex animal jars for observation. The time taken
to expel the bead from the anus was expressed in minutes for
each mouse. Mice with unexpelled beads after 45 min were
necropsied to determine bead location. Data presented are
mean ( SEM for each group of mice (n ) 6-10).
(5) Schiller, P. New Use of Dipeptide Derivatives WO 98/28327, J uly
2, 1998.
(6) Schiller, P. W.; Nguyen, T. M. D.; Weltrowska, G.; Wilkes, B.
C.; Marsden, B. J .; Lemieux, C.; Chung, N. N. Differential
stereochemical requirements of µ vs δ opioid receptors for ligand
binding and signal transduction: Development of a class of
potent and highly δ-selective peptide antagonists. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89 (24), 11871-11875.
(7) Marsden, B. J .; Nguyen, T.M-D; Schiller, P. W. Spontaneous
degradation via diketopiperazine formation of peptides contain-
ing a tetrahydroisoquinoline-3-carboxylic acid residue in the
2-position of the peptide sequence. Int. J . Pept. Protein Res. 1993,
41 (3), 313-316.
(8) Rose, C.; Vargas, F.; Facchinetti, P.; Bourgeat, P.; Bambal, R.
B.; Bishop, P. B.; Chan, S. M. T.; Moore, A. N. J .; Ganellin, C.
R.; Schwartz, J .-C. Characterization and Inhibition of a Chole-
cystokinin-Inactivating Serine Peptidase. Nature 1996, 380,
403-409.
(9) Breslin, H. J .; Miskowski, T. A.; Kukla, M. J .; Leister, W. H.;
De Winter, H. L.; Gauthier, D. A.; Somers, M. V. F.; Peeters, D.
C. G.; Roevens, P. W. M. Design, Synthesis, and Tripeptidyl
Peptidase II Inhibitory Activity of a Novel Series of (S)-2,3-
Dihydro-2-(4-alkyl-1H-imidazol-2-yl)-1H-indoles. J . Med. Chem.
2002, 45 (24), 5303-5310.
(10) von Geldern, T. W.; Hutchins, C.; Kester, J . A.; Wu-Wong, J .
R.; Chiou, W.; Dixon, D. B.; Opgenorth, T. J . Azole Endothelin
Antagonists. 1. A Receptor Model Explains an Unusual Struc-
ture-Activity Profile. J . Med. Chem. 1996, 39, 957-967.
(11) Gordon, T. D.; Singh, J .; Hansen, P. E.; Morgan, B. A. Synthetic
approaches to the ‘Azole’ peptide mimetics. TetrahedronLett.
1993, 1901-1904.
(12) Byk, G.; Lelievre, Y.; Duchesne, M.; Clerc, F. F.; Scherman, D.;
Guitton, J . D. Synthesis and conformational analysis of peptide
inhibitors of farnesyltransferase. Bioorg. Med. Chem. 1997, 5
(1), 115-124.
(13) Ross, M. Grimmett Imidazole and Benzimidazole Synthesis;
Academic Press: San Diego, California; 1997; pp 151-165.
(14) During the course of our work, a couple related benzimidazoles
(e.g. 4g) were reported by Lazarus et al. Compound 4g was
reported possessing high affinity for the δ OR with δ functional
antagonist activity (pA2 7.67), based on the MVD assay. We
found 4f to possess modest agonist activity based on the δ GTPγS
OR assay. Balboni, G.; Salvadori, S.; Guerrini, R.; Bianchi, C.;
Santagada, V.; Calliendo, G.; Bryant, S. D.; Lazarus, L. H. Opioid
pseudopeptides containing heteroaromatic or heteroaliphatic
nuclei. Peptides 2000, 21 (11), 1663-1671.
Mou se F eca l P ellet Ou tp u t Test.26 Animals were trans-
ferred to indirect bedding cages and allowed to acclimate for
24-48 h. Test compound(s) or vehicle was administered ip (0.1
mL per 10 g body weight). The number of fecal pellets expelled
was quantified every hour for up to 6 h. Data are presented
as cumulative number of pellets over time and have been
expressed as mean ( SEM for each group of mice (n ) 10).
The overall effect of drug treatment was determined by
quantifying the area under the curve using the trapezoidal
rule.
