Arylacetamides as peripherally restricted kappa opioid receptor agonists

Article abstract of DOI:10.1016/S0960-894X(00)00519-9

Analogues of the kappa (κ) opioid receptor agonist, ICI 199441, were prepared. K(i) values for these analogues at the cloned human κ opioid receptor ranged from 0.058 to 25 nM. Trifluoromethylaryl derivatives were potent analgesics when administered subcu

Full text of DOI:10.1016/S0960-894X(00)00519-9

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2567±2570
Arylacetamides as Peripherally Restricted Kappa Opioid Receptor
Agonists
Virendra Kumar,* Michael A. Marella, Luz Cortes-Burgos, An-Chih Chang,
Joel A. Cassel, Je€rey D. Daubert, Robert N. DeHaven, Diane L. DeHaven-Hudkins,
Susan L. Gottshall, Erik Mansson and Alan L. Maycock
Adolor Corporation, 371 Phoenixville Pike, Malvern, PA 19355, USA
Received 10 August 2000; accepted 7 September 2000
AbstractÐAnalogues of the kappa (k) opioid receptor agonist, ICI 199441, were prepared. Ki values for these analogues at the
cloned human k opioid receptor ranged from 0.058 to 25 nM. Tri¯uoromethylaryl derivatives were potent analgesics when admi-
nistered subcutaneously in the rat and were more peripherally restricted than the parent compound, ICI 199441. # 2000 Elsevier
Science Ltd. All rights reserved.
Introduction
During the past several years a number of groups have
synthesized kappa (k) opioid agonists and demonstrated
that these compounds were antinociceptive agents that
lacked the adverse side e€ects of constipation, respira-
tory depression, and drug abuse and dependence liabil-
ities associated with mu (m) opioid agonists such as
morphine.1a c Most of these compounds were developed
as centrally active analgesics. However, preclinical and
clinical studies with centrally active k agonists have
revealed that these agents have their own pro®le of side
e€ects, including sedation, diuresis, and dysphoria.2,3
More recently the presence of opioid receptors in the
peripheral nervous system has been described, and it has
been postulated that activation of these receptors may
produce antinociception without inducing centrally-
mediated side e€ects. Our strategy is to synthesize per-
ipherally restricted
k
receptor agonists as anti-
Chemistry
nociceptive agents that are improvements over the
peripheral k agonists that have been reported in the lit-
erature.1c To achieve this goal, we have chosen the
centrally active k agonist ICI 199441 (1),4 an analogue
of U-50488,5 as a template and initially we have focused
on modi®cations of the arylacetamide moiety to give
compounds of general structure 2.
The syntheses of these compounds required the diamine
(3), which was prepared following the literature procedure6
from (S)-(+)-phenyl glycine. Condensation of the dia-
mine with various aryl acetic acids was accomplished in
the presence of N-hydroxybenzotriazole hydrate (HOBT),
1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydro-
chloride (EDCI), and N,N0-diisopropylethylamine to
give compounds of general structure 2 (Scheme 1).7
The nitro analogues (2a±d) were prepared following the
above procedure, and reduction with either Raney
*Corresponding author. Tel.: +1-610-889-5779; fax: +1-610-889-
2203; e-mail: vkumar@adolor.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00519-9