Da ta An a lysis. Data from the mouse glass bead expulsion
test were analyzed by the Kruskal-Wallis Test (nonparametric
ANOVA) followed by Dunn’s multiple comparisons test. AUCs
from the mouse fecal pellet output test were statistically
analyzed with a one-way ANOVA followed by the students
modified t test with the Bonferroni correction. P < 0.05 was
considered to be statistically significant in all cases.
(15) Schiller, P. W.; Nguyen, T. M. D.; Berezowska, I.; Weltrowska,
G.; Schmidt, R.; Marsden, B. J .; Wilkes, B. C.; Lemieux, C.;
Chung, N. N. The TIPP opioid peptide family: development of
a new class of highly potent δ-receptor antagonists with extraor-
dinary δ-selectivity. Pept. Chem. 1992, Proc. J pn. Symp., 2nd,
1992, Meeting Date(Pub. 1993), 337-340.
(16) Halgren, T. A. Merck molecular force field. J . Comput. Chem.
1996, 17, 490-642.
(17) Balboni, G.; Salvadori, S.; Guerrini, R.; Negri, L.; Giannini, E.;
J insmaa, Y.; Bryant, S. D.; Lazarus, L. H. Potent δ-opioid
Ack n ow led gm en t. We appreciate the assistance of
William J ones for obtaining HR mass-spectral data and
J ames Gainor for assistance with nomenclature. We also
would like to thank J ohn Carson, Patricia Andrade-

5020 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 21
Breslin et al.
receptor agonists containing the Dmt-Tic pharmacophore. J .
Med. Chem. 2002, 45, 5556-5563.
(18) Bryant, S. D.; George, C.; Flippen-Anderson, J . L.; Deschamps
J . R.; Salvadori, S.; Balboni, G.; Guerrini, R.; Lazarus, L. H.
Crystal structure of dipeptides containing the Dmt-Tic pharma-
cophore. J . Med. Chem. 2002, 45, 5506-5513.
(19) SYBYL, version 6.8.1, Tripos Inc. (1699 South Hanley Rd., St.
Louis, MO 63144).
(20) Guzman, A.; Quintero, C.; Muchowski, J . M. Alkylation of R-tert-
butoxycarbonylamino ketone enolate anions. A useful synthesis
of R-alkyl-R-amino ketones, 2-acylpyrrolidines, and 2-acylpip-
eridines. Can. J . Chem. 1991, 69 (12), 2059-2063.
(21) Codd, E. E.; Shank, R. P.; Schupsky, J . J .; Raffa, R. B. Serotonin
and norepinephrine uptake inhibiting activity of centrally
acting analgesics: structural determinants and role in anti-
nociception. J . Pharmacol. Exp. Ther. 1995, 274(3), 1263-
1270.
ovary cells stably expressing the human B2 bradykinin receptor.
International Immunopharmacology 2001,
965.
1
(5), 955-
(23) Collier, H. O.; Dinneen, L. C.; J ohnson, C. A.; Schneider, C. The
abdominal constriction response and its suppression by analgesic
drugs in the mouse. Br. J . Pharmacol. 1968, 32 (2), 295-
310.
(24) Haley, T. J .; McCormick, W. G. Pharmacological effects produced
by intracerebral injection of drugs in the conscious mouse. Br.
J . Pharmacol. 1957, 12, 12-15.
(25) Rush, B. D.; Ruwart, M. The role of accelerated colonic transit
in prostaglandin-induced diarrhea and its inhibition by prosta-
cyclin. Br. J . Pharmacol. 1984, 83, 157-159.
(26) Coutinho, S. V.; Plotsky, P. M.; Sablad, M.; Miller, J . C.; Zhou,
H.; Bayati, A. I.; McRoberts, J . A.; Mayer, E. A. Neonatal
maternal separation alters stress-induced responses to vis-
cerosomatic nociceptive stimuli in rat. Am. J . Physiol. Gas-
trointest. Liver Physiol. 2002, 282 (2), G307-G316.
(22) Zhang, S.-P.; Wang, H.-Y.; Lovenberg, T. W.; Codd, E. E.
Functional studies of bradykinin receptors in Chinese hamster
J M030548R