2568
V. Kumar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2567±2570
nickel/hydrazine hydrate8 or PtO2 gave the reported
anilines9 (2e±h). Mesylation of these anilines with
methanesulfonyl chloride/triethylamine resulted in the
bis-sulfonamide derivatives, 2i±l (Scheme 2).
e€ects and lipophilicity of the tri¯uoromethyl group.
Encouraged by these ®ndings, the analogues 2m±o were
prepared from the diamine (3) and from tri¯uoromethyl
phenyl acetic acids (Scheme 1).
A number of derivatives of the intermediate 3,4-
dichloroaniline (2e) were planned to develop the struc-
ture±activity relationship for peripheral k agonists.
However, this aniline turned out to be sensitive to the
bases used. During base catalyzed reactions of 2e such
as acylations, sulfonation, and alkylations, the desired
compounds were isolated as minor components while
the predominant product from each of these reactions
was 5,6-dichloro-2-oxo-indole (4).
Biological Results
In Vitro:11 Kappa (k) opioid receptor anity was
determined by the displacement of bound [3H]U69,593
and mu (m) and delta (d) opioid receptor anities were
obtained by displacement of bound [3H]diprenorphine
(Table 1) using membranes prepared from cells expres-
sing the cloned human opioid receptors. The selectivity
for the opioid receptors was de®ned by the ratio of the
Ki values. The intermediate nitro and amino compounds
(2a±h) exhibited high anities for k receptors with Ki
values in the subnanomolar to low nanomolar range.
However, the receptor selectivity was compromised
depending upon the positional attachment of the sub-
stituents. For example, compound 2c with 4-NO2 on the
phenyl ring was less selective for the k receptor than
compounds 2a (2-NO2) and 2b (3-NO2). The amino
compounds had similar activities in k receptor binding
relative to their precursors. However, compound 2e was
also the least selective among the opioid receptors.
It was presumed that due to the presence of dichloro
on the aniline, the intermediate nitrogen anion might
favor the internal cyclization by expelling the diamine
portion of 2e to give 4 instead of reacting with electro-
philes.
To circumvent this problem, a surrogate for the 3,4-
dichlorophenyl acetic acid was sought. In a closely
related series of k agonists, the ICI group has reported10
that the replacement of both chlorines of the phenyl
acetyl of ICI 197067 with tri¯uoromethyl resulted in
compounds with enhanced binding anity at the k
opioid receptor, despite increased electron withdrawing
The peripheralization of these amines as bis-sulfon-
amide derivatives resulted in compounds 2i±l that were
2- to 463-fold less potent at the k receptor than ICI
199441. Among the bis-sulfonamides, 2j was the most
potent at the k receptor (Ki=0.096 nM); however, its
receptor selectivity relative to the d opioid receptor was
less than 50-fold (Table 1).
Scheme 1.
Scheme 2.

V. Kumar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2567±2570
2569
The search for a surrogate of 3,4-dichloro substituents
on the phenyl acetyl group resulted in the preparation
of the tri¯uoromethyl compounds 2m±o. These com-
pounds not only maintained potent k receptor anity
but also retained the selectivity pro®le relative to the
other opioid receptors (Table 1). The 4-tri¯uoromethyl
derivative (2o) was the most potent with a Ki value of
0.06 nM for the k receptor, which is comparable to that
of ICI 199441 (1).
These compounds lack systemic bioavailability, but are
active when applied directly at the site of injury, and
therefore may be useful as topical analgesic agents.
In the tri¯uoromethyl series of compounds (2m±o), the
4-tri¯uoromethyl compound (2o) was more potent when
given by the sc route than the 2- and 3-substituted
tri¯uoromethyl derivatives (2m and 2n, 60-fold and 21-
fold respectively). The compound 2o was also nearly 6-
fold more active in this assay than ICI 199441 (1), thus
indicating that both electron withdrawing properties
and lipophilicity of the substituents play a signi®cant
role in enhancing the analgesic activity.
In Vivo:11 Compounds were evaluated in vivo for their
ability to inhibit formalin-induced ¯inching in rats.
Selected compounds that were ecacious when admi-
nistered intrapaw (antagonism of ¯inching 580% at
300 mg) were titrated to determine A50 values. Otherwise
data are reported as % inhibition values at the indicated
dose (Table 2). Compounds were also evaluated in the
same assay by the sc route at an initial dose of 10 or 30 mg/
kg and % inhibition or A50 values were determined (Table
2). The 3- and 4-anilines (2g and 2h) were less active than
the corresponding 2-amino derivatives (2e and 2f). In con-
trast, the bis-sulfonamide derivatives (2i±l) of these amino
compounds were equally active in this assay. Similarly, the
tri¯uoromethyl compounds (2m±o) showed equal ecacy
in the formalin assay when given intrapaw.
The compounds (2d, 2f, 2m, 2n, and 2o) that showed
signi®cant activity in the formalin assay following sc
administration were evaluated for sedation, a centrally
mediated side e€ect. For this purpose, the rotarod per-
formance assay was chosen. A peripheral restriction
index, as an indicator of the CNS liability of com-
pounds, was calculated by the ratio of the rotarod ED50
compared to the A50 to inhibit formalin-induced ¯inch-
ing by the sc route (Table 2). The intermediates 2d and
2f had a peripheral restriction index of 4 and 6, respec-
tively, and the centrally active ICI 199441 (1) had a
ratio of 3, indicating improvement in peripheral restric-
tion for compound 2f. Among the tri¯uoromethyl deri-
vatives, the 3- and 4-substituted compounds (2n and 2o)
had peripheral restriction indices of 9 and 13, respec-
tively, 3- to 4-fold higher than ICI 199441. However, the
2-tri¯uoromethyl analogue (2m) produced sedation in
However, the di€erence in analgesic activity of these
compounds became evident when compounds were
given by the sc route in the same assay. Bis-sulfon-
amides (2i, 2k, and 2l) were inactive despite being e€ec-
tive analgesics when administered locally to the paw.
Table 1. In vitro binding anities to cloned human opioid receptors
2
R
Yield
(%)
Kappa binding Ki (nM)a
Selectivity for k
m Ki/k Ki
Selectivity for k
d Ki/k Ki
[3H]U69,593
b
b
b
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
2-NO2
3-NO2
4-NO2
87
95
93
60
55
70
47
73
94
49
42
68
82
67
93
0.41 (0.18±0.93)
0.065 (0.046±0.093)
1.1 (0.37±3.0)
0.091 (0.058±0.14)
0.086 (0.029±0.26)
0.58 (0.20±1.7)
0.93 (0.38±2.30)
0.93 (0.43±2.0)
8.6 (7.0±11.0)
0.096 (0.085±0.11)
6.0 (3.0±12.0)
>1000
>1000
52
>1000
>1000
410
2-NO2, 3,4-dichlorob
2-NH2, 3,4-dichlorob
430
260
580
140
b
2-NH2
3-NH2
4-NH2
>1000
>1000
>1000
>1000
320
>1000
>1000
>1000
>880
>800
>1000
>1000
>1000
>1000
33
>1000
>1000
>1000
>1000
>1000
b
b
2-N(SO2CH3)2
2-N(SO2CH3)2, 3,4-dichloro
3-N(SO2CH3)2
4-N(SO2CH3)2
2-CF3
25.0 (22.0±30.0)
0.13 (0.93±0.18)
0.064 (0.03±0.14)
0.058 (0.046±0.073)
3-CF3
4-CF3
3,4-Dichloro (ICI 199441) (1)
0.054 (0.039±0.076)
>1000
600
aValues are the means of three to six experiments and 95% con®dence intervals are given in parentheses.
bKnown compounds (ref 9).

2570
V. Kumar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2567±2570
Table 2. In vivo pro®le of peripherally restricted k agonists
2
R
Formalin ¯inching
50, mg (i.paw)a
Formalin ¯inching
50, mg/kg (sc)a
Rotarod
ED50, mg/kg (sc)a
Peripheral
restriction index
A
A
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
2-NO2
3-NO2
4-NO2
29 (12±69)
95% @ 300b
Not tested
5.3 (2.2±11)
7 (3.5±11)
65 (27±192)
58% @ 300b
52% @ 300b
17 (4.8±62)
2-NO2, 3,4-dichloro
2-NH2, 3,4-dichloro
2-NH2
0.24 (0.18±0.49)
Not tested
1.5 (0.85±2.2)
0.97 (0.37±2.3)
8.9 (3.2±20)
4
6
3-NH2
4-NH2
2-N(SO2CH3)2
2-N(SO2CH3)2, 3,4-dichloro
3-N(SO2CH3)2
4-N(SO2CH3)2
2-CF3
3-CF3
4-CF3
3,4-Dichloro (ICI 199441) (1)
21% @ 30b
Not tested
28 (9.7±65)
91% @ 300b
95% @ 1000b
98% @ 300b
97% @ 300b
0.82 (0.32±1.7)
0.74 (0.24±1.69)
15% @ 10b
30% @ 30b
0.54 (0.21±1.23)
0.02 (0.006±0.04)
0.009 (0.002±0.02)
0.05 (0.02±0.09)
71% @ 3c
0.18 (0.06±0.46)
0.12 (0.04±0.34)
0.14 (0.06±0.34)
9
13
3
aNumbers in parentheses are the 95% con®dence intervals.
bData are the % antagonism of ¯inching at the doses indicated.
cData are the % decrease from baseline at the doses indicated.
the rotarod consistent with central activity, demon-
strating sedation with compounds having electron
withdrawing and lipophilic groups at this position.
References and Notes
1. (a) Barber, A.; Gottschlich, R. Exp. Opin. Invest. Drugs
1997, 6, 1351. (b) Szmuszkovicz, J. In Progress in Drug
Research; Jucker, E., Ed.; Birkhauser Verlag: Basel, 1999; Vol.
52, pp 167±195. (c) Szmuszkovicz, J. In Progress in Drug
Research; Jucker, E., Ed.; Birkhauser Verlag: Basel, 1999; Vol.
53, pp 1±51.
2. Giardina, G.; Clarke, G. D.; Grugni, M.; Sbacchi, M.;
Vecchietti, V. Il Farmaco 1995, 50, 405.
3. Scopes, D. I. C. Drugs Future 1993, 18, 933.
Thus, replacement of the 3,4-dichlorophenyl portion of
ICI 199441 with 3- or 4-tri¯uoromethylphenyl acetyl
resulted in compounds with reduced CNS liabilities.
Work is in progress to develop the SAR of the periph-
erally restricted k opioid receptor agonists using
compounds 2n and 2o as templates.
4. Costello, G. F.; Main, B. G.; Barlow, J. J.; Carroll, J. A.;
Shaw, J. S. Eur. J. Pharm. 1988, 151, 475.
5. Szmuszkovicz, J.; VonVoigtlander, P. F. J. Med. Chem.
1982, 25, 1125.
Conclusions
A series of novel arylacetamides was prepared as per-
ipherally restricted k opioid receptor agonists based on
the centrally active agonist, ICI 199441, with k receptor
binding anities from 0.05 to 25 nM. Bis-sulfonamides
(2i, 2k, and 2l) were antinociceptive in the formalin-
induced ¯inching assay in rats when administered
locally (intrapaw) but not systemically (sc). These com-
pounds may have potential utility as topical or local
antihyperalgesics. Replacement of 3,4-dichlorophenyl
with 3- or 4-tri¯uoromethylphenyl resulted in com-
pounds 2n and 2o, with enhanced analgesic activity in
rats and improved peripheral restriction indices three
and four times higher than ICI 199441. The develop-
ment of the SAR around compounds 2n and 2o is in
progress to further improve the peripheral restriction of
the series.
6. Costello, G. F.; James, R.; Shaw, J. S.; Slater, A. M.;
Stutchbury, N. C. J. J. Med. Chem. 1991, 34, 181.
7. Compounds were puri®ed on a silica gel column and con-
verted to either HCl or CH3SO3H salts. All the new com-
pounds (2i±o) reported here gave satisfactory H NMR, MS
(FAB), and elemental analyses.
8. Chang, A.-C.; Takemori, A. E.; Ojala, W. H.; Gleason, W.
B.; Portoghese, P. S. J. Med. Chem. 1994, 37, 4490.
9. Weerawarna, S. A.; Davis, R. D.; Nelson, W. L. J. Med.
Chem. 1994, 37, 2856.
10. Barlow, J. J.; Blackburn, T. P.; Costello, G. F.; James, R.;
Le Count, D. J.; Main, B. G.; Pearce, R. J.; Russel, K.; Shaw,
J. S. J. Med. Chem. 1991, 34, 3149.
1
11. DeHaven-Hudkins, D. L.; Cortes-Burgos, L.; Cassel, J.
A.; Daubert, J. D.; DeHaven, R. N.; Mansson, E.; Nagasaka,
H.; Yu, G.; Yaksh, T. J. Pharmacol. Exp. Ther. 1999, 289, 494
and references therein